GLOBAL CASSAVA RESEARCH ANO DEVELOPMENT The Cassava Econmy of Asia Adaptmg to Economic Change Documentatwn of cassava demand studies conducted m Asia m collaborauon With natwnal programs Draft verswn prepared for TAC for the Nairobi meetmg june 21 July 1 1987 May 1987 CIAT Centro lnternacwnal de Agncultura Tropical GLOBAL RESEARCH CASSAVA AND DEVELOPMENT The Cassava Econorny of Asia Adaptmg to Econornic Change by o Dr John Lynarn with collaboration frorn the followmg persons with respect to the Chapters mdicated Dr Boon¡lt T!tapiwatanakun Kasetsart Umversity Tha!land Dr Delane Welsch Umversity of Mmnesota Malaysia The Chapter entitled 'Trends and DistnbutiOn of Chmese Cassava Productwn and Use was prepared by Dr Bruce Stone Internatlonal Food Pohcy Research Instltute May 1987 CIAT Centro InternaciOnal de Agncultura Tropical Preface The accompanying manuscr1pt on the Cassava Economy of Asia represents a work still 1n progresa The study is essentially complete in its ma]or findings but the work has not yet been shaped into a consistent whole Distribution at th1s stage is done in order to share at an early stage the findings of the study with those interested in understanding the current status and future potential of the cassava crop The report should therefore be read as a draft the introductory chapter is not included here and the animal feed section for the China chapter was not ready in time for inclusion Also some of the f1gures are sitll lacking in the text The study has adopted a country-by-country approach to the analysis of the cassava economy in Asia It will hopefully be apparent from the study that this approach was correct as the differences between the various countries are large indeed The study covers all the maJor cassava producing countries in the region except Vietnam for which access was restricted The study relies almost exclusively on secondary data sources The only primary data collection involved a cost survey of chipping and pelleting factories in Tha1land A dependence on ex1sting data source has often left areas where further detail would have been valuable especially in production issues Nevertheless Asian countries have relatively well developed data systems which allowed a significant level of detail in the analysis although the data base for cassava is far weaker than that for the principal grains The study was carried out by John Lynam the economist in the CIAT Cassava Program except for the chapter on China which was dony by Dr Bruce Stone of the International Food Policy Research Institute Dr Lynam was aided in this task by Dr Boon]it Titapiwatanakum of Kasetsart Un1versity who oversaw the cost survey of the cassava processing plants in Tha1land Dr Delane Welsch of the University of Minnesota was hired as a consultant for the early phases of the proJeCt to help in data collect1on and initial plann1ng of the subject material The author visited all the countr1es and the principal production zones but not extensive period of t1me was devoted to more in-depth stud1es 1n the countries W1th the current study as a plann1ng base there are now plans to undertake more micro-leve! studies which w1ll support CIAT s overall research effort on cassava in Asia The current volume should therefore be seen as an integral part of CIAT s research effort in the reg1on and as such the contans and results will be subJeCt to revision as more information is developed about the crop in Asia An 1ndependent researcher may have approached the subJect differently and in some 1nstances may have put emphas1s on d1fferent issues in the conclusions However what has been more valuable for CIAT is the process inherent in the study The study prov1des only a snapshot in t1me of an ongoing exercise focused on a fuller 1ntegration of this type of research into research on cassava production and processing technology in Asia Having been forced to develop hypotheses probe data sources and understand markets and policies the CIAT Cassava Program has itself deepened its understanding of cassava in the region an understanding on which it now can bu1ld I II The Cassava Economy of As1a Adapting to Economic Change Taking Root A History of Cassava in Asia (Chapter in preparation) 1 2 3 4 5 Ind1a 1 2 Introduct1on and early distribution The integration of cassava into Asian food The advent of steamship trade and the r1se Shifting comparative advantage Malaysia Thailand Cassava a crop in permanent transition Kerala and Tamil Nadu Production a Production trends and distr1bution b Cassava production systems 1 Kerala 2 Tam1l Nadu e Yields economies of tapioca pearl Indonesia and d Costs of production and labor utilizaiton e Technology Markets and Demand a A synthesis of production and utilization 1 Kerala 2 Tam1l Nadu 3 Other States b Cassava for direct human consumption e The dried chip market d The starch market e Pricing and market efficiency 3 Conclusions III Trends and Distribution of Chinese Cassava Product1on and Use (by Bruce Stone IFPRI) 1 Production a Production trends and distr1but1on b Cassava production systems e Yields d Costs of production and labor util1zation e Technology development 2 Markets and demand a A synthesis of product1on and util1zat1on b Cassava for direct human consumpt1on e The starch market d The domest1c and export animal feed market 3 Conclusions IV Indonesia 1 Production a Production trends and distribution b Cassava production systems e Yields d Costs of production and labor utilization e Technology development 2 Markets and Demand a A synthesis of production and utilization b Cassava for direct human consumption e The starch market d Gaplek in feed markets e Pricing and market efficiency 3 Conclusions V Malaysia 1 Production a Production trends b Cassava production systems e Yields d Costs of production and labor utilization e Technology development 2 Markets and Demand a A synthesis of production and utilization b The domestic and export market for starch e The domestic animal feed market d Pricing and market efficiency 3 Conclusions VI Philippines 1 Production a Production trends and utilization b Cassava production systems e Yields d Costs of production and labor utilization e Technology development 2 Markets and Demand a A synthesis of production and utilization b Cassava for direct human consumption e The starch market d The dried chip market e Pricing and market efficiency 3 Conclusions VII Thailand 1 Production a Production trends b Cassava production systems e Yields d CQsts of production and labor utilization e Supply response f Technology development 2 Markets and Demand a A synthesis of production and utilization b The cassava pellet export market 1 Price formation 2 Profitability of the cassava pellet industry e The cassava starch market d The animal feed market 3 Conclusions VIII World and Regional Markets for Cassava Products IX 1 Protectionism and Substitution Decline in the World Starch Trade 2 Protectionism and Substitution The Rise in Trade in Cassava Feedstuffs a Demand for cassava in the EEC b The Asian regional market for cassava feedstuffs 3 Conclusions A Comparative Analysis of Cassava Production and Utilization in Tropical Asia 1 A Comparative Analysis of Production 2 A Comparat1ve Analysis of Consumption 3 Marketing and Price Formation 4 Cassava s Future Role in Asia li INDIA Cassava within the Rural Economies of Kerala and Tamil Nadu India is a vast diverse sub-continent where over three-quarters of the 684 million people (1981 Census) live in the rural and sector where their welfare is subject to the vagaries of the annual monsoons Consequently a maJor concern of agricultural policy has been developing the capacity of the country to feed itself and this 1n turn has resulted in a commitment to attaining self-sufficiéncy in food grain production This goal was achieved in the mid-1970' s essentially by focusing on development of the more productive agricultura! regions (Sarma 1982) Self-suff1ciency while indicating a termination in imports is nevertheless a relative concept because it implies that consumption is limited to production availability rather than determined by demand factors The central government has attempted to control the resultant price fluctuations by intervening in grain marketing to manage demand The government operates a public food distribution system at subsidized prices to ensure that a certain mínimum level of universal distribution of food grains is achieved independent of income levels As Sarma has noted This (self-suff1ciency) strategy which was confined to certain crops and areas with assured irrigat1on also resulted in the widening of interpersonal and 1nterregional dispar1ties The social justice objective in terms of reducing unemployment or underemployment and alleviating poverty in rural areas remained largely unfulfilled (p 24) The cassava-growing areas in the south of India have been such a region which has remained largely outside the area of impact of the "green revolution technology Although cassava is very much a regional crop in India this is also true of all other crops except rice Analyzing cassava in southern Ind1a thus provides some ins1ght into rectifying the d1sparities between regions in India PRODUCTION Product1on Trends and Distribut1on Cassava is very much a reg1onal crop in India two states Kerala and Tamil Nadu make up 97/ of cassava product1on in India (Table 2 1) On a country wide basis cassava makes only a small contribution to total calorie supplies with production being more or less equivalent to some of the minor coarse grains such as barley or the small m1llets However in the south of the country cassava ranks second to rice as the maJor calorie producing crop Given the range of temperature and rainfall conditions in India this type of regional specialization in crop product1on would be expected for non-irr1gated crops Accord1ng to the official data series area planted to cassava in India increased slowly from the mid-s1xties to the mid-sevent1es reaching a peak area of 392 thousand hectares in 1975-76 (Table 2 1) Since then cassava area has declined quite markedly reaching a level of 310 thousand hectares in 1981-82 The trends in area are due princ1pally to changes in II -2- Table 2 1 Ind1 a Trends 1n Area Product1on and Y1eld for the Country and the MaJOr Produc1ng States 1964-1981 Ind1a Kera la Tam1l Nadu Crop Year Are a Product1on Y1eld Are a Product1on Y1eld Area Product10n Y1 (OOOha) (000 t) ( t/ha) (OOOha) (000 t) (t/ha) (000 ha) (000 t) (t 1964-65 240 o 3 033 o 12 6 209 o 2 763 o 13 2 25 o 243 o 9 1965-66 271 o 3 467 o 12 8 230 o 3 095 o 13 5 35 o 339 o 9 1966-67 290 o 3 817 o 13 2 245 o 3, 410 o 13 9 39 o 377 o 9 1967-68 335 o 4 520 o 13 5 298 o 4,198 o 14 1 30 o 285 o 9 1968-69 359 o 4 636 o 12 9 298 o 4 081 o 13 7 55 o 527 o 9 1969-70 353 o 5 214 o 14 8 296 o 4 666 o 15 8 44 o 513 o 11 1970-71 353 o 5 216 o 14 9 294 o 4 617 o 15 7 47 o 567 o 12 1971-72 353 7 6 025 9 17 o 303 3 5,429 3 17 9 42 6 545 o 12 1972-73 363 2 6 317 4 17 5 304 8 5 629 4 18 7 50 o 629 5 12 1973-74 368 2 6 420 9 17 1 306 4 5 659 5 18 5 51 7 681 6 1~ 1974-75 387 6 6 325 9 16 3 317 9 5,625 1 17 7 52 7 564 9 1C 1975-76 392 o 6 638 3 16 9 326 9 5 390 2 16 5 50 1 1 115 8 2¿ 1976-77 385 8 6 375 o 16 5 323 3 5 125 5 15 9 48 o 1 128 2 2~ 1977-78 358 3 5 688 3 15 9 289 7 4 188 6 14 5 52 8 1 310 3 2~ 1978-79 361 5 6 050 1 16 7 289 9 4 226 3 14 6 54 o 1 682 o 31 1979-80 365 3 5 952 2 16 3 290 3 4 223 6 14 5 58 1 1 591 4 2i 1980-81 320 8 5 868 1 18 3 243 3 4 097 8 16 8 53 3 1 539 3 21 1981 82 310 2 5 267 4 17 9 241 8 4 073 o 16 8 42 3 1 324 8 31 Source Bullet1n on Commerc1al Crop Stat1St1cs and Agr1cultural S1tuat1on 1n Ind1a M1n1stry of Agr1culture II -3- cassava plantings in Kerala Cassava has been widely planted in Kerala since at least the turn of the century In the 55-year per~od from 1920 to 1975 cassava area in Kerala expanded at a relatively slow and uneven rate of 1 3% per annum (Table 2 2) Since 1975 cassava area has declined rapidly to the same level as the early sixties On the other hand area planted to cassava in Tamil Nadu has remained relatively constant at around 50 thousand hectares since the late 1960 s Production trends are more d~fficult to evaluate since the basis on which yield has been estimated has been changed twice In 1963 yield levels in Kerala were revised sharply upward from a trend of 7 t/ha to a rising yield trend starting at 12 t/ha In 1979 a crop cutting survey was instituted ~n Kerala and Tamil Nadu and what had been a rising trend in yields in Kerala was revised downward In Tamil Nadu on the other hand yield estimates were dramatically increased Given these revisions in yield estimates production trends which follow from the area and yield estimates are somewhat meaningless What can be said with some degree of confidence is that production in Kerala has declined markedly since 1975 at an annual rate of about 5% per annum Cassava production in Tamil Nadu in the same period has shown a slight increase The dominant question that arises is the reason behind the declining area and production of cassava in Kerala Cassava production systems Kerala Kerala is one of the most populous rural areas in the tropics Population densities in some distr~cts exceed 1000 people per square kilometer About 81% of the populat~on reside in the rural area according to the 1981 census while a little less than half of the work force are directly involved in agriculture However a more accurate reflection of the populat~on pressure is that while average farm size is only O 49 of a hectare only one third of the work force in the agricultural sector have access to land Moreover over 70% of the population who do own land have less than half a hectare (Table 2 3) As a consequence of this population pressure land use is very intensive Excluding forest reserves and non-agricultural uses 87-' of available land is cultivated The cropping intensity index in Kerala in 1977/78 was 132 percent well above the average for India as a whole However this figure is more remarkable when it is considered that two-thirds of cultivated area is under permanent tree crops Thus for area under annual crops the cropping intensity index is 192 percent that is a substancial portien of the land under annual crops is double or triple cropped Cassava is the most important annual crop in Kerala after rice making up 38% of the net area sown to annual crops Two factors expla~n why cassava has achieved such importance in so intensive an agricultural system First the non-irrigated upland areas are characterized by lateritic so~ls which are low in inherent soil fertil~ty especially phosphorus and are quite acidic Cassava in comparison to most other annual crops is well adapted to such soils even with relatively minimal amounts of fertilizer Second cassava gives very h~gh carbohydrate yields under these cond~tions With average yields around 15 t/ha only triple II -4- Table 2 2 Ind1a Growth 1 n Area Planted to Cassava 1n Kera la 1920-1980 Area Crop Year (000 ha) 1920-21 164 1925-26 170 1930-31 194 1934-36 175 1940-41 183 1944-45 197 1952-53 205 1955-56 222 1960-61 245 1965-66 260 1970-71 294 1975-76 327 1980-81 243 Source Pan1kar et al 1977 and Government of Kerala Stat1st1cs for Plann1ng D1rectorate of Econom1cs and Stat1st1cs Trlvan- drum var1ous years II -5- Table 2 3 lnd1a Percentage D1str1but1on of Farms by S1ze 1n Kerala 1970-71 S1ze of 01str1but10n Hold1ng of Hold1ng (ha) (%) Below O 04 18 7 o 04 - o 25 37 2 o 25 - o 50 15 6 o 50 - DO 13 3 1 00 - 2 DO 9 7 2 00 - 3 DO 3 2 3 DO - 4 DO 4 More than 4 00 o 9 Total 100 o SOURCE Stat1st1cs for Plann1ng 1980 Government of Kerala 1980 cropping of rice under irn.gation gives higher dry weight yields in the state While rice is grown on the irrigated bottomland cassava is grown on the sloping upland areas On these upland soils cassava competes pr~marily with tree crops for land and it is the general concensus that cassava is being displaced by higher value tree crops However for the principal tree crops increased plantings of rubber and cashewnut are more than offset by declining area of coconut and black pepper (Table 2 4) The crop or crops that are displacing cassava remain unclear from the aggregate data but the strongest hypothesis still remains some combinat~on of tree crops Cassava production systems in Kerala are relatively simple compared to countries such as Indonesia This is partly due to the constraints on potential intercrops imposed by soil conditions Annual rainfall in the state averages about 3000 mm and varies from about 2000 mm ~n the south to 3800 mm in the north There is a long dry period from December to March when little rain at all is received The rains start in April-May when 60-65% of the cassava crop is sown (Hone 1973) The monsoons arrive in full force in June-July From 35-40% of the crop is planted in September-October when the rains have fallen off but befare the start of the dry season in December Land preparation is done completely by hand and any green vegetation in the plot is concentrated in the soil below where the cassava stems are to be sown The stakes are sown vertically at populations of 10 to 12 thousand per hectare In such intensive systems weed control is fairly meticulous and when farmyard manure or wood ash is available it is incorporated in the same form as the green manure Some chemical fertilizer is certainly used on cassava in Kerala although there is conflicting data to suggest just how extensive this use is Certainly potassium fertil~zer consumption is a much higher percentage of total fertilizer consumption in Kerala than in India as a whole (33 3/ of consumption as compared to 11 4% in the whole country) Cassava (and tree crops) has a higher potassium requirement than gra~n crops A National Council of Applied Economic Research survey in 1975/76 found that 83% of cassava area in Kerala was fertilized but that only 19 kg/ha of nutrients were applied to the area fertilized Desai (1982) has found this survey to substantially overestimate aggregate fertilizer consumption in Kerala He provides estimates for India as a whole suggesting that in 1976/77 38 2% of cassava area was fertilized at a rate of 33 kg/ha The limited data available thus suggests that there ~s some fertihzation of cassava but at very low rates of applicat~on The cassava roots are harvested at about 10 months with the bulk of the crop being harvested in the dry period from December to February The percentage of the crop that is sold off the farm is open to some question A relatively dated report (Tapioca Market Expansion Board 1972) est~ates that about 40% of production enters market channels (Table 2 S) This would appear a bit low considering that cassava is such a pervasive consumption item in Kerala that about two-th~rds of households in Kerala do not grow cassava and that household consumption surveys show higher consumption levels for purchased cassava than own product~on (Table 2 6) II - 7- TABLE 2 4 India Area under Principal Tree Crops in Kerala 1970-80 Rubber Crop Year Coconut (000 ha) Black Pepper (000 ha) Less than 2 has (000 ha) Total (000 ha) Cashewnut (000 ha) 1970-71 1971-72 1972-73 1973-74 1974-75 1975-76 1976-77 1977-78 1978-79 1979-80 719 1 730 3 745 4 744 8 748 2 692 9 695 o 673 5 660 6 664 5 117 5 116 3 116 3 118 2 108 2 110 6 108 7 101 o 80 5 107 2 68 5 71 7 74 1 771 79 4 81 9 85 5 88 4 91 3 n a Source Government of India Bulletin of Commercial Crop Directorate of Econom~cs and Statistics Ministry various years 203 1 n a 208 8 n a 213 1 n a 217 5 103 2 221 3 104 9 224 4 109 1 230 6 113 3 233 4 127 o 235 9 n a n a n a Statist~cs of Agriculture II - 8- Table2 5 Ind1a Percent of Farm Product1on Commerc1al1zed 1n Var1ous D1str1cts of Kerala State 1971 Percent D1 Strl ct Commerc1al1zed Tnvandrum 46 8 Qu1l on 32 2 A lleppey 33 9 Kottayam 28 5 Ernakulum 16 9 Tr1 chur 53 4 Pa 1 ghat 776 Malappuram 42 6 Kozh1kode 38 2 Cannonore 23 o Kera la 39 3 Source Tap1oca Market Expans1on Board 1972 \ Table 2 6 lnd1a Consumpt1on of R1ce and Cassava by lncome Strata and by Source of SupplV Rura 1 Kera la 1977 (kg/household/week) Annual R1ce Cassava Household Total Own Open Tota 1 Own Open lncome Consumpt1on Rat 10n Product1on Market Cons ump t 1 on Product1on Market (Rupees) (kg) (kg) (kg) (kg) (kg) (kg) (kg) Less than 600 8 40 5 65 - 2 75 12 90 o 40 12 50 601-1200 9 43 6 39 - 3 04 11 31 2 96 8 35 1201-2400 13 47 7 70 177 4 00 15 46 4 13 11 33 "' 2401-3600 13 89 6 67 1 11 6 11 12 66 4 33 8 33 H 3601-4800 12 00 4 90 2 00 5 10 6 70 4 50 2 20 H More than 4800 13 42 5 14 5 71 2 57 3 29 3 29 SOURCE George 1979 The perversity of the latter is due to the positive relation between income and land ownership in Kerala and the shift from cassava to rice at higher incomes 40% is then probably a minimum estimate of marketed surplus of cassava in Kerala The most common marketing practice is for farmers to sell the standing cassava crop to purchase agents for a lump sum payment The agents do not necessarily harvest straight away but must harvest before the start of the rains Farmers as well gradually harvest the crop themselves selling in small lots by the roadside or in local markets When marketing of the fresh root is problematic particularly in the north of Kerala the roots are peeled sliced and dried as chips during the principal harvest per~od in the dry season Wholesale merchants and weekly markets serve as assembly points for roots and chips Tamil Nadu The other maJor cassava producing zone is ~n the western part of Tamil Nadu where production is principally concentrated in Salem District Production systems for cassava are considerably different from those in Kerala and this arises from a change ~n the limiting product~on constraint from soil factors in Kerala to moisture availability in Tamil Nadu Rainfall in the major production area of Salem Distr~ct averages 820 mm per year This average however masks a very high variation with annual ra~nfall in the last ten years ranging from 550 mm to 1250 mm There is a five-month dry season from January to May when rainfall averages no more than 14 mm in the whole period This limited rainfall is in many cases supplemented by irrigation Farm s~ze for cassava farmers in Tamil Nadu is somewhat larger than that in Kerala A sample of 70 cassava farmers in Salem D~strict found an average farm size of 2 6 hectares with an average area sown to cassava of 7 5 ha (Uthamalingam 1980) The larger farm size reflects in part the much drier conditions in Tamil Nadu and the relative scarc~ty of irrigation water Cassava is grown almost strictly as a cash crop in these cropping systems and competes for land principally with cotton and to a lesser extent rice and sugar cane Cassava s role in these cropping systems is defined by its access to a ready market (the industrial starch market) and cassava' s efficiency in water use Over 85% of the irrigation water is provided by wells and the farmer must plan his cropping pattern around expected rainfall and available water stored in the wells When irrigation water ~s in short supply farmers turn from rice and sugarcane to cassava or cotton depending on output prices According to the sample of 70 farms in Salem D~strict 907 of the farms grew cassava under irrigation The crop cutt~ng survey in all of Tamil Nadu found that 727 of the plots were grown under irrigation The irrigated crop ~s planted at the end of the ra~ns in January Up to four or five irrigations are needed for establishment Frequency of irrigation afterwards depends on water availab~lity in the wells and the arrival of premonsoon showers in June On average 20 ~rrigations are given at an interval of 15 to 20 days II -11- The rainfed crop is sown at the start of the southwest monsoon in August The crop is assured of no more than f1.ve months of rainfall before the start of the dry season in January which is followed by the pre-monsoon showers in June-July A ra1.nfed crop is often grown on as little as 500 mm of rainfall The urigated crop is usually harvested after 8 to 10 months while the rainfed crop requires 12 months before l.t can be harvested Land preparation relies on bullocks and for the irrigated crop the land is ploughed four or five times before forming either beds and channels or ridges and furrows Plant populatl.on is approximately 10 000/ha Stakes are sown vertically and normally six or seven weedings are done during the course of the crop year Fertilizatl.on or manuring l.S a common practice for cassava l.n Tamil Nadu especially for the irrigated crop The crop-cutting survey found that 74% of the cassava plots were either fertilized or manured using either animal manure or a vegetable compost The farmer survey l.n Salem found an average application of 18 5 t/ha of farmyard manure or 15 1 t/ha of compost Manuring is often comb1.ned with application of compound fertilizer Moreover cassava is usually planted in rotation with other crops and w1.ll often take advantage of residual fertility from fertilizer appll.cation on prior crops However where cassava is grown in successive years in the same plot there is a marked tendency for yield to drop A typical trend is 35 t/ha in the first year 24 t/ha l.n the second and 17 t/ha in the third (Tapioca Experiment Station Salem District prívate communication) In contrast to Kerala most of the cassava is harvested and marketed by farmers only a small percentage is sold standing in the lot In the Salem farm sample 87-' of the cassava was marketed dl.rectly by farmers The reason for this is the very decentralized nature of the cassava starch processing industry The industry consl.sts of upwards of 500 relatively small-scale plants distributed throughout the district Coordination of harvesting by the farmer and process1.ng of the fresh roots at the factory are eas1.ly managed without the need of middlemen or large expendl.tures on transport Yields By world standards cassava yields in India are h1.gh Yields in the 1980-81 crop year averaged 16 8 t/ha in Kerala and 28 9 t/ha in Tam1.l Nadu With the generally intensive level of cultural pract1.ces used in Kerala and Tamil Nadu this h1.gh yield is not surprising The difference in yields between Kerala and Tamil Nadu is due essentially to the poorer soils in Kerala and the use of l.rrigation and associated higher 1.nput levels l.n Tamil Nadu The author is unaware of any farm-level data on distribution of cassava yields in Kerala and therefore of any estl.mates of yield variance across farms in the state The dl.strict-level data suggest a slight tendency for y1.elds to be higher l.n the southern and central parts of the state and lower in the north Thus the 1980-81 crop estimates suggest average yields of 15 t/ha in the four southern d1.stricts and of 11 t/ha l.n Kozhikode and 12 t/ha in Malappuram suggest little variation in yields implicat~on for across farm variation in the north This limited data across the state but has l~ttle In Tamil Nadu a crop cutting survey in 7 d~stricts in the state found a significant variation in farm-level yields (Table 2 7) The yield d~stribution was skewed toward the lower side of the mean and as well exhibited a very extended upper tail that is a more or less typical distribution for farm-level cassava yields apart from the very high mean Over 15% of the plots had yields of over 37 t/ha with a maximum yield of 84 2 t/ha Tamil Nadu provides a perfect example of the yield potential of cassava when grown under very favorable production conditions Part of the reason why national cassava yields in other parts of Asia never approach such levels is that cassava is usually grown under more marginal agro-climatic conditions Yet even within a highly productive region such as Tamil Nadu over a quarter of the farmers are getting less than 15 t/ha Such typical yield distributions lie at the heart of production research what factors explain the difference in yields at the low and high end of the distribution and to what extent are these factors a function of farmer management or a function of more or less uncontrollable biolog~cal and edapho-climatic factors facing the farmer? The issue is cr~tical to understanding the substancial y~eld gap for cassava between the experiment station and farm level and how closely experimental yields translate into farm-level yields Costs of production and labor util~zation In such densely populated rural areas and in such intensive product~on systems as exist in southern India the expectation is that relative to other cassava production areas wage rates will be low labor input per hectare will be high inputs that substitute for land will be appl~ed at high levels and labor costs will be a lower portion of total costs The available data suggest per hectare labor inputs of 265 days for irrigated systems in Tamil Nadu 139 days for rainfed systems in Tamil Nadu (Uthamalingam 1980) and 116 days for product~on systems in Kerala (N~nan 1984) The breakdown of labor activities for Tamil Nadu shows that weeding is the principal labor requirement and makes up 60% of total labor demand with inputs in rainfed systems requiring about half that in ~rrigated systems (Table 2 8) Labor for harvesting forms the next maJor component in both systems followed by land preparation In Kerala on the other hand land preparation is by far the principal source of labor demand again reflecting the non-use of any sort of alternative power source in preparing the land Labor use for weeding is far below that employed in Tamil Nadu either in irrigated or rainfed systems Thus moisture for weed growth is not a factor influencing labor input The key difference is the use of h~red female labor in Tamil Nadu whereas in Kerala especially on farms of less than one hectare most of weeding is done by family labor almost solely men Labor input in cassava systems in India is lower than Indones~a but significantly h~gher than labor input in Thailand and the Ph~l~pp~nes Th~s result is expected g~ven the relative that ~n Malays~a II - 13 - Tab1e 2 7 lnd1a Y1e1d D1str1butlon from Crop Cutt1ng Survey Tam1 1 Nadu 1979-80 (287 farms) Y1e1d Strata (t/ha) o- 7 5 7 5-15 o 15 0-22 5 22 5-30 o 30 0-37 5 37 5-45 o 45 0-52 5 52 s-6o o 60 0-75 o 75 0-90 o Average Y1e1d = 24 5 t/ha Standard Dev1at1on = 14 1 t/ha Max1mum Y1e1d = 84 2 t/ha lrr1gated Y1e1d = 27 4 Un1rr1gated Y1e1d = 15 6 Percentage DIStrlbUtlon 13 14 16 25 16 8 5 2 o 3 SOURCE Unpubl~shen resu1ts of crop cutt1ng survey Tam1 1 Nadu II - 14 - TABLE 2 8 India Labor Use in Cassava Production Systems in Tamil Nadu 1978-79 and in Kerala 1976-77 Tamil Nadu Kerala Irrigated Rainfed Rainfed Activity Men Women Men Women M en (days/ha) (days/ha) (days/ha) (days/ha) (days/ha) Preparatory Cultivation 27 2 11 9 54 Seeds and Sowing 15 2 3 6 6 S S 3 lO Manuring S 4 7 l a Irrigation 25 3 Weeding 96 7 91 9 27 Harvesting 30 6 28 l 22 Miscellaneous 1 8 1 9 2 Total 103 7 161 6 53 S 85 o 115 a Included in weeding Source Uthamalingam 1980 Ninan 1984 II - 15 - differences in the land-labor ratios in the cassava growing regions of the different countries Moreover labor costs are a lower proportion of total production costs in India as compared to the latter three countries In Tamil Nadu labor makes up only 357 of variable production costs and less than 20% of total costs This is due to the large expenditures on fertil~zer and land rental A comparison of production costs between Kerala and Tamil Nadu (Table 2 9) shows that per ton costs are h~gher in Kerala than Tamil Nadu The difference is due in large part to differences in y~eld levels part~cularly when it is considered that rainfed systems in Tamil Nadu are of only marginal ~mportance Moreover when average yields reported for the state are used in place of the study's sample yields the d~fference becomes even more marked Nevertheless the flow of cassava ~s from Kerala to Tamil Nadu and not v~ce versa This is due to the very seasonal nature of cassava supply in Tamil Nadu and the fact that the opportunity cost of irrigated land when there is suffic~ent water is much higher than is reflected ~n average rental rates Technology Development Not only ~s there very limited potential for expanding area in cassava in southern India but competition from other crops has actually resulted in declin~ng area planted to cassava in Kerala There is an obvious demand for technology that would lead to increases in cassava yields The question arises since the product~on systems are so intensive and cultural practices are of such a high leve! whether there is a sign~ficant yield gap to exploit? This issue is at the heart of the work of the Central Tuber Crops Research Institute (CTCRI) in Kerala Under the Indian Council of Agricultura! Research the ~nstitute assumes principal responsibility for research on cassava in Ind~a Most of their work is focused on cond~tions in Kerala where research has been carr~ed out since 1963 Independent research on cassava ~s carried out ~n Tamil Nadu at the Tamil Nadu Agricultura! University in Corimbatore and the Tapioca Experiment Stat~on establ~shed in 1971 ~n Salem D~strict as part of Horticultura! Department of Tamil Nadu Th~s d~vision in activit~es allows research to focus on the very d~fferent product~on systems of Kerala and Tam~l Nadu Moreover India has had the longest period of continuous ~esearch on cassava in Asia The search for yield increasing technology in Kerala has focused on essentially four principal factors (a) improved high-y~elding variet~es (b) so~l fertil~ty management (e) control of African cassava mosaic virus and (d) ~ntercropp~ng systems The two pr~nc~pal constraints on increased product~vity are perceived to be soil factors and the v~rus d1sease G1ven the high level of cultural practices in the state overcoming these two constra~nts would probably not lead in themselves to much higher yield levels MaJor increases in per hectare product1v~ty would have to comb~ne as well improved varieties and intercropping with the problem in the later being the ident1fication of an adapted legume crop II - 16 - Table 2 9 Ind1a Cost of Product1on of Cassava 1n Tam1l Nadu and Kera 1 a 1978-79 Cost Item Van ab 1 e Costs Preparatory Cult1vat1on Seeds and Sow1ng Manures and Manur1ng Irr1gat1on Weed1ng Pl ant Protect1 on Harvest1ng Interest on Work1ng Cap1tal Total Vár1able Cost F1 xed Cos ts Rental Value of Land Deprec1at1on Interest on F1xed Cap1tal Total F1xed Cap1tal Total Costs Y1eld (t/ha) Var1able Cost per Ton Total Cost per Ton Tam1l Irr19ated (Rupee/ha) 273 o 220 5 1,101 6 300 1 477 6 237 7 274 1 2 884 7 1 776 4 210 7 387 5 2 374 6 5. 259 3 22 96 123 9 229 7 Source Uthamal1ngam 1980 Hone 1973 Nadu Ra1nfed (Rupee/ha) 180 4 222 o 529 2 228 2 177 5 140 4 1 477 7 989 7 147 8 228 4 1 365 9 2 843 6 10 74 137 6 265 2 Kera la Ra1 nfed ( Rupee/ha) 466 6 221 1 687 6 79 8 349 5 17 o 200 6 212 3 2 234 5 1 880 o 4 114 5 13 63 163 9 301 9 II - 17 - During the early years of CTCRI when a germplasm bank was being assembled one selection from Malaysia M-4 was released and found wide acceptability with farmers This var~ety has since set the standard and developing hybrids to replace M-4 has been a difficult task Only five hybrids have been released since the inception of the institute R-165 H-97 and H-226 in 1970 and R-2304 and H-1687 in 1977 A fert~l~ty tr~al carr~ed out at the experimental stat~on arguably gives sorne indication of potential y~eld gain w~th these varieties (Table 2 10) Average y~elds of M-4 at intermediate fertilizer levels are at about the state average of 15 t/ha indicating l~ttle gain to be achieved by agronomic practices The hybrid R-2304 yielded 24 t/ha at intermediate fertil~zer levels and 32 t/ha at relatively high fertilizer levels Because most cassava grown in Kerala is consumed as a bo~led root quality characteristics are very important This has probably been one of the principal factors limiting the wider adoption of the hybrids These quality characteristics ~nclude RCN content short cooking time (due to limited fuel resources of households) softness with cooking (apparently related to the ratio of amylose to amylopectin) good consistency (high starch content) and to a more minar extent wh~teness of the flesh (H-1687 for example is yellowish due to a high carotene content) M-4 ~s recognized to have good culinary quality and for these properties to be stable across locations and through the growing season The result is usually a price discount for roots from the hybrids for example farm prices of O 90 rupees/kg for M-4 versus O 75 rupees/kg for R-1687 (field notes 1982) Thus a 25% y~eld advantage is almost canceled by a 20i pr~ce discount Besides higher yielding ability and root quality characteristics the other majar breeding objective is f~eld tolerance to cassava mosaic virus M-4 though brought from Malaysia where the disease does not exist has relatively high field tolerance as do almost all the released hybrids Tolerance does not imply immunity with this disease and tolerant varieties must be combined with adequate selection of clean planting material since this is the princ~pal means of spreading the d~sease Unlike in West Africa where the disease is easily spread by the white fly vector effective white fly infection in India is only 2 to 5% The final two breeding obJectives are short maturity and plant type compatible with intercropping systems The latter is complementary to the research on intercropping systems Most of the cassava in Kerala is grown in monoculture due in large part to the lack of adaptation of potent~al commercial ~ntercrops to the lateritic soils The ~nstitute is having sorne success in promoting peanuts as a suitable intercrop with cassava Moreover since cassava ~s planted continuously for many years in the same plot ma~ntaining soil organic matter is difficult Long term fert~lny trials have shown that applying farm yard manure with fertilizer gives a signif~cantly h~gher yield than fert~l~zer alone and that manure appears to be necessary in maintaining yield levels over t~me (CTCRI 1980 and 1982) Increasing cassava production ~n southern India is dependent on increasing yields These yield increases in turn depend on the development of h~gh-yielding varieties that do not sacrif~ce quality for yield and that are tolerant to cassava mosaic v~rus The ~mproved varieties in turn ~ply heavier demands on soil fertility and thus higher rates of fert~lizer application Although the research obJectives are Table 2 10 Ind1a Cassava Root Y1eld of D1fferent Var1et1es 1n a Fert1l1zer Tr1al NR Comb1nat10ns (kg/há of N and K20) . _ .. Var1et1es 50 50 50 100 50 150 75 75 75 150 75 225 100 100 100 150 100 200 100 250 Mean H-165 22 67 23 01 22 88 24 24 22 84 26 47 28 30 25 08 23 87 27 93 24 73 H-2304 24 07 25 99 25 27 27 84 30 42 28 64 32 16 32 96 32 43 31 41 29 12 H-1687 19 29 19 04 21 47 19 62 20 13 22 96 26 05 26 39 25 31 25 02 22 53 M-4 15 18 14 76 15 66 16 95 16 10 15 83 18 62 18 66 17 48 18 62 17 79 Mean 20 30 20 70 21 32 22 16 22 16 22 37 23 47 26 28 24 77 25 74 Source Central Tuber Crops Research Inst1tute Annual Report 1978-79, Tr1vandrum 11 - 19 - quite straight forward after twenty years of consistent breeding effort CTCRI has found the progress to be slow in part because substant1al effort at the beginning had to be devoted to more basic stud1es since little basic research had been done on cassava up to that point in time in part because the1r varietal evaluation system requires approx1mately ten years from cross to potential release of a new variety and possibly in part because the recombination of all desired characters at adequate levels has a low probability producing a requisite hybr1d The efforts upto th1s point in time suggest that a goal of average farm-level yields of 25 t/ha is a feas1ble objective If the goal is worth pursuing depends in turn on the prospective outlook for utilization of the cassava crop Markets and Demand Kerala and Tamil Nadu present very different market structures (Table 2 11) In Kerala the market for fresh cassava for human consumption dom1nates wh1le in Tamil Nadu virtually all of the roots are processed into starch or tapioca pearl (see Appendix 2 1 for a discussion of the data sources used to construct the supply and utilization table) There is evidence of sorne trade between the two states but this appears to be relatively small and the flow is in only one direction from Kerala to Tamil Nadu Cassava markets in the two states appear to react independently of each other a feature re1nforced by the periodic controls on exports of cassava by the Kerala State government The focus therefore will be on the evaluat1on of Kerala and Tamil Nadu as two relatively independent markets Cassava for Direct Human Consumption Cassava as a direct food source achieves substantial weight 1n only the food economy of Kerala As might be expected in rural economies where population pressure on land is high per capita food consumption levels are low About 70~ of average 1ncomes are spent on food with the principal component being rice on which 30% of total income is spent (Table 2 12) In the rural areas over 6% of average income 1s spent on JUSt cassava In such economies food consumption is directly dependent on income levels and as can be seen 1n Table 2 13 food calorie distribution is symmetric to income distribut1on Average daily caloric intake is just over 2000 calor1es Using the relatively gross standard of 2100 calories as the minimum daily requirement Table 17 shows as muchas 35% of the population in rural areas and 507 in the urban areas falling below minimum requirements Because of the work and activ1ty patterns of the poor in rural areas calorie shortages can be considered to be chronic Cassava plays a key role 1n the calorie nutr1tion of the population of Kerala Cassava 1s at least as important (Nat1onal Sample Survey 28th Round) or more important (Kumar 1979) than rice for the low-income strata in rural areas Rice is however the preferred food and consumption increases markedly with income However at least for the 81% of the population in the rural areas cassava consumption shows a slight increasing trend across income strata (Table 2 14) Even though per capita consumption levels are high as compared to Indonesia for example the National Sample Survey would ind1cate sorne limited capac1ty by rural consumers to increase cassava consumpt1on w1th increases in income although with everything else equal most of that increase 1n 1ncome would go to increased rice consumption II - 20 - Table 2 11 Ind1a Product1on and Ut1l1zat1on of Cassava Roots by State 1977/78 Domest1c Ut1l1zat1on Human Consum~t1on Ammal S tate Product1on Export Fresh Dned Starch Feed (000 t) (000 t) (000 t) (000 t) (000 t) (000 t) Kera 1 a 4189 22 2437 619 499 Tam1l Nadu 1310 126 1162 y Andra Pradesh 137 123 Other 52 47 Ind1a 5688 22 2610 619 1784 11 Includes 109 thousand tons of roots and ch1ps 1mported from Kerala Source CIAT est1mates Waste (000 t) 503 131 14 5 653 II - 21 - Table 2 12 Ind1a Average Consumer Expend1ture Pattern Item Cereals R1ce Cassava Grams and Pulses Vegetab 1 e 01l M1lk and Da1ry Products Meat F1 sh Eggs Other Food Items Total Food Fuel and L1ght Cloth1ng Rent Other Non-Food Total Non-Food Total Rural Amount Percent ( Rupees) (:)') 18 14 17 70 3 53 o 72 1 12 1 82 2 52 11 75 39 60 2 97 2 63 o 10 10 05 15 75 55 35 32 8 32 o 5 4 1 3 2 o 3 3 4 5 21 2 71 5 5 4 4 8 o 2 18 2 28 5 100 o Kerala 1973 74 Urban Amount Percent ( Rupees) (1) 18 10 26 3 17 26 25 o 1 57 2 4 1 21 1 8 1 72 2 5 3 93 5 7 3 42 5 Q 15 59 24 2 45 74 67 8 3 60 5 2 2 55 3 7 1 26 1 8 14 78 21 4 22 19 32 2 58 93 100 o Source Government of Ind1a the Nat1ona1 Sample Survey 28th Round 1973/74 II - 22 - Table 2 13Jndla Calonc Consumpt1on by lncome Strata 1n Kerala 1971-72 Per Cap1 ta Rural Urban Monthly % D1stnbut1on Per Cap1ta % D1stribut1on Per Cap1 ta Expend 1 tu re of Households Ca lon e of Households Calone (Rupees) Consumpt1on Cons ump t 1 on 0-15 3 1 893 3 3 953 15-21 5 9 1229 7 6 1079 21-24 4 6 1716 5 7 1575 24-28 8 5 1466 6 9 1490 28-34 13 o 1900 12 1 1787 34-43 9 5 2320 14 5 1989 43-55 15 6 2603 14 2 2289 55-75 18 6 2900 10 9 2700 75-100 9 2 3614 7 3 3060 More than 100 12 3 4293 17 6 3907 Average 100 o 2023 100 o 2103 Source Stat1St1cs for Plann1ng 1980 Government of Kerala II - 23 - Table214. Ind1a Monthly Per Cap1ta Consumpt1on of Cassava and R1ce by Income Strata 1973/74 Cassava R1ce Income Strata Rural Urban Rural Urban (Rupees/capl ta) (kg/caplta) (kg/caplta) (kg/caplta) (kg/caplta) 0-13 5 04 1 96 13-15 8 33 o 20 1 75 3 60 15-18 4 63 12 50 3 42 1 67 18-21 7 60 3 23 3 18 2 95 21-24 6 49 3 05 4 34 4 23 24-28 5 14 5 59 4 98 4 06 28-34 7 49 3 06 5 06 5 60 34-~ 6 48 4 10 6 05 5 59 43-55 7 79 4 04 7 26 7 81 55-75 7 20 4 73 8 43 7 32 75-100 6 86 3 24 10 44 9 90 100-150 7 35 2 02 11 88 8 81 150 200 11 16 1 65 15 37 9 63 Greater than 200 5 43 1 50 18 67 10 50 Average 6 99 3 64 7 33 7 23 Source Government of Ind1a, The Nat1ona1 Samp1e Survey 28th Round Nat1onal Sample Survey Organ1zat1on 1973/74 II - 24 - Because of the limited incomes in Kerala a low-cost-per-calorie food such as cassava plays a principal role as a supplement to the higher cost rice A principal issue is whether promoting technical change in cassava production and the resultant lower prices will lead to bridging the calorie deficit In the maJor cassava producing district of Trivandrum cassava prices tend to be substantially lower and rice prices higher than in other districts The survey of Kumar in Trivandrum suggests that cassava consumption levels are substantially higher and rice consumption slightly lower than the average for Kerala (Table 2 15) However for the poorer income strata total calorie consumption is substant~ally higher than for the state average for this stratum In areas such as the survey area where average annual consumption reaches 172 kg there is probably not much potent~al for further increases in cassava consumption but changing the rice-cassava price relationship in other parts of Kerala would on the basis of this very limited comparison lead to increases in cassava consumpt~on and increased calorie consumption Shah (undated) has argued that attempts to increase the production of low cost high calorie foods with a view to bridging the calorie gap by themselves may prove inadequate' because preferences for food qual~ties other than just calories bias consumption even in the low income groups to more costly foods Food consumption patterns across ~ncome groups as described above would indeed confirm that food quality is important but as well that for the poor where price differences are sufficiently large cassava can constitute up to two thirds of total calorie intake that is the poor are very responsive to changes in relative prices of substitutes The central government has in part incorporated the quality argument in its system of public food distribution The foodgrain distribution system has played a maJor role in the food economy of Kerala since 1964 when food shortages in India led to food zoning and curtailment of private interstate trade The system depends on a comprehensive system of ration or fair price shops at which consumers are given quotas for foodgrains and prices are set well below open market prices However consumption requirements are well above the ration quota and consumers must purchase their additional requirements from the open market The availability of ration rice has a marked influence on rice and cassava consumption patterns A study by George (1979) found that consumption of ration rice was relatively constant across income strata (Table 2 6) although this finding is based on household income Kumar (1979) found that ration rice consumption increased with ~ncome when expressed on a per capita basis However whereas the higher income strata were able to complement th~s allotment with rice from open market purchases and at the highest income levels from own production the lower income strata supplemented the ration rice with very high levels of cassava consumpt~on most of which was purchased (George 1979) Nutrition of the poor thus depended principally on ration rice allotments and cassava purchases as was also found by Kumar Wheat is also available through the ration shops but George (1979) found that rural households consumed only a small quantity of wheat When their rice quota was exhausted consumers preferred to purchase cassava II - 25 - Table 2 15 Ind1a Monthly Rural Consumpt1on of Cassava and R1ce by Income Strata Kumar Survet Open Market In come Stra ta Cassava Rat10n R1ce R1ce Total R1ce (RuEees/CaElta) (k g/ ca21 ta) (kg/caElta) (k g/ ca21 ta) (kg/caElta) 0-15 19 95 1 60 69 2 29 15-24 17 68 2 29 1 46 3 75 25-34 16 13 2 51 2 04 4 55 35-49 16 09 2 67 2 06 4 73 50-74 14 35 3 46 1 64 5 10 Greater then 75 11 4 19 3 55 2 35 5 90 Average 14 13 2 89 1 98 4 87 11 For Kumar sample there are two observat1ons only Sources Kumar 1979 Government of Ind1a 1973/74 Nat1onal Sam2le Survey Cassava R1ce (kg/ca21ta) ( kg/ca21 ta) 6 27 1 88 6 47 3 83 6 70 5 03 7 18 6 17 7 20 8 43 7 16 12 08 6 99 7 23 II - 26 - from the open market than wheat from the ration shops Wheat purchases from the ration shops accounted for only about one-third of the total wheat allotment for the total sample and were the lowest in the low income household (p 33) Given the preference for rice a principal determinant of the demand for cassava will be ration rice allotments The f~rst factor to cons~der is whether ration rice consumption is influenced by demand factors Two studies (George 1979 and Kumar 1979) conclude that ration rice consumption ~s not influenced by demand factors but purely by supplies available that is all that is available would be consumed As levy procurement of r~ce within Kerala dropped to neglig~ble levels the ration system in Kerala came to rely almost completely on allotments from the Central Pool of the Food Corporation of India (FCI) Moreover these allotments now account for over half of rice supplies in Kerala (Table 2 16) and whereas such allotments should introduce a certain stability in r~ce supplies they are in fact the major cause of variability ~n rice availability in the state The author knows of no study which analyzes the determinants of state allocation of ration rice by the FCI but obviously there are other criteria than just maintenance of per capita consumption levels over time There is little choice but that cassava will cont~nue to be a principal component of a food strategy in Kerala and in particular cassava can be used to provide a certain flexibility in the operation of the food ration system in the state The dried chip market A peeled dry chip similar to gaplek in Indonesia is produced in Kerala The market principally provides an alternative outlet for cassava during the principal harvest period from December to April which coincides with the dry season The chips are principally produced and assembled in the northern districts with Calicut Trichur and Changanachery being the principal assembly centers Data on the markets for cassava chips are v~rtually non-existent What can be said is that this market is not as large nor as well-integrated as the gaplek market ~n Indonesia Most consumera ~n Kerala have relatively direct access to fresh roots and most field observations would suggest a consumer preference for fresh over dr~ed cassava The one and relatively dated source (Tapioca Market Expansion Board 1972) on processed cassava consumption suggests very limited consumption levels with an average annual per capita consumption of 9 S kg of dried product Indications are that the dried chip market for human consumption will remain very l~mited As is apparent in Indonesia a well functioning dr~ed chip market provides an element of price stability to the fresh root market especially where the major portian of planting and harvesting takes place at relatively restricted times of year The chip market acts as a storage mechanism for cassava during the low season and provides a price floor during the peak harvest period In Kerala the other major market for cassava chips is for processing into starch and glucose especially glucose Fresh roots produce a h~gher quality starch (Meuser et al 1978) but chips are used in the starch industry in Kerala because they are II - 27 - Table 2 16 Ind1a R1ce Product1on Rat1on R1ce Take-off and R1ce Ava1lab1 llt1es 1n Keral a 1971-1980 R1 ce 1 Rat1on Card Tota 1 Product1on _j Take-off Suppl1es Year (000 t) (000 t) (000 t) 1971 857 844 1701 1972 892 874 1766 1973 908 764 1672 1974 830 786 1616 1975 814 539 1353 1976 879 937 1816 1977 828 1380 2208 1978 854 872 1726 1979 848 570 1418 1980 N A 812 N A 1/ R1ce product1on 1s on a m1lled bas1s by crop year Source Government of Kerala Stat1st1cs for Plann1ng and Government of Ind1a Bullet1n on Food Stat1st1cs cheaper on a starch basis and help to ma1ntain operation outside the peak harvest season However if roots were available at the price and quantity desired the starch industry would operate exclus1vely on roots This particular outlet then does not provide a certain demand on which to develop an expansive dried ch1p market The other pr1ncipal option in developing a dried cassava market is the export market India exported l1mited quantities of cassava chips to Europe between 1957 and 1964 The largest export level reached in this period was 72 thousand tons in the 1958-59 crop year Exports virtually ceaaed until 1977 when exports to the EEC were resumed (Table 2 17) This reopening of export shipments was brought on by a substantial price fall in dried cassava in Kerala in 1977 which brought prices in line with f o b prices in Thailand (Figura 2 1) Through the early part of the 1970 s Upto 1977 cassava pr1cea in India were normally well above Thai prices and exports were not profitable From the beginning of 1977 through mid-1981 Indian prices remained in line with Thai prices and exporta continued at a rate of about 20 thousand tons a year India fortunately enjoyed a rising international price for caasava during this period and pr1ces in Kerala very closely tracked f o b Thai prices from early 1977 through mid-1981 at which point Indian prices could not match a falling international price In 1982 India again effectively dropped out of the export maket Export levels of 20 to 30 thousand tons result in high shipping costa and does not allow incentives for investment 1n more efficient marketing and procesaing capacity -- although there is some compensation in that India is closer than competitors to European markets At this stage Kerala does not have the production base to develop an effect1ve export market and simultaneously meet domestic requirements nor will India ever be in the position of being a large exporter of cassava products However a significant increase in yield levels could lead to further development of th1s nascent industry which would in turn provide incentives for further market 1ntegration the setting of a stable floor price and in turn lower and more stable prices for fresh cassava for food The starch market The market for cassava for starch production is divided between a fully integrated industry based on small-to-medium scale plants in Tamil Nadu and a relatively fragmented starch industry in Kerala consisting of two large-scale plants 3 medium-scale and 50 small-scale plants The principal constraint on expansion of this industry is supply of raw material to run the plants The 1ndustry in Kerala probably operates at no more than 50% capacity Factor1es here must compete with cassava for the fresh market and during at least part of the year must offer a lower price for cassava roots than pe~tains on the fresh market in order to rema1n competitive with production in Tamil Nadu Thus in 1981 a major starch factory in Kerala paid 260 rupees/t for roots which compared to farm level prices in Tamil Nadu of between 280 to 360 rupees/t and farm gate prices for the fresh market in Kerala of 400 rupees/t (field observations 1982) The farmer price would only cover variable production costs for the farmer and representa a pr1ce at which farmers would sell roots of low qual1ty or where 1dentif1cat1on of other market outlets was a constraint Further II - 29 - TABLE 2 17 India Imports by the EEC of Cassava Chips from India 1975-1985 Year 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 Source Quantity (tons) o o 7 949 37 182 26 799 11 915 24 215 3 037 10 23 40 NIMEXE Analytic Tables for Foreign Trade II - 30 - :-- 1' - .- "~ Ir..,¡- ~--r r, J --- ~" ~-~-~,... .- -- ¡- t -·- J -= -1 .. ll. ... í 1::' ..... 1 l...l l.J. -·-· 11¡-1 1:::!1 .. 1 • ._,, ·-· ! u í 1 l .. !::' l 1 l - ...1 - .. US$ lll"! Pel¡,:;t Ft 1 :e 1,...1 6 JP ;¡~,~-·L 1 1 Indr.:~ 11 1 1 ! r r¡ 1 r 1:' j 1 11 1 11 ~·' 11 1 1¡1 1 1 1 l ~1 1 {ll 1l 1 1 11<1'1 rJ 1 1 ~ 1 1 1 r 1 •• J I 1 11 • 1 1 1 1 ,, 1¡ ~ 1 ~ ! 1 ' 1 1 ,. 1 111,/io ¡1 ' 1 1 1 1 / 1 f 1 1 ' 1 1 J~ 1 l lf 1 1 '" 1 t 1 ) 1 ! l II - 31 - development of the starch industry ~n Kerala requires that prices in the fresh food and starch markets be brought closer in line Unlike the chip export market the root market for starch ~s already probably large enough to set an effective pr~ce floor should that ever be necessary As it is declining production trends and r~sing cassava pr~ces impl~es that the starch industry in Kerala will remain mor~bund The cassava root market for starch in Tamil Nadu functions as a single integrated market The starch industry here nevertheless operates at between 45 to 60% capac~ty Competition in Tam~l Nadu does not come on the demand side with alternative market outlets but rather from the supply side where cassava must compete with a substantial number of crop alternatives for ~rrigated land Root prices to the farmer are in turn determ~ned principally by the sale price of starch since roots make up approximately 80~ of the total cost of starch or sago production (Table 2 18) The cost and operating structure of the starch and sago industry shown in Table 2 18 suggests a relatively competitive small-to-medium scale industry where annual returns on fixed ~nvestment of from 17 to 31~ provide a normal return on investment considering the general capital scarcity that character~zes the Indian economy With further increases in farm production capacity there is little doubt that a dropping cassava pr~ce would motivate further investment ~n processing capacity The end market for sago and starch is not well documented The market for both~ is centered ~n the more northern states The end use of starch is principally in the textile industry especially Bombay Here cassava starcli competes with maize starch wh~ch ~s preferred over cassava starch apparently because of the h~gher viscos~ty and sells at a premium to cassava starch The cassava pearl or sago on the other hand is used str~ctly in food uses and the largest market appears to be Bengal particularly Calcuta Uses range from a festival food to a f~ller for rice Ex-factory prices of sago in 1978-79 of 1 55 rupees/kg compare favorably to rice prices of 2 2 rupees/kg The potential consumption of starch and sago in India is not known but traders knowledgeable about the industry suggest that demand ~s no constraint at forseeable product~on levels Pricing and market efficiency Price determinat~on and market allocation between competing uses are governed at least in Kerala essentially by factors which ~nfluence the demand for fresh cassava for human consumption The starch ch~p and export markets serve to set someth~ng of a price floor by absorbing any surpluses at the most competitive price at the t~me Because of the very marked seasonality of harvest such surpluses occur seasonally during the year as well as period~cally from year to year Because the fresh human consumption market makes up such a large part of total production compared for example to Java -- any changes ~n e~ther cassava supply or fresh root demand will create substantial instability in supplies going to alternat~ve markets Due to th~s factor and the very severe constraint on expansion in product~on area the development of these alternative markets has been very fragmented II - 32 - Tab le 2 18 Ind1a Annua 1 Cos ts of Product1on of Starch and Tap1oca Pearl 1n Tam1l Nadu, 1978-79 Starch Ta~10ca Pea rl Small Large Small Large Cost Item Factory Factory Factory Factory (Ru~ees) (Ru~ees) ( Ru~ees) (Ru~ees) Vanable Costs Cassava Roots 465,611 690 303 497 227 989 237 Temporary Labor 25 294 39,236 43,826 78 011 Fuel 5 060 11,492 El ectn c1 ty 4 292 7,624 4 687 9 240 Coconut 011 2 955 4 864 Gunny Bags 23 891 36,035 25 602 50 436 Interest on Work1ng Cap1tal 23 039 36 605 33 333 69 067 Total Var1able Costs 542,127 809,803 612 689 1 212 346 F1xed Costs Permanent Labor 9,091 11 277 7 237 12 908 Off1 ce Overhead 2 171 4 181 2 040 3,825 Deprec1at1on Bu1ld1ngs 2,174 2,870 1 703 2,695 Machwery 6 832 10 285 5 003 10 617 Interest on F1xed Cap1tal 15,937 22 910 13 295 19 618 Taxes 3 250 4 000 2 756 3 786 Total F1xed Costs 39 455 55 523 32 034 53 449 Total Costs 581,583 865' 326 644 723 1 265 795 Annua 1 Output ( tons) 431 6 652 8 411 8 822 o Total Cost per Ton 1347 1326 1566 1540 Output Pr1ce per Ton 1333 1333 1556 1555 Value of By Products per Ton 85 93 72 72 Source Ulthamal1ngam 1980 II - 33 - Although cassava consumption and prices are obviously influenced by rice availability and prices there are no stud~es which measure the degree of this influence Planning and investment in rice production cassava production and ration rice distribution in Kerala are critically dependent on such a study Price series provide the only data which shed light on the ~nteraction between the rice and cassava markets and here several inexplicable trends become apparent One special difficulty in analyzing price series ~s separating out the effects of inflation in the general price level Since the consumer budget is weighted so heavily by food purchases the consumer price index will reflect changes in food prices more than other products These tend to be somewhat volat~le anyway but in India upto 1977 food zoning heavily restricted interstate trade in food grains Food price levels thus varied by state and using the consumer price index for India as a whole to deflate prices in any particular state will probably not be reflect~ve of price ~nflation in that particular state For this reason the consumer price ~ndex in Trivandrum was used to deflate all prices in Kerala Dur~ng the decade of the 1970's real retail rice price rose till 1974-5 and then fell dramatically (Table 2 19) due to increases in rat~on rice availability Retail cassava pr~ces on the other hand remained relatively constant through the period resulting in rice becoming relat~vely cheaper to cassava While the marketing margin for fresh cassava in Kerala is proportionally low compared to marg~ns in other countries the margin has masked much higher variabil~ty ~n cassava prices at the farm and wholesale levels (Table 2 20) At the farm and wholesale levels comparable though not as marked trends to those that have occurred in the retail rice market have occurred In particular there is a falling real cassava price at a time (1976-78) when production was declining rapidly This would support a marked influence of rice prices and availabilit~es on cassava prices In 1979 the brief l~nkage to international prices caused cassava prices to rise The dominant issue then is what has been happening with rice availabilities? Through the decade of the 1970's rice production ~n Kerala was relatively stable (Table 2 16) The component of variabil~ty in rice supplies in Kerala was the availability of rat~on rice What is ~nexpl~cable with the ava~lable data is the low rice prices in 1978 and 1979 Since food zoning and restrict~ons on interstate trade of food grains were eliminated in 1977 it ~s possible that there have been flows of rice into Kerala from other states brought by private traders and sold on the open market However even the limited ev~dence on open market availabilities suggest that such supplies were not much changed in the years 1978 and 1979 (Table 2 21) and that eliminating food zoning has had no impact on rice supplies in Kerala Rice prices in Kerala have been traditionally higher than in the other Indian states (eg retail rice prices in 1981 ~n Kerala were 3 3 Rs/kg compared to 2 4 Rs/kg in Tam~l Nadu) and wh~le the liberalization of trade flows should bring prices more in line the mechanism to do th~s has to be ~ncreased availabil~t~es Thus while it is not clear why rice pr~ces have declined and in turn put a damper on cassava prices that should otherw~se have been rising in response to declin~ng production This allowed cassava prices to become compet~t~ve in the world market for a period of five years To the extent II - 34 - Table 2 19 Ind1 a Constant1 Reta1l Pr1ces of R1ce and Cassava 1n Kerala 1970-1979 Year R1ce Cassava R1ce/ Open Ma rket/ (Rupee/kg) (Rupee/kg) Cassava Rat1on R1ce 1970 2 87 55 5 2 1 5 1971 2 78 57 4 9 1 4 1972 3 04 55 5 5 1 6 1973 3 47 58 6 o 1 8 1974 3 84 56 6 8 2 6 1975 3 53 54 6 5 2 7 1976 3 02 62 4 9 N A 1977 2 73 58 4 7 N A 1978 2 43 55 4 4 N A 1979 2 33 61 3 8 N A 1 Pr1ces deflated by consumer pr1ce 1ndex 1n Tr1vandrum 1975 = 100 Source Government of Kerala 1980 George 1979 II - 35 - Table 2 20 Ind1a Avera9e Pr1ces of Fresh Cassava Roots at the ~arm Wholesale and Reta1l Level 1970-80 Farm-level )j Wholesale )j Reta 11 )j Year Nom1nal Real Nom1nal Real Nom1nal Real (Rupee/t) (Rupee/t) (Rupee/t) (Rupeett Rupee/t) (Rupee/t) 1970 N A N A 209 386 300 550 1971 214 391 222 407 310 570 1972 235 406 240 415 320 550 1973 309 446 311 449 400 580 1974 384 423 397 437 510 560 1975 400 400 391 391 540 540 1976 398 449 391 441 550 620 1977 325 376 323 373 500 580 1978 316 353 326 363 490 590 1979 398 411 410 424 590 610 1980 N A N A 443 N A N A N A Y Deflated by consumer pr1ce 1ndex 1n Tr1vandrum 1975 = 100 Source Government of Kerala Stat1st1cs for Plann1ng D1rectorate of Econom1cs and Stat1st1cs, Tr1vandrum var1ous years II - 36 - Table 2 21Indla Ava1lab1l1ty of R1ce 1n Three MaJar Markets 1 n Kera 1 a 1970-81 Year Jan-Mar Apr-June July-Sept Oct-Dec Total (000 t) (000 t) (000 t) (000 t) (000 t) 1970 21 o 10 7 5 5 4 4 41 3 1971 7 2 12 1 9 4 11 3 40 o 1972 25 7 25 7 15 3 15 3 82 o 1973 112 9 8 8 5 12 2 41 7 1974 8 6 9 6 8 4 4 7 31 3 1975 4 2 8 3 11 3 4 5 28 3 1976 4 3 12 4 7 8 10 9 35 4 1977 12 6 12 5 11 7 9 7 46 5 1978 12 o 13 9 8 7 11 2 45 8 1979 8 1 10 6 5 5 7 1 31 3 1980 8 o 5 1 5 o 13 1 31 2 1981 102 8 6 3 3 24 9 47 o Source Government of Ind1a Bullet1n on Food Stat1st1Cs D1rectorate of Econom1cs and Stat1st1cs M1n1stry of Agr1culture var1ous years II - 37 - that increased rice supplies can be assured this would have the greatest impact on nutrition in Kerala What is clear however is that there are no such assurances Maintaining low pr~ced cassava for the human consumption market prov~des a critica! element of stability in food supplies What is needed however is better integration with alternative markets which can handle surpluses when rice supplies are adequate What this requires is a larger production base and this can only be achieved with further ~ncreases in yields Conclusions Cassava serves a majar if somewhat d~stinct economies of Kerala and western Tamil Nadu role in the agricultura! In Kerala interna! r~ce product~on ~s stagnant and there is an increasing portien of the upland area being planted to higher value tree crops Food suppl~es thus rely critically on rice allocations from the central pool and more recently apparent privately-traded inflows from outside the state However in maintaining or improving the food intake and nutrition of the low income strata the options are increases in rice rationing off-take or more plentiful and cheaper cassava Compared to rice where an increase in the peor s ration allotment implies an increase for everyone cheaper cassava could target directly on the peor and would not involve subsidies from the public treasury -- theses subsidies are born by the Food Corporation of India and not the Kerala State government (George 1985) The design of a food and nutrition policy in Kerala is heavily dependent on the prognosis for rice production in India as a whole both given that food zon~ng is a policy of the past and that rice stocks in the central pool have increased in the mid-1980 s Nor should pol~cy makers appear insensitive by suggesting that the peor should JUSt eat cassava Pure pragmatism suggests that the calorie intake of the peor is crit~cally low and that cassava can be as cheap a means as any of increasing calorie intake In Tamil Nadu on the other hand a potential growth industry much like the case of Indonesia ex~sts in the starch and tapioca pearl market The industry is constrained by lack of raw material for processing and for farmers there is no restrictions on finding market outlets for their production Prices are in most respects relatively stable and any increases in yields will directly improve farmer incomes The ~ssue then is how much higher farm level yields can be ra~sed in these two states over the relatively high level which farmers already achieve Such increases will almost certainly depend on higher yielding varieties The research of the CTCRI suggests that there is scope for doing this in Kerala An issue wh~ch CTCRI is very conscious of is that the qual~ty characteristics of these improved varieties shall have te remain high since cassava is essentially consumed in a fresh form In Tamil Nadu on the other hand there are no such restr~ctions other than that the yield gap to be exploited there appears to be much ~aller Southern India represents ene of the few situat~ons in Asia (Java is the other) where the only frontier for cassava to exploit is the y~eld frontier II - 38 - Appendl.X 2 1 A synthesis of production and utilization The uncertainty surrounding the cassava production estimates and the paucity of data on cassava consumption 1.n 1.ts various end uses makes the development of a consistent supply and distribut1.on series a speculative enterprise The exercise will be attempted by first separating Kerala and Tamil Nadu then reviewing the available consumption data for each state and finally integrating these est1.mates with the production estimates The result provides the basis for the evaluation of cassava markets and demand in southern India Kerala An analysis of cassava utilizat1.on must begin with an est1.mate of human consumption of fresh roots Several estimates ex1.st but as can be seen in Table 2A 1 there is a substantial range in these estimates Given that Kumar' s sample introduces a substantial upward bias in the cassava consumption estimate -- consumpt1.on is higher in the southern districts in rural areas and in the lower income strata -- the striking feature is the difference between the estimates from food balance sheets and those from sample surveys The George and Kumar samples have upward biases in their estimates of per capita consumption The National Sample Survey is probably the best structured sample and thereby estimate of consumption levels Since fresh human consumption is considered the largest single market for cassava the difficulty arises of how to account for the difference between the consumer sample estimate and that derived from production estimates in the food balance sheets Dried cassava chips are also produced in Kerala principally in the northern districts and primarily in the period October to April These chips go into various end uses Dried cassava can be prepared in the home and eaten especially when fresh cassava is not available Cassava flour is also produced by grinding the ch1.ps At least one factory operates in Malappuram exactly for this purpose The flour is in turn used to produce f1.ne noodles Often the flour is produced in the home Also large starch factories also buy chips for process1.ng particularly for glucose production Finally from 1955 to 1966 cassava chips were exported After that exports ceased until just recently and since 1977 Ind1.a has again been exporting modest amounts of cassava chips Statistics on production and utilization of cassava chips are practl.cally non-existent The Tapioca Market Expansion Board provides the single estimate of household consumption of processed cassava products and estimates an annual consumption of 9 5 kg per capita of dried cassava It can only be assumed that cassava flour is included in this figure Cassava chip exports were initiated again in 1977 after a lull of about 10 years Exports remain small and 1.rregular Imports into the European Commun1.ty from India were 7 949 t in 1977 37 182 t in 1978 26 799 t in 1979 and 11 915 t in 1980 Chips purchased by the starch factories are assumed to be included in starch production figures This leaves only potential exports of dried cassava to other states Data on transport through selected checkposts for the period May 1975 to May 1976 give the following figures II - 39 - Table 2A 1 Ind1a D1fferent Est1mates of Per Cap1ta Consumpt1on of Fresh Cassava 1n Kera 1 a Annua 1 Sample Sample Per Cap1ta So urce S1ze Structure Per1od Consumpt1on Kumar 43 households Tr1vandrum D1str1ct Feb-Sept 1974 171 9 George lOO households Nat1onal Sample Survey 890 households Tap1oca Market unknown Expans 10n Board U N Dept of Econom1c and Soc1al Affa1rs Govt of Kerala Rural Only Bottom 50% of Income S trata Two V1l1ages Rural Only Complete S tate Rural and Urban All but One D1str1ct Rural and Urban Food Balance Tables Food Balance Tables Nov 1977 114 7 Qct 1973-June 78 3 1974 1971 56 5 1961/62-1970/71 208 4 1974 276 Sources Kumar 1979 George 1979 Government of Ind1a 1973/74 Government of Kerala 1972 U N Department of Econom1c and Soc1al Affa1rs 1975 Government of Kerala 1977 Tapioca chips Dry Tapioca Quantity (M T ) N A 90 150 Value (100 000 rupees) 78 80 44 34 At the Kozhikode wholesale market the price for cassava chips in this period was 62 rupees/100 kg which impl~es a volume of tapioca ch~ps of 12 710 t On the other hand the per ton price for dr~ed cassava implied by the above value and volume figure is 49 rupees/t a figure undervalued by at least a factor of ten A selection of either the volume or value figure is arbitrary Processing the chips into starch is possible but 90 thousand tons is a bit excess~ve in relation to starch production capacity in Tamil Nadu Moreover assembly of th~s volume is a bit large compared to more recent international export volumes It is therefore assumed that 90 thousand quintals (100 kg) were exported to Tamil Nadu implying a total export volume for the two products of 21 725 t Starch is the other major consumption form of cassava in Kerala The industry is reckoned to run at undercapacity and to be a much more minor producer than Tamil Nadu A listing of reported starch plants -- (Table 2A 2) although not necessarily a complete listing-- and their estimated annual production gives a starch production figure of approximately 57 thousand tons An alternative unpublished estimate for 1977/78 is 110 808 t of starch (State Planning Board private communicat~on) The latter figure would imply a much larger industry than is commonly reckoned The final entry in the accounting of cassava utilization ~n Kerala is root export to Tamil Nadu Most reports on the starch industry in Tamil Nadu cite imports of cassava roots from Kerala The roots princ~pally come from Trichur district in the north Estimates of these exports are few Hone (1974) presents an estimate of 400-800 thousand tons and cites a figure that licenced exports of up to 400 thousand tons are permitted This is a remarkable volume considering that road transport is relatively scarce and expensive--transport costs add as much as 40~ to root purchase price in Kerala A transport price of 150 rupees per ton was cited (f~eld notes 1982) compared to a wholesale root price in Trichur of 519 rupees in 1981 The higher cost of root production in Kerala together with the transport cost is bound to make cassava roots from Kerala competitive only outside the pr~ncipal harvest season in Tamil Nadu Moreover cassava production in Trichur district is one of the lowest in Kerala producing 114 thousand tons in 1980/81 A more reasonable est~ate ~s probably in the range of 50 to 75 thousand tons A synthesis of these various consumption estimates is presented ~n Table 2A 3 for the year 1977 Comparing the consumption aggregate to the 1977/78 production figure that is after the production series had been radically revised downward due to the crop cutting survey reveals that abou~ a mill~on tons still remain unaccounted for Wastage in an economy such as Kerala with the small distances to market and the well developed marketing serv~ces ~s probably small but may be assumed to be ~n the ne~ghborhood of 10 to 12% At this point there is no more justificat~on for revising the consumption figure upward as for revising the production figure downward Assuming that the human consumption figure is underestimated and putting the remainder in that category would imply a per II - 41 - Table 2A 2 Ind1a Est1mated Capac1ty and Output of Starch Plants 1n Kera la Plant Lekshm1 (Qu1lon) Tap1oca Products (Tr1chur) Mode Chem1cal Sago (Qu1lon) Pemba Starch (Qu1lon) 50 small scale plants Total Capac1ty (t of starch/day) 80 t 100 t 10 t 10 t 3 t Product1on Est1mate ( t/year) 15 125 17 500 1 500 1 500 21 500 57 125 Source Report of the Sub-Comm1ttee of the Tap1oca Market Expans1on Board Department of Food Government of Kerala Tr1vandrum 1972 II - 42 - Table 2A 3 Ind1a Est1mates of Product1on and Ut1l1zat1on of Cassava 1n Kera 1 a 1977/78 Est1mate Useage (t) Human Consumpt1on-Fresh 1,854 850 1 Human Consumpt1on-Dr1ed 225 045 2 Starch 110,808 3 Internat1onal Export-Ch1ps 7 950 4 Interstate Export-Ch1ps 12,700 5 Interstate Export-Roots 75 000 6 Waste 502 630 Total Ut1l1zat1on Product10n Convers1on Rate 1 o 2 75 4 5 2 75 2 75 1 o 1 o Fresh Root Est1mate ( t) 1 854 850 618 875 498 636 21,860 34 925 75 000 502 630 3 606 776 4 188 600 Sources 1 Nat1onal Sample Survey 1973/74 2 Tap1oca Market Expans1on Board 3 Kerala State Plann1ng Board 4 Renshaw 1983 5 Govern- ment of Kerala Stat1st1cs for Plann1ng 6 Est1mate II - 43 - capita consumption level of 103 kg/year Compared to the other sample est~mates this is not unreasonable but certainly suggests that earlier estimates of per cap1ta consumption from food balance sheets were substantially overest1mated generally by more than 100Á Tamil Nadu The market for cassava in Tamil Nadu as compared to Kerala is dom1nated by demand for industrial uses as opposed to food uses The starch and tapioca pearl industry centered in Salem District is considered to be the maJor end user of cassava in Tam1l Nadu There are 611 starch factories 1n Tamil Nadu 497 of which are located in Salem District and the other 114 of which are located in Dharampuri South Arcot and Coimbatore districts (Salem Starch and Sago Manufacturers s Cooperative private communication and Uthamalingam 1980) Utilizat1on of cassava roots would then follow from the operational character1st1cs of these plants Uthamalingam (1980) selected a sample of 30 starch and pearl factories in Salem town and 1n outlying rural areas The operational structure is given in Table 2A 4 There are 228 pearl factor1es and 269 starch factories in Salem and assuming a distribution of 75% small-scale and 25% large-scale leads to an average annual output per factory of 499 t This annual average starch output thereby implies an annual production level of 248 thousand tons in Salem District and an add1tional 57 thousand tons in the three adjacent districts Uthamalingam (1980) provides alternative estimates based on the quant1ty shipped by railway and that purchased by the Salem Sago and Starch Merchants Associat1on (Table 2A 5) These are only about one-th1rd of the above estimates The rail shipments obv1ously do not include the starch consumed locally -- a food habits survey by the Prote1n Foods Association of India suggests significant local consumption of pearl -- or that transported by road and therefore provides only a minimum est1mate of production and an idea of variation of production from year to year The estimate based on per factory output impl1es root ut1lization of 992 thousand tons in Salem and 228 thousand tons in the adJacent districts assuming the relatively high conversion rate reported in Tamil Nadu of 4 1 Most reports suggest that food usage of the cassava root 1s relatively minimal in Tamil Nadu The 1973/74 National Sample Survey reports an average annual rural consumption of cereal substitutes of 4 1 kg/year for the whole state It is probable that this figure includes only cassava but it is not certain what percentage would be root and what would be processed cassava Since the only reported consumption in Tam1l Nadu is for rural areas it is probable that this figure only includes root consumpt1on This would imply a total food consumption of 125 thousand tons The recapitulation of the consumption together w1th an assul~U'd 10% wastage gives a total figure of 1 514 thousand tons wh1ch compares favorably with the production estimate of 1 682 thousands tons in 1978/79 and 1 591 thousand tons in 1979/80 A small change in the starch convers1on rate could account for any difference The production and consumption data would appear to be more or less consistent at least since the 1977/78 crop year II -44- Table 2A 4 Ind1a Character1st1cs of Starch and Pearl Factor1es 1n Salem D1str1ct Tam1l Nadu 1978/79 11 Starch Pearl Sma11 Large Small Large Root Input (t) 1 629 6 2 416 1 1 635 3 3 287 3 Starch Output (t) 431 6 652 8 411 8 822 o Convers1on Rate (r) 26 5 27 2 25 2 25 o Average Operat1on Per1od 135 144 175 184 ( days) 11 In Salem D1str1ct there are 269 starch factor1es and 228 tap1oca pearl faetones Source Uthamal1ngam 1980 II - 45 - Table 2A S Ind1a Annual Ra1l Sh1pments of Starch and Pearl from Salem and Purchases by the Salem Sago and Starch Merchant s Assoc1at1on 1970-1977 Ra1l Sh1pments Assoc1at1on Purchases Year Pearl Starch Pearl Starch t t t t) 1970 52 589 39 553 N A N A 1971 55 171 28 987 N A N A 1972 41 133 41 488 N A N A 1973 22 249 41 102 N A N A 1974 18 871 42 822 N A N A 1975 44 774 45 827 N A N A 1976 36 394 30 656 38 605 29 583 1977 55 702 35 081 55,095 26 596 Source Uthamal1ngam 1980 II - 46 - Other States For the sake of completeness Andhra Pradesh is the only other state w~th anywhere close to a significant production volume Product~on in this state was 88 2 thousand tons in 1979/80 and 171 O thousand tons in 1980/81 This volume is comparable to about 10% of the production of Salem District Cassava is a rainfed crop in Andhra Pradesh and is principally grown in East Godavar~ District The cassava root is used exclusively in a small cassava pearl industry located in the district III ~Trends and D1str1bUt1on of Ch1nese Cassava Product1on and Use 1820-1984 (A technolog1cal and econom1c exam1nat1on of h1stor1cal development and future potent1al \/ / o Bruce Stone A paper prepared for the Internat1onal Center for Trop1cal Agr1culture Palm1ra Colomb1a Internat1onal Food Pol1cy Research Inst1tute Hash1ngton D C December 1986 TRENOS ANO OISTRIBUTION OF CHINESE CASSAVA PROOUCTION ANO USE 1820 - 1984 Product1on trends and d1str1but1on No off1c1al nat1onal data ser1es for cassava 1n the Peoples Republ1c have been publ1shed by Ch1nese author1t1es It 1s poss1ble to obta1n est1mated ser1es from the Food and Agr1cultural Organ1zat1on of the Un1ted Nat1ons 1 Such ser1es are based on assumed annual 1ncrements 1n harvested area for most years and somewhat less regular but a s1m1lar monoton1cally non-decreas1ng set of est1mates for product1on Y1elds appear to be der1ved from the rough area and product1on est1mates by calculat1on The only f1gure among these wh1ch appears to have come from a Ch1nese source 1s the 3 m1ll1on ton product1on f1gure c1rca 1980 prov1ded unoff1c1ally asan undated est1mate to the 1982 CIAT delegat1on by one of the agr1cultural sc1ence 1nst1tutes v1s1ted 1n Guangdong Earl1er work 2 has concluded that the ent1re FAO ser1es for root and tuber crops bears l1ttle relat1on to the aggregate ser1es publ1shed s1nce 1979 by Ch1nese stat1st1cal author1t1es 3 It 1s now also clear that the FAO le g FAO Supply Ut1l1zat1on Tapes 1984 Standard1zed Commod1ty Balance Tape 1984 Rome Product1on Yearbook Tape 1984 Rome 1985 Rome 1985 FAO 1985 and FAO 2sruce Stone An Exam1nat1on of Econom1c Data on Cassava Product1on Ut1l1zat1on and Trade a paper prepared for the Internat1onal Center for Trop1cal Agr1culture (CIAT) Internat1onal Food Pol1cy Research Inst1tute Wash1ngton D C August 1983 3e g He Kang et al Zhongguo Nongyebu [M1n1stry of Agr1culture of ~h1na] {eds ) Zhongguo Nongye N1anJ1an 1980 [Agr1cultural Yearbook of Ch1na 1980] (Be1J1ng Nongye Chubanshe [Agr1cultural Publ1sh1ng House] 1980) and Zhongguo GUOJ1a TongJ1JU [State Stat1st1cal Bureau] Zhongguo TongJ1 N1anJ1an - 1983 [Stat1st1cal Yearbook of Ch1na 1983] (Be1J1ng TongJ1 Chubanshe [Stat1st1cal Publ1sr1ng Hruse] 1083) \. N' Known Cassava Grow1ng Reg1ons of the People s Republlc of Ch1na (see text for detalls) .¡..Ocl ~\ (iJ..f' ... ,... ... _ea.... o ~ 0<'1 0\.1 ou ..... _.,. o H ;!?/."" \.)L:S - .. M .. d -1 ~ ¡ ! i -·"'"' j 'i~ ~¡'-> He .,. ~\-1'\ ~'§ Pr -.. o~ ~ '::.\ 5'f ~ ¿:}8 .. .. .,. ... *Locat1on as descr1bed by CIAT delegat1on 1982 as the most south~1estern t1p of Ha1nan Island - 2 ser1es for cassava ~se confl1ct w1th off1c1ally publ1shed ser1es for one of the two pr1nc1pal grow1ng reg1ons and w1th scattered nat1onal est1mates for 1nd1v1dual years found elsewhere 1n Ch1nese publ1cat1ons S1nce 1984 the FAO has taken account of sorne of the recent 1nformat1on 1n formulat1ng current root and tuber crop est1mates for publ1cat1on 1n FAO Product1on Yearbooks But much recent 1nformat1on has not been reflected 1n FAO ser1es and add1t1onal work 1s requ1red to obta1n a rel1able 1mpress1on of long term trends for 1nd1v1dual crops 1nclud1ng cassava Accord1ng to Ch1nese sources 4 cassava had been 1ntroduced 1nto Ch1na from South Amer1ca v1a nanyang [the South Seas or Pac1f1c Ocean] by 1820 although 1t 1s not clear whether 1t entered Guangdong Prov1nce d1rectly from the West or whether 1t was 1ntroduced 1nd1rectly follow1ng reg1onal cult1vat1on 1n Sr1 Lanka Ind1a or Indones1a By far the ma1n Ch1nese produc1ng area 1s the extreme south below the Trop1c of Cancer (23 5°N) espec1ally Guangdong 4L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong [Cassava Cult1vat1on and Use] (Guangzhou Guangdong KeJ1 Chubanshe [Guangdong Sc1ent1f1c and Techn1cal Publ1sh1ng House] 1981) author s preface and p 4 Cassava 1s conf1rmed to have been grown 1n Ch1na for more than 100 years 1n Zhongguo Kexueyuan 0111 YanJ1Usuo J1ngJ1 0111 YanJlush1 [Ch1nese Academy of Sc1ences Inst1tute of Geography Econom1c Geography Research Room] Zhongguo Nongye 0111 Zonglun [A General Treat1se on Ch1na s Agr1cultural Geography] (BelJ1ng Kexue Chubanshe [Sc1ent1f1c Publ1sh1ng House] 1980) p 129 1820 was also the 1ntroduct1on date ment1oned dur1ng a spr1ng 1982 delegat1on from the Internat1onal Center for Trop1cal Agr1culture (CIAT) and recorded 1n James H Cock and Kazuo Kawano Cassava 1n Ch1na unpubl1shed tr1p report CIAT Palm1ra Colomb1a June 1982 p 1 However Mushu Za1pe1 yu L1yong clearly 1nd1cates that 1820 1s the earl1est record of cassava cult1vat1on so far uncovered the 1ntroduct1on date may well have been earl1er 3 Prov1nce and Guangx1 Zhuang Autonomous Reg1on Of the two product1on has typ1cally been greatest 1n Guangdong Cassava 1s also cult1vated 1n FuJlan Yunnan Hunan Gu1zhou and Ta1wan Prov1nces but much less extens1vely and to a very m1nor extent 1n Hube1 J1angx1 ZheJlang and S1chuan Sorne est1mates of prov1nc1al cult1vated area gleaned from Ch1nese sources are arranged 1n Table 1 Wh1le cassava had been 1ntroduced 1nto Guangdong and Guangx1 by the f1rst half of the 19th century and a book devoted to cassava plant1ng methods had been publ1shed as early as 1900 the f1rst cult1vat1on record 1n FUJlan 1s 1920 and 1n Ta1wan 1929 Introduct1on dates for most other prov1nces were cons1derably later Hunan 1941 Gu1zhou 1942 ZheJlang 1954 and J1angx1 1959 Cult1vat1on of cassava 1n Yunnan though potent1ally beg1nn1ng earl1er was est1mated at only two thousand hectares 1n 1960 Most farmland 1n these prov1nces fall w1th1n what 1s descr1bed 1n Ch1nese sources as the expans1on area north of the Trop1c of Cancer and south of 30°N There 1s exper1mental cult1vat1on of cassava even north of 30°N w1th the northernmost plant1ngs at the Hebe1 Forestry Sc1ence Inst1tute at 39°20 N These exper1ments began dur1ng the fam1ne years 1n 1960 and 1961 1n Hube1 Anhu1 J1angsu Shaanx1 Shandong L1aon1ng S1chuan and Hebe1 wh1ch const1tute the f1rst record of cassava related act1v1t1es 1n these prov1nces 5 Cassava 5L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong author s preface and pp 4 9 and 10 4 - Table 1 Area Sown w1th Cassava 1n Ch1na and MaJar Ch1nese Cassava-Grow1ng Prov1nces 1943-1984 1943 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 Notes Ch1na Guangdong Guangx1 FUJlan Ta1wan 365 3 (470 530) ( 350) 33 4 * <149 * <201 167 3 * <223 * <236 ( 200) ~195 ~158 ~159 (thousand hectares) 41 5 37 6 48 5 41 3 67 5 62 6 93 o 104 3 132 6 118 8 127 9 104 4 >6 7 ( 183 5/158 7) 153 4 154 3 158 5 102 2 70 3 73 7 124 7 145 6 129 6 124 5 107 9 100 8 131 9 110 5 74 6 131 o 156 o 207 8 190 4 175 2 120 6 94 o 8 o 9 o 10 4 10 7 10 6 10 9 12 3 11 9 13 o 17 2 18 2 20 2 19 8 20 5 21 o 22 o 25 o 25 9 24 7 24 6 24 6 24 3 26 8 21 8 22 2 22 3 19 5 17 o 14 9 13 9 9 9 5 8 5 2 Hunan ZheJ 1 ang Yunnan Gu1zhou J1angx1 2 o o 6 ( o 3) Empty data cells 1nd1cate that the stat1st1cal 1nformat1on 1s not ava1lable and do not denote zero values Parentheses enclose rough est1mates for the 1nd1cated or nearby years The appl1cable years for parenthes1zed est1mates were not stated 1n the source Other prov1nces where farmers grow cassava 1nclude Hube1 and S1chuan but sown area 1s m1nor Ta1wan Prov1nce lS now normally not 1ncluded 1n nat1onal aggregated stat1st1cs for the People s Sources Guangx1 Ta1wan Republ1c of Ch1na although separate data entr1es for Ta1wan are not unusual among PRC stat1st1cal compend1a Ta1wan 1s probably 1ncluded 1n the 1961 nat1onal f1gure however * These f1gures probably overest1mate off1c1ally recorded plant1ngs by 20-40 thousand hectare See Table 7 Guangx1 J1ngj1 N1anJ1an B1anj1bu [Guangx1 Econom1c Yearbook Ed1tor1al Oepartment] (eds ) Guangx1 J1ngJ1 N1an]1an 1985 [Guangx1 Econom1c Yearbook 1985] (Nann1ng Guangx1 J1ngJ1 N1anj1an B1anj1bu 1985} pp 531 and 593 The 1976 f1gure was conf1rmed 1n Guangx1 Nongye 0111 B1anx1ezu [Guangx1 Agr1cultural Geography Ed1tor1al Board] (eds ) Guangx1 Nongye 0111 [Guangx1 Agr1cultural Geography] (Nann1ng Kexue Chubanshe [Sc1ent1f1c Publ1sh1ng House] 1980} p 76 The lower f1gure for 1962 1s from L1ang Guangshang (ed ) 1 Mushu Za1pe1 yu L1yong (Guangzhou Guangdong KeJ1 Chubanshe 1981) p 9 Republ1c of Ch1na Execut1ve Yuan 01rectorate-General of Budget Account1ng and Stat1st1cs Stat1st1cal Yearbook of the Republ1c of Ch1na 1985 (Ta1pe1 Republ1c of Ch1na 1985} p 281 The 1952 54 f1gures were added from Republ1c of Ch1na 01rectorate-General of Budget Account1ng and Stat1st1cs Stat1st1cal Yearbook of the Republ1c of Ch1na 1982 (Ta1pe1 Republ1c of Ch1na 1982} p 115 Ch1na and other Prov1nces The 1978 f1gure 15 from Zhongguo Kexueyuan 0111 YangJ1usuo J1ngj1 0111 YanJ1USh1 [Ch1nese Academy of Sc1ence Inst1tute of Geography Econom1c Geography Research Laboratory] Zhongguo Nongye 0111 Zonglun [A General Treat1se on Ch1nese Agr1cultural Geography] (Be1J1ng Kexue Chubanshe 1980) p 129 The 1981 f1gure 1s from James H Cock and Kazuo Kawano Cassava 1n Ch1na unpubl1shed tr1p report - 7 Table 2 Cassava Product1on Area and Y1eld 1n Guangx1 Zhuang Autonomous Reg1on 1950-1984 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 Product1on Area (Gra1n Equ1valent) (Fresh Root) Tons Tons (Hectare~) 30 045 39 365 41 870 36 635 42 535 35 365 58 280 91 000 165 205 140 330 88 045 115 855 189 260 152 335 160 225 167 835 84 435 173 715 162 120 216 750 235 990 211 295 262 270 206 545 170 765 260 425 187 065 141 865 258 295 312 645 481 215 484 280 468 255 326 680 241 180 150 225 196 825 209 350 183 175 212 675 176 825 291 400 455 000 826 025 701 650 440 225 579 275 946 300 761 675 801 125 839 175 422 175 868 575 810 600 1 083 750 1 179 950 1 056 475 1 311 350 1 032 725 853 825 1 302 125 935 325 709 325 1 291 475 1 563 225 2 406 075 2 421 400 2 341 275 1 633 400 1 205 900 41 507 37 567 48 493 41 340 67 453 62 647 93 013 104 320 132 567 118 840 127 913 104 353 183 547 153 433 154 307 158 520 102 220 70 300 73 667 124 733 145 600 129 613 124 480 107 900 100 847 131 900 110 473 74 567 131 020 155 993 207 760 190 387 175 173 120 640 94 001 Y1eld (Gra1n Equ1valent) (Fresh Root) T/Ha T/Ha o 724 1 048 o 863 o 886 o 631 o 565 o 627 o 872 1 246 1 181 o 688 1 llO 1 031 o 993 1 038 1 059 o 826 2 471 2 201 1 738 1 621 1 630 2 107 1 914 1 693 1 974 1 693 1 903 1 971 2 004 2 316 2 544 2 673 2 708 2 566 3 619 5 239 4 317 4 431 3 153 2 823 3 133 4 362 6 231 5 904 3 442 5 551 5 156 4 964 5 192 5 294 4 130 12 355 11 004 8 6ó9 8 104 8 151 10 535 9 571 8 467 9 872 8 467 9 513 9 857 10 021 11 581 12 718 13 365 13 539 12 829 Notes Cassava product1on and y1eld data are often quoted 1n Ch1nese stat1st1cal sources en a gra1n equ1valent bas1s S1nce 1964 the convers1on to gra1n equ1valence for all root and tuber crops has meant d1v1d1ng the fresh we1ght by f1ve although th1s would undervalue cassava sweet potatoes and taro relat1ve to most cereal crops 1n terms of calarles per un1t we1ght It 1s assumed that the product1on and y1eld data 1n the source for th1s table appeared 1n Guangdong 6 - Internat1onal Center for Trop1cal Agr1cultural Research (CIAT) Cal1 Colomb1a June 1982 pp 1 2 The 1961 f1gure 1s from L1ang Mushu Za1pe1 yu L1yong p 9 Th1s source also stated that nat1onal cassava sown area rema1ned around 5 m1ll1on mu dur1ng the 1960s (300-367 000 hectares assum1ng 4 5-5 5 m1ll1on mu ) The f1gure for Hunan ZheJ1ang and J1angx1 comb1ned was g1ven as around 5 000 mu {333 ha ) 1n each year of the 1960s The overest1mates for Guangdong for 1965 1970 1975 1978 1979 and 1982 84 are from Table 7 A 1981 overest1mate of 201 thousand hectares was also calculated The 1979 and 1982-84 est1mates are relat1vely clase approx1mat1ons The 1965 1970 1975 and 1978 f1gures probably overest1mate by at least 20-40 thousand hectares See Table 7 The 1943 and 1972 f1gures are from L1ang Mushu Za1pe1 yu L1yong p 9 and the 1981 est1mate 1s from Cock and Kawano Cassava 1n As1a p 1 seems te enJOY sorne very m1nor farmer cult1vat1on 1n S1chuan but probably not elsewhere w1th1n the exper1mental area In fact 1t 1s not yet clear from the est1mates of nat1onal Guangdong and Guangx1 cult1vat1on assembled 1n Table 1 that cassava expans1on efforts have resulted 1n s1gn1f1cant 1ncreased plant1ngs outs1de of those two prov1nces In the absence of a rel1able nat1onal cassava product1on ser1es the best approx1mat1on would be te synthes1ze product1on ser1es for Guangdong and Guangx1 Fortunately complete 1950-84 ser1es for Guangx1 were publ1shed 1n 1985 (Table 2) These data though not necessar1ly w1thout flaws prov1de the best understand1ng of year te year movements 1n cult1vat1on and y1elds A glance at Table 2 w1ll - 8 - gra1n equ1valent form The or1g1nal data have therefore been mult1pl1ed by f1ve to calculate fresh root we1ght Source Guangx1 JlngJl N1anJ1an B1anJ1bu (eds ) Guangx1 J1ngJ1 N1an]1an 1985 (Nann1ng Guangx1 J1ngJ1 N1anJ1an B1anJ1bu 1985) pp 531-532 and 593 conf1rm that the 35-year per1od encompasses cons1derable var1at1on 1n both Our1ng the 1950s sorne government 1n1t1ated efforts were undertaken to expand cult1vat1on of cassava wh1ch was v1ewed as a crop capable of prov1d1ng cons1derable bulk and calor1c content per un1t area One cannot rule out the poss1b1l1ty however that a port1on of the 1mpl1ed 1ncrease 1n cult1vat1on reflected prev1ously unreg1stered cassava areas eventually 1ncluded 1n stat1st1cal coverage espec1ally dur1ng the format1on of agr1cultural producers cooperat1ves (1954 56) and the people s communes (1958) Elsewhere6 1t has been demonstrated that most of the 1mpl1ed growth 1n total root and tuber crop area s1nce 1952 1s l1kely to be real the actual f1gures rema1n1ng 1n all probab1l1ty w1th1n about 5 percent (below) the off1c1al data The cons1derable 1ncrease 1n cassava area 1n 1958 parallels an even larger reported 1ncrease for all root and tuber crops Wh1le 1958 was ayear of extreme stat1st1cal d1stort1on cast1ng doubt on 6Bruce Stone Crop Product1on An Analys1s of Ch1nese Data on Root and Tuber The Ch1na Quarterly September 1984 pp 594 630 - 9 - the magn1tude of the 1ncrease the 1mpl1ed growth was no greater than that of 1956 much of wh1ch may have been real 1958 was also a year 1n wh1ch great efforts were made to 1ncrease foodcrop product1on by whatever means poss1ble Root and tuber crops 1nclud1ng cassava were correctly 1dent1f1ed as the eas1est means to effect a short term leap 1n bulk food product1on It 1s d1ff1cult however to accept the 1mpl1ed 1958 1ncrease 1n average y1eld toar unprecedented level espec1ally 1n v1ew of the (except for sweet and wh1te potatoes more modest) expans1on of area planted w1th other food crops and ma1ntenance of y1elds 1n that year In sum wh1le 1t appears that the total Guangx1 foodcrop data (exclud1ng cassava) have been adJusted 1n the 1985 Guangx1 Econom1c Yearbook for the stat1st1cal d1stort1on typ1cal of 1958 publ1shed mater1als 1t 1s qu1te poss1ble that those for cassava may not have been part1cularly 1n the y1eld category The decl1ne 1n 1959 area however followed by sorne recovery 1n 1960 are undoubtedly real although 1t 1s 1mposs1ble to ver1fy the exact f1gures Inflated reports of m1raculous gra1n product1on success 1n 1958 led author1t1e~ to 1ncrease area sown w1th econom1c crops 1n 1959 at the expense of staples 7 When the truth became clear (1958 had been a good but not spectacular year) 1t was too 7L1 Choh-m1ng The Stat1st1cal System of Commun1st Ch1na (Berkeley Un1vers1ty of Cal1forn1a Press 1962) Kenneth R Walker Food Gra1n Procurement and Consumpt1on 1n Ch1na (Cambr1dge Cambr1dge Un1vers1ty Press 1984) N1cholas R Lardy Agr1culture 1n Ch1na s Modern Econom1c Development Cambr1dge Cambr1dge Un1vers1ty Press 1983 - 10 - late to correct spr1ng plant1ng Sorne cornpensat1on would have been rnade w1th 1959 fall planted cassava however and 1n 1960 1n v1ew of poor harvests for all foodcrops the prev1ous year The y1eld decl1ne 1n 1960 1s cons1stent w1th w1despread natural d1sasters throughaut Ch1na est1rnated ta be the worst 1n the twent1eth century These were samewhat less severe 1n Guangx1 than 1n sorne ather prav1nces but y1elds af ather Guangx1 foad crops repartedly de l1ne by a we1ghted average of 9 percent dur1ng 1960 and 1961 8 Spr1ng planted cassava 1n part1cular 1s subJect ta 1nsect darnage dur1ng the seedl1ng per1ad and 1n the fall typhoon darnage The law area f1gure far 1961 1s cans1stent w1th bath poar stat1st1cal caverage dur1ng the per1ad and s1gn1f1cant rural d1slacat1on assoc1ated w1th the 1960-61 farn1ne thraughaut Ch1na wh1ch rnay have part1ally extended 1nta Guangx1 The large 1ncrease 1n cassava area 1n 1962 fallawed by subs1dence dur1ng the follaw1ng few years 1s alsa expla1nable 1n terrns af react1on to the 1960-61 farn1ne Geograph1c coverage rnay not have been cons1stent throughout the ser1es Q1nzhou Spec1al D1str1ct was transferred frorn Guangx1 to Guangdong 1n 1955 then back ta Guangx1 1n 1965 Q1nzhou 1ncludes the ent1re current Guangx1 coast and extends north frorn the current prov1nc1al border to the Yu R1ver then angles southwest towards the 8Guangx1 JlngJl NlanJlan B1anJ1bu [Guangx1 Econorn1c Yearbook Ed1tor1al Board] Guangx1 Jln9Jl N1anJ1an, 1985 [Guangx1 Econorn1c Yearbook 1985] (Nann1ng Guangx1 J1ngJ1 N1anJ1an BlanJlbu 1985) p 530 - 11 - border w1th V1etnam In 1976 area sown w1th foodgra1ns 1n Q1nzhou covered 461 333 hectares Area planted w1th root and tuber crops 1n the western d1str1ct of Guangdong c1rca 1957 (1nclud1ng Q1nzhou Spec1al D1str1ct and ZhanJlang Prefecture) cons1sted of 28 3 percent of total area sown w1th foodcrops (exclud1ng soybeans) a l1ttle less than 5 percent of wh1ch was planted w1th cassava and mao potatoes These reports suggest that someth1ng on the arder of 6 thousand hectares of cassava were transferred from Guangx1 to Guangdong 1n 1955 then (potent1ally more extens1ve cassava area) back to Guangx1 1n 1965 Th1s could expla1n the counter trend movements of cassava area 1n the Guangx1 ser1es for 1955 and 1965 Data osc1llat1ons dur1ng the succeed1ng decade (1966-77) are less understandable as a funct1on of nat1onw1de econom1c developments and may be pecul1ar to cassava orto Guangx1 Hypotheses for expla1n1ng these osc1llat1ons 1nclude the lagged effect of earl1er shocks echoed v1a the rotat1on system (see below) and per1od1c reclamat1on 1n1t1at1ves In Guangx1 cassava 1s often grown dur1ng the early years of a reclamat1on proJect 1n arder to earn sorne econom1c return befo1e reclamat1on 1s complete When the qual1ty of farmland construct1on and f1eld preparat1on perm1ts cassava 1s often phased out to make way for more h1ghly valued crops 9Bruce Stone Crop Product1on An Analys1s of Ch1nese Data on Root and Tuber pp 612-615 9 - 12 - The low planted area f1gures for 1967 and 1968 and part1cularly the h1gh average y1eld est1mates for those years are espec1ally anomalous Although fert1l1zer use accelerated dur1ng the 1960s w1despread appl1cat1on to cassava as early as 1967-68 1s very unl1kely One 1s consequently mot1vated to hypothes1ze about a stat1st1cal qu1rk e g 1ndependent product1on and area est1mates w1th the latter underest1mated due to stat1st1cal confus1on typ1cal of the early years of the Cultural Revolut1on per1od (1966-77) Even exclud1ng 1967 and 1968 the data 1nd1cate a marked 1ncrease 1n y1elds from an average of 4 5 tons per hectare (1950 66) to 9 O tons per hectare (1969 77) or 10 3 tons per hectare (1969 84) Sorne of th1s 1ncrease per un1t product1v1ty 1s expla1nable 1n terms of 1n1t1at1on of fert1l1zer appl1cat1on and cult1vat1on of cassava on state farms w1th plent1ful access to fert1l1zers But state farms 1n Guangx1 occup1ed only 20 thousand hectares (1982) and large port1ons of th1s total were devoted to cult1vat1on of gra1n crops and sugar cane 10 It seems unl1kely therefore that 1ncreased fert1l1zer use alone can fully expla1n th1s y1eld 1ncrease In the absence of def1n1t1ve 1nformat1on what could expla1n a sudden doubl1ng of average y1elds 1n the mld-196Qs? One hypothes1s would emphas1ze techn1cal change Much of the 1mportant select1on and breed1ng work was undertaken 1n the late 1950s and early 1960s 10zhongguo GuoJla TongJlJU Zhongguo TonqJ1 N1an]1an 1983 pp The South Ch1na Trop1cal Crops Research Academy bred or selected many of the well-known var1et1es under current product1on represent1ng s1gn1f1cant 1mprovement 1n aggregate speed and quant1ty of root product1on dur1ng the 1959-62 per1od The South Ch1na Agr1cultural Sc1ence Academy 1n Guangzhou bred or selected for multlpl1cat1on and d1ssem1nat1on several other h1gher y1eld1ng var1et1es dur1ng the 1957-62 per1od 11 Part1cular attent1on pa1d to cassava dur1ng th1s per1od may also have produced 1mportant results 1n 1mprov1ng f1eld cult1vat1on techn1ques Another hypothes1s would suggest that cassava cult1vat1on on somewhat better land was 1n1t1ated dur1ng th1s per1od The Cultural Revolut1on decade (1966-77) was marked by a pol1cy of local self suff1c1ency 1n gra1n product1on and escalat1on of quota del1ver1es In sorne cases quotas were spec1f1ed 1n terms of part1cular crops needed by the state In other cases quotas were spec1f1ed only 1n terms of we1ght of staples leav1ng the cho1ce of crops to each collectlvlty of farmers Although farmers rece1ved compensat1on for quota del1ver1es tax Land taxes pr1ces were notor1ously low 1nvolv1ng an 1mpl1clt amount1ng to roughly 5 13 percent of output dur1ng th1s per1od depend1ng on locat1on were also payable 1n k1nd Taxes and quotas were therefore obl1gat1ons to be d1scharged w1th commod1t1es ach1ev1ng the h1ghest bulk y1eld per un1t area Although fresh we1ght of root and tuber crops was d1v1ded by 4 for these 11L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong pp 77 78 - 14 - account1ng purposes through 1963 and by 5 thereafter cassava may have been cult1vated and even fert1l1zed by a w1der var1ety of local1t1es 1n South eh1na w1th the express purpose of exped1t1ously d1scharg1ng these obl1gat1ons 12 The determ1nants of var1at1on dur1ng the f1nal perlad (1978-84) are somewhat eas1er to 1dent1fy w1th conf1dence The steady growth 1n y1elds 1s almost certa1nly related to an 1ncrease 1n manufactured fert1l1zer nutr1ent appl1cat1on Although average appl1cat1on levels for cassava are not known w1th prec1s1on nutr1ent appl1cat1on w1th1n eh1na as a whole tr1pled between 1976 and 1984 and doubled between 1978 and 1984 culm1nat1ng w1th an average rate of 120 6 kg /ha of sown area Eff1c1ency of ut1l1zat1on also 1ncreased dur1ng the per1od Although the average level 1n Guangx1 was somewhat lower 1t grew even more rap1dly than the nat1onal average between 1976 and 1982 (to 110 2 Kg /ha then stagnated 1n 1983 (112 4 Kg /ha ) and 1984 (109 7 Kg /ha ) parallel1ng y1eld progress 1n Guangx1 13 12For further d1scuss1on of these 1ssues see Bruce Stone eh1na s 1985 Foodgra1n Product1on Target lssues and Prospects 1n Anthony M Tang and Bruce Stone Food Product1on 1n the People s Republ1c of eh1na IFPRI Research Report no 15 (Wash1ngton De Internat1onal Food Pol1cy Research Inst1tute 1980) pp 147-149 13sruce Stone Ch1nese Fert1l1zer Appl1cat1on 1n the 1980s and 1990s lssues of Growth Balance Allocat1on Eff1c1ency and Response 1n US eongress Jo1nt Econom1c eomm1ttee (eds ) eh1na s Economy Looks to the Year 2000 vol 1 The Four Modern1zat1ons (Wash1ngton D e U S Government Pr1nt1ng Off1ce 1986 pp 453 496 and State Stat1st1cal Bureau PRe Stat1st1cal Yearbook of eh1na 1985 (Hongkong and Be1]1ng Econom1c Informat1on and Agency and eh1na Stat1st1cal Informat1on and eonsultancy Serv1ce 1985) p 283 15 - Appl1cat1on of manufactured fert1l1zers to cassava 1s l1kely to be much below the average level for all crops 1n Guangx1 except on state farms but scattered survey reports 14 conf1rm that on farmers f1elds near cassava research 1nst1tut1ons 1n South Ch1na y1elds wh1ch are comparable to the recent Guangx1 prov1nc1al averages are only obta1nable w1th fert1l1zer appl1cat1on or under good so1l and cl1mat1c cond1t1ons atyp1cal of most Ch1nese cassava grow1ng areas One of the survey respondents however also 1nd1cated that the cassava research 1n Ch1na had made s1gn1f1cant progress 1n develop1ng 1mproved var1et1es and low-cost cultural pract1ces a decade earl1er Yet the predom1nant var1et1es planted 1n the 1980s were among those selected (or bred) dur1ng the late 1950s and early 1960s (see below) The r1se and fall 1n cassava area dur1ng the 1978 84 per1od 1s attr1butable to a number of factors the most powerful of wh1ch has been the r1se and fall of opportun1t1es for export to the European Commun1ty W1th EC pressure on Tha1land (the dom1nant and low cost suppl1er) to reduce exports dur1ng the late 1970s Ch1nese exports responded to the opportun1ty Wlth rap1d growth 1n 1979 1980 and 1981 14 Delph1 Survey for the Assessment of Potent1al Y1elds of Cassava c1rculated to cassava breed1ng 1nst1tut1ons 1n Ch1na and elsewhere by J S Sarma Internat1onal Food Pol1cy Research Inst1tute 1986 The respondent who ment1oned var1etal and cultural 1mprovement a decade ago was L1u YlngJlng of the South Ch1na Inst1tute of Botany 1n Guangzhou - 16 - (Table 3) befare s1m1lar pressure eventually forced a decelerat1on beg1nn1ng 1n 1982 (w1th 1981 fall sown cassava) 15 Other c1rcumstances contr1but1ng to th1s respons1veness 1nvolve changes 1n rural 1nst1tut1ons s1nce 1978 79 farmers have been allowed more control over cropp1ng and management dec1s1ons but are also afforded less market secur1ty from the government as a guaranteed buyer At the same t1me very poor locat1ons typ1cal of many Ch1nese cassava grow1ng areas have been released from tax and quota obl1gat1ons wh1le the government 1n response to substant1al success 1n accelerat1ng nat1onal foodcrop product1on growth began emphas1z1ng h1ghe1 qual1ty 1n farm procurement 1tems compared w1th the cons1derable prev1ous per1od emphas1s on cheaper bulk1er products such as most root and tuber crops and the lowest qual1ty grades of cereal crops These cons1derat1ons coupled w1th the overall l1beral1zat1on of econom1c act1v1t1es 1n rural areas expla1ns the fall 1n cassava area toa 1984 level below that typ1cal of the pre- 1978 per1od The decl1ne 1n sown area cuts across most gra1n crops throughout Ch1na but 1s part1cularly noteworthy 1n proport1onal terms 1n the case of crops typ1cally grown 1n poorer farmlands and character1zed by low pr1ces and weak markets such as sorghum wh1te potatoes bean crops and no doubt cassava (Table 4) In Guangdong and Guangx1 although unsu1table for such a warm mo1st cl1mate 15sruce Stone An Analys1s of Ch1nese Data on Root and Tuber Crop Product1on pp 623-625 Bruce Stone An Exam1nat1on of Econom1c Data on Cassava Product1on Ut1l1zat1on and Trade 1n Ch1na pp 16 22 To European Commun1ty Only (metr1 e tons) 1963 20 977 1964 33 393 1965 72 676 1966 57 077 1967 53 173 1968 28 015 19&9 1 324 1970 4 984 1971 14 859 1972 16 070 1973 8 083 1974 4 111 1975 4 211 1976 7 253 1977 999 1978 1 327 1979 51 449 1980 335 989 1981 606 589 1982 440 181 1983 15 222 1984 143 000 Notes and Sources - 17 - Table 3 PRC Cassava Exports 1963 1984 Dr1ed Cassava Cassava Tap1oca Cassava Starch Share of EC net Total Cassava Imports Exports (percent) (metr1c tons) (metr1c tons) (metr1c tons) o 2 4 000 11 429 o 2+ 4 000 11 429 o 2+ 7 000 6 500 2 000 o o+ 1 000 2 000 o o 1 000 1 000 1 o: 51 000 5 800 2 060 6 9 336 000 20 500 2 500 9 1: 607 000 10 000 1 500 5 4 445 000 14 000 1 500 o 4 460 000 2 7 1 314 285 TotalCassa> Exports 1n Fresh Root Egu1valent~ (metr1c tor 60 657 11 948 7 403 183 522 1 067 070 1 788 073 1 343 397 1 314 285 European Commun1ty data for dr1ed cassava 1mports from Ch1na and other countr1es are comp1led from EUROSTAT and NIMEXE Analyt1c Tables for Fore1gn Trade (wh1ch are 1n clase agreement) Total dr1ed cassava cassava tap1oca and cassava starch export data are from Fooc and Agr1culture Organ1zat1on of the Un1ted Nat1ons Supply Ut1l1zat1on Accounts Tape 1984 Rome 1985 The fresh root equ1valents of all cassava exports aggregated together appear 1n FAO Standard1zed Commod1ty Balance Tape 1984 Rome 1985 The 1983 and 1984 data must be regarded as open to sorne quest1on and may be rev1sed 1n future compend1a Table 4 Area Sown w1th MaJar Cereals Bean Crops Roots and Tubers 1n Ch1na 1976 85 Sweet Other and Only Only Cereals Wh1te Sweet Wh1te & Bean Total R1ce Wheat Corn So:tbeans M1llet Sorghum Pota toes Pota toes Pota toes Cro12s Foodgra1ns (thousand hectares) 1976 36 217 28 417 19 228 6 691 4 501 4 329 10 366 10 994 120 743 1977 35 526 28 065 19 658 6 845 4 477 3 759 11 229 10 841 120 400 1978 34 421 29 183 19 961 7 144 4 271 3 456 11 796 6 800 5 000 10 355 120 587 1979 33 873 29 357 20 133 7 247 4 173 3 173 10 952 10 355 119 263 1980 33 879 29 228 20 353 7 227 3 872 2 693 10 153 9 829 117 234 1981 33 295 28 307 19 425 8 023 3 888 2 610 9 621 9 789 114 958 1982 33 071 27 955 18 543 8 419 4 039 2 783 9 370 6 916 2 454 9 283 113 463 ..... CXl . 1983 33 137 29 050 18 824 8 414 4 087 2 707 9 402 6 840 2 562 8 426 114 04 7 1984 33 179 29 577 18 537 7 286 3 797 2 384 8 988 6 426 2 562 9 136 112 884 1985 32 070 29 218 17 694 7 718 8 571 108 845 Sources Most data were converted from Ch1nese un1t f1gures or were calculated from data appear1ng 1n State Stat1st1cal Bureau (SSB) PRC Stat1st1cal Yearbook of Ch1na 1985 (Hong Kong and BelJlng Econom1c Informat1on and Agency and Ch1na Stat1st1cal Informat1on and Consultancy Serv1ce Centre (CSICSC) 1985) p 253 1985 data were added from SSB PRC Ch1na A Stat1st1cal Surve:t 1n 1986 (BelJlng CS!CSC 1986) p 37 1982 84 f1gures for sweet potatoes and for wh1te potatoes are from He Kang et al Zhongguo Nongye N1anJ1an B1anJ1 We1yuanhu1 [Ch1nese Agr1cultural Yearbook Ed1tor1al Comm1ttee] (ed ) Zhongguo NongJ~e N1an]1an 1983 [Agr1cultural Yearbook of Ch1na 1983] (BelJlng Nongye Chubanshe [Agr1cultural Publ1sh1ng House] 1984) p 40 He Kang et al Zhongguo Nongye NlanJlan 1984 (BelJlng Nongye Chubanshe 1985) p 88 He Kang et al Zhongguo Nongye N1an]1an 1985 (BelJlng Nongye Chubanshe 1986) pp 147 148 The est1mates for sweet and wh1te potatoes 1n 1978 are from Bruce Stone An Analys1s of Ch1nese Data on Root and Tuber Crop Product1on The Ch1na Quarterly September 1984 p 628 - 19 - wheat had been cult1vated for 1mport subst1tut1on purposes W1th relaxat1on of th1s uneconom1c emphas1s on wheat sown area decl1ned 1n the two prov1nces Less drast1cally area sown w1th several other food crops such as paddy sweet potatoes sorghum and m1llet also fell 1n favor of econom1c crops espec1ally sugarcane (Tables 5 and 6) After 1979 1s 1t poss1ble to conf1rm that the trends 1nd1cated for Guangx1 are representat1ve nat1onally? Even w1thout nat1onal data the add1t1on of ser1es for Guangdong would prov1de a reasonable proxy Unfortunately cassava ser1es for Guangdong are unava1lable but a very rough approx1mat1on may be d1scerned from Table 5 The left hand column 1s compr1sed of f1gures quoted for Guangdong spec1f1cally The center column 1s der1ved from data appear1ng 1n the 1984 and 1985 Guangdong Stat1st1cal Yearbooks These data are not est1mates of cassava area ~ se but are formed by deduct1ng data for sugar cane peanuts sesame JUte kenaf and tobacco from f1gures for total area planted w1th econom1c crops The est1mates 1n parentheses to the r1ght more closely approx1mate cassava plant1ngs 1nasmuch as area sown w1th all o1l crops all f1bers and med1c1nal herbs have also been deducted from the econom1c crop area along Wlth sugarcane and tobacco on the bas1s of recent Agr1cultural Yearbook of Ch1na volumes to arr1ve at the res1duals Dur1ng the recent decade at 1east cassava has been class1f1ed as an econom1c crop 1n product1on stat1st1cs rather than as a foodcrop and the calculated res1dual should be predom1nantly 20 - ~ :!! ¡;; " 8 " ~ "' ~ " " " "' ~ " " - ~ ~ ~ - - ~ "' ~ :: - " - ~ ~ " o " ~ ~ ~ o " "' g " o (¡! "' " ~ - ~ - "' !1 :: ;;; ~ "' "' "' - " ~ o ~ " - o - " "' "' ~ ~ -- = :: - -" - ~ ~.o - !1: "' ~ !'! -" - :¡: ~ "' :;; " ~ " " - " - - - ~ - "' " ~ - - - ~ -.i"' - ~ o "' o " z ~ " 3 o "' - !! - - ~ - ~ - - - o :; - E ;i!:: " o o ~ ~ ~ o É l;' " "' 8 ~ " "' 8 8 ::; .: ~ " - ~ ~ .. ~ o o --- o " ~ - - - "' ¡,¡ ;; g " ~ :: :; ~ - "' "' "' :;; :: ~ - :; - :;; "' t:; " "' :0 ;;: :;; :: 8 o " - ~ " ~ " :; ;:; :: "' :;¡ ~ - - "' - "' ;:: :;; " "' :;; ~ " o ::; g - - :;; - ~ o - - - ::: o - "' - ~ :: ~ " - - " - - ;; ::: - - - - ~ " "' o o ~ - " :;; -" "' ::: ;¡; ;;; :;; -" :;: ::: :;; "' - :;: ~ "' - " ¡:¡ ::; - - - "' "' " - " - - - ::: - " " " - " " S "' " ~ o " ::: 8 :;: :;; ~ :;; " ~ - ~ ~ - ::: " -~ - - -- " ::: - ~ - - - - - ;; .. " - ~ "' 8 o " "' - ~ -;:: :;; " ;:: "' :0 " - " ; - ~ - ~ :: :: ~- " ::: - - - - - ~ - -- " :;; - g " " :;; " g " - " ::: :; ~ - - ~ " ;¡ -- ~ - "' ~ -~ "' - - - - - "' - :0 .. - z ..j "' ~ ::: - -, - ., " ~ 1 ~ :: ~= ~¡ " - -1 = ~~- o ,_ "" ~ ~~- ~ -" - Q ;j" - --- -~ --- "' :0 o;J " ~¡;:, 1 " ... 00 o .,.._ ...... = o-.,.._ ... __ ,_ ~­;;:~.,..--- ..... o .................. «> -oo.., -- .... o "'oo--:4 ~-­o- ::o·~~ = 21 o o o o ::: go~~~ ............ ...., ~ ..,_ -- - - - o 1 J- ' -= " = -1 1 1 - 22 - compr1sed of but should overest1mate area planted w1th cassava The est1mate 1n the r1ght hand column 1s der1ved by deduct1ng publ1shed Ch1nese est1mates for area sown w1th cassava 1n Guangx1 (1961) Ta1wan (1961) FUJlan (1961) Yunnan (1960) Gu1zhou (1961) and Hunan ZheJlang and J1angx1 (c1rca 1960s) from a pub11shed 1961 nat1onal f1gure The calculated f1gure substant1ally exceeds the res1dual based overest1mates of cassava area 1n Guangdong for surround1ng years 1n a per1od when cassava area 1n other Ch1nese prov1nces was undoubtedly small These data are ev1dently 1n confl1ct An exam1nat1on of 1950s Ch1nese mater1al prov1des an 1mpress1on that 1950s cassava area 1n Guangdong was greater than that 1mpl1ed by the resldual-based overest1mates 1n the center column of Table 7 Guangx1 cassava area 1n 1957 all Guangx1 fa~land planted for example was around one-quarter of w1th root and tuber crops If the same proport1on were relevant for Guangdong 1957 cassava area would total more than 300 thousand hectares But whereas 36 21 percent of Guangx1 root and tuber crop product1on cons1sted of crops other than sweet potatoes th1s f1gure was only 13 percent for Guangdong and 1ncluded cassava taro wh1te potatoes and mao potatoes pr1mar1ly the f1rst two categor1es 16 St1ll 1957 Guangdong cassava area cou1d eas11y have been 1n the range of 100 200 thousand hectares 16see data and Ch1nese sources c1ted 1n Bruce Stone Analys1s of Ch1nese Data on Root and Tuber Crop Product1on 616 An PP 609 - 23 Table 7 Est1mates of Area Sown w1th Cassava 1n Guangdong Prov1nce 1943 1984 Guangdong Cassava area est1mates 1n Ch1nese sources Res1dual based est1mates of other econom1c crops 1n Guangdong (thousand hectares) Nat1onal est1mate m1nus Guangx1 Yunnan FUJlan Ta1wan Gu1zhou Hunan, ZheJlang & J1angx1 1943 33 4 1952 25 1957 57 1961 240 1962 25 1965 149 1970 201 1972 167 3 1975 223 1978 236 1979 (215) 1980 237 1981 200 (201) 1982 243 (195) 1983 188 (158) 1984 206 (159) Sources Data appear1ng 1n the left- and rlght-hand columns are based on Table 1 except that the Ta1wan Prov1nce f1gure deducted along w1th those from other prov1nces from the nat1onal est1mate for 1961 (10 000 ha ) was taken from the same source as the nat1onal f1gure L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 9 Data appear1ng 1n the center column are based on data from Guangdongsheng TongJlJU [Guangdong Prov1nce Stat1st1cal Bureau] (ed ) Guangdongsheng TongJl N1anJ1an 1984 [Guangdong Prov1nce Stat1st1cal Yearbook 1984] (X1anggang X1anggang JlngJl Daobao Shechuban [Hong Kong Econom1c Reporter Publ1sh1ng House] 1984) pp 113-114 and Guangdongsheng TongJlJU Guangdongsheng TongJl N1anJ1an 1985 [Guangdong Prov1nce Stat1st1cal Yearbook 1985] (X1anggang X1anggang JlngJl Daobao Shechuban 1985) pp 107-108 Sown area data for sugarcane peanuts sesame JUte kenaf and tobacco were deducted from total area sown w1th econom1c crops Data for rapeseed and other a1lcrops other f1bers and med1c1nal herbs have also been deducted from the f1gures appear1ng 1n parentheses on the bas1s of Zhongguo Nongyebu [Ch1nese M1n1stry of Agr1culture] Zhongguo Nongye N1anJ1an 1980 1962 1983 1984 and 1985 (BelJlng Nongye Chubanshe [Agr1cultural Publ1sh1ng House]--r981 1983 1984 1985 and 1986) 24 - Dur1ng the 1950s cassava was treated expl1c1tly as shule1 [1nclud1ng both tuber crops and tuberous roots] wh1ch 1n turn were class1f1ed as l1angsh1 [staple food crops) occas1onally as part of m1scellaneous gra1ns By the m1d-1970s however 1t 1s clear that cassava was excluded from shule1 and l1angsh1 stat1st1cs and 1ncorporated as a sub category or as a res1dual w1th1n J1ngJ1 zuowu [econom1c crops) The trans1t1on date has not been clearly determ1ned although 1964 and 1976 have been suggested as cand1dates In v1ew of the trends exh1b1ted for Guangx1 1n Table 2 and the forego1ng d1scuss1on attempt1ng to resolve the confl1ct 1mpl1ed 1n Table 7 1t seems l1kely that the 1950s econom1c crop stat1st1cs appear1ng 1n the Guangdong Prov1nce Stat1st1cal Yearbooks though recently publ1shed are unl1kely to have been adJusted for 1nclus1on of cassava hence the center column cannot be used as a proxy for cassava area for the 1950s nor probably for 1962 From 1965 onward however these res1duals may well prov1de the best 1nd1cat1on of trends 1n (though not exact est1mates of) Guangdong cassava area s1nce cassava 1s l1kely to dom1nate the category It should be noted however 1n v1ew of econom1c l1beral1zat1on s1nce 1979 that the d1vergence of th1s res1dual ser1es and actual cassava area 1s l1kely to have 1ncreased espec1ally s1nce the decl1ne 1n export opportun1t1es 1n the early 1980s pp 600 604 17 25 - Unfortunately desp1te the ava1lab1l1ty of an off1c1al cassava ser1es for Guangx1 and a rough approx1mat1on of trends for Guangdong 1t 1s st1ll not poss1ble to be def1n1t1ve about nat1onal trends for Ch1na It 1s clear that cassava was planted on less than 100 thousand hectares 1n the m1d-1940s r1s1ng qu1ckly to perhaps around 250 thousand hectares by 1957 and 355 thousand hectares (exclud1ng Ta1wan) by 1961 dur1ng the fam1ne Total plant1ngs on the Ch1nese ma1nland probably subs1ded to roughly 300 thousand hectares by 1965 and were certa1nly not much lower 1n 1972 when plant1ngs 1n Guangdong and Guangx1 alone totalled 292 thousand Off1c1al area sown w1th cassava 1n the two southern prov1nces seems to have r1sen to 370 thousand hectares 1n 1979 perhaps peak1ng 1n 1980 at 410-420 thousand hectares subs1d1ng to 390 tha and 370 tha 1n 1981 and 1982 and plummet1ng to 275 tha and 250 tha 1n 1983 and 1984 But whether cassava area rose apprec1ably outs1de of these two southern prov1nces s1nce the early 1960s 1s not clear The (undated) total of 350 thousand hectares g1ven to the CIAT delegat1on by Ch1nese cassava breeders 1n spr1ng 1982 would 1mply that 1t has not wh1le the (undated) Inst1tute of Geography est1mate (around 500 thousand hectares) publ1shed 1n 1980 suggests e1ther cons1derable expans1on 1nto other prov1nces or more aggress1ve est1mates of non- f1eld cult1vat1on Barr1ng the unl1kely event of relat1vely even d1str1but1on among other ment1oned prov1nces off1c1ally recorded plant1ngs of 120 190 thousand hectares outs1de of Guangdong and Guangx1 1mpl1ed by the Inst1tute f1gure and the prov1nc1al est1mates - 26 - would surely have been ment1oned by the breeders or 1n cassava- related publ1cat1ons wh1le the 350 thousand hectare f1gure though purportedly 1nclud1ng an est1mate for cassava on pr1vate plots does not even appear to cover probable plant1ngs 1n the two southern prov1nces Part of the problem 1s that cassava area 15 undoubtedly more d1ff1cult to est1mate than that of mo5t f1eld crops s1nce cons1derable proport1ons are grown on pr1vate plot5 on narrow str1p5 adJacent to roads and f1elds on h1lly and 1ncompletely cleared land not yet or normally cons1dered farmlands and on t1ny corners not even counted among pr1vate plot 5tat1st1cs There 1s even sorne 1llegal cult1vat1on under trees on state rubber plantat1ons for example 18 The ln5t1tute of Geography f1gure probably 1ncorporate5 a more aggre551Ve e5t1mate ba5ed on sorne survey ev1dence of the5e k1nds of plant1ngs wh1ch 1n large part elude off1c1al stat1st1cal coverage All that can be cla1med w1th near certa1nty 15 that nat1onal ca55ava plant1ng reached another maJor peak 1n the late 19705 or early 1980s and then decl1ned rap1dly w1th the subs1dence of opportun1t1es for 1nternat1onal trade 1ncreas1ng l1beral1zat1on of rural econom1c act1v1t1e5 and a probable cut back 1n the government s role 1n ca5sava market1ng p 621 - 27 - Nat1onal product1on trends are even less d1scern1ble The only ava1lable f1gure for recent product1on 1s 3 m1ll1on tons prov1ded to the CIAT delegat1on 1n spr1ng 1982 19 although l1ke the 350 thousand hectare f1gure prov1ded at the same t1me 1t may well be an underest1mate The best 1nd1cat1on of nat1onal y1eld trends 1s undoubtedly the Guangx1 ser1es 1n Table 2 w1th sorne reservat1ons about a few of the years such as 1967 and 1968 The nat1onal average 1rnpl1ed by the f1gures g1ven to the CIAT delegat1on 1s 8 6 tons per hectare suggest1ng that average y1elds 1n Guangdong and elsewhere are lower than 1n Guangx1 But th1s cornpar1son too cannot be taken too l1terally s1nce the four to f1ve tons per hectare 1981 Guangdong average suggested by such an exerc1se 1rnpl1es too great a d1vergence between Guangx1 and Guangdong part1cularly 1n v1ew of greater general ava1lab1l1ty of fert1l1zer 1n the latter prov1nce W1th1n these two southern prov1nces sorne of the pr1nc1pal cassava grow1ng areas can be 1dent1f1ed The f1rst record of Ch1nese cassava cult1vat1on was 1n 1820 1n Gaozhou County part of ZhanJlang Prefecture 1n southwestern Guangdong 20 Gaozhou 1s not a coastal county and earl1er cult1vat1on 1s ent1rely poss1ble In the 1950s there 1s cont1nued record of cassava 1n ZhanJlang Prefecture where uplands const1tuted 27 5 percent of cult1vated land a greater 19James H Cock and Kazuo Kawano Cassava 1n Ch1na unpubl1shed tr1p report Internat1onal Center for Trop1cal Agr1culture Palm1ra Colomb1a June 1982 p 1 20L1ang Guangs~ang {ed ) r1ush~ Za pe1 JU L1yong p 4 - 28 - proport1on than 1n other Guangdong Prefectures Su1x1 County and the ZhanJlang c1ty suburbs (where uplands compr1sed 12 percent) 1n the center of the prefecture and Xuwen County on the southern t1p of the Le1zhou Pen1nsula are ment1oned 1n 1950s l1terature on cassava but the crop may have been grown more generally throughout the gra1n def1c1ent Le1zhou Pen1nsula and 1n the uplands adJacent to the J1anJ1ang Pla1n where m1scellaneous gra1ns (80 9 percent of wh1ch were root or tuber crops) compr1sed 44 percent of staple foodcrop product1on 1n 1955 Throughout the ZhanJlang Prefecture and enclosed mun1c1pal areas root and tuber crops (valued at one-fourth fresh we1ght) const1tuted only 28 percent of staple crop product1on wh1ch occup1ed 95 percent of sown area Sweet potatoes were the pr1nc1pal root crop however w1th cassava and mao potatoes compr1s1ng a l1ttle less than 5 percent of root and tuber crop product1on 21 But cassava cult1vat1on clearly was not l1m1ted to southwestern Guangdong 1n the 1950s There 1s also record 1n the Econom1c Geography of South Ch1na (1959) of cassava and taro be1ng grown 1n the mounta1nous uplands surround1ng the Su1 and X1 R1ver Valleys 1n West Central Guangdong notably HualJl Guangn1ng S1hu1 Gaoyao and Deq1ng Count1es all 1n Zhaoq1ng Prefecture Cassava was not spec1f1cally ment1oned 1n the d1scuss1on of Ha1nan lsland but has 21sun J1ngzh1 (ed ) Huanan D1chu J1ngJ1 D1l1 [Econom1c Geography of South Ch1na] (BelJlng Kexue Chubanshe [Sc1ent1f1c Publ1sh1ng House] 1959) Translated 1n Jo1nt Publ1cat1ons Research Serv1ce August 24 1969 no 14954 pp 137-138 and 178 179 When these stat1st1cs were gathered the reg1on 1ncluded the Q1nzhou ~pec1al D1str1ct encompass1ng known cassava grow ng a ea~ ~~e~ as Hepu County and the Be1ha1 suburbs - 29 - been grown there at least s1nce 1912 when a well-known Malays1an var1ety was 1ntroduced 1nto Dan X1an rubber plantat1ons Accord1ng to 1951 stat1st1cs roots and tubers accounted for 38 5 percent of gra1n consumpt1on 1n pla1ns areas of the Island and 69 8 percent 1n h1lly d1str1cts paddy r1ce prov1d1ng most of the rema1nder 1n both cases 22 In Guangx1 cassava was generally d1str1buted 1n the XunJ1ang and L1u]1ang Valleys (east central Guangx1) character1zed by relat1vely barren drought prone land Yet y1elds of 7 5 15 O tons per hectare were c1ted It was used as food feed and to produce starch for cotton yarn 1n the c1ty of Wuzhou 1n east central Guangx1 on the Guangdong border where Guangx1 s f1rst starch factory was opened 1n 1952 Cassava was also w1dely planted 1n southeastern Guangx1 and along the southern coast espec1ally Hepu County and the suburbs of Be1ha1 on the southeast coast But although Be1ha1 and Wuzhou rema1ned maJor centers by the m1d-to-late 1950s cassava starch factor1es and consequently expanded cassava cult1vat1on had spread w1dely 1n the Autonomous Reg1on 1nclud1ng N1ngm1ng 1n the southwest Barna Yaozu Autonomous County toward the northwest and Wum1ng 1n the center of the Reg1on 23 In Yunnan cassava cult1vat1on 1n 1960 was recorded 1n Hekou Yaozu Autonomous County 1n the south 22op c1t pp 137 138 and p 201 transfer below See deta1ls of var1etal c1t pp 258 and 333-334 Guangx1 J1ng]l N1an]1an Guangx1 J1ng]1 N1anJ an 1985 p 192 30 - along the V1etnarnese border 1n Dehong Da1zu J1ngpozu Autonornous Prefecture 1n the west along the Burrnese border and elsewhere 24 By 1972 Zhaoq1ng Prefecture had taken over as the pr1nc1pal cassava grow1ng reg1on of Guangdong account1ng for 57 thousand hectares or 33 9 percent of the prov1nc1al f1gure for that year ZhanJlang Prefecture was next w1th 33 thousand hectares or 19 5 percent The rerna1n1ng 77+ thousand hectares were d1str1buted throughout Guangdong 1nclud1ng Ha1nan Island and Shaoquan Me1x1an Shantou Foshan and Hu1yang Prefectures Sorne of these secondary reg1ons 1ncreased cassava plant1ngs rap1dly 1n the late 1970s Cassava area 1n Me1x1an Prefecture for exarnple 1n the northeast cerner of the prov1nce grew frorn 10 800 hectares 1n 1977 te 40 000 hectares 1n 1978 25 In spr1ng of 1982 a delegat1on of cassava breeders frorn the Internat1onal Center for Trop1cal Agr1culture (CIAT) v1s1ted a nurnber of cassava grow1ng areas 1n Guangdong 1nclud1ng Ba1sha County and Ha1kou Mun1c1pal1ty on Ha1nan lsland three state farrns 1n ZhanJlang Prefecture and Dongguan County (Hu1yang Prefecture) en the Pearl R1ver Delta Sorne 1rnpress1on of area trends en the Delta can be obta1ned frorn stat1st1cs for Dongguan Cassava plant1ngs decl1ned frorn 8 600 ha (1957) te 4 600 ha (1977) w1th rnuch of the decl1ne occurr1ng 1n the 1970s Cassava area then fell even more rap1dly te 24L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 9 251bld - 31 - 3 157 4 ha 1n 1978 then 3 100 ha (1981) and 2 816 8 (1982) But on the other s1de of the Delta 1n Ta1shan (Foshan Prefecture) cassava was not grown on a large scale unt1l recently And Fucheng Commune (w1th1n Dongguan County) cassava area fell from 500 to 367 hectares between 1980 and 1981 but recovered to 434 ha 1n 1982 26 Y1elds observed by the CIAT delegat1on were generally 1n the 6 to 8 ton/ha range but 20-25 tons/ha was cla1med for sorne state farms and exper1ment stat1ons 27 Average y1elds for Dongguan County on the Delta were 11 73 tons/ha 1n 1978 and 15 76 tons 1n 1982 Fucheng Commune w1th1n Dongguan County cla1med around 15 tons/ha 1n 1980 14 43 tons/ha 1n 1981 and 17 75 tons/ha 1n 1982 28 In Guangdong generally w1th 1200-1800 mm of annual ra1nfall y1elds on farmer s f1elds w1th poor so1ls have been est1mated by one Ch1nese breeder to fall typ1cally between 5 to 7 tons per hectare and between 10 to 13 tons under good cl1mat1c cond1t1ons and so1l cond1t1ons Throughout Southern Ch1na (800-2000 mm/yr annual ra1nfall) y1elds are est1mated by another breeder to be 5 to 9 tons per hectare on poor so1ls and 15 30 tons/ha (avg 20 tons/ha ) under good cond1t1ons W1thout fert1l1zer or 1rr1gat1on however poor so1l y1elds were reported to be 3 to 6 tons/ha (average 4 tons) and for good so1ls 26cock and Kawano Cassava 1n As1a op c1t The 1957 1977 and 1981 f1gures for Dongguan County are from p 13 The 1978 and 1982 data the Fucheng Commune data and the 1mpress1ons for the 1970s and for Ta1shan are from Prof Graham Johnson Dept of Anthropology and Soc1ology Un1vers1ty of Br1t1sh Columb1a correspondence Sept 19 1983 27cock and Kawano Cassava 1n Ch1na p 1 28Graham Johnson op c1t 32 - w1th good weather 12 to 18 tons/ha In Zhaoq1ng and Shaoiuan Prefectures {1450-1700 mm/yr avg ra1nfall) farmers y1elds w1thout fert1l1zer and 1rr1gat1on were reporteó by an agronom1st spec1al1Z1ng 1n cassava to average 6 4 tons/ha under poor cond1t1ons and 11 2 tons/ha under good cond1t1ons W1th fert1l1zer but w1thout 1rr1gat1on these averages rose to 11 69 tons/ha and 19 7 tons/ha w1th ranges of around 4 tons/ha Average y1elds on research stat1ons run 2 to 10 tons per hectare h1gher than those quoted above for farmers f1elds 29 These data 1n sum would seem to suggest that most cassava 1n Guangdong 1s grown on poor land espec1ally uplands and unt1l recently rarely rece1ved much fert1l1zer Total cassava area has fallen dur1ng the past decade or so on better lands such as those typ1cal of the Pearl R1ver Delta (w1th scattered temporary except1ons dueto the short l1ved EC export opportun1t1es) lead1ng to sorne decl1ne 1n the average qual1ty of farmland grow1ng cassava Th1s decl1ne has been more than counterbalanced by the 1ncrease 1n fert1l1zer appl1cat1on to cassava 1n recent years such that average y1elds {though not necessar1ly total product1on) have 1ncreased sharply The h1gher cassava y1elds on state farms and for prlvate and cooperat1ve farm1ng 1n the Pearl R1ver Delta locat1ons l1ke 29oelphl survey responses sent to J S Sarma (IFPRI) for Shaoquan and Zhaoq1ng Prefectures by Huang X1 of the Inst1tute of Drought Gra1n Crops Guangdong Prov1nce Academy of Agr1cultural Sc1ences Guanzhou June 28 1986 for Guangdong by L1u YlnQJlng of the South Ch1na Inst1tute of Botany Ch1nese Academy of Sc1ences Guangznou June 30 1986 and for South Ch1na Academy of Trop1cal Crops Pesearch Dan X1an Ha1nan Island June 20 1986 33 Dongguan County are part1ally expla1nable 1n terms of greater access to (and more attract1ve relat1ve pr1ces for) manufactured fert1l1zers as well as to often better so1l and h1gher standards of agronomy But an add1t1onal 1mportant factor relates to var1etal adopt1on An espec1ally small port1on of cassava grown on state farms and on the Delta 1s l1kely to be ut1l1zed for d1rect human consumpt1on so there 1s l1ttle reason for managers and farmers to cult1vate the lower y1eld1ng sweeter var1et1es character1zed by low cyan1de and h1gher prote1n content as well as greater overall palat1b1l1ty (see below) The argument 1s at least part1ally relevant for Zhaoq1ng and & Shao~uan Prefectures wh1ch are becom1ng one of Guangdong s maJar reg1ons for process1ng 1ndustr1es ut1l1z1ng cassava and for s1m1lar reasons east central and southern Guangx1 1 h1stor1cally among the pr1nc1pal cassava-growlng areas w1th1n the Autonomous Reg1on Cassava product1on systems Cassava 1n Ch1na 1s grown both extens1vely and 1n small plots and scattered plant1ngs Extens1ve cult1vat1on 1s most notable on but by no means conf1ned to state farms and 1s pr1nc1pally assoc1ated w1th starch product1on the domest1c an1mal feed market and exports Outs1de the state farm sector w1th the formal d1ssolut1on of the communes 1n favor of the household product1on respons1b1l1ty system 1t 1s safe to assume that extens1ve cult1vat1on has decl1ned somewhat s1nce the early 1980s However 34 - Graham Johnson has po1nted out 30 that rural reforms have 1n sorne 1nstances strengthened rather than weakened cooperat1on 1n South Ch1na so 1t cannot be assumed that extens1ve cult1vat1on 1n the old cooperat1ve sector has d1sappeared S1nce the format1on of agr1cultural producers cooperat1ves (1954 56) and the people s communes (1958) collect1ve lands const1tut1ng the vast ma]or1ty of Ch1nese farmlands have been cult1vated communally However the 54 thousand communes have normally not been the pr1nc1pal cult1vat1on un1t More often smaller un1ts the 719 thousand br1gades or most commonly the 56 m1ll1on product1on teams have cult1vated as cooperat1ve groups A product1on team normally cons1sted of around th1rty farm fam1l1es (an average of 139 people) that pooled usually cont1guous land and shared cult1vat1on respons1b1l1t1es 31 The pr1nc1pal farm un1t var1ed geograph1cally 1n s1ze but by the late 1970s averaged around 8 6 hectares 1n Guangdong and 8 9 hectares 1n Guangx1 and certa1nly less 1n the very densely populated Pearl R1ver Delta of Guangdong 32 30Graham E Johnson Ch1nese Agr1culture Sorne vol 55 no 3 (Fall) 1982 The Product1on Respons1b1l1ty System 1n Examples from Guangdong Pac1f1c Affa1rs PP 430 449 31zhongguo GuoJ1a TongJ1JU [State Stat1st1ca1 Bureau of Ch1na] Zhongguo TongJ1 N1an]1an 1983 [Stat1st1ca1 Yearbook of Ch1na 1983] (Be1J1ng TongJl Chubanshe [Stat1st1ca1 Pub11sh1ng House] 1983) p 147 p 148 0111 YanJlUsuo Zhongguo Nongye 0111 Zonglun 35 - S1nce the early 1980s however cult1vat1on of collect1ve lands 1s no longer a communal respons1b1l1ty but has been delegated to several spec1al1zed households Normally 1t 1s the part1cularly sk1lled farmer who 1s entrusted w1th respons1b1l1ty for farm1ng collect1ve lands But 1n relat1vely advanced communes or 1n suburban areas non agr1cultural act1v1t1es w1th h1gher 1ncome earn1ng potent1al attract the most able workers As1de from collect1ve lands 1nd1v1dual farm fam1l1es ma1nta1n pr1vate plots of normally O 03 O 05 hectares wh1ch are used pr1mar1ly for fam1ly product1on of food 1tems espec1ally vegetables and l1vestock products (and consequently fodder for the latter) Although no est1mates are ava1lable for cassava cult1vat1on on such lands the 1mportance of cassava as a sw1ne feed the cons1derable 1mportance of sw1ne 1n the l1vestock economy of South Ch1na and the dom1nance of fam1ly-owned and managed sw1ne w1th1n the sw1ne husbandry sector suggest that pr1vate plot cult1vat1on of cassava 1n South Ch1na 1s not tr1v1al In add1t1on to formally establ1shed pr1vate plots ass1gned to each fam1ly there appears to be cult1vat1on of cassava on an even more fragmentary bas1s on narrow str1ps adJacent to roads and f1elds on steep h1lls1des and other areas not formally counted among cult1vated lands and 1llegally 1n econom1c forests reclamat1on areas and other lands managed by the state The latter may be d1st1ngu1shed however from planned cult1vat1on on such lands by the State Farm and Reclamat1on Bureau \lh1le land 1s be1ng cleared and 36 recla1med cassava 1s often grown as an 1ntermed1ate crop for a few years unt1l 1t 1s d1scont1nued when f1eld transformat1on progress allows cult1vat1on of the pr1nc1pal crop 33 F1nally cassava 1s planted as a f1eld crop on state farms There 1ts cult1vat1on 1s espec1ally extens1ve and 1s character1zed by h1gh standards of agronomy and abundant appl1cat1on of modern 1nputs part1cularly fert1l1zers V1s1tors 1nterested 1n cassava are often brought to state farms to v1ew extens1ve cult1vat1on and h1gh y1elds but state farm plant1ngs rema1n a small proport1on of total cassava area Cult1vated area on state farms 1n Guangdong var1ed between only 60 and 64 thousand hectares from 1981 to 1984 and rema1ned at 20 thousand hectares 1n Guangx1 In 1984 state farm sown area 1n Guangdong was only 86 900 hectares or less than 1 8 percent of tho prov1nc1al total of wh1ch 72 200 hectares were planted w1th cereals beans sweet and wh1te potatoes o1lcrops and sugarcane leav1ng a res1dual of 14 700 hectares wh1ch could have been planted w1th cassava vegetables green manure other fodder crops or other southern 1ndustr1al crops such as s1sal hemp In Guangx1 state farm sown area was only 17 400 hectares or less than O 5 percent of the reg1onal total of wh1ch the res1dual category 1nclud1ng cassava 33Bruce Stone An Analys1s of Ch1nese Data on Root and Tuber Crop Product1on The Ch1na Quarterly September 1984 p 621 L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 36 Bruce Stone An Exam1nat1on of Econom1c Data on Ch1nese Cassava Product1on Ut1l1zat1on and Trade 37 - compr1ses but 3 300 hectares 34 Thus pr1vate and collect1ve plant1ngs dom1nate cassava area 1n Ch1na Ava1lable 1nternat1onal data on cassava ut1l1zat1on 1n Ch1na 1s unrel1able but 1t 1s clear that an1mal (espec1ally sw1ne but also cattle f1sh and s1lkworm) feed 1s assoc1ated w1th each of the cassava product1on systems Exports and starch product1on as well as less trad1t1onal 1ndustr1al and process1ng uses are assoc1ated w1th collect1ve product1on and the state farms wh1le d1rect human consumpt1on 1s assoc1ated w1th pr1vate product1on and the collect1ve sector 1n poorer areas Mach1ne cult1vat1on 1s assoc1ated w1th a port1on of the extens1ve plant1ngs between 100 m and 300 m above sea 1 e ve 1 Between 300m and 1 000 m cassava 1s grown 1n rotat1on w1th dryland crops as far as 30°N Most cassava 1n Ch1na 1s un1rr1gated but the cl1mate prov1des adequate mo1sture 1n most years and locat1ons Th1s 1s espec1ally true 1n the south where fall-planted cassava 1s common 35 Cassava 1s cult1vated year round 1n South Ch1na w1th the pr1nc1pal plant1ngs concentrated 1n spr1ng and fall The plant1ng mater1al may be e1ther freshly cut stakes or stored mater1al Storage 1s pract1ced by cutt1ng long stakes wh1ch may e1ther be left 1n the sun 1n bundles or placed under trees Cutt1ngs are fa1rly 34Ch1na Agr1cultural Yearbook Ed1tor1al Board Ch1na Agr1cultural Yearbook 1985 (Be1J1ng Agr1cultural Publ1sh1ng House 1986) pp 114 and 185-186 35L1ang Guangshang 1 ed 1 11us'1u La1pe1 {U L ;,ong p 36 38 - short (10 15 cm) w1th m1n1mal select1on Plant1ng 1s fa1rly deep (up to 10 cm and hor1zontal) Germ1nat1on var1es cons1derably by locat1on but 1s frequently very poor and strands are not un1form Land preparat1on 1s generally acceptable and 1s done manually by draft an1mal or tractor-d1awn 1mplements 36 Spr1ng cassava (e g 1n the Guangzhou area) 1s typ1cally planted between January and March and harvested 1n the fall after at least 8 months espec1ally from October although for fodder purposes cutt1ngs may be taken cont1nuously over an extended per1od of t1me The spr1ng and summer seasons cons1derably a1d leaf and stem growth of sprlng-planted cassava and fall arr1ves opt1mally for starch format1on Y1elds of spr1ng planted cassava tend to be large but are less rel1able s1nce typhoons 1n fall occas1onally cause damage Furthermore low temperatures 1n spr1ng extend the budd1ng and sprout1ng per1od and thus the r1sk of 1nsect damage But spr1ng planted cassava f1ts well 1nto South Ch1nese 1ntercropp1ng and rotat1on systems fac1l1tat1ng the ach1evement of as many as three crops per year 1nclud1ng one of cassava 37 Fall- and Wlnter planted cassava 1s common 1n the most trop1cal areas w1th harvests start1ng the follow1ng fall The peak per1od for both plant1ng and harvest1ng 1s September to November Fall-planted 36cock and Kawano Cassava 1n Ch1na p 7 37rhe d1scuss1on of spr1ng and fall planted cassava 1s pr1mar1ly from mater1al appear1ng 1n L1ang Guangshang (ed ) Mushu La1pe1 yu L1yong pp 10 1 and 33 34 39 - ca55ava 15 pract1cable from around Gaozhou County (21°56 N ZhanJlang Prefecture Guangdong Prov1nce) 5outh where temperature5 average about 22 7°C annually and the lowe5t average January temperature5 exceed 15°C The5e area5 al5o enJOY 1304-1718 mm of ra1nfall per year and 1941-2455 hour5 of 5unl1ght h1gher than more northerly reg1on5 e5pec1ally dur1ng the w1nter ho5p1table cond1t1on5 for fall plant1ng thereby prov1d1ng more Of cour5e fall-planted and 5prlng-planted ca55ava are not mutually exclu51ve QlJlng Br1gade for example 1n D1anba1 County (w1th1n the coa5tal zone ly1ng along the South Ch1na Sea well te the 5outh of Gaozhou) planted 25 thou5and hectare5 of ca55ava 1n 1972 approx1mately one-th1rd fall- planted two-thlrd5 5pr1ng planted A pr1nc1pal advantage of fall-planted ca55ava 15 the potent1al for avo1d1ng typhoon damage Th15 1s part1cularly 1mportant on the Le1zhou Penln5ula and Ha1nan l5land ln5ect damage te the 5prout5 15 al5o lower s1nce cr1cket populat1on5 decl1ne rap1dly 1n fall and the 5prout1ng perlad 15 collap5ed w1th sprout5 and root5 beg1nn1ng w1th1n a week after plant1ng Fall planted ca55ava can be more conven1ently l1nked w1th ser1culture s1nce leaves are prov1ded more opportunely w1thout 1nfluenc1ng root y1eld W1th the longer 5eason ca5sava planted 1n fall fac1l1tates fuller ut1l1zat1on of product1on capac1ty 1n local 5tarch factor1es and 1s conven1ent for on-farm l1vestock development The pr1nc1pal drawbacks are the slower w1nter growth and the 1nconven1ence of the longer sea5on for rotat1on and mult1ple cropp1ng Thus even 1n the far 5outh 1f the cropp1ng 1nten51ty 15 h1gh ca55ava 15 apt te be planted 1n 5pr1ng \J1th - 40 - v1rtually all cassava north of 22°N andan 1mportant port1on of the rema1nder planted 1n spr1ng the maJOr1ty of cassava 1n Ch1na 1s l1kely to be spr1ng planted The Ch1nese are well aware of the necess1ty of rotat1on and 1ntercropp1n~ for cont1nued cassava cult1vat1on They est1mate that y1elds decl1ne by 20 30 percent 1n a second consecut1ve year of cassava cult1vat1on and by 30 40 percent for three consecut1ve years 38 The CIAT delegat1on noted however that cassava 1s grown as a monocrop 1n sorne areas 39 South Ch1nese rotat1on systems are complex and var1ed those 1nclud1ng cassava are no except1on F1gure A presents notable 2-year through 6-year rotat1on systems for cassava and other dryland food crops In newly recla1med areas cassava 1s often grown for one or two years among Jade cass1a (Ch1nese c1nnamon) mounta1n apr1cot bamboo tong o1l tea o1l rubber trees or 1n other econom1c forests Ch1nese l1terature po1nts out the 1mportance of rotat1on of cassava w1th green manure crops 1n econom1c forests to avo1d eros1on Cassava 1s normally the pr1nc1pal crop 1n a small number of exceed1ngly peor local1t1es anda very few state farms As Table 5 and 6 1nd1cate the most 1mportant crop 1n South Ch1na 1s unquest1onably paddy r1ce compr1s1ng 63 percent of sown area 1n Guangdong 1n 1984 and 59 percent 1n Guangx1 Paddy f1elds occupy 63 38L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 40 39cack ~nd Kawano Ca sava 1n Ch1na p 8 41 F1gure A Cassava Rotat1on Systems 1n Ch1na 2 year systems cassava - upland r1ce sweet potatoes cassava - peanuts sweet potatoes spr1ng peanuts fall-planted cassava - fall harvested cassava spr1ng soybeans 3-year systems cassava - sugar cane sugar cane cassava peanuts wheat - upland r1ce sweet potatoes 4-year systems cassava mung beans sweet potatoes sugar cane - sugar cane 5-year systems peanuts wheat sugar cane upland r1ce sugar cane - sugar cane- 6-year system cassava sugar cane sugar cane - soybeans sweet potatoes upland r1ce rad1shes - peanuts sweet potatoes Notes and Sources L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong and Kawano Cassava 1n As1a p 8 the authors noted often grown w1th legume crops predom1nantly peanuts p 40 In Cock that cassava was - 42 - percent of cult1vated land 1n Guangx1 and are s1rn1larly dorn1nant 1n Guangdong Sweet potatoes are second 1n order of planted area 1n Guangdong and cornb1ned w1th wh1te potatoes totalled 10 percent of sown area Peanuts (6 percent) and sugar cane (5 percent) rank th1rd and fourth probably followed by cassava at around 3 percent Soybeans rna1ze bast f1bers and tobacco are also grown and unt1l 1ts de-ernphas1s 1n recent years wheat area exceeded cassava plant1ngs In Guangx1 rna1ze 1s second at 11 percent of sown area followed by soybeans and sweet potatoes (5 percent each) sugar cane and peanuts (3 5 percent each) and green rnanure crops as a group (2 5 percent) Cassava at 2 1 percent 1s sl1ghtly below vegetables and rnelons as a group When cassava area peaked 1n 1980 1ts share was 4 3 percent rank1ng f1fth beh1nd r1ce rna1ze soybeans and sweet potatoes and h1gher than all econorn1c crops 40 Y1elds Most ava1lable 1nforrnat1on on cassava y1elds was prov1ded 1n the sect1on on product1on trends and d1str1but1on In that sect1on 1t was suggested that the cons1derable 1ncrease 1n average y1elds dur1ng the latter 1960s (Table 2) was due to var1etal 1rnprovernent and to sorne extent 1rnprovernent 1n cultural pract1ces wh1le y1eld growth s1nce the late 1970s has been pr1nc1pally the result of 1ncreased fert1l1zer appl1cat1on to cassava cornplernented by sorne 1rnprovernent 1n var1et1es and cult1vat1on techn1ques Mean cassava y1elds throughout Ch1na 4Drable 5 and 6 and 0111 YanJ1Usuo 8 6 tons/ha 1n 1980) approx1mate the average for íh1na Agr1cultJrQl Yearbook 1985 pp Zhongguo Nongye 0111 Zonglun pp 77 79 114-126 43 - the rest of the world but are somewhat h1gher than mean y1elds 1n the rema1nder of As1a Mean y1elds 1n Guangx1 (13 1 tons/ha 1981-84 average) however are somewhat h1gher than the 1nternat1onal average and the h1ghest y1elds from f1eld cult1vat1on 1n Ch1na (average 20-25 tons/ha w1th a max1mum of 30 tons/ha or more) are comparable to the very h1ghest y1elds 1n the world 41 But Ch1nese cassava 1s also grown on peor so1ls w1th no fert1l1zer or 1rr1gat1on where average y1elds have been character1zed 1n the 3 to 8 ton range The average f1gures c1ted above suggest that those peor cond1t1ons are more typ1cal of Ch1nese cassava cult1vat1on than the state farm or Pearl R1ver Delta pr1vate and cooperat1ve farm1ng exper1ence However survey results suggest that even on peor so1ls w1thout 1rr1gat1on fert1l1zer appl1cat1on can 1ncrease y1elds on both research stat1ons and operat1ng farms by an average of at least 6 tons per hectare Y1eld d1fferences among farms are due not only to d1fferences 1n so1l fert1l1ty cl1mat1c cond1t1ons adopted var1et1es and appl1ed fert1l1zers but to substant1al d1fferences 1n management as well Farmers 1n sorne areas use unselected plant1ng mater1als g1v1ng very peor stands and low y1elds On pr1vate plots management var1es more than on collect1ve lands w1th1n a s1ngle v1c1n1ty but the level of agronomy 1s often fa1rly h1gh 42 41 1bld from James H p 1 and 8 Delph1 Survey responses Cock June 24 1983 Table 2 and correspondence 42cock and Kawano Cock June 24 1983 Cassava 1n Ch1na correspondence from James - 44 - Among the responses of three Ch1nese cassava breeders surveyed low y1eld potent1al of ex1st1ng var1et1es and unava1lab1l1ty of fert1l1zers were both l1sted by each respondent as 1mportant constra1nts on farmers y1elds But the survey results also suggest that output market1ng problems storage and process1ng d1ff1cult1es and general lack of product1on 1ncent1ves may restr1ct appl1cat1on of labor and fert1l1zers to cassava 1n sorne areas 43 Although there 1s cons1derable var1at1on 1n the qual1ty of cult1vated var1et1es Ch1na has several popular var1et1es such as South Ch1na 205 prov1d1ng reasonably h1gh and stable y1elds It 1s the prov1s1onal conclus1on of one 1nternat1onal breeder that l1ke Tha1land 1n the recent past and Malays1a currently r1g1dly selected CIAT clones could outy1eld the best Ch1nese cult1vars only sl1ghtly Th1s contrasts w1th Indones1a and the Ph1l1pp1nes where the best local var1et1es are more eas1ly dom1nated 44 Peor fert1l1zer response and 1nadequate extens1on were l1sted as a secondary constra1nt on y1elds as was 1nadequate mo1sture 1n sorne areas The 1982 CIAT delegat1on noted that fert1l1zer appl1cat1ons were not generally l1nked to so1l analyses or recommendat1ons made on the bas1s of exper1mental results Each of the surveyed breeders appeared to agree that pests and d1seases were relat1vely un1mportant 43oelphl Survey results 44Kazuo Kawano Tr1p Report to Ch1na (18-24 January 1986) unpubl1shed tr1p report prov1ded 1n correspondence from Kawano Apr1l 14 1986 45 1n l1m1t1ng cassava y1elds The 1982 CIAT delegat1on also found that although pests and d1seases were not chem1cally controlled they appeared to be of very low 1nc1dence and harvest losses from such sources were concluded to be m1n1mal The most commonly observed d1sease was Cercospora leaf spots and dur1ng the dry months Tetranychus m1tes are reported to be a problem 45 Costs of product1on and labor ut1l1zat1on The 1982 CIAT delegat1on was told that labor use var1ed from 100 man days per hectare w1th mechan1cal land preparat1on to 270 days w1thout mach1nes and total product1on costs were est1mated at $550 US per hectare 170 days may be somewhat excess1ve for manual land preparat1on but although the total of 270 days per hectare 1s h1gher than 1n sorne As1an countr1es 1t 1s not unprecedented The total cost f1gures are l1kely to have come d1rectly from the product1on accounts of ene or more Guangdong state farms where workers are pa1d set wages or from a small sub-group of more prosperous cassava grow1ng collect1ves wh1ch happened to have kept good records and where y1elds are h1gh Most of the 1mpl1ed cost per man-day of around $2 US would be labor A proJect prospectus for an agr1cultural cred1t appl1cat1on to the World Bank 1nvolv1ng cassava cult1vat1on 1mpl1ed a return to labor of $1 25 US per day Much of the labor 1nvolved 45cock and Kawano Cassava 1n Ch1na p 7 - 46 espec1ally where cassava 1s fert1l1zed 1s for hand weed1ng s1nce herb1c1des are not used 46 Much of the non-labor costs on state farms would cons1st of fert1l1zer appl1cat1on The h1ghest per hectare appl1cat1on rates encountered by the CIAT delegat1on 1n 1982 were 20 tons of organ1c manures 375 k1lograms of superphosphate (45 68 kg of P2o5) and 150 k1lograms of mur1ate of potash (37 5 kg of K2o) 47 Such rates are l1kely to have ex1sted only on state farms w1th plent1ful access to fert1l1zers and/or few alternat1ve uses Impl1ed per hectare reta1l value of th1s level of manufacturad fert1l1zer use alone would have totalled $ US 48 On collect1ve lands w1th plent1ful access to fert1l1zers use of manufactured products 1s less lav1sh but organ1c manure use w1th assoc1ated h1gh labor requ1rements 1s very substant1al In Fucheng Commune of Dongguan County on the Pearl R1ver Delta average y1elds of 21-22 5 tons per hectare on 400 hectares of cassava were ach1eved w1th 225 k1lograms of ammon1um sulfate per hectare But 1n add1t1on three organ1c manure appl1cat1ons were undertaken 1nvolv1ng total per hectare use of 3 tons of sw1ne and cattle manure 3 4 5 tons of human n1ght so1l and 15 tons of green manure (pr1mar1ly legumes) m1xed w1th 22 5 tons of so1l On the Huashan State Farm 1n L1ngshan County Guangx1 per 46¡bld pp 7-8 correspondence from John Lynam Program December 22 1983 Stone An Exam1nat1on af on Ch1nese Cassava Praduct1on Ut1l1zat1on and Trade 4 7 Cae k and Ka1~ano Cassava 1n Ch1na p 7 48 CIAT Cassava Econom1c Data pp 6-9 - 47 hectare appl1cat1ons of 255 k1lograms of ammon1um sulphate and 15 tons of organ1c manure y1eld1ng 19 62 tons per hectare were est1mated to prov1de 141 k1lograms of n1trogen 79 k1lograms of phosphor1c ac1d and 180 k1lograms of n1trogen 49 One of the 1986 Ch1nese survey respondents prov1ded a comb1ned per hectare est1mate of farmer fert1l1zer use on poor so1l cassava lands 1n Guangdong of 150 k1lograms assoc1ated w1th average y1elds of only 5 tons per hectare wh1le another respondent based on Ha1nan Island (Guangdong) 1mpl1ed that no manufactured fert1l1zers were used on cassava by farmers regardless of so1l cond1t1ons 50 It 1s very unl1kely that much fert1l1zer has been appl1ed to cassava on d1stant collect1ves and pr1vate plots Th1s 1s due to low farmgate cassava pr1ces a weak cassava market 1n many areas (see below) and to the h1gher pr1ces and d1ff1cult access assoc1ated w1th fert1l1zer purchase unless such purchase 1s l1nked to sales to government procurement organ1zat1ons of farm goods 1n part1cular state demand Pr1vate plot product1on of cassava employ1ng household labor and Wlthout manufactured fert1l1zer use could be conducted for purposes of home consumpt1on and hog feed at very low 1mpl1ed return to labor However w1th the low y1elds assoc1ated w1th most product1on such returns could be well under $1 US per day and may have been susta1nable only as a funct1on of Ch1nese labor market 49L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 86 50oelp1l Survey responses - 48 - restr1ct1ons W1th 1ncreas1ng l1beral1zat1on of economlc act1v1t1es 1n the 1980s labor opportun1ty costs have r1sen substant1ally 1n suburban and wealth1er rural farm areas As export opportun1t1es have decl1ned these healthy econom1c movements have undoubtedly worked aga1nst cassava cult1vat1on 1n such areas Opportun1ty costs would be less affected 1n poorer and more d1stant farm areas but the state s decl1n1ng market1ng role 1s less apt te be v1gorously replaced by pr1vate market development 1n such areas Technology development Publ1cat1cn of L1ang T1ngdong s Zhong Mufanshu Fa [Cassava Plant1ng Methods] 1n 1900 was a benchmark 1n the 1nlt1at1cn of a formal process of cassava technology 1mprovement 1n Ch1na wh1ch could span t1me and space As 1nd1cated 1n the f1rst sect1on cassava spread te FuJlan and Ta1wan 1n the 1920s roughly 100 years after 1ts f1rst known cult1vat1on 1n ne1ghbor1ng Guangdong Introduct1on 1n Hunan and J1angx1 1n the early 1940s may have been the f1rst example of del1berate trans-prov1nc1al d1ssem1nat1on by Ch1nese sc1ent1f1c 1nst1tut1ons The Peoples Republ1c agr1cultural sc1ence establ1shment gave attent1on to cassava as a bulky relat1vely drought res1stant crop wh1ch could be grown en poor so1ls and st1ll prov1de growth 1n ava1lable calor1es per un1t of farmland w1th sorne advantages 1n y1eld stab1l1ty Alternat1vely 1t could also furn1sh raw mater1als for 1ndustry Th1s or1entat1on toward bulky cheaper food 1tems and 1ndu tr1al crops was well w1th1n a trad1t1on establ1shed early 1n the - 49 h1story of most soc1al1st governments and st1ll cont1nues to d1st1ngu1sh the pattern of food product1on and ava1lab1l1ty although to a decreas1ng extent over t1me ln the Sov1et Un1on Eastern European countr1es and North Korea as well as 1n Ch1na V1etnam and other soc1al1st nat1ons more su1ted to cassava product1on 51 Although d1ssem1nat1on of cassava was emphas1zed throughout the 1950s broaden1ng cult1vat1on 1n the two southern prov1nces and 1n1t1at1ng 1t 1n ZheJlang and J1angx1 cassava research began to show results 1n the late 1950s Between 1957 and 1962 the Agr1cultural Sc1ence Department s Gra1n Crops Laboratory of the South Ch1na Academy of Agr1cultural Sc1ence 1n Guangzhou (23°8 N) selected 10 var1et1es from a pool of 30 for d1ssem1nat1on at least s1x of wh1ch have been extens1vely cult1vated 1nclud1ng ZaJlao [Hybr1d] no 4 and 'ftnn1 X1ye [Indones1an th1n leaf] exh1b1t1ng 11 percent and 23 percent y1eld 1mprovements over w1dely planted Hongwe1zhong [Red Ta1l Var1ety] and M1anbao Mushu [Bread Cassava] ZaJlao no 1 and Nanwan Mushu [South Bay Cassava] y1eld1ng 70 86 percent of Hongwe1zhong but exh1b1t1ng other des1rable character1St1cs such as super1or ed1b1l1ty h1gher starch rates and/or y1eld stab1l1ty Although breed1ng obJectlves for cassava have broadened cons1derably s1nce the 1950s h1gher root y1elds and 1mproved ed1b1l1ty rema1n as central 51Sh1geru Ish1kawa Ch1na s Food and Agr1culture A Turn1ng Po1nt Food Pol1cy 2 (May 1977) p 93 Bruce Stone Ch1na s 1985 Foodgra1n Product1on Target lssues and Prospects 1n Anthony M Tang and BrJc~ Stcne Food ProdJct1on 1n the Peooles Reoubl1c of Ch1na Research Report no 15 (Wash1ngton D C Internat1onal Food Pol1cy Research Inst1tute 1980) pp 92 96 - 50 - Table 9 Cassava Root Nutr1t1onal Content (percent) Van ety M1anbao Mushu 101 [Bread Cassava 101] Naom1 Mushu 102 [Glut1nous R1ce Cassava 102] Mala1huang 103 [Malay Yellow 103] Wenchang Hongx1n 104 [Wenchang Red Heart 104] Maom1ng Ba1x1n 105 Water Starch Soluble Content Rate Sugar 64 o 29 2 1 29 63 o 29 o 2 15 63 2 31 3 1 46 62 4 30 5 1 26 [Luxur1ant & famous Wh1te Heart 105]60 6 32 6 1 54 Ha1nan Hongx1n 211 [Ha1nan(Island) Red Heart 211] a Huguang ~lngJlng 210 [Huguang Green Stem] Hongwe1zhong 201 [Red ta1l var1ety 201] (Inn1 X1ye 202 [Indones1an Th1n Leaf 202] jinn1 Daye 203 [Indones1an B1g Leaf 203] Nanyang Q1ngp1 204 [South seas Green sk1n 204] Nanwan Mushu 205 [South Bay Cassava 205] Huanan 206 [South Ch1na 206] Huanan 207 [South Ch1na 207] ZlJlngzhong 208 [Purple stem var1ety 208] Fanyu ZlJlng 209 [Fanyu (County)Purple Stem 209] Average of all var1et1es 67 o 57 6 710 65 4 66 o 66 o 66 o 59 o 64 8 70 1 61 8 64 2 26 8 36 8 23 7 27 7 28 2 28 8 28 1 35 6 29 6 21 5 23 o 28 8 1 85 1 23 2 22 2 03 1 69 2 87 1 85 1 93 1 00 3 43 2 02 1 86 Prote1n Fat F1ber o 61 o 20 o 74 o 81 o 20 o 80 1 09 o 15 o 72 1 55 o 21 o 84 1 04 o 13 o 68 o 50 o 21 o 71 1 40 1 14 o 63 o 59 o 32 o 68 o 73 o 13 o 76 o 92 o 14 o 61 o 60 o 17 o 72 1 13 o 17 o 64 o 99 o 16 o 71 o 88 o 12 o 74 o 47 o 19 o 90 o 86 o 15 o 88 o 89 o 17 o 74 Sources - 51 - L1ang Guangshang (ed ), Mushu Za1pe1 yu L1yong [Cassava Cult1vat1on and Use] Guangzhou Guangdong Kezh1 Chubanshe [Guangdong Sc1ent1f1c and Techn1cal Publ1sh1ng House] 1981) p 108 foc1 of the Ch1nese breed1ng program 52 South Ch1na 201 1s also known as Hongwe1zhong or Dongguan Hongwe1 [Dongguan Red Ta1l] A h1gh y1eld1ng cult1var w1th h1gh cyan1de content 1t 1s the most popular var1ety for flour product1on Cult1vated on pla1ns h1lly tracts and mounta1nous uplands th1s var1ety covers 70-80 percent of cassava area 1n many Guangdong and Guangx1 Prefectures It 1s also exper1mentally cult1vated 1n the Yangz1 Valley South Ch1na 202 orYlnn1 X1ye was 1ntroduced from Indones1a 1n 1956 by the South Ch1na Agr1cultural Sc1ence Department 1n Guangzhou It typ1cally outy1elds Hongwe1 by a small marg1n but has the h1ghest cyan1de content of popular var1et1es and 1s thus also used 1n process1ng 1ndustr1es pr1mar1ly for flour and starch product1on Plant1ngs are concentrated on the Aox1 State Farms There has also been successful exper1mental cult1vat1on 1n NanJlng South Ch1na 205 or Nanwanmushu was the shortest of the s1xteen lead1ng cult1vars tested and 1s famous for w1thstand1ng the August 17 typhoon 1n 1963 It comb1nes y1eld stab1l1ty w1th h1gh potent1al 52Llang Guangshang (ed ) Mushu Za1pe1 yu L1yong pp 10 and 77 Much of the succeed1ng d1scuss1on on var1et1es and 1nst1tut1ons 1s based on pp 77 80 and Table 9 w1th a few add1t1ons from Cock and Kawaro Cassava 1n As1a - 52 and 1s good for flour and espec1ally starch product1on where 1t s1gn1f1cantly outperforms other popular var1et1es As Table 9 Q 1nd1cates Huguang ~lngJlng [Huguang Green Stem] or South Ch1na 210 and South Ch1na 206 have by far the h1ghest starch rates per un1t we1ght but Nanwanmushu s respectable rate coupled w1th h1gher y1eld potent1al make 1t a clear leader 1n starch per un1t of harvested area Follow1ng Nanwanmushu South Ch1na 206 207 andllnnl X1ye feature the h1ghest starch content per un1t area South Ch1na 205 1s an 1nternat1onally recogn1zed cult1var w1th s1m1lar character1St1cs to those of the Vassour1nha var1ety of Braz1l and the Ph1l1pp1nes The greatest area of Nanwanmushu concentrat1on 1s Zhongshan Dongguan and other count1es 1n the Pearl R1ver Delta but 1t 1s planted w1dely throughout Guangdong South Ch1na 101 or M1anbao Mushu 1s also known as Mala1hong [Malay Red] s1nce 1t was 1ntroduced onto rubber plantat1ons 1n Dan X1an from Malays1a 1n 1912 The var1ety comb1nes y1eld stab1l1ty w1th low cyan1de content and reasonably h1gh y1eld potent1al and 1s recogn1zed as Ch1na s best tast1ng cult1var Plant1ngs are concentrated on Ha1nan Island espec1ally 1n Dan X1an Wenchang and Baot1ng Count1es but bread cassava 1s also grown 1n most areas of Guangdong and has been exper1mentally cult1vated 1n Hebe1 Prov1nce farther north than any other var1ety (39°20 N) lts characterlst1cs are relat1vely s1m1lar to those of A1p1n Valenc1a of Southeast As1a South Ch1na 104 or Wenchang Hongx1n [Wenchang Red Heart] 1s the h1ghest y1eld1ng var1ety among the better tast1ng (sweeter) cult vars It has the h1ghest prote1n content of the 16 lead1ng - 53 - var1et1es also features low cyan1de concentrat1ons reportedly outy1elds M1anbao Mushu by 22 percent but 1s not typ1cally preferred to the latter for d1rect consumpt1on South Ch1na 104 1s planted predom1nantly 1n Wenchang and Q1ongshan Count1es on Ha1nan Island w1th l1ttle cult1vat1on elsewhere Among other palatable var1et1es Maom1ng Ba1x1n [Maom1ng Wh1te Heart] or South Ch1na 105 from Maom1ng Mun1c1pal Area near Guangdong s Le1zhou Pen1nsula and Nuom1 Mushu [pol1shed glut1nous r1ce cassava] or South Ch1na 102 are worthy of ment1on Both outy1eld M1anbao Mushu by 10 11 percent w1th substant1ally greater super1or1ty 1n more northern areas Both are sweet and low 1n cyan1de content w1th South Ch1na 102 lowest of the s1xteen prom1nent var1et1es A var1ety known as 6068 1s also famous for 1ts excellent eat1ng qual1t1es and 1s planted on around 10 000 hectares desp1te 1ts modest y1elds In sum the South Ch1na Trop1cal Crops Research Academy concentrated not only on select1on and d1ssem1nat1on of cult1vars featur1ng h1gher and more stable root y1elds and 1mproved ed1b1l1ty but has focused breed1ng attent1on 1n comb1n1ng those character1st1cs and 1n1t1ated research on starch content By focus1ng on faster as opposed to str1ctly h1gher root y1elds the Academy also brough to cassava breed1ng 1n th1s early per1od the beg1nn1ngs of a qu1ntessent1ally Ch1nese or1entat1on breed1ng to f1t rotat1onal patterns and mult1ple cropp1ng sequences 54 - \J1th the catastroph1c fam1nes of 1960-61 centered 1n North Ch1na and the Yangz1 Valley efforts to spread cassava cult1vat1on northward 1ntens1f1ed cons1derably The focal 1nst1tut1on 1n th1s effort was the ZheJlang Prov1nce Sub-troplcal Crops Inst1tute 1n P1ngyang (27°38 N) Between 1962 and 1964 the 1nst1tute 1ntroduced 31 var1et1es from Guangdong Guangx1 and FuJlan 1nclud1ng Hongwe1 Nanwanmushu Inn1 Daye Sh1be1~1ngJ1ng [stone tablet green stem] and ZaJlao nos 1-6 But as Table 10 1nd1cates there has been exper1mental cult1vat1on much further north although the South Ch1na Trop1cal Crops Research Academy has 1nd1cated that good growth and y1elds are cons1stently obta1ned only up to around 26°N wh1ch cuts across southern Hunan Gu1zhou J1angx1 and FUJlan As1de from the above-ment1oned 1nst1tut1ons sorne cassava related research 1s reportedly conducted 1n each of the prov1nces w1th1n wh1ch cassava has been 1ntroduced In South Ch1na other relevant 1nst1tut1ons are the Guangx1 Prov1nce As1an Trop1cal Crops Research Inst1tute 1n Nann1ng the South Ch1na Crop Research Inst1tute and the South Ch1na Inst1tute of Botany w1th1n the Ch1nese Academy of Sc1ences the Inst1tute of Drought Res1stant Gra1ns and the Upland Gra1ns Department 1n the Guangdong Agr1cultural Sc1ence Academy and the South Ch1na Agr1cultural College all 1n Guangzhou However cassava research 1s not reputed to be a s1gn1f1cant current focus of any of the Guangzhou 1nst1tut1ons Cassava research and development 1n Ch1na 1s 1ncreas1ngly sh1ft1ng 1ts focus from the or1g1nal narrowly def1ned goals of 55 - Table 10 Results of Cassava s North M1grat1on Cult1vat1on Exper1ments Exper1ment1ng Un1t Northwest Agr1cul tural Sc1ence Academy Hube1 Oashahu Farm Anhu1 Prov1nce Crops Inst1tute NanJlng Botan1cal Inst1tute Ch1na Root and Tuber lnst1tute Shaanx1 Prov1nce Gra1ns Crops Inst Shandong Prov1nce Crops Inst1tute Luda (Oal1an) no 1 Farm Hebe1 Prov1nce Forestry Sc1ence Inst1tute Locat1on Var1ety (N lat1tude) 30° A B O 32 04 A B C 38 20 A B Plant1ng Date Apr 25 Apr 21 Apr 12 Apr 15 May 6. May 7 Apr 15 May 6 Apr 21 Notes A= Naom1mushu [Glut1nous R1ce Cassava] B= M1anbaomushu [Bread Cassava] C= Inn1 X1ye [Indones1an Th1n Leaf] O= Mala1huang [Malay Yellow] Harvest Total Grow1ng Date Oays Nov 25 216 Nov 22 216 Nov 3 206 Nov 5 205 Oct 24 172 Oct 23 170 Oct 24 193 Oct 23 171 Oct 24 187 Fresh Root Y1 e 1 d (tons/ha ) 33 o 18 75 30 o 20 325 23 25 24 4' 37 5 45 o 5 775-17 7 22 5 12 75 19 5 37 5 45 o Sources L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong [Cassava Cult1vat1on and Use] Guangzhou Guangdong Kezh1 Chubanshe [Guangdong Sc1ent1f1c and Techn1cal Publ1sh1ng House] 1981) p 26 - 56 1mprov1ng y1eld and ed1b1l1ty The ma1n 1mprovement efforts st1ll 1nclude ed1b1l1ty but also emphas1ze cult1vat1on techn1ques espec1ally cassava s relat1on to other crops 1n var1ous systems and the comb1ned development of cassava and non-crop rural act1v1t1es Breed1ng obJect1ves also 1nclude early plant1ng early r1pen1ng and rap1d matur1ty goals as well as d1sease res1stance h1gh y1elds and h1gh starch and prote1n content 53 Research and development goals related to cult1vat1on techn1ques feature 1mprovement 1n rotat1on synerg1es seasonal cult1vat1on 1ntercropp1ng and ach1evement of two or even three r1pen1ngs per year Bean crop and cassava rotat1ons and 1ntercropp1ng are of part1cular 1nterest as techn1ques for develop1ng so1l strength The 1982 CIAT delegat1on observed that cassava was often 1ntercropped w1th gra1n legumes 1n more 1ntens1vely cult1vated areas and est1mated that y1elds of both crops were probably reduced by only 15-30 percent result1ng 1n relat1vely eff1c1ent land use w1th good so1l conservat1on propert1es 54 S1nce 1979 non crop agr1culture has been emphas1zed 1n Ch1na part1ally correct1ng for the substant1al pre 1979 stress on food crops espec1ally staples Consequently a recent goal for cassava development has been to 1ntegrate cassava w1th forestry an1mal husbandry ser1culture aquaculture and rural s1del1nes for 53L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 10 54Ibld correspondence from James H Cock Cassava Program D1rector CIAT June 24 1983 57 - cooperat1ve product1on Invest1gat1on of add1t1onal and even novel 1ndustr1al uses 1s also of 1ncreas1ng 1nterest Survey respondents among Ch1nese cassava breeders and agronom1sts 55 appeared opt1m1st1c about the potent1al for growth 1n farmers y1elds dur1ng the next 4 and 14 years Respondents were 1nstructed to base the1r assessments on ex1st1ng var1et1es and those currently under development but the1r est1mates d1ffered cons1derably They were also opt1m1st1c about the prospects for 1ncreas1ng that potent1al v1a a doubl1ng of research expend1tures related to cassava w1th the most conservat1ve assessments prov1ded by the representat1ve of the 1nst1tut1on where most research on cassava 1s conducted In h1s v1ew farmers y1elds on ooor so1ls could 1ncrease from currently 3 6 tons per hectare to 4-8 tons by 1990 and 5-9 tons by 2000 or 5-10 tons and 6-12 tons respect1vely w1th a doubl1ng of research expend1tures W1 th good so 11 and cl1mat1c cond1t1ons farmers y1elds could 1ncrease from currently 15-30 tons/hectare w1th fert1l1zer to 18-35 tons by 1990 and 20-40 tons by 2000 or 25 35 tons and 35-45 tons w1th a doubl1ng of research resources lt 1s clear that y1elds can 1mprove espec1ally 1n Guangdong v1a greater access to manufactured fert1l1zers analys1s and extens1on related to 1ts opt1mal use and to proper select1on of plant1ng mater1als Fert1l1zer pr1c1ng d1str1but1on and analyt1c systems are undergo1ng cons1derable structural change 1n Ch1na 55Delph1 Survey responses - 58 Proper resolut1on of rema1n1ng and newly emerg1ng d1ff1cult1es w1ll be 1nstrumental 1n ach1ev1ng y1eld progress through growth 1n fert1l1zer use 56 It also appears that there may be sorne l1m1ted potent1al exploitable w1th further 1nternat1onal exchange of genet1c mater1als 57 State farms are technolog1cal leaders 1n cassava cult1vat1on though not for most staple crops and careful select1on of plant1ng mater1als and quest for 1mproved cult1vars are ev1dent on state farms Y1eld progress on several state farms 1n recent years has allowed cont1nued prof1tab1l1ty of cassava cult1vat1on desp1te decl1n1ng pr1ces Th1s means that new 1mproved var1et1es can move rap1dly 1nto full scale product1on 1n Ch1na What may be called for are 1nst1tut1onal l1nks wh1ch can br1ng state farm developments 1nto the pr1vate and collect1ve economy more exped1t1ously A new var1ety must undergo reg1onal test1ng for three years The results are presented to the prov1nc1al seed comm1sS10n wh1ch may then recommend the var1ety to seed product1on compan1es for mult1pl1cat1on Work on 1ntercropp1ng and rotat1onal systems 1s someth1ng Ch1nese researchers do part1cularly well and 1s l1kely to lead to sorne further 1mprovements Sorne of these may not 1mmed1ately 56For deta1ls see Bruce Stone Ch1nese Fert1l1zer Appl1cat1on 1n the 1980s and 1990s lssues of Growth Balance Allocat1on Eff1c1ency and Response 1n U S Congress Jo1nt Econom1c Comm1ttee (eds ) Ch1na s Economy Looks Toward the Year 2000, vol 1 The Four Modern1zat1ons (Wash1ngton D C U S Government Pr1nt1ng Off1ce 1986) pp 453 496 57cock and Kawano Cassava 1n Ch1na Ch1na (18 26 January 1986) Ka11ano Tn p Report to - 59 - 1ncrease cassava y1elds per se but may 1mprove the attract1veness of plant1ng cassava and thus arrest 1ts decl1ne 1n area What 1s s1ngularly m1ss1ng for cassava as well as for many other crops 1s soc1o econom1c research 1n cassava areas part1cularly poorer enes Lack of agro-econom1c data and analys1s for assess1ng constra1nts l1m1t1ng farmers y1elds 1s recogn1zed by the South Ch1na Trop1cal Crops Academy 58 F1nally w1th the reduct1on 1n export opportun1t1es and the curta1led government role 1n market1ng development of demand and market 1nst1tut1ons are of part1cular 1mportance for cont1nued expans1on of cassava product1on and use These 1ssues w1ll be undertaken 1n the follow1ng sect1ons MARKETS ANO DEMAND A synthes1s of product1on and ut1l1zat1on As 1nd1cated above product1on stat1st1cs for cassava 1n Ch1na are h1ghly fragmentary except for Guangx1 Zhuang Autonomous Reg1on for wh1ch data are complete though even for Guangx1 quest1ons of rel1ab1l1ty and comparab1l1ty rema1n Ut1l1zat1on data however are almost wholly unava1lable w1th the except1on of the 1nternat1onal trade data comp1led from European Commun1ty Analyt1c Tables for Fore1gn Trade appear1ng 1n Table 3 Government procurement data for cassava assuredly ex1st but have not been made ava1lable 1n Ch1nese 58oelph1 Survey response from Tan Xuecheng breeder - 60 - stat1st1cal compend1a on market1ng and trade Product1on data from cassava flour and starch factor1es as well as from other 1ndustr1al processors are certa1nly generated but are not of suff1c1ent 1mportance to appear among nat1onal stat1st1cal ser1es 1n the relat1vely deta1led Guangdong Prov1nce Stat1st1cal Yearbooks and the Guangx1 Econom1c Yearbook 1985 although the latter conta1ns a s1ngle column of dlscuss1on of the starch market 1n wh1ch cassava 1s ment1oned As a reg1onally concentrated crop cassava has not turned up among publ1shed results from nat1onal farm surveys Even L1ang Guangshang s cassava-spec1f1c publ1cat1on Mushu Za1pe1 yu L1yong [Cassava Cult1vat1on and Use] prov1des not a s1ngle stat1st1c on aggregate ut1l1zat1on In the past 1t has been clear that FAO est1mates of cassava use were all based on constant percentages of est1mated product1on 59 For example the FAO Supply Ut1l1zat1on Accounts Tape 1981 ev1dently 1ncorporated the follow1ng percentages feed use (25 percent) waste (5 percent) food use (67 percent) process1ng (3 percent) use for tap1oca (70 percent of process1ng) starch use (30 percent of process1ng) 60 S1nce the product1on ser1es was mechan1cally generated from v1rtually no stat1st1cal base the ut1l1zat1on ser1es were 1nev1tably unrel1able even 1f the percentage shares were roughly correct Conversely regardless of the accuracy of the product1on est1mates 59Bruce Stone Cassava Product1on the ut1l1zat1on shares have assuredly not been An Exam1nat1on of Econom1c Data on Ch1nese Ut1l1zat1on and Trade pp 13-22 60Food and Agr1culture Organ1zat1on of the Un1ted Nat1ons Supply Ut1l1zat1on Accounts Tape 1981 Reme 1982 - 61 constant over t1me w1th feed and process1ng use 1ncreas1ng 1n 1mportance at the expense of d1rect human consumpt1on Moreover shares for feed and process1ng would exceed the shares 1mpl1ed by the 1981 Ut1l1zat1on Tapes even for the 1960s 61 Asan exam1nat1on of Tables 11 and 12 w1ll reveal FAO ut1l1zat1on ser1es for Ch1na are now generated 1n a more compl1cated fash1on but h1stor1cal product1on area and y1eld f1gures are 1dent1cal to those appear1ng on the older tapes As1de from the 1nternat1onal trade ser1es wh1ch relates well to and 1s probably based on the EC Analyt1c Tables for Fore1gn Trade FAO ser1es are st1ll generated from an extremely weak stat1stlcal bas1s wh1ch probably cons1sts of no more than the partner country trade data and the s1ngle product1on f1gure c1rca 1980 prov1ded to the 1982 CIAT delegat1on In these recent FAO ser1es such as Supply Ut1l1zat1on Accounts Tape 1984 released at the end of 1985 unprocessed feed 1s set at 10 percent throughout the 1961 83 per1od and waste 1s dropped from 5 percent on prev1ous tapes to 3 percent for the ent1re per1od D1rect food consumpt1on est1mates have become trended values decl1n1ng from 72 O percent of product1on 1n 1962 to 67 O percent 1n 1979 (Table 12) Processed uses have become monot1cally non-decreas1ng trended values beg1nn1ng somewhat arb1trar1ly at 15 O percent 1n 1962 and r1s1ng to 20 O percent 1n 1979 of wh1ch dr1ed cassava (ch1ps ~nd 61stone An Exam1nat1on of Econom1c Data on Ch1nese Cassava Th1s paper was prov1ded to both CIAT and the FAO Stat1st1cal OlVlSlOn s Bas1c Data Un1t 1n 1983 and prov1ded part of the bas1s for subsequent adJustments 62 Table 11 FAO Est1mates of Ch1nese Cassava Product1on Are a and Y1eld 1961-1984 Harvested Area Product1on Y1eld 1982 Ta~e 1984 Ta~e 1982 Ta~e 1984 Ta~e 1982 Ta~e 1984 Ta~e (1000 hectares) (1000 metr1c tons) ( tons per hectare) 1961 80 940 11 750 1962 85 1000 11 765 1963 85 950 11 176 1964 90 1000 11111 1965 90 1100 12 222 1966 95 95 1100 1100 11 579 11 579 1967 100 100 1200 1200 12 000 12 000 1968 120 120 1400 1400 11 667 11 667 1969 130 130 1500 1500 11 538 11 538 1970 140 140 1600 1600 11 429 11 429 1971 150 150 1800 1800 12 000 12 000 1972 160 160 1900 1900 11 875 11 875 1973 170 170 2000 2000 11 765 11 765 1974 170 170 2000 2000 11 765 11 765 1975 180 180 2100 2100 11 667 11 667 1976 180 180 2200 2200 12 222 12 222 1977 190 190 2200 2200 11 579 11 579 1978 200 200 2300 2300 11 500 11 500 1979 200 200 2500 2500 12 500 12 500 1980 226 226 3000 3300 13 274 14 602 1981 236 230 3120 3500 13 232 15 217 1982 235 3600 15 319 1983 240 3800 15 833 1984 So urce FAO Supply Utll1zat1on Accounts Tape 1981 Ro me 1982 FAO Supply Utll1zat1on Accounts Tape 1984 Reme 1985 - 63 Table 12 FAO Est1mates of Ch1nese Cassava Product1on and Use 1961-1983 Product1on of wh1ch Feed Waste Food Processed of wh1ch 1nput to ChlES & Pellets TaElOCa Starch (1000 tons) 1961 940 94 28 668 140 90 20 30 1962 1000 100 30 720 150 100 20 30 1963 950 95 28 666 160 110 20 30 1964 1000 100 30 699 171 120 21 30 1965 1100 110 33 756 201 150 21 30 1966 1100 110 33 740 217 160 22 35 1967 1200 120 36 807 237 180 22 35 1968 1400 140 42 959 259 200 24 35 1969 1500 150 45 1014 291 230 26 35 1970 1600 160 48 1099 293 230 28 35 1971 1800 180 54 1246 320 250 30 40 1972 1900 190 57 1330 323 250 33 40 1973 2000 200 60 1384 356 280 36 40 1974 2000 200 60 1380 360 280 40 40 1975 2100 210 63 1467 360 280 40 40 1976 2200 220 66 1519 395 300 50 45 1977 2200 220 66 1519 395 300 50 45 1978 2300 230 69 1606 395 300 50 45 1979 2500 250 75 1675 500 400 55 45 1980 3300 330 99 1466 1405 1300 60 45 1981 3500 350 105 1545 1500 2000 65 45 1982 3600 360 108 1512 1620 1500 75 45 1983 3800 380 114 1606 1700 1700 78 45 Notes and Sources FAO Supply Ut1l1Zat1 on Accounts Tape 1984 Ro me 1985 To reach quant1t1es of processed products extract1on rates of 35 percent for ch1ps and pellets (dr1ed cassava) 22 percent for taploca and 18 percent for starch are appl1ed 1n FAO data 64 - pellets for feed e1ther for domest1c use or export) starts at 2/3 of the processed amount 1n 1962 and r1ses to 80 O percent 1n 1979 Cassava 1nput to starch product1on beg1ns at 20 O percent of the processed arnount 1n 1962 and decl1nesto 9 O percent 1n 1979 The absolute quant1t1es 1n FAO data forrn a step funct1on rerna1n1ng constant for f1ve-year per1ods then 1ncreas1ng by 5 thousand tons 1n a s1ngle year then rerna1n1ng constant aga1n for f1ve years Cassava 1nput to tap1oca product1on cornpr1ses the rerna1nder w1th absolute quant1t1es r1s1ng 1n s1rn1lar rnonot1cally non-decreas1ng fash1on but w1th shares decl1n1ng sl1ghtly to 11 percent by 1979 FAO data appear 1n other forrnats but the stat1st1cal base or lack thereof rerna1ns the sarne For exarnple the Standard1zed Cornmod1ty Balances Tape 1984 (Rome 1985) 1ncludes ser1es for ava1lab1l1ty (product1on rn1nus exports) food (d1rect food consurnpt1on plus cassava 1nput to tap1oca process1ng) and other uses (waste plus cassava 1nput to starch process1ng) Because of the rnass1ve 1ncrease 1n exports 1n 1979-81 the post 1979 FAO ser1es exh1b1t sorne pecul1ar1t1es Dr1ed cassava 1nput on the Supply Ut1l1zat1on Tape 1ncreases frorn 20 O percent to 42 6 percent of product1on frorn 1979 to 1980 (Table 12) for exarnple and the prograrn synthes1z1ng these ser1es generated large negat1ve nurnbers for other uses 1n 1980 and 1984 on the Standard1zed Cornrnod1ty Balance Tape Nevertheless these ser1es represent sorne 1rnprovernent 1n cred1b1l1ty over the 1981 82 tapes The waste percentage has been lowered (to what 1s probably the rn1n1mum parametr1c value used by FAO) The est1rnated product1on shares of processed cassava have been 65 - ra1sed very substant1ally and exh1b1t a r1s1ng trend 1nclud1ng sl1ghtly r1s1ng then stagnat1ng absolute quant1t1es for starch product1on and a mass1ve accelerat1on 1n dr1ed cassava to parallel the appearance of lucrat1ve export opportun1t1es 1n the 1980s Food uses exh1b1t a plaus1ble decl1n1ng share of cassava product1on and the FAO trade data now 1ncludes the overwhelm1ngly 1mportant movements 1n the dr1ed cassava trade s1nce 1979 But 1t must be remembered that there 1s no actual stat1st1cal bas1s for these ut1l1zat1on shares save a very 1nd1rect one based on the fore1gn trade data and all ser1es are essent1ally der1ved from the almost wholly unrel1able product1on est1mates Of course 1t 1s much eas1er to cr1t1c1ze than to suggest super1or alternat1ves s1nce l1ttle quant1tat1ve 1nformat1on from Ch1na 1s ava1lable But 1t may be reasonable to suggest that several of the 1mprovements s1nce the 1981-82 tape d1d not go far enough Ch1na has developed a cons1derable reputat1on for low food waste As others have prev1ously 1nd1cated th1s reputat1on may be somewhat exaggerated 62 But w1th a large proport1on of the cassava crop allocated to same farm an1mal feed and h1gh labor appl1cat1on per hectare one may reasonably expect that at least cassava waste 1n Ch1na 1s qu1te low The 1982 CIAT delegat1on observed that the pr1mary use of cassava was as an1mal feed Of course the1r sample was b1ased toward more product1ve farms though they v1s1ted sorne very poor 62e 9 Compos1t1on Vaclav Sm1l Prospects Ch1na s Food Ava1lab1l1ty Requ1rements Food Pol1cy (May 1981) pp 67-77 66 - communes where cassava was the pr1nc1pal human food source any of the state farms 1mmed1ately b1ased the sample on such a br1ef tr1p Based on Table 1 and other f1gures prov1ded above state farm ca~sava plant1ngs could not have exceeded 3 5 percent of Guangx1 cassava area 1n 1984 although probably totall1ng 5-10 percent of product1on In Guangdong the proport1ons could be sl1ghtly h1gher but state farm cassava 1s clearly a m1nor share of the total However the C!AT delegat1on found cassava pr1mar1ly grown for an1mal feed on communes as well as on state farms Accord1ng to the extens1ve surveys (also b1ased toward more product1ve farms) conducted by NanJlng Un1vers1ty students superv1sed by John Loss1ng Buck between 1929 and 1933 18 percent of the output of sweet potatoes (generally a food preferred by Ch1nese to cassava) was employed as an1mal feed 1n the reg1on The proport1on was almost half 1n the more product1ve areas of eastern Guangdong Only 60 percent of the taro crop was used for human food 63 S1nce the 1930s sw1ne stocks and gra1n and sugar product1on have 1ncreased more rap1dly than the human populat1on 1n the reg1on (Table 13) and per cap1ta 1ncomes have 1ncreased 01lseed and soybean product1on has decl1ned 1n Guangx1 but 1n Guangdong product1on 1ncreased at about the rate of populat1on growth over the 5 decade per1od g1ven that 1ncluded 1930s f1gures are somewhat prone to overest1mat1on Cattle stocks decl1ned over the 1970s 1n Guangdong but due to the1r smaller numbers and d1et preference for leaves and grasses over roots th1s 63John Loss1ng Buck Land Ut1l1zat1o~ 1n Ch1na (Atlas and Study) (Nank1ng Nank1ng Un1vers1ty 1937) Atlas pp 82 an~ 67 Table 13 Growth Ind1ces for Human Populat1on L1vestock and Gra1n Sugarcane Peanut and Soybean Product1on 1n Guangdong and Guangx1 1930s-1984 Human populat1on Sw1ne stocks Cattle & buffalo stocks Small rum1nant stocks Foodgra1n product1on Sugarcane product1on Peanut product1on Soybean product1on Cassava product1on Notes 1979 84 Guangdong Guangx• (1952-1957 avg =100) 162 a/ 280 b/ 74 e; 15 e/ 171 - 246 285 d/ 182 ~/ 181 257 261 310 181 691 138 757 Average Guangdong Guangx1 (1930s=100) 174 178-199 1631 168 156 221 205-249 69 469 ~/ Based on a we1ghted average of m1dyear to approx1mate a m1dyear 1955 f1gure f1gures f1gures for 1954 and 1957 1979-84 data are year end Ql Based on a m1dyear 1955 f1gure A we1ghted average of m1dyear 1953 m1dyear 1955 anda year end 1957 1s sl1ghtly lower f/ Based on year-end 1984 and 1957 f1gures ~/ Based on 1953-56 average The 1ndex number based on 1957 alone 1S 199 ~/ Based on 1952-56 average The 1ndex number based on 1957 alone 1 S 94 Sources Bruce Stone An Exam1nat1on of Econom1c Data on Ch1nese Cassava Product1on Ut1l1zat1on and Trade paper prepared for the Internat1onal Center for Trop1cal Agr1culture (CIAT) IFPRI Wash1ngton D C August 1983 Table 11 Data have been supplemented from Guangx1 J1ngJ1 N1anJ1an B1anJ1bu Guangx1 J1ngJ1 N1anJ1an 1985 pp 519 530 532 and 594 and from State Stat1st1cal Bureau PRC Stat1st1cal Yearbook of Ch1na 1983 1984 and 1985 -- -- - 68 - decl1ne would have less effect on the allocat1on of the cassava root 1tself than would the sw1ne stock growth rate Accord1ng to a 1980 survey of 15 914 households an average of 94 4 k1lograms of meat (mostly pork) 35 6 k1lograms of gra1ns and 126 k1lograms of vegetables were produced on pr1vate plots Although hog feed1ng reg1mens 1n Ch1na have been concentrate poor h1stor1cally the fatten1ng process would st1ll requ1re around 82 k1lograms of concentrate per hog and the requ1rement has been r1s1ng w1th greater peasant autonomy adJusted purchase pr1ce structure and grow1ng acceptance that extremely concentrate-poor d1ets are uneconom1c 64 In Guangdong and Guangx1 a s1zable proport1on of th1s concentrate cons1sts of cassava taro and sweet patato Of the three cassava would be the crop w1th the h1ghest proport1on allocated for feed One may conclude that even for domest1cally ut1l1zed cassava 20-25 percent (for feed use plus dr1ed cassava from 1961 79 1s probably too small a proport1on for feed and the trend must have been r1s1ng more rap1dly over the per1od than assumed by FAO When one cons1ders that from 1980 82 dr1ed cassava exports must have const1tuted 30 60 percent of what the 1982 CIAT delegat1on was told was nat1onal product1on and that exports may st1ll exceed 30 percent of annual output even the current FAO feed proport1ons of 50 55 percent ( dr1ed cassava plus feed ) may be too low 64see Stone Ch1na s 1985 Foodgra1n Product1on Target PP 99 103 The 1980 survey appeared 1n X1nhua [New Ch1na News Agency] news bullet1n June 16 1981 - 69 Table 14 Development of Starch Product1on 1n South Ch1na 1952-1984 Number of Starch Requ ned Proport1on of Total DEeratwg Facton es Product1on Fresh Root Cassava OutEut Guangx1 Guangdong Guangx1 Guangx1 Guangx1 (metr1c tons) 1952 1 282 ( 1 500) ( 1) 1959 12 275 ( 68 000) ( 10) 1962 29 1972 56 10 000 (40 60 000) (3-14) 1983 284 59 400 ( 242 500) ( 15) 1984 240 49 000 ( 200 000) ( 17) Notes and Sources F1gures 1n parentheses are calculated est1mates The FAO extract1on rate of 18 percent was used for the 1950s data to calculate fresh root equ1valent assum1ng also that all Guangx1 starch was produced from cassava (Actually small amounts of corn are a1so used ) For 1ater years an extract1on rate of 24 5 percent was used ba5ed on the 5tatement that starch content of dr1ed cassava 1s more than 70 percent (Guangx1 J1ngJ1 N1anJ1an B1anJ1bu 1985) [Guangx1 Econom1c Yearbook Ed1tor1a1 Board] Guangx1 J1ngJ1 N1anJ1an 1985 [Econom1c Yearbook of Ch1na 1985] (Nann1ng Guangx1 J1n9J1 N1anJ1an B1anJ1bu 1985) p 192) If the FAO-adopted dry1ng factor of 35 percent 1s used th1s 1mpl1es a 5tarch extract1on rate of more than 24 5 percent wh1ch 15 poss1ble e5pec1ally 1n v1ew of sub5tant1al cas5ava 5elect1on and breed1ng 1n Ch1na for h1gh 5tarch content The 1982 ClAT delegat1on observed extract1on rates of 25 29 percent w1th 5-10 percent re51due5 for an1mal feed (Cock and Kawano Cas5ava 1n Ch1na p 8) It 1s not clear why the FAO-adopted extract1on rate for tap1oca (22 percent) 15 h1gher than for starch and exh1b1ts as much as a 4 percent d1fference 51nce tap1oca product1on normal1y fol1ows from starch product1on thereby ach1ev1ng a very sl1ghtly lower extract1on rate (correspondence from John K Lynam Cassava Program Centro Internac1onal de Agr1cultura Trop1cal (CIAT) December 22 1983 ) The proport1on allocated to starch product1on 1s probably also cons1stently underest1mated by FAO Data assembled 1n Table 14 suggest that 1f the Guangx1 record can be taken as representat1ve of both southern prov1nce5 ut1l1zat1on of cassava for starch product1on dur1ng the 1960s and 1970s const1tute not 10 20 percent of all cassava used for process1ng as assumed by FAO (2-3 percent of product1on) but closer to 10 percent of total product1on and 70 - potent1ally h1gher 1n several low product1on years Assum1ng the adopted extract1on rates and the Guangxl ser1es are roughly correct and that starch produced from raw mater1als other than cassava was 1ndeed very m1nor 1n Guangx1 then the starch 1ndustry cla1med more than 15 percent of fresh root product1on 1n the Autonomous Reglan 1n 1983 and 1984 The proport1on for Guangdong 1s probably somewhat lower but appears to be r1s1ng at present All 1n all 1f forced to est1mate current ut1l1zat1on of Ch1nese cassava m1ght run 60-65 percent for feed (1nclud1ng dr1ed cassava plus fresh feed exports and domest1c use) 15 20 percent for the starch 1ndustry 2-4 percent for tap1oca productlon and as l1ttle as 1-3 percent for waste leav1ng somewhere around 10-20 percent for d1rect human consumpt1on As suggested 1n earl1er papers and as FAO seems to accept 1t 1s qu1te poss1ble that the 3 m1ll1on ton c1rca 1980-81 product1on f1gure 1s an underest1mate but the product1on trend for the last few years 1s almost certa1nly downward The Guangx1 starch product1on f1gure l1sted somewhat arb1trar1ly for 1972 1s based on the statement that starch product1on 1n Guangx1 rema1ned at around 10 000 tons dur1ng the 1960s and 1970s (Guangx1 J1ngJ1 N1anJ1an 1985 p 192) Most data 1n the table appeared 1n 1b1d The number of starch factor1es operat1ng 1n Guangx1 1n 1962 and 1n Guangdong 1n 1972 are from L1ang Guangshang (ed ) Mus hu -- Za1pe1 yu L1yong [Cassava Cult1vat1on and Use] (Guangzhou Guangdong KeJl Chubanshe [Guangdong Sc1ent1f1c and Techn1cal Publ1sh1ng House] 198qf p ~ The proport1on of total GuangXl cassava product1on was calculated from data appear1ng 1n th1s table and 1n Table 2 71 Cassava for d1rect human consumpt1on The prev1ous sect1on has concluded that cassava for d1rect human consumpt1on probably compr1ses only 10 20 percent of current product1on There appear to be four pr1nc1pal categor1es of d1rect human consumpt1on of cassava 1n Ch1na consumpt1on related to ethn1c m1nor1t1es where cassava has a trad1t1onal d1etary role consumpt1on related to forest cult1vat1on 1n remate areas consumpt1on assoc1ated w1th exceed1ngly poor and/or rlsk-prone farm1ng areas consumpt1on related to part1cular cu1s1ne and espec1ally seasonal preparat1ons These four categor1es are not mutually exclus1ve but seem to character1ze the d1rect human consumpt1on demand for cassava L1ttle recent ethnograph1c 1nformat1on on m1nor1t1es 1n South Ch1na seems to be ava1lable but taro and cassava are known to be 1mportant food 1tems among the Yao m1nor1ty 1n northern Guangdong 65 The Mao people of Tha1land are also hab1tual consumers of cassava Mao people 1n South Ch1na were l1kew1se reported to eat cassava and \ mao potatoes dur1ng the 1950s 66 Even among Han Ch1nese (93 3 percent of Ch1na s populat1on) home-processed cassava flour 1s often used as a th1ckener 1n southern Ch1nese soups and 1n mak1ng spec1al cakes at fest1val t1mes such as New Year s Eve 1n FUJlan for example 67 65Buck Land Ut1l1zat1on 1n Ch1na (Atlas) p 98 66sun J1ngzh1 (ed ) Bureau PRC Stat1st1cal Huanan D1chu JlngJl D1l1 Yearbook of Ch1na, 1985 p State Stat1st1cal 195 67cock and Kawano Cassava 1n Ch1na Bureau PRC Stat1st1cal Yearbook of Ch1na, p 11 1985 State Stat1st1cal p 195 - 72 - Poorly developed and poorly 1ntegrated markets are almost a def1n1ng character1stlc of develop1ng countr1es and Ch1na 1s no except1on In Ch1na market development was further retarded by a number of factors F1rst for a th1rty year perlad c1v1l war and World War II comb1ned to destroy normal market act1v1ty 1n many areas of Ch1na Although Guangdong and Guangx1 were spared to a much greater extent than North Ch1na the Northeast and the Yangz1 Valley they were not unaffected by war and nearby cassava-grow1ng prov1nces such as Yunnan and Hunan were d1rectly 1nvolved as was FuJlan located d1rectly across the stra1ts from colon1al Ta1wan For example transport veh1cles and draft an1mals were purchased or commandeered for the war effort War t1me 1nflat1on sent market1ng back toa sem1 barter era and cred1t fac1l1t1es were severely affected In the 1950s cond1t1ons stab1l1zed but the government soon began to take over large segments of market1ng act1V1t1es W1th gra1n cr1ses 1n 1953 and 1955 and the d1ff¡cult1es the government was exper1enc1ng w1th procurement of foodstuffs for c1t1es gra1n trad1ng became a state monopoly 1n 1954 and by 1955 each un1t of land 1n Ch1na was ass1gned a f1xed quota of (usually) gra1n to be del1vered to state purchas1ng organ1zat1ons at low f1xed pr1ces Taxes were also pa1d 1n k1nd but gra1n del1very obl1gat1ons d1d not end there After reta1n1ng a prov1nc1ally determ1ned per cap1ta quant1ty to meet 1mmed1ate food feed and seed needs of rural farms and households and even after tax and quota obl1gat1ons were met 80-90 percent of all surplus gra1n was also to be sold to the state Not only was pr1vate gra1n trad1ng 1llegal and most gra1n 1n excess of a modest - 73 - standard for heme consumpt1on soaked up by government purchas1ng organ1zat1on but pr1vate traders were des1gnated as class enem1es The state for 1ts part was hav1ng enough trouble prov1d1ng for urban and army consumpt1on as well as reserv1ng one-two m1ll1on tons per year to export for fore1gn exchange For the most part only relat1vely prom1nent rural areas exper1enc1ng natural d1sasters rece1ved rel1ef gra1n More remate and most very peor areas were left on the1r own w1thout access to gra1n suppl1es from the outs1de After the fam1nes 1n 1960-61 and espec1ally dur1ng the Cultural Revolut1on per1od (1966-76} th1s s1tuat1on was 1nst1tut1onal1zed as a pol1cy of local self-suff1c1ency w1th d1sastrous 1mpl1cat1ons for ga1ns from spec1al1zat1on and trade and for exceed1ngly poor r1sk- prone areas h1stor1cally dependent on trad1ng and non agr1cultural act1v1t1es to garmer enough to eat W1th procurement problems pers1st1ng the government further restr1cted non-farmlng act1v1t1es and made m1grat1on 1llegal 1n arder to l1m1t the state s urban obl1gat1ons but thereby b1nd1ng many farmers even more closely to peor and rlsk-prone agr1culture 68 68see Bruce Stone Relat1ve Foodgra1n Pr1ces 1n the People s Republ1c of Ch1na Extract1ve Rural Taxat1on Through Publ1c Monopoly 1n John W Mellar and Ra1sudd1n Ahmed (eds ) Agr1cultural Pr1ce Pol1cy for Develop1ng Countr1es (Balt1more Johns Hopk1ns Un1vers1ty Press 1987) and Bruce Stone Ch1nese Soc1al1sm s Record on Food and Agr1culture Problems of Commun1sm vol 35 no 5 (Sept -Oct ) 1986 pp 63-72 See also Tang and Stone Food Product1on 1n the People s Republ1c of Ch1na Kenneth Walker Foodgra1n Procurement and Consumpt1on 1n Ch1na (Cambr1dge Cambr1dge Un1vers1ty Press 1984) and N1cholas Lardy Agr1culture 1n Ch1na s Modern Econom1c Development (Cambr1dge Cambr1dge Un1vers1ty Press 1983) - 74 - It 1s not d1ff1cult to 1mag1ne that w1th th1s 1nst1tut1onal framework cassava at least 1n the south had a part1cularly 1mportant role to play Cassava was an 1deal crop for 1nsur1ng m1n1mum levels of consumpt1on because 1t 1s a relat1vely drought res1stant stable y1eld1ng eas1ly stored crop prov1d1ng h1gh calor1c levels per un1t area and performs well relat1ve to alternat1ve crops even under poor agronom1c pract1ce and so1l cond1t1ons As a crop cult1vable on forest lands and h1lls1des 1t was also 1deal for susta1n1ng reclamat1on teams 1n remete areas W1th the rap1d 1ncreases 1n South Ch1nese r1ce product1on dur1ng the past decade (Table 5 6 and 13) the 1980s legal1zat1on of pr1vate gra1n trad1ng and guaranteed state food del1ver1es for areas concentrat1ng on the product1on of econom1c crops cassava s spec1al 1nst1tut1onally 1nduced 1mportance has been decl1n1ng However cassava 1s st1ll grown 1n exceed1ngly poor areas 1n South Ch1na for essent1ally the same reasons food secur1ty and easy prov1s1on of needed calor1es under 1nopt1mal cond1t1ons It should be emphas1zed for example that seven count1es 1n Guangdong and e1ght 1n Guangx1 averaged per cap1ta collect1ve d1str1buted 1ncome 1n 1977 of less than 50 yuan ($20-25 U S at concurrent off1c1al rates) 69 Wh1le th1s category excludes 1mportant 1ncome sources such as pr1vate plot and s1del1ne product1on and sorne 1n-k1nd payments from collect1ve work 1t 1s 1nd1cat1ve of the amount of cash ava1lable for farmers 69Nongyebu Renm1n Gongshe Guanl1JU [M1n1stry of Agr1culture Bureau of People s Commune Management] Y1J1uq1q1 zh1 Y1J1Uq1J1Un1an Quanguo Q1ongx1an Q1ngx1ng [The Cond1t1on of the Nat1on s Poor Coun b 197 - 197DJ X11hua !Jetao Lde-1 Ch1na Monthly] no 2 1981 pp 117-120 - 75 - from the1r pr1nc1pal assets 1n very poor local1t1es 70 The number of count1es fall1ng below th1s lowest benchmark 1ncreased to 11 1n Guangdong 1n 1978 but decl1ned to 7 1n 1979 (1n Guangx1 8 1n 1978 and 6 1n 1979) In Guangdong the very poorest reg1ons appear to be 1n the northeast such as Wuhua and Longchuan Count1es and on Ha1nan Island 1n the South 1nclud1ng the known cassava area of Basuo (Dongfang County) In Guangx1 such count1es seem to be clustered 1n the north and west for example Du an Yaozu Autonomous County Luocheng Donglan and Napo Count1es as well as Barna Yaozu Autonomous County where cassava 1s known to be w1dely cult1vated 71 But w1th the except1on of the exceed1ngly product1ve Pearl R1ver Delta no part of South Ch1na can be excluded as a reg1on where d1rect consumpt1on of cassava 1s not 1mportant for sorne segment of the poorer rural populat1on Areas were cassava 1s an 1mportant d1rect calor1e source need not be remate Even w1th1n the Ha1kou Mun1c1pal Area on Ha1nan Island 11 percent of cult1vated area 1n the Yong S1ng Townsh1p for example 1s planted w1th cassava two-th1rds of wh1ch 1s consumed d1rectly as a staple 72 Th1s 1s because only 4 70D1str1buted collect1ve 1ncome averaged around two-th1rds of the total 1nclud1ng pr1vate plot and s1del1ne 1ncome dur1ng those years accord1ng to a State Stat1st1cal Bureau (SSB) survey of 10 282 households (Zhongguo GUOJ1a TongJlJU Zhongguo TongJl N1anJ1an, 1981 pp 431) But th1s may have excluded 1n-k1nd d1str1but1on of product1on from collect1ve lands For a full d1scuss1on of Ch1nese d1str1but1on data and 1ts problems see E B Vermeer Income D1fferent1als 1n Rural Ch1na The Ch1na Quarterly vol 89 (March) 1982 pp 1-21 71Nongyebu Renm1n Gongshe GuanllJU Q1ngx1ng X1nhua Yuebao no 2 1981 Cassava 1n ~r na 1977 1979 Quanguo Q1ongx1an percent of the farmed area 1s su1table for r1ce cult1vat1on the rema1nder be1ng rocky h1lls1des upon wh1ch fru1t tree hort1culture 1s be1ng attempted Cassava plant1ng prov1des an econom1c hedge aga1nst heavy market dependence The Starch Market What l1ttle quant1tat1ve 1nformat1on 1s ava1lable on starch product1on 1n Guangdong and Guangx1 has been recorded 1n Table 14 H1stor1cally a s1gn1f1cant share of f1nanc1ng for capac1ty construct1on andan 1mportant share of sales del1ver1es have been assoc1ated w1th overseas Ch1nese espec1ally 1n nearby Hong Kong and Maca u In 1952 the Wuzhou Charcoal Industry started Guangx1 s f1rst starch factory (J1ul1an Crude Starch Factory later renamed the Wuzhou Mun1c1pal Starch Factory) w1th f1nanc1al ass1stance from the government and from overseas Ch1nese Its san31aopa1 [Tr1angle Brand] cassava starch was exported from Wuzhou 1n east central Guangx1 to Hong Kong Macau Southeast As1a Japan and the M1ddle East S1nce the m1d to late 1950s Be1ha1 1n the far south Barna Yaozu Autonomous County 1n the northwest X131ang Farm 1n the east Wum1ng Overseas Ch1nese Farm 1n central Guangx1 N1ngm1ng Overseas Ch1nese Farm 1n the southwest and other farm1ng areas set up f1xed scale factor1es 73 The des1gnat1on Overseas Ch1nese Farm 1s an 1nd1cat1on that overseas Ch1nese f1nanc1al resources are 1nvolved 1n the commune s development 73Guangx1 J1ngJl N1an31an B1an31bu Guangx1 J1n931 N1an]1an 1985 - p 192 - 77 - In Guangdong cassava starch product1on may have begun even earl1er but at least by the early 1970s 56 factor1es had been set up 1n the prov1nce and hongpa1 [Red Brand] cassava starch from the Dongguan Flour and Starch Factory on the Pearl R1ver Delta was sold w1dely 1n Southeast As1a and Eastern Europe 74 Dur1ng the 1950s 1960s and 1970s 1t seems that product1on econom1es and the pr1ce structure concertedly favored cassava as a raw mater1al for starch product1on s1nce desp1te the prov1nc1al self-sufflclency 1mperat1ves for the per1od Guangdong and Guangx1 exported starch not only to Hong Kong Macau and fore1gn countr1es but to other Ch1nese prov1nces as well W1th l1beral1zat1on of rural econom1c act1v1t1es s1nce the late 1970s small scale starch process1ng plants have been establ1shed espec1ally as townsh1p and v1llage enterpr1ses By 1983 the total number of starch factor1es 1n Guangx1 had 1ncreased sharply to 284 though w1th comb1ned f1xed assets of only 25 m1ll1on yuan 75 But e1ther product1on econom1es no longer so clearly favored the use of cassava as a raw mater1al or cassava product1on 1n other prov1nces was expand1ng to meet the1r demands for starch Th1s comb1nat1on of overdevelopment of product1on capac1ty and loss of part of the 1nterprov1nc1al market brought about a contract1on 1n the South Ch1nese starch 1ndustry 1n 1984 In Guangx1 the number of enterpr1ses decl1ned by 17 percent and product1on fell by 16 percent (Table 14) However part of th1s decl1ne may be dueto 1ntens1f1ed 74L1ang Guangshang (ed Mushu Za1pe1 yu L1yong p 9 75Guangx1 J1ngJ1 N1an]1an 1985 p 192 compet1t1on from nearby Zhaoq1ng and Shao~an Prefectures 1n Guangdong where starch product1on has been 1ncreas1ng rap1dly 76 A var1ety of 1ndustr1es use cassava starch 1n Ch1na the most trad1t1onal be1ng the cotton yarn 1ndustry wh1ch prov1ded demand for the f1rst Guangx1 factory 1n Wuzhou 77 But the Wuzhou and Be1ha1 factor1es have expanded and d1vers1f1ed to use cassava starch as a bas1s for glucose product1on In 1984 Guangx1 produced 7 800 tons of glucose pr1mar1ly for the candy 1ndustry 80 percent of th1s total was produced 1n the Wuzhou and Be1ha1 factor1es the latter export1ng to Hong Kong Tha1land and other countr1es The Wuzhou factory has also 1n1t1ated tr1al product1on of denatured starch and w1th purchase of techn1cally super1or equ1pment from Japan has 1ncreased 1ts extract1on rate by more than 5 percent 78 In Guangdong the Dongguan Factory has also d1vers1f1ed and now produces glucose brewer s yeast and w1ne 79 As early as 1972 1t exported cassava leaf starch to Japan and to England large quant1t1es of glucose part1ally based on m1llet as well as cassava 80 In Shaoluan and Zhaoq1ng Prefectures 1n add1t1on to 76oelphl survey response comments by Huang X1 agronom1st Inst1tute for Dryland Gra1n Crops Guangdong Prov1nce Academy of Agr1cultural Sc1ence Guangzhou June 28 1986 77sun J1ngzh1 Huanan J1ng]1 D1chu pp 258 and 333-334 78Guangxl J1ngJ1 N1an]1an 1985, p 192 79correspondence from Graham Johnson Department of Anthropology and Soc1ology Vancouver September 19 1983 Professor of Anthropology Un1vers1ty of Br1t1sh Columb1a 80L1ang Guangshang (ed ) Mushu Za1pe1 yu L1yong p 9 - 79 - cassava starch factor1es a number of other process1ng 1ndustr1es have been establ1shed wh1ch ut1l1ze cassava 1nclud1ng a monosod1um glutamate factory molasses plants brewer1es and feed-processlng plants 81 81oelphl survey response from Huang X1 June 28 1986 IV A Multi-Ma.rket INDONESIA 1 Ca.ssa.va. Economy In the 1960 s Indonesia. a.nd especia.lly Java. wa.s portra.yed as the epitome of the food crisis fa.cing Asia. The blea.k prospects for increa.sing a.gricultura.l production on a. very restricted fa.rm-size base were most eloquently a.rticula.ted in the a.gricultura.1 involution thesis of Clifford Geertz (1963) in which a. degra.ding resource base wa.s a.ccelera.ted by the increa.sing impoverishment of the a.gricultura.l popula.t~on The low point for the a.gricultura.l sector wa.s a.rgua.bly rea.ched in 196 7 when per cap Ha. rice a.va.~la.bility rea.ched its lowest level in the deca.de a. situa.tion compounded by a. tight interna.tiona.l rice ma.rket a.nd severe foreign excha.nge constra.ints However during the next deca.de rice production grew by 4 2% per a.nnum a.llowing per ca.pita. consumption levels to increa.se from 91 to 123 kg per yea.r In the 1978-84 period growth ~n rice production a.ccelera.ted even further to 6 7% per yea.r High y~elding rice va.r~eties investment in irrigation systems and subsid~zed fert~lizer prices resulted in dramat~c ~ncreases in rice yields the principal source of growth in production A revita.lization of r~ce production together with the sound ma.na.gement of sharp increases in oil revenues resulted in an annual GDP growth rate of 7 6% throughout the 1970's Indonesia had broken out of the low-income trap by focusing on domestic needs together w~th sound investment of export revenues Rice has been the centerpiece of agricultura.l policy in Indonesia in the post-war period Rice is the princ~pal source of farmer income the major food source the dominant expenditure itero in the consumera budget and therefore the major component in consumer price indices Any policies directed to farmer incomes rural employment nutritional objectives food security or control of inflation had to consider rice (Dorosh 1986) The policy thrusts in rice ~n the la.st two decades has had two princ~pal dimensiona First through the BIMAS program there has been a concerted effort to create a profitable environment for adoption of yield-~ncreasing rice technology A massive extension effort focused on the irrigated sector combined w~th subsidized fertilizer a.nd production cred~t have led to rapid adoption of improved technologies The second component has been management of domest~c r~ce prices through BULOG ( the national logistics agency) through support price operat~ons control over imports and development of a buffer stock scheme Both these policies impinge on secondary carbohydrate crops such as ca.ssava In the first insta.nce credit and extension systems are focused on the irr~gated sector with few resources available for upland crops In the second place rice prices have a la.rge influence on the demand for secondary staples such as cassava and maize The 1980 s nevertheless has witnessed sorne tendency toward a more comprehensive and thus diversified approach to food and agricultural policies as witnessed by the ~nvolvement of BULOG ~n the ma~ze and soybea.n sectors l./ This cha.pter draws heavily on the work of the Food Research Institute Stanford University Ma.ny parts of the chapter amount to summaries of the research found in Falcon et al The Cassava Economy of Java and it ~s hoped the citations are numerous enough to reflect this debt While agricultura! growth on a very limited farm-size base was achieved through a focus on raising rice yields on Java a complementary strategy was area expansion on the low populated outer islands This involved providing incentives for people to move off Java and gave rise to the transm~gration projects Indonesian economic plann~ng remains committed to transmigration to the outer islands and while the initial per family settlement costs appear high the autonomous secondary migration that is now apparent in some of the older projects on Sumatra appear to support this policy of developing the agricultura! frontier in Indonesia Ironically cassava has remained outside the purv~ew of agr~cultural policy in Indonesia and yet the crop has played a s~gnificant role in underpinning key policy obJectives (see Falcon et al p 165-69) This ~nvisibility to policy-makers ~s interesting for a crop that is grown throughout Indonesia that has played a key role in transmigration projects that histor~cally has been an important export crop and that is the second most important calorie source in the diet It is a mark of cassava s inherent productivity and versatility that it has flourished without government support However as policy focus shifts to upland crops particularly maize and where there is substitution between maize and cassava on both the supply and demand side then there ~s a need to bring cassava into the policy framework Markets and Demand Indonesia is the premier example of a well ~ntegrated cassava economy The multi-use characteristics of cassava are fully exploited Cassava ~s consumed as food both in a fresh and dry form it is exported and a significant portien is processed into starch (Table 4 1) Moreover a significant d~fference in utilization patterns exists between Java and the outer islands On Java utilization forms are fa~rly balanced between fresh roots for human consumption gaplek and starch On the outer islands on the other hand fresh root consumption is by far the largest consumpt~on form a not surprising fact given the lack of infrastructure and a principal focus on subsistence consumption Understand~ng how cassava production is allocated to these various markets each with relatively different growth potential will aid in developing a more effective planning frame for cassava in the Indones~an agricultura! sector Cassava for direct human consumption The food economy of Indonesia is based on rice While less preferred than rice cassava is the second most important carbohydrate source according to Susenas data (Table 4 2) although it still makes up no more than 10% of average calorie ~ntake The successful extension in irrigated areas of the high yielding rice varieties resulted in increasing per capita availabilities of the grain during the last decade and a half Trends in cassava consumption are more difficult to interpret The food balance estimates follow production trends and suggest a distinct increase ~n consumption s~nce 1973 on the other hand the Susenas estimates suggest more or less stable consumption over the decade (Table 4 3) What is clear TABLE 4 1 Indonesia Utilization Direct Food Consumption Fresh Roots Gap le k Gaplek Flour Starch Gap le k Exports Waste Total Ut1l1zation Source 4 1 Supply and Ut1lization of Cassava (on a Fresh Root Basis) on Java and the Outer Islands 1978 Java Off-Java Indonesia (000 t) (000 t) (000 t) 1 928 5 1 201 8 3 130 3 2 679 o 492 9 3 171 9 80 o 80 o 3 064 3 1 076 8 4 141 1 294 o 630 o 924 o 529 9 105 7 635 6 8 575 7 3 507 2 12 082 9 TABLE 4 A 1 Indonesia Utilization Java Direct Food Consumption Fresh Roots Gaplek Gaplek Flour Starch Gaplek Exports Waste Sub-total Production Off-Java Direct Food Consumption Fresh Roots Gaplek Starch Gaplek Exports Waste Sub-total Production Sources See text Supply and utilization Estimates for Cassava 1978 Primary Data Estima tes 20 3 kg/cap 9 4 kg/cap 446 180 t 98 lSO t 20 2 kg/cap 3 1 kg/cap 21S 3SO t 209 642 t Implied Fresh Root Use (000 t) 1 928 S 2 679 o 80 o 2 476 3 294 o S29 9 7 987 7 1 070 6 492 9 1 076 8 630 o 102 2 3 372 S AdJusted Fresh Root Use (000 t) 1 928 S 2 679 o 80 o 3 064 3 294 o S29 9 8 S7S 7 9 484 8 1 201 8 492 9 1 076 8 630 o lOS 7 3 S07 2 3 S07 2 IV - 3 - is that cassava continues to maintain a secondary but yet 1mportant role in the Indonesian food economy with this 1mportance lying more in distr1bution of cassava consumption rather than in aggregate averages Cassava is consumed principally in the form of fresh roots and gaplek with these two forms being prepared 1n a variety of forms 1n the home There is a marked regional variation in consumption patterns of both fresh roots and gaplek Although per capita consumption levels for cassava are the same for Java as the outer islands fresh consumption is much more important off-Java probably due to the less seasonal nature of root production and the greater difficulty in drying Gaplek consumpt1on is concentrated in the eastern part of Java where so1l and ra1nfall are more marginal (Figure 4 t) while fresh consumption on Java is relat1vely more evenly distributed - The locus of cassava consumpt1on is very much in the rural sector due not only to the bulk of the population residing in rural areas but also to the much higher per capita consumpt1on of cassava 1n these areas There is a signif1cant change in consumption of non-preferred staples between rural and urban areas (Table 4 2) Gaplek and maize are rarely consumed 1n an urban setting and yet are qu1te important in rural areas Fresh cassava consumption while higher 1n rural areas nevertheless is still at s1gnificant levels 1n urban areas even given the problems of market1ng such a perishable commodity Unnevehr (1982) estimates that in rural areas about two-thirds of fresh cassava and one-half of gaplek are subsistence consumption Count1ng urban consumption only 37% of fresh cassava that is utilized for human consumption is marketed Probably the most important component influencing the distribution of cassava consumption is income Gaplek consumption shows a consistently declin1ng trend with income (F1gure 4 2) Gaplek is a non-preferred food principally consumed by the poor Fresh cassava consumption at least in rural areas increases markedly with increasing 1ncome at low levels of income levels off at med1um income levels and declines slightly at high income levels The overall tendency is for total cassava consumption (excluding starch) to decline with income Approximately 40?' of the population in Indonesia consumes less than 1900 calories per day (Table 4 4) Th1s group 1s obviously constrained by income in the amount of food which they can purchase and thus must make more use of cheap calorie sources The poorer 1ncome groups pr1ncipally in the rural areas substitute cassava and maize for the more expensive but more highly preferred rice (Figure 4 2) Cheap cassava allows the lower income segmenta of the population to ach1eve a h1gher calorie intake w1th their lim1ted food budget than they would have been able to achieve w1th just rice Cassava is thus a potentially key commodity in pol1c1es focusing on nutr1tion and the related issue of rice import management 'l:) The importance of cassava in the d1et and the relat1vely ub1quitous distribution of fresh root consumption implies that quality characteristics cannot be sacrificed in a varietal development program TABLE 4 2 Indonesia Annual Per Capita Rural and Urban Consumption of Starchy Staples 1976 and 1978 1976 1978 Commodity Total Rural Urban Total Rural Urban (kg) (kg) (kg) (kg) (kg) (kg) Indonesia Rice 111 2 110 S 114 3 109 2 109 2 109 2 Corn 9 9 11 9 o 7 11 4 14 o 1 o Cassava fresh 26 2 29 9 9 S 20 2 22 9 8 8 Cassava gap le k 6 4 7 9 o 2 7 3 8 8 o o Java Rice 103 3 102 4 107 3 99 8 98 8 104 o Corn 11 S 14 o O S 1S 1 17 7 1 o Cassava fresh 21 6 24 9 6 7 20 3 22 9 7 8 Cassava gap le k 8 o 9 7 o 1 9 4 11 4 o (} Off Java Rice 124 8 124 4 126 6 130 o 130 o 119 6 Corn 7 o 8 3 1 1 S 7 6 8 1 6 Cassava fresh 34 2 36 S 14 4 20 2 22 4 10 4 Cassava gap le k 3 8 4 6 o 3 3 1 3 6 o o Source Dixon John A Food Consumption Patterns and Related Demand Parameters in Indonesia A Review of Available Evidence 1982 TABLE 4 3 Commodity Rice Maize a/ Cassava - Sweet pota toes Indonesia Comparison of Food Balance Sheet and Susenas Estimates of Annaul Per Capita Consumption 1969/70 1976 1978 FBS Susenas IV FBS Susenas V FBS Susenas (kg) (kg) (kg) (kg) (kg) (kgl 107 3 103 2 116 2 111 2 123 4 109 2 19 1 22 o 18 3 9 9 27 2 11 4 53 9 41 1 76 o 42 2 74 o 38 5 17 4 8 8 16 o 10 8 13 4 5 7 VI a Cassava is expressed in fresh root equivalent to fresh root equivalent using a 1 2 5 ration dried forms are converted Source Dixon John A "Food Consumption Patterns and Related Demand Parameters in Indonesia A Review of Available Evidence 1982 TABLE 4 4 Indonesia Total Calorie Intake by Income Strata Estimated from the Susenas V Survey 1976 Monthly Expenditures Per Capita Less than Rp 2 000 Rp 2 000- 2 999 Rp 3 000- 3 999 Rp 4 000- 4 999 Rp 5 000- 5 999 Rp 6 000- 7 999 Rp 8 000- 9 999 Rp 10 000-14 999 More than Rp 15 000 Share of Total Population (%) 15 3 23 8 19 5 13 6 8 8 9 4 4 2 3 8 1 6 Calories Per Capita Per Day (Kilocalories) 1 381 1 870 2 034 2 084 2 288 2 533 2 794 3 066 3 284 Average 2 064 Source Dixon John A Food Consumption Patterns and Related Demand Parameters in Indonesia A Review of Available Evidence 1982 IV - 4 - The role of cassava w1thin an overall nutrit1on policy follows from an analysis of demand parameters Estimates of income elast1cities by Dixon (1982) show that among the poorer income strata there is a sign1ficant increase 1n cassava consumption both as fresh and gaplek with increases in income (Table 4 S) Such changes in cassava consumpt1on could come from real increases in income or from changes 1n the r1ce price since expenditure on rice makes up such a large part of the consumer budget Substantial substitution between caloric staples would be expected depending on relative prices and in fact elasticity estimates suggest substantial responsiveness to price changes Timmer (1980) reports a cross price elasticity of fresh cassava with rice ~7 O 77 showing a very marked effect of rice prices on cassava consumption - Cassava s role 1n the Indonesian food economy wh1le not central is nevertheless critical to the support of that proport1on of the populat1on facing a risk of not meet1ng their calor1c needs from rice suppl1es This populat1on 1s essentially defined by low incomes and in years of poor rice harvest their nutr1tional status can be put further at risk by rising rice prices The government' s policy has been to try to maintain stable rice prices and this task is vested in the government grain market1ng agency BULOG which attempts to stabilize r1ce pr1ces through rice imports and to a more limited extent through wheat imports BULOG was a1ded in this effort in the last decade and a half by the widespread adoption in the irrigated areas of the high-yield1ng r1ce var1eties Nevertheless rice imports have almost consistently exceeded one million tons up to 1980 and have occasionally reached two mill1on tons At these levels Indonesia can account for as much as a th1rd of the world export market having a pronounced affect on world r1ce prices and therefore the fore1gn exchange costs necessary to meet import requirements Since 1980 imports have been around half a mill1on tons although levels rose to 1 2 m1llion tons in 1983 As the benefits of the new rice technolog1es start almost certa1nly to plateau Indones1a will again be faced with high 1mport requirements in a world rice market that is very thin To resolve th1s dilemma Indonesia has 1ncreas1ngly turned to wheat imports which are cheaper and where Indonesia forms a minar percentage of the world market However Indonesia has on the whole failed to consider the potent1al role of the secondary staples cassava and maize Total consumption of both of these commodities has essent1ally been static over the past decade and a half implying a declining contribut1on to total caloric consumption since rice consumption has risen dramatically Since there are real supply-side constra1nts on meeting future nutr1t1onal objectives with r1ce s1nce the locus of wheat consumpt1on is pr1ncipally 1n urban areas and s1nce cassava and maize are already important staples for the rural poor a strategy to increase production of these crops at lower pr1ces (that is techn1cal change) would contribute directly to increased calorie }_/ Dixon (1982) on the other hand could find no s1gnificant cross price elast1c1t1es but based his est1mat1on only on Java whereas Timmer s was based on Indonesia as a whole TABLE 4 5 Indonesia Price and Expenditure Elasticities for Rice and Cassava by Income Strata on Java 1976 Commodity Expenditure Elasticity Rice Urban Rural Fresh Cassava Urban Rural Gap le k Urban Rural Price Elasticity Rice Urban Rural Fresh Cassava Urban Rural Gap le k Urban Rural Low o 329 o 831 o 094 o 849 n e o 833 -0 31 -1 28 1 27 -1 09 n e -2 49 Note n e means not estimated Expenditure Group Medium o 107 o 485 -0 275 o 117 n e -1 018 -0 56 -0 45 o 14 -0 82 n e -2 06 High -0 121 o 133 -0 654 -0 627 n e -2 90 n e o 18 n e -0 67 n e -2 18 Average o 194 o 560 -0 131 o 276 n e -0 616 -0 48 -o 84 o 44 -0 81 n e -1 86 Source Dixon John A Food Consumption Patterns and Related Demand Parameters in Indonesia A Review of Available Evidence 1982 IV - S - consumption of the most vulnerable pupulation By integrat1ng cassava 1nto overall food policy BULOG would have considerable more flexib1l1ty in managing r1ce imports and prices However because of the overall inelastic1ty in food demand for cassava this flexibil1ty is dependent on some diversification in end markets That is diversifying end uses as the production base expands not only provides a certa1n market stability for farmers but as well ensures alternat1ve food supplies when rice is in short supply The starch market Starch is the largest single market (on a root equivalent basis) for cassava in Indonesia A cassava starch industry has existed on Java since the turn of the century Pr1or to World War II and independence this industry was based principally on plantations and was geared princ1pally to export The recovery from the damage 1ncurred during the war precipitated a shift from foreign to domestic ownership and from export to domestic markets Indonesia is currently the largest producer of cassava starch in the world and essentially all the production is destined to domestic markets Unlike othe¡1countr1es in Asia there is virtually no production of starch from maize - The structure of the cassava starch industry is characterized by great diversity Starch factories are spread throughout Java and Sumatra but with a particular concentration in West Java Location of the starch industry is primarily dependent on access to a ready water supply to a sufficient concentration of root production to adequate transport infrastructure and to non-seasonality of root supply These factors have until recently given the edge to West Java as the center of starch production However as transport infrastructure has improved on Sumatra part1cularly in Lampung starch production has expanded rapidly This has been enhanced by the less seasonal supply of roots on Lampung From v1rtually no production in the early 1960's the starch industry on Lampung has expanded rapidly especially in the 1970 s to become the second largest starch-producing province after West Java Diversity is also a character1stic of the scale of processing Rudimentary household processing techniques co-exist with large-scale capital intensive factories w1th a significant range of plant sizes between these two extremes Nelson (1984) has recently analyzed the economics of starch production in Indonesia At 1980 prices all processing modes were found to be profitable (Table 4 6) The large mills were found to be most profitable but only because the tax incidence was much less than on household product1on and medium-scale factories To mot1vate investment the government has instituted tax holidays for three to six years for large-scale firms This together with a subs1dy on d1esel fuel and exemption from duty for imports of processing equipment g1ve a distinct advantage ~o insuring the profitability of the large scale plant However from a soc1al point of view Nelson finds that the household A single starch/corn o1l plant It princ1pally relies on maize in operation in 1984 Indocorn is operat1ng in Indones1a imports for its operat1on and was not TABl.E 4 6 Indonesia Cost Item Variable Costs Cassava Roots Labor Fuel Working Capital Taxes Miscellaneous Sub-Total Fixed Costs Depreciation Capital Costs Administration Sub-Total Total Costs Revenue Starch Processing Costs per Ton by Scale of Processing Unit 1980 Household (Rp/t) 123 737 21 357 663 5 405 9 520 3 661 164 343 2 950 3 790 6 740 171 083 178 940 Processing Technique Medium-Scale (Rp/t) 123 737 6 757 3 049 2 858 12 627 3 156 152 184 8 444 13 290 4 330 26 064 178 248 178 940 l.arge-Scale (Rp/t) 110 882 2 234 7 386 6 292 2 108 15 045 143 947 9 218 19 134 2 495 30 847 174 794 184 395 Source Nelson Gerald Implications of Developed Country Polic~es for Developing Countries The Case of Cassava 1982 IV - 6 - production generates both the highest level of social profit as well as the most employment Nelson further reports that household starch production has expanded rapidly in the 1970's motivated by 1ncreased capacity utilization w1th the introduction of mechanical graters The few figures on starch suggest that production has 1ncreased rapidly through the 1970's (Table 4 7) This growth was characterized by significant increases in household production on Java and very rapid growth of large-scale processing on Lampung The starch market was both large and growing providing quite strong demand for cassava roots Root production at least on Lampung responded accordingly The factors that were driving this increased demand for cassava starch are less well documented Concensus seems to exist that the largest end use for starch is as krupuk a cr1spy wafer consumed as a snack food Nelson reports that this industry takes as much as 65% of total starch production -- this implies an annual per capita consumption figure of 2 9 kg -- while the rest goes into other food processing industries (157) the textile industry (10%) and glucose production (3%) The only complementary data comes from the SUSENAS consumer budget surveys The 1976 survey reports an average annual per capita consumption level of starch of 1 4 kg on rural Java and O 1 kg in urban areas of Java (Dixon 1984) However Dixon considers this to be a significant underestimate because it does not include direct purchases of krupuk or other bakery products using starch He suggests that a more reasonable per cap1ta estimate for Java's is 2 4 kg for rural areas and 1 O kg for urban areas i e an average of 2 1 kg These estimates however appear to discount the data from the 1978 survey for krupuk consumption which suggests per capita consumption levels of krupuk alone of 2 S kg in rural Java and 6 6 kg 1n urban Java Per capita starch consumption may be as high as S kg per cap1ta (see Appendix 4 1) which means that cassava starch 1s a more important food item than 1s often considered Starch is the dominant end market for cassava in Indonesia moreover the limited evidence on demand suggests that th1s market will continue to grow for a signif1cant period into the future Most of this growth comes from the use of starch as a food source with consumpt1on in this case being skewed toward the h1gher income strata D1xon (1984) estimates income elastic1t1es for krupuk of 1 56 in rural areas and 1 35 in urban areas S1gnificantly consumption patterns for cassava starch skewed as they are toward the rich are the mirror image of those for gaplek which are highly skewed toward the poor Product differentiation and market segmentation allows cassava in this case to serve two very distinct roles as a basic secondary staple for the poor and as something of a luxury food for h1gher income groups A feature of the cassava starch industry 1n Indonesia compared to that of some other countries 10 Asia is that there is no effective competit1on from maize starch even though maize 1s a major crop in Indonesia The situation is further confounded by the fact that ma1ze is at least 1ntermittently exported at world pr1ces while gaplek wh1le also exported competes at the higher pr1ce levels set in the European Community Maize should thus be more compet1t1ve as a raw material source TABLE 4 7 Indonesia Provine e West Java Central Java East Java Total Java Lampung North Sumatra Riau Other Provinces Total Indonesia Source Falcon et al 1984 Estimated Production of Starch 1974 and 1979 Production 1974 1979 (mt) (mt) 188 220 239 220 126 020 149 180 33 300 57 780 347 540 446 180 27 750 150 750 15 900 24 lOO 30 900 30 900 9 600 9 600 431 690 661 530 The Cassava Economy of Java IV - 7 - for starch productJ.on than cassava However in the particular case of IndonesJ.a starch substJ.tution J.s limited by quality factors and in particular course sun-dried starch is necessary in preparJ.ng krupuk the domJ.nant market The fine flashdried starch cannot be used in krupuk unless mJ.xed with the coarser starch Thus maize starch is constrained to competing in the much smaller industrial market with cassava starch produced J.n the larger factories and g1ven the scale econom1es in wet milling maize could not establish a large enough market to JUStify a factory Nevertheless the competition between maize and cassava becomes a factor in the recent interest in the production of hJ.gh fructose sweetners (HFS) Indonesia has over the past decade consistently increased its imports of sugar to the point that imports now total between 500 to 700 thousand tons a year Not only are imports increasing but Indonesia maintains high internal sugar prices to support producers on the one hand and to limit consumption on the other hand A policy directed at self-sufficiency in sugar is limited by the availabil1ty of land su1table for sugar cane and the competJ.tion between rice and cane for this land Therefore producing high fructose sweetners from either maize or cassava in upland areas holds some attraction However the substitution of liquid high fructose sweetners for sugar occurs over only a limited range of end uses of sugar The largest market direct human consumption has lim1ted possibilities for substitution at this stage of market development Development of the HFS market depends on exploit1ng industrial uses especially food processing and bottled beverages Estimates on the size of this market are based on scanty data two sources put the potent1al consumption at between 220 and 500 thousand tons per year (Argento and Wardrip 1983 Tate and Lyle 1981) Nevertheless this market is expected to grow at a estimated rate of 54 through the rest of the century (Pearson 1984) Indonesia has already committed itself to producing h1gh fructose sweetners A cassava-based factory is already in operation 1n Malang on Java L1censes for the construct1on of 4 more factories have been issued to bring total production capacity to 110 thousand tons of HFS Nevertheless two basic factors will largely determine the future of th1s industry First the econom1c viability of high fructose sweetner production will necessar1ly rest on the maintenance of the high domestic price level for sugar DomestJ.c wholesale prices for sugar in 1984 were Rp 575 per kg (US$0 57) compared to a world market price of US$0 26 per kg (Pearson 1984) Second licens1ng procedures and subsidies on capital in- vestments will be critical in determining whether sweetner production is based on cassava or maize This is because maize plants are based on very large capital investments whereas this is not necessary for cassava The econom1c advantage of one crop over the other is difficult to project WJ.th any degree of certaJ.nty but the most complete cost analysis to date is that of Pearson (1984) Pearson concluded that ma1ze would be a lower cost alternativa than cassava in HFS production due to three principal tenets First there are significant economies of scale in the maize wet m1lling process while 1n cassava these are m1nimal Second the price d1.stortions in the world market for cassava relative to maize are assumed to pers1.st and will in turn influence domestic profitabill.ty Third the domestic marketing system and/or BULOG are able to assemble the supplies necessary to maintain a large-scale maize plant in operation BULOG s control over imports may provide the supply stability necessary for continuity of operation Nevertheless planning of the HFS industry has been based on cassava for several practica! reasons First HFS production based on cassava is profitable under present domestic sugar prices as set by BULOG Second expansion of cassava production does not depend on yield increases as is the case for maize but can be based on further area expansion in the off islands especially those with good infrastructure as 1.n south Sumatra A supply response is much more assured in the cassava case Third capital requirements for HFS production are significantly less in the cassava case as a HFS production line can be added to existing cassava starch factor1.es as was done in the Malang case Conversely the smaller scale maize well milling plant was not profitable at existing sugar prices (Pearson 1984) A focus on small-scale cassava plants allows a more evolutionary and less risky approach to market development since production can initially be based on relatively small scale plants that have alternative product lines and not on majar capital investments in large-scale maize wet milling plants The key factor in the choice between maize and cassava is the relative price of the raw material Pearson bases his analysis on relative pr1.ces 1.n the world market that is a relative price of dried cassava to maize of 92 However as portrayed in Figure 4 3 only very rarely during the 1970's and 1980 s has relative prices of the two crops been that high Cassava usually trades at a significantly larger discount to maize in Indonesia and is often at the break-even price ratio of 64 calculated by Pearson for cassava to compete with large scale maize wet-milling plants The reasons for this larger price discount are (1) maize prices are often not in line with world market prices (Dorosh 1986) and (2) world cassava prices have often been below the US$110/t figure used in the analysis Because of the EEC import quota the prospect is for f o b cassava prices to be below this level in the medium term future (see Chapter VIII) Basing HFS production on cassava allows s1.gnificantly more flexibility 1.n market development than does maize The prof1.tability of cassava-based HFS does not depend on the economies of scale necessary for ma1.ze-based HFS to be profitable This allows greater flexibility in investments in capac1.ty and in plant location For cassava-based HFS factories can be located in cassava production areas and based on starch slurries from the direct root processing or alternatively can be located next to maJar market areas and use processed starch as a raw material Relative transport costs and control over raw material costs will determine the choice Maize wet mill1.ng plants on the other hand will probably be located near to consumption points that is Jakarta and will depend on steady supplies of maize from majar storage facilities or imports A single large-scale wet milling plant operating for 300 days per year requires about 275 thousand tons of maize per annum This greatly exceeds el.ther annual export or import volumes over the past two decades and is far above total annual IV - 9 - maize sales by BULOG Moreover maize-based HFS will be competLng with the animal feed industry for raw material supplLes most of which LS currently supplied to the concentrate industry from BULOG stocks which are often imports (Table 4 8) Cassava 1 s potentLal role in this industry will thus be based on BULOG s sugar price policy and on the future abLlity of the Indonesian maize economy to generate and assemble significant surpluses of this commodity (see Dorosh et al for such an assessment) In summary the cassava starch market remains very dynamic and represents the largest end use for cassava in IndonesLa (Table 4 1) With the high income elasticity for krupuk the potential in the high fructose sweetner market and any increases in the textile paper or plywood industrLes the demand for starch wLll continue to Lncrease There is some indicatLon that demand is outstripping supply since Ln both 1982 and 1983 IndonesLa had to import over 50 thousand tons of starch each year (Table 4 9) These are very s1gnificant volumes which were primarily caused by below trend production levels 1n those two years but are nonetheless indicat1ve of the relatLve size and importance of the starch market in Indones1a Gaplek in Feed Markets Gaplek forros an integral part of cassava production and market systems 1n Indonesia When properly dried gaplek is a stable commodity and provides the farmer the option of harvesting and storing his cassava especially when there is a time premium on harvest1ng the cassava to plant the next crop Moreover gaplek since it can be stored and transported provides a means of integrating cassava markets Finally gaplek has multiple uses it can be used directly for human consumption can be ground into flour for noodle product1on or can be a raw material source for feed concentrate production or even for manufacture of low quality starch and its derivatives such as glucose or fructose sweetners Gaplek 1s currently used principally for human food especially by the lower income consumers in rural areas Indonesia is also a consistent although h1ghly variable exporter of gaplek to the European Community This export market serves the very important function of sett1ng a pr1ce floor under domestic prices for gaplek and in turn cassava 1n general (Unnevehr 1982) The export market 1s effective 1n setting this pr1ce floor even though this market rarely accounts for more than 10h of cassava production Only twice since 1970 have gaplek exports exceeded 400 thousand tons (Table 9) and export levels more generally oscillate between 150 and 350 thousand tons Internal gaplek prices have 1n general followed the general ris1ng trend in world prices (Figure 4 4) w1th exports being partLcularly responsive to the devaluation of the rupiah in 1978 A simLlar devaluat1on in 1983 did not produce such a response due to a tight domest1c market This apparent tightening of domest1c markets is espec1ally evident in Lampung where the gaplek export market was the engine of growth for the cassava 1ndustry in the fust half of the 1970 1 s Gaplek exports from Lampung stagnated after 1975 and have declined markedly sLnce 1981 The gaplek industry has had d1.ff1.culty competing with the expanding starch TABLE 4 8 Indonesia Maize Sales by BULOG to Feedmills Origen Domes tic Average Sales Year Total Sales Imports Procurement Price (tons) (%) (?') (Rp/Kg) 1977-78 17 299 72 28 50 1978-79 44 455 73 27 120 1979-80 36 835 21 79 90 1980-81 72 308 15 85 105 1981-82 147 162 lOO 110 1982-83 224 653 97 3 135 1983-84 46 110 9 91 130 Source Mink Stephen Corn in the Indonesian Livestock Economy 1984 TABLE 4 9 Year 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 Indonesia International Trade in Cassava Starch and Gaplek 1970-84 Cassava Starch Gaplek Exports a Exports Imports Total Java Lampung (000 t) (000 t) (000 t) (000 t) (000 t) 1 3 337 9 264 7 70 S 1 3 4S8 3 36S 7 86 9 1 1 343 S 240 7 100 8 1 3 16 1 7S 3 42 2 32 9 7 S 394 9 190 o 198 3 o 1 303 3 89 1 206 7 S 8 148 6 9 8 138 2 183 2 37 S 142 o o 1 o 6 307 8 98 2 193 9 1 o o 2 709 6 495 3 191 7 2 4 14 2 386 1 219 8 160 6 3 o 1 o 372 6 159 6 194 2 53 9 211 3 143 o 54 8 1 6 63 9 256 9 179 7 72 4 S 2 o 3 38S 2 n a n a a Includes gaplek meal Source Central Bureau of Statistics Exports Imports van.ous years industry on Lumpung even when world prices were recently relat1vely high Th1s declining trend was exacerbated by the poor crop years in 1982 and 1983 The tightening of export supplies of gaplek have made the voluntary quotas formalized with the EC in 1982 rather superfluous The quota was set at 500 thousand tons in 1982 rising to 825 thousand tons by 1986 when the agreement ended Compared to the Thai quota which decl1ned over the period the Indonesian agreement was very much largesse but in principal only There is very l1ttle potential for meet1ng the quota volumes even with the 1983 devaluation The advantages of the latter were negated by a bad crop year and the 1984 fall in the world price brought on by the effect of the quota on the Thai cassava industry Netherless the current level of the gaplek export market undervalues its importance An export price floor set in the EC not only earns Indonesia a significant economic rent but also serves to maintain price incentives should future production growth increase New cassava production technology or further transport infrastructure development on Sumatra could bring about such growth and the export market could serve to buffer farmer prices were production growth significant The short term problem with current strong domestic markets for cassava is to maintain sufficient pelleting and export capacity to insure the world price linkage The med1um term problem is to insure that a sufficiently large quota in the EC market is maintained to allow the cassava industry to expand without significant price instabil1ty Certainly should there be any renegotiation of the quota agreement the negotiations should balance the short-term constraints on exportable surpluses with the longer term gains from maintenance of export flexibility The maintenance of this world price export floor for gaplek however would be expected to inhibit the development of gaplek as a carbohydrate source in domestic mixed feed production If gaplek prices are set in the EC and maize prices are linked to the world coarse grain market gaplek prices would be expected to be out of line with maize in domest1c feed rations (see for example World Bank 1984) This argument however holds less often than not If a competitive ratio of relative prices of maize and gaplek is taken as 70 then gaplek should have been very competitive with maize through much of the 1970 s and 1980 s (Figure 4 3) As explained above the principal reason why price relatives have favored cassava is that domestic maize prices are not well l1nked to the international market and are often above implicit export prices (Dorosh 1986) Least-cost feed formulation models demonstrate that gaplek was competitive in poultry rations at 1984 prices (Table 4 10) However what is suprizing is that gaplek does not displace more maize at this price ratio of 52 This is due to the high internal price for soybean meal (Nelson 1986) Since 1982 BULOG has been the so le importer of soybean meal and s1nce Indonesia has no soybean crushing facil1t1es most soybean meal is imported Moreover soybean meal prices have been kept high to motivate a shift to domestic protein sources such as copra meal However in 1983 when BULOG cut soybean meal imports in half to save foreign exchange feed mills imported rapeseed and sunflower seed meals which were not under BULOG control Two additional factors militate against gaplek TABLE 4 10 Feed Component Ma1ze Cassava Chip Soybean Me al Fish Meal Kapok Meal Indones1a Least Cost Feed Ration for Poultry at 1984 Prices Feed Price Compos1tion (Rp/Kg) w 134 45 7 70 9 6 335 21 4 575 7 5 89 14 2 Source CIAT use in balanced feed rations First there is a preference for ma~ze because of its carotene content which gives the eggs and poultry meat a yellower color Second BULOG can be relied on for maize supplies when these are not available on the local market especially since the maJor mills are located near to major urban areas especially Jakarta Since most gaplek surpluses on Java are generated in the eastern part of the island and since internal transport costs are relatively high marketing channels to the feed industry have not developed The balanced feed/commercial livestock sector is not as well developed as similar industries ~n such countries as Thailand or the Philippines This is principally due to a relatively late start as the first feed factories were only established in 1972 However the other structural features of this industry are very similar Growth in mixed feed production has been spectacular rising from essentially no industry ~n 1972 to an estimated 400 thousand tons in 1982 (Alfred e Toepfer Company private communication) About 85 to 90% of production is poultry rat~ons and the commercial poultry industry has grown in close association with the feed sector (Table 4 11) This growth in the poultry/mixed feed industry has been motivated by increasing demand for meat and eggs precipitated by rising per capita incomes during the 1970 s In sum a viable poultry/mixed feed industry has been established in Indonesia with prospects for very s~gnificant future growth as is reflected in the high income elastic~t~es for animal products (Table 4 12) A factor that may be a constraint on growth in the poultry ~ndustry and by implication for the mixed feed industry is the pres~dential decree limiting the size of layer units to 5000 birds and of broiler operat~ons to 750 head per week The objective of the decree is the maintenance of a labor intensive poultry industry and a more equitable distribution of income opportunities The principal effect will be on costs of eggs and poultry meat since larger producers are usually able to ach~eve higher feed conversion rates and fewer losses -- although with effective extension programa and access to inputs there is no necessary reason why this should continue Mink (1984) estimates the result of such a shift to small producers will be an annual reduction of 35 000 tons in demand for carbohydrate sources The potential role of cassava in the balanced feed market thus dependa on a number of interrelated factors First the continuing growth in the starch market and maintenance of direct food consumption will limit potential surpluses and bid cassava away from the feed market unless there is a significant increase in product~on Second the Indonesian feed industry requires some experience in the appropriate handling of cassava in mixed feed rations and in developing gaplek marketing channels to Jakarta A similar lag existed in using cassava in the Thai feed industry but th~s inertia has now been overcome Third any major increase in cassava feed use will require more certainty ~n supply of soybean meal and some rationalization of protein prices Finally as has happened in Thailand maize will form the principal carbohydrate source in feed rat~ons but cassava can come in and out of the diet depending on relative prices Currently between 450 (World Bank 1984) and 700 thousand (Mink 1984) tons of maize are used as animal feed in Indonesia representing about 15% of TABLE 4 11 Indonesia Growth in Poultry Population and Industr1al Feed Production 1970-82 Poultry Population Poultry Feedstuffs Year 1979 1971 1972 1973 1974 197S 1976 1977 1978 1979 1980 1981 1982 a Village Chickens (000) 66 30S 71 S7S 88 700 97 4S7 100 721 112 S93 123 S20 122 798 126 741 127 918 134 693 14S 678 143 2S8 Commercial Layers (000) 474 1 291 1 68S 2 234 3 499 3 69S S 18S 7 001 11 S99 1S 412 21 6S8 27 837 41 6SS a Commercial Broilers (000) 4 030 8 032 Combined figure for commercial layers and broilers Layer Broiler (OOOt) (000 t) na n a n a n a na n a n a n a na n a n a n a n a n a 141 6 86 4 168 2 102 6 203 S 124 2 241 9 147 6 n a n a n a n a Source Poultry population is from Mink Stephen Corn in the Indonesia Livestock Economy 1984 and Feed Production is from Hertropf Joachim The Feed Industry in Overseas Countries 198S TABLE 4 12 Indonesia Product Eggs Chicken Meat Por k a Directora te General Source Mink Stephen Income Elasticities for Animal Products Data Source SU SENAS 1 6 2 2 1 4 of tivestock Corn in the Indonesia tivestock Economy' 1984 DGLS a 1 2 1 3 1 o TABLE 4 l3 Indonesia Gaplek Marketing Margins from Farm to Pelleting Factory 1980 Assembly Agent Farmer price Moisture loss Transportatl.on Profit Wholesaler Assembler sale prl.ce Transportation and loading Moisture loss Profit Purchase Agent Wholesaler sale price Management fee Profl.t Factory-Gate Price Total Margin (7 of Factory Price) Trenggalek (Rp/Kg) 34 o 4 5 5 o 1 5 45 o 6 o 3 o 1 o 55 o 21 o (38 2-b) Java Gunung Kidul (Rp/Kg) 45 o 2 o 1 o 48 o 5 5 1 5 1 o 56 o 11 o (19 6,() Kediri (Rp/Kg) 45 o 2 o 1 5 1 5 50 o 3 7 o 3 1 o 55 o 10 o (18 2%) Sumatra Lampung (Rp/Kg) 22 o 1 8 3 o 3 2 30 o 4 8 2 7 7 5 45 o 1 o 3 o 49 o 27 o (55 1i) Source Java l.S from Falcon et al Lampung is from World Bank for MaJor Food Crops 1983 The Cassava Economy of Java 1984 Indonesia Policy Options and Strategies TABLE 4 14 Indonesia Fresh Root Marketing Margins from Farm to Starch Mill 1980 Assembly Agent Farmer price Harvesting Porterage Transportation and Loading Moisture loss Profit Factory-Gate Price Total Margin (% of Factory Price) Garut (Rp/Kg) 20 o 1 o 3 o 4 2 o 4 1 4 30 o 10 o (33 3%) Java Kediti (Rp/Kg) 18 o 3 2 o 7 1 1 23 o S O (21 7"/) Sumatra Lampung (Rp/Kg) 9 9 6 7 2 2 1 2 20 o lO l (SO S%) Source Java is from Falcon et al The Cassava Economy of Java 1984 Lampung is from World Bank Indonesia Policy Options and Strateg~es for Major Food Crops 1983 IV - 12 - the total ma~ze crop As depicted in Figure 4 3 cassava is periodically competitive with maize ~n balanced feeds The feed ration industry ~s perfectly adaptable to such short-term response to changes in price and availabilities As the domestic feed industry expands it will be arguable whether the feed industry or export market provides the most benef~cial price floor for cassava Pricing and Market Eff~c~ency The Indonesian cassava economy representa in many ways the ideal development of the crop that is cassava is deployed w~thin diverse and complex cropp~ng systems across a range of agroclimatic conditions and is fully utilized in a broad spectrum of end uses Such full explo~tation of the production and utilization potential of the cassava crop relies fundamentally on well functioning markets and ~n particular on integrated markets in which pr~ces serve to allocate cassava between the range of end uses That is farmers are receiving a price for their cassava roots that reflects its best end use in the country Such a situation requires that cassava prices be linked spatially across the country and l~nked vertically across different forms The development of such linkages for a highly perishable bulky commodity is difficult and is dependent on the existence of either a highly developed transport refrigerated storage and marketing system (eg vegetables in the U S ) or processing of the roots to a stable storable commodity s~nce the f~rst does not exist in Indonesia the role of gaplek can be singled out as crucial to well integrated cassava markets in the country Unnevehr (1984a) (1984b) has analyzed market integration and price transmission on Java and what follows is drawn directly from that research The key to her analysis ~s the concept that cassava prices within Java are set by domestic supplies of staple foodstuffs and demand for cassava products subject to a lower bound set by export parity the local demand curve for cassava has two portions -- a downward sloping domest~c curve and a perfectly elastic export floor (Unnevehr 1984a) A demand curve was estimated to test for this kink When East Java pr~ces were at export parity the correlation with world market prices was O 95 Gaplek prices at the East Java port Surabaya in the 1971-79 period were at export parity 79% of the time Th~s demonstrates the effective operation of the price floor and the fact that the export market was a principal determinant of domestic prices throughout this period This is seen in Figure 4 4 charting Thai and Indonesian gaplek prices Effective price transmission and adequately linked markets ~mphes relat~vely competit~ve price formation throughout the country This however does not ~mply that all farmers face the same price s~nce transport and marketing costs will differ depending on location relative to markets and the level of development of transport infrastruture In fact marketing and transport costs make up a very sign~ficant portian of the wholesale or retail price for both fresh roots and gaplek Assembly costs of fresh roots for starch plants and gaplek for pellet~ng plants are relatively high compared to the eventual farm leve! price (Tables 4 13 and 4 14) On Lampung assembly costs alone consume half of the factory price paid for roots and 557 of the pr~ce paid for gaplek This s~gnif~cantly F•gure !.i 3 lndonesaa. · ~a t1o of Gaoh:-< :rr to Maaze Prace an Sura.ba:!:§a 1 1 j 1 ~ 1 J 1 1 1 ,r; ;. 1 1 11 n ~ 1 J 1 1 1 1 1 1 ¡ 1 ![ 1 i ¡ 1 1 J ~ rp- ,--. 1 l::.J-11- : 1¡1 J 1 1 11 1 1 ; 1 1 1 1 "1 ~ 1 ~ 11 reduces pr~ce incentives for farmers since the complete marketing marg1n (farmer to retail) for money alternativa grain crops on Lampung is only around 20 to 30% (Word Bank 1983) The effective operation throughout Indonesia of the export price floor under domestic cassava prices however depends critically on spatial integration of the various cassava markets Such integration relies on two components first integration between fresh root and gaplek prices and second between gaplek prices in different markets throughout the country In terms of the linkage between fresh root and gaplek prices variation in fresh root pr1ces explained over 90 percent of the variation in gaplek prices in 7 of 19 markets on Java and over 80 percent of the variation in 18 of the 19 markets (Unnevehr 1982) Not only were gaplek and fresh root prices strongly linked but there was also a strong l~nkage of gaplek prices between markets across Java and this l~nkage was principally due to the operation of the export price floor Thus when domestic prices were at export parity the correlation coeffic~ent of gaplek prices in the 19 d1fferent markets was greater than or equal to O 90 for 106 of 171 potencial pa1rs On che other hand when domestic prices were above export parity only prices 1n 27 pairs of markets were correlated at the level of O 90 (Table 4 1S) When domestic prices were at export parity domestic price variation of gaplek was due almost completely to variation in the export price (Unnevehr 1982) Since there was a generalizad price linkage both between markets and between roots and gaplek the operation of an effective price floor was demonstrated for Java as a whole When domestic prices rose above export parity price variation was much more influenced by regional supply and demand conditions for cassava Moreover internal transportation costs tended to lower the export floor for more remate markets increasing the influence of local supply and demand conditions Thus the number of months the prices at 19 internal markets were at export parity varied from 32 to 70% of the time all less than the 78% at Surabaya Nevertheless what is remarkable is how often domest1c prices have been at che pr1ce floor In the period 1971 to 1979 monthly prices 1n ma]or markets were at export parity between a third to four-f1fths of the time Production 1n chis period grew at an annual rate of approximately 2 8% at a time when population growth was 2 O% and income growth was S 3% Normal growth in food demand for cassava (assuming a combinad income elasticity of O 1) and the rapid growth in starch production should have put some upward pressure on cassava prices Moreover never more than 1S/ of domestic production was exportad and the figure was usually less than 10% Surpluses at export prices thus were never that large Part of the reason was that there was a general upward trend in export prices However the other maJar factor affect1ng cassava prices is the domest1c price of rice and over chis period the real price of rice fell substant1ally (Figure 4 S) due to the impact of improved rice technology and import policy Timmer (1980) finds a cross-price elasticity between cassava and rice of O 77 indicating significant decreases in cassava TABLE 4 15 Indonesia Correlation Greater than or Equal to o 80 o 85 o 90 o 95 Total Possible Pairs Gaplek Price Correlations Among 19 Producing Area Markets Number of Markets Correlated When Prices Were A hove Export At Export Price Floor Price Floor 102 149 63 137 27 106 2 32 171 171 SOURCE Unnevehr Laurian Cassava Marketing and Price Behavier on Java 1982 consumption for a decline in rice prices During the period of rapid expansion in rice supplies the cassava export market served a critical funct1on of providing an effective price floor and thus maintaining incomes of cassava farmers As Indonesia exploits most of the yield gain poss1ble from the rice technology domestic rice prices and rice imports are again likely to become important policy issues Cassava because of this price linkage to r1ce allows additional flexibility in meeting r1ce pr1ce policy objectives In the future improving cassava production may be a far less expensive means of maintaining rice prices than rice imports Any cost reductions in transport or scale economies in assembly will tend to favor cassava over other crops On the other hand to assembly costs must be added processing costs Both the gaplek and starch processing industry has been found to be socially efficient (Nelson 1982) Less than a quarter of the export parity price for both starch and pellets is consumed by processing costs (Table 4 16) The cassava process1ng industry is relatively dynam1c and as well permits a significant degree of diversity Labor intensive household starch production co-exists with capital intensive large scale factories All are profitable although government tax and capital credit policies tend to favor the large-scale plants when the household units are socially more eff1c1ent and employ significantly more labor (Nelson 1982) Cassava marketing systems in Indonesia have evolved in response to transport infrastructure development and changes in market demand There has been almost no intervention by government agencies apart from the tax credits for large scale processing plants and the import tax on starch As the evidence suggests cassava markets function very eff1c1ently in Indonesia given the constraints imposed by infrastructure There is not only little need for government involvement in cassava markets but unlike rice any such intervent1on in a commodity w1th multiple markets would be counter-productive without a comprehens1ve policy and this would be difficult to attain Unlike many other countries in Asia Indonesian cassava markets reflect national supply and demand conditions w1th a buffer provided by the export market Further development of cassava in Indonesia w1ll be relatively easy given such a well functioning marketing system PRODUCTION Demand for cassava remains very dynam1c in Indonesia especially as markets have continued to diversify and cassava demand is not dependent on just food demand for fresh roots and gaplek Potential markets in the area of high fructose sweetners and balanced feeds remain untapped due to lock of sufficient product1on and Indonesia has not come close to meeting the import quota set in the EEC With such a strong demand situation the questions naturally turn to production and the means of 1ncreasing an already significant growth rate Production trends and distribut1on Cassava was introduced into Indonesia through early Portuguese trade with the Spice Islands but did not become well established as a major crop until the mid to late 1800's The spread of cassava was promoted by the TABLE 4 16 Indonesia Production Social Costs and Profits in Starch and Gaplek 1980 Starch Gaplek Medium Large Household Scale S cale Chips Pellets (000 Rp/t) (000 Rp/t) (000 Rp/t) (000 Rp/t) (000 Rp/t) Export Parity Price 178 9 178 9 184 4 81 9 Root Costs 122 S 122 S 110 9 58 9 Processing Costs 39 2 45 o 66 7 S 2 Social Profit 17 2 11 4 6 8 17 8 Source Nelson Gerald Implications of Developed Country Policies for Developing Countries The Case of Cassava 1982 82 S 58 9 18 8 4 8 Dutch as a fam~ne reserve Also by the turn of the century the Dutch had developed a large cassava starch industry on Java directed towards export which also provided incentives for expansion of cassava production By the mid-1960's area sown to cassava on Java reached a peak of 1 4 million hectares and has since declined (Table 4 17) Since 1975 cassava area on Java has been relatively stable at an even one million hectares Cassava area on the off-islands remained static through the 1960 s and early 1970 s Only in the later part of the 1970 s has area in the off-islands shown a significant increase due to the transmigration proJects and the expansion of the gaplek trade and starch on Lampung The distribution of cassava production in Indonesia to a large extent corresponda with the distribution of population About 70% of the cassava is produced on Java Java is followed by Sumatra which accounts for a little over 10% The rest of the production is distributed throughout the other islands (Table 4 18) Cassava is thus grown throughout Indonesia almost wholly in upland areas and has established itself as a maJar palawija (secondary upland food) crop in Indonesia Over the decade of the seventies cassava production grew at annual rate of 2 7% per annum in Indonesia However this production growth was marked by very different rates of growth between regions On Java cassava production grew at an annual rate of 1 8% while off-Java the growth rate was S 2% Even on Java growth occurred only in Central and Eastern Java while production was stagnate in Western Java By far the most rapid rate of growth occurred in Lampung on Sumatra where production grew at a 12 2% annual rate tripling in the space of a decade The faster rate of growth on the off-islands than on Java would be expected particularly given the severe land constraint on Java versus the outer islands and the policy to settle populations on the outer islands The 1 8% growth rate in production on Java in the 1970 s was due to a decli-g7 ~n area of O 9% per year and an annual increase in y~elds of 2 87 - Historically yields on Java had been static at a little over 7 t/ha since the 1920's (Rache 1983) and only s~nce 1973 have yields levels shown a consistent rising trend The natural question is what are the factors that have precipitated this relatively sudden and rapid rise in yields? A corrollary however would be the identification of the factors that have kept yields on Java much lower than other maJar producing countries in Southeast Asia that is about half the yield levels in India and Tha~land The intensity of production systems on Java and the favorable agro-climatic conditions would suggest sim~lar or higher y~eld potent~al These issues shall be explored in the follow~ng two sections Production growth on the outer islands during the 1970 s showed a distinctly different pattern to that on Java The principal factor respons~ble for the 5 2% production growth rate was the 3 27 annual expansion in area This is similar to the population growth rate off-Java of 3 0% ~n the 1971-80 period However most of this expans~on was 'il See Rache (1983) for a discussion of factors contributing to declining area planted to cassava TABLE 4 17 Indonesia Cassava Area Production and Yields Java and Indonesia 1951-81 Are a Product~on Yields (milhon ha) (million tons) (tons/ha) Java and Java and Java and Madura Indonesia Madura Indonesia Madura Indones~a 1951 75 87 S 3 7 1 7 1 8 2 1952 77 93 S 1 7 S 6 6 8 1 1953 87 1 04 6 S 9 o 7 S 8 7 1954 87 1 07 6 4 9 6 7 4 9 o 1955 88 1 08 6 S 9 4 7 4 8 7 1956 90 1 12 6 4 9 1 7 1 8 1 1957 99 1 22 7 2 10 1 7 3 8 3 1958 1 08 1 34 8 1 11 3 7 S 8 4 1959 1 19 1 46 9 o 12 7 7 6 8 7 1960 1 14 1 42 8 6 11 4 7 S 8 o 1961 1 14 1 48 8 4 11 2 7 4 7 6 1962 1 14 1 45 8 1 11 4 7 1 7 9 1963 1 28 1 56 8 7 11 6 6 8 7 4 1964 1 26 1 58 9 1 12 3 7 2 7 8 1965 1 40 1 75 9 7 12 6 6 9 7 2 1966 1 17 1 51 8 3 11 2 7 1 7 4 1967 1 18 1 52 8 3 10 8 7 o 7 1 1968 1 16 1 50 8 8 11 4 7 6 7 6 1969 1 14 1 47 8 2 10 9 7 2 7 4 1970 1 09 1 40 8 o 10 5 7 3 7 S 1971 1 10 1 41 8 1 10 7 7 4 7 6 1972 1 13 1 47 7 9 10 4 7 o 7 1 1973 1 06 1 43 8 1 11 2 7 6 7 8 1974 1 16 1 51 9 6 12 9 8 3 8 S 1975 1 02 1 41 9 3 12 3 9 1 8 7 1976 1 00 1 35 8 8 12 2 8 8 9 o 1977 99 1 36 9 1 12 5 9 2 9 2 1978 1 01 1 38 9 S 12 9 9 4 9 3 1979 1 02 1 44 9 9 13 8 9 7 9 6 1980 1 00 1 41 9 8 13 7 9 8 9 7 1981 99 1 40 9 9 13 7 10 o 9 8 Source Falcon et al The Cassava Economy of Java 1984 concentrated on Sumatra and particularly in Lampung Area and production expansion thus appeared to be related more to expanding infrastructure and market possibilities than to expanding population However expanding area was not extensive in nature since cassava yields as well rose at a rate of 2 0% per annum on the outer islands Thus trends in cassava product~on in Indonesia over the past decade have been favorable particularly given the severe land constraint on Java where the bulk of the cassava is produced Nevertheless cassava production on the outer islands is growing much faster due in part to the unexploited land resources there This creates something of a dichotomy in any further expansion of cassava which as will be seen in the succeeding analysis is reinforced by other majar differences in production systems between Java and the outer islands Cassava production systems Cassava production systems in Indonesia unlike other majar cassava producing countries in Asia are complex Complexity in this case introduces diversity and across Indonesia there ~s substantial variation ~n product~on systems based on agro-climatic conditions land availability and market access (Table 4 19) Unfortunately there has been only ene majar attempt to study these production systems in depth and as a result this section will by necessity principally summarize the research of Rache (1983) in his analysis of cassava cropping systems in three regions of Java Moreover because of the differences in land/labor ratios between Java and the outer islands production systems on Java w~ll be considered independently of those off-Java The complexity of cassava production systems on Java derives from ~ntercropping and rotation systems and from double-cropping with rice in certain land types Because median farm size on Java is only O 4 hectares farmers seek to optimize returns to this limited resource Over half of cassava grown on Java is intercropped (Table 4 20) with the principal ~ntercrops being maize and upland rice and ~n West Java legumes such a peanuts and soybeans In certain areas clase to urban areas where fresh market prices are suff~ciently high cassava in monoculture will follow rice on irrigated land particularly where there is not suffic~ent water for a second rice crop Finally although cassava will in most cases not complete for land with rice it will have to compete for labor and capital resources so that appropriate timing of cassava cultural practices is a majar factor in production systems Agro-climatic conditions particularly rainfall distribution so~l type and soil fertil~ty together with irrigation availability are determ~n~ng factors in the choice of cassava cropping system Rainfall is adequate for cassava all over Java but in certain rainfed areas is lim~ting for other crops Thus as rainfall reliability declines from west to east (Figure 4 6) cassava production tends to be concentrated more in the eastern part of Java and on the island of Madura (Figure 4 7) even though cassava is grown throughout Java apart from the irrigated areas of the northern pla~ns IV - 17 - So~l type topography and the eroded state of soils define the other major constra~nt on adaptation of upland crops Soils with maJor fertility acidity or toxicity problems such as Ult1sols are princ~pally found on the outer islands The principal soil constraints on Java are h~ghly eroded unterraced hills~des Such areas tend to be most common in the south-central coastal zone an area where cassava production is most highly concentrated Whereas rainfall d~stribution pr~nc~pally affects timing and whether one or two intercrops can be planted land type determines the range of crops that can be grown At the extreme where soils are highly eroded cassava ~s the crop of last resort / In general as soil and rainfall constraints become more severe first legumes leave the intercropping system followed by upland rice and finally maize leaving cassava as the sole crop on highly eroded soils Where soil and rainfall are not limiting all of these crops can be included in one system as shown ~n Figure 4 8 However generally upland rice is the principal ~ntercrop in the wetter western part of Java wh~le maize is the principal intercrop in the central and eastern reg~ons In most systems the land is prepared before the start of the heavy rains normally around October or November The upland rice and/or maize are planted f~rst and after establ~shment ~n two to four weeks cassava is planted Where soil conditions are not limiting this system provides effective ground cover until cassava reaches full canopy wh~ch in turn aids in controlling erosion under the h~gh rainfall conditions of Java The resource structure of the systems vary substantially (Table 4 21) Labor use is high even in those areas where bullocks are used in land preparation and inter-row cultivation Fert~lizer use tends to be higher in the more productive land types principally because more respons~ve crops are planted in the ~ntercrop system and relatedly such systems probably g~ve the h~gher marginal return to fertilizer use Cassava yield levels thus vary substantially between systems Over 70% of cassava is planted in the major rainy period from September to January (F~gure 4 9) This introduces two principal constraints on cassava production systems First th1s coincides with the major rice planting season which creates competition for labor resources Second the crop must be harvested and the land cleared by the start of the next rains Where cassava is dr~ed into gaplek the harvest must be earl~er to take advantage of the dry season In those systems were cassava follows a rice crop timing is crucial since the crop has only six to eight months before harvest Nevertheless the longer matur~ty of the cassava complements the harvest~ng pattern for rice (Figure 4 10) The maJor port~on of the cassava harvest occurs ~n the June-October per~od after the pr~ncipal r~ce harvest ~nsuring a more stable supply of carbohydrate sources This tends to coincide with the dry. period so that cassava roots can be processed into gaplek where markets for fresh cassava are not assured Roche (1983) presents evidence which suggests that cassava cont~nues to grow and add root we~ght during the dry season -- th~s would not be the case were soil moisture limiting Farmers thus face a trade-off between timely harvest for either gaplek dry~ng or early land preparation and eventual cassava y~eld Table 4 18 Cassava Distribution by Island and Per Capita Production 1980 Percentage of Per Capita Province/Region Production Total Production '- (000 t) (%) (kg/cap) Java 9 795 8 71 4 107 3 Jakarta 4 o o 6 West Java 1 975 3 14 4 71 9 Central Java 2 970 7 21 6 117 1 Jogyakarta 655 7 4 8 238 9 East Java 4 190 2 30 5 143 6 Sumatra 1 601 5 11 7 57 2 Lampung 984 4 7 2 212 9 Ka liman tan 303 4 2 2 45 1 Sulawesi 581 7 4 2 56 1 Nusa Tenggara Timu 852 9 6 2 313 3 Other 591 o 4 3 71 2 Total 13 726 3 100 o 93 1 Source Central Bureau of Statistics IV - 18 - Where cassava pr1ncipally supplies starch factories or urban markets there 1s a demand for more continuous supplies of roots However staggered planting is only possible where rainfall is suffic1ent to support the intercropping system during most of the year such as in West Java or where land types are suited only for pure stand cassava In general providing for more continuous supplies of cassava roots 1s heavily constrained by rainfall distribution and the complexity of the cropping system on the small farms of Java Moving from Java to the outer islands the factors which determine cassava production systems change dramatically rainfall distribution soils farm size and markets all change quite significantly The initial striking difference is in rainfall distribution In general the outer 1slands have a more continuous supply of rainfall than Java On Sumatra Kalimantan and to a slightly lesser extent Sulawesi the major portien of area is suitable for continuous cropp1ng as compared to only 20% of the area of Java (neglecting the irrigated areas) Interest1ngly per capita production of cassava in Indonesia is h1ghest in those areas -- Java and Nasa Tenggara -- where there is a significant part of the area with constraints on water availability dur1ng the year (Table 4 18) Soils in general also vary markedly between Java and the outer islands Whereas rainfall is not as limiting on the outer 1slands soils in these areas impose much more severe constrains on cereal and legume crops although not on cassava The soils are in general ultisols being quite acidic of a low fertil1ty status and occas1onally having relatively high levels of exchangeable aluminium Because of these so1l problems together with the erodability on slopes much of th1s land area has been classified as marginal for cereal and legume crops Cassava however is well adapted to these soils but continuous cropping of such so1ls requires appropriate crop and soil management to maintain productivity levels Cassava production systems on the outer islands have in many ways been cond1t1oned by the dictates of the transmigrat1on schemes Before the advent of the transmigration schemes much of cassava on the outer islands was grown in a shift1ng agricultural system Such a system was very extensive particularly since the abandoned fields returned to alang-alang (Imperata cylindrica) rather than the or1ginal forest fallow The transmigrat1on schemes superimposed a fixed farm size structure over the original shifting system Farmers were in general given 3 S hectares to exploit and apart from the Lampung area the settlement areas were chosen where the soils were not ult1sols Farmers however could not effectively ut1lue the whole 3 S hectares On the one hand labor- intensive cropping patterns were brought from Java to an area where labor needs relied solely on family availability and there was no bullock power On the other hand infrastructure was limited and there was no effective market even were surpluses to be produced Until sufficient infrastructure was developed such as happened on Lampung there was little 1ncentive to sow over O 6 to 1 O hectares sufficient to meet family food needs TABlE 4 19 Characteristics of the Five Major Cassava-Producing Regions of Java and Madura Cassava as a percent of total major food crops harvested Range of cassava yields (tons/ha) Official data 1977-79 Field surveys 1979/80 Level of soil erosion Principal intercrop with cassava Principal end use of cassava Direct human consumption of cassava Quantities Form Source Falcon et al West Java 15% 10-12 6-20 High Upland rice legumes Starch Low Fresh Central Java 18% 9-11 5-12 High Corn Gap le k sales Low to moderate Fresh gaplek The Cassava Economy of Java 1984 South- Central Java 35% 7-9 2-10 Severe Upland rice corn Staple food High Gap le k East Java 14-' 10-11 10-40 Modera te to high Corn Gap le k sales staple food Modera te to high Gap le k Madura 24% 7-9 4-8 Modera te Corn Staple food gaplek sales High Fresh Gaplek TABLE 4 20 Farm Size o 1-0 3 ha o 3-0 5 ha o 5-0 75 ha o 75-1 o ha 1 0-2 o ha 2 o + ha ALL FARMS Farms Containing Intercropped Annual Crops as Percentages of all Farms on Which These Specific Crops Were Harvested 1973 Percentages of Farms Harvesting Intercropped Cassava Upland Rice Maize (%) (%) (k) 52 9 57 7 51 1 53 3 61 5 51 5 54 8 64 6 52 7 55 6 67 7 53 5 56 6 69 2 44 2 54 4 66 3 52 4 54 2 63 1 54 4 Source Ro che Fredrick Cassava Production Systems on Java 1983 TABlE 4 21 Indonesia Resource Structure of Cassava Cropping Systems in Three Survey Sites 1980 Garut Gummg Kidul Intercropped Intercropped Intercropped Inputs and Jculated on a per ton basis these per ton costs (except in one cas~f - are in general equal to or less than the 1980 farm-level price - Cassava-based systems generate sufficient prof~t to cover the market costs of the factors of production a fact of some significance in such intensive systems The maintenance of normal proht levels for cassava is reflected in both tbe importance of the cassava as a cash crop and its relative stability in the cropping systems of Java and Southern Sumatra over the last several decades Technology development Since the constraints on cassava yields are both not fully understood and vary substantially across Indonesia a research program to develop yield-increasing cassava technology needs both a clase linkage to farmer production systems and a quite extensive testing system Moreover raising cassava yields will have to be done within intercropping systems and it will not be possible to heav~ly sacrifice yields of other crops in increasing cassava yields especially that of upland rice Finally yield potent~al will be heavily c~rcumscribed by climatic and soil conditions so that any yield gap analysis will have to be defined in terms of locat~on and land system Such a research focus requires a certain critical level of resources yet research resources for palawija crops have traditionally been lim~ted as most resources have been devoted to rice Agricultura! research is relatively centralized in Indonesia and comes under the responsibility of the Agency for Agricultura! Research and Development (AARD) AARD is divided into seven majar research centers of which cassava comes under the Central Research Institute for Food Crops These central research institutes are in fact a coordinating body for a set of regionally based research centers of which there are seven under the Central Research Institute for Food Crops Cassava research in Indonesia is centered in the Root Crop lmprovement Program which is under the Bogar Research Institute for Food Crops There is some consideration of plans for decentralizmg research decision-making and making the seven research ~nstitutes semiautonomous which could mean that cassava research could be done in more of these ~nstitutes However currently cassava research is centered at Bogar which focuses on more basic research Thus all of the cassava breed~ng research is done at Bogar Agronom~c research and advanced selection of clones are done at some of the other research centers 2_/ The one exception is one of the terraced h~lls~de systems in Garut Since this system was unfert~lized monoculture cassava this probably indicates land of inferior quality and therefore of a lower opportunity cost Nevertheless a constant rental value was appl~ed to all systems thereby probably overestimating costs for this system The 1980 price was exactly equal to the average (on a deflated price basis) of the period 1969-81 This is based on the rural market pr~ce series for fresh roots for Java and Madura published by the Central Bureau of Statistics IV - 23 - Cassava technology development in Indonesia in the postwar period has principally focused on varietal development and fert~ll.zer trials Two var~eties Adira I and II were released in 1978 Adira I has a lower HCN content shorter maturity higher starch content and about the same yield potent~al (35 t/ha) as Adira II Ad~ra I ~s apparently grown quite widely on Lampung (Roberto Soenaryo private communication) but ~ts adoption on Java has not been w~despread Understanding why farmers have not adopted Adira I could offer valuable insights into whether the problem is the var~ety or its extension Clearly in Indonesian cassava systems yield ~s only one criterion among many that will mot~vate farmer adoption Rache (1983) argues that the most immed~ate avenue to ~ncreasing cassava yields is through a comb~nation of the Adira I var~ety and appropriate fertilization In the longer term more f~nely tuned varietal development together with integrated fert~lization rotation seed management and intercropp~ng practices designed for homogenous land systems will probably be the principal means to achieving significant increases in cassava yields Certa~nly the object~ve w~ll be a stable continuous cropping system in upland areas with cassava as a s~gnificant component Another consideration is whether a dist~nction should be made in a casssava research strategy for Java versus the outer islands Resolut~on of th~s issue to a large extent will depend on whether research is decentralized and on land policy and the availability of labor-sav~ng technology in the transm~gration schemes Currently cassava and other food crop production on the outer islands depends on the very labor-intens~ve production systems developed on Java Farmers usually cannot utilize all the land allocated to them because of the lack of labor and/or tenant markets (Wardhani 1976) Research in the outer islands to date has focused primarily on further ~ntens~fication of ~ntercropping systems w~th pr~ncipal focus on resolving part~cular soil constra~nts A broader setting of research objectives might consider whether higher farmer incomes could be achieved with a cont~nued focus on just land productivity or whether the focus should be on technologies that requ~re less intensive labor use leading to the cultivation of more land A focus on less-labor intensive cropping systems for the outer islands would reinforce cassava s role as a cash crop at least ~n those areas where infrastructure is sufficiently well developed However the important role of cassava as a food crop where it is principally consumed ~n the fresh form should not be sacrificed A s~ngular focus on mechanization and var~eties for the industr~al starch market would favor pr~marily the plantation systems without attendant benef~ts for food consumption of cassava A research strategy for cassava in Indonesia significant level of diversification and a clearly defined l~nkage between requires a product~on constraints and end use In part~cular in any var~etal development the focus should be on the development of dual-purpose varieties where food quality parameters are maintained in the selection process This is critica! to the maintenance of price integration which has been so important to the growth of cassava in Indonesia IV - 24 - Conclusions Growth in the Indones1a economy has been impressive aver the decade of the 1970's continuing through to 1982 GDP growth averaged 7 6% per annum in the 1970's and was above that mark in 1980 and 1981 These growth rates were well above the average for either industrial or developing countr1es Only in 1982 did the economy start to be affected by the internacional economic recession and GDP growth fell to 2 3% rebounding to around 4% the following year The decline in oil prices and demand for agricultural exports led to a significant decline in the foreign exchange reserve position culminating 1n a devaluation of the rupiah in 1983 and 1986 and tighter controls on imports Future growth in the Indonesian economy is highly dependent on what happens in the petroleum export market nevertheless the economy is projected to grow by 5~ per year through the rest af the decade (World Bank 1984) Such signif1cant growth in incomes have a marked impact on food demand Estimated annual per capita consumption of r1ce increased from 107 kg in 1970 to 145 kg in 1983 Fortunately rapid demand growth corresponded with the rapid adoption of short stature rice technology and rice production almost doubled in this period even with very m1nor change in the land area planted to rice Nevertheless Indonesia remained a maJar net importer of rice importing as much as 2 million tons in 1980 Growth in production of rice is expected to slow somewhat through the end of the decade as the growth rate in yields declines Nevertheless Indones1a is expected to remain at or near self-sufficiency in rice while continu1ng to maintain some capacity to import when production deviates from trend (World Bank 1984) Indonesia has been relatively successful in attaining self-sufficiency in the production of basic foodstuffs and in maintaining relatively stable consumer prices especially for rice While the government has been successful in meet1ng two of its food policy ob]ect1ves impact on ra1s1ng farmers incomes the th1rd pr1ncipal food policy obJective has been less widespread This is because the income generation from the new r1ce technologies was d1rected almost exclusively toward the irrigated sector The benefits from the new rice technology have been inequitably d1stributed between regions and since the bulk of the population continues to depend on agriculture for their income continued neglect of the upland areas will further increase these d1sparities Two principal concerns should govern policy toward the upland sector The first is the relative priority between development of the upland areas on Java and those on the outer islands Java accounts for 47% of Indones1a s GDP 62% of the population and only 7% of the land area The soils on Java are relatively fertile transport infrastructure 1s relat1vely well developed and very labor intensive production systems~ave evolved to suit the extremely small average farm size On the outer islands on the other hand the soils tend to be infertile and h1ghly acidic and infrastructure is not as highly developed Land is relat1vely plentiful The population distribution between Java and the outer islands creates a situation where both land and labor resources are underutilized and the transmigration proJects were established to remedy this imbalance IV - 25 - Between 1971 and 1980 approximately 2 1 million migrants resettled Ln the outer islands of which one million were resettled through the transmigratLon program ThLs program had a sLgnificant impact on agricultura! employment Of the 1 8 mLllion increase Ln agricultura! employment in this period 1 4 mülion was off Java (World Bank 1982) Certainly any increase in area planted to crops will have to come on the outer islands and the government LS currently attemptLng through agricultura! research estate development and the transmigration proJects to establish a base for future growth on the outer islands The second issue LS the choice of crops where technology can be expected to raise productivity and markets are suffLciently expansive to absorb the increases in production thereby leadLng to increases Ln farmer income Certainly cassava must be considered as a principal choice for both Java and the outer islands Maize is an alternatLVe choice on Java and tree crops -- and maize in selected areas -- are an alternative on the outer Lslands Cassava could have a signifLcant potential impact given a higher committment of resources to support research on the crop As a crop for development of the upland areas cassava has several advantages Most importantly the cassava marketing system Ln Indonesia is probably the best developed in Asia wLth the possible exceptLon of the larger but more specialLzed system in Thailand Prices efficiently allocate cassava between regions across different end uses and over time Moreover an effective price floor is provided by the gaplek export market EffLcient markets together with the multiple end uses for cassava partLcularly the high consumption of gaplek and fresh cassava by the poor allows the Lntroduction of improved production technology to achieve the dual policy obJectLve of increasing farmers' Lncomes and improving calorie intake of the rural poor Moreover, the rapidly growing starch market with potential under current policies for the development of high fructose sweetners provides scope for the absorption of significant increases in production and any further surpluses could be exported at least upto the 825 thousand ton quota Nevertheless the very uncertain situation in the EC market for cassava pellets will continue to affect the Indonesian cassava economy Lf not in lower import quotas after 1986 then in the impact on world prLces and the impact that lower world prices will have on Indonesia farmers There LS sorne opinion (World Bank 1984) that IndonesLan will be in a surplus positLon in both maize and cassava by the end of the decade with little hope of absorbing these production increases in dornestic markets For cassava the report overlooked the large and dynarnic starch market but certainly any maJor productivity increases will probably result in internal prices remaining effectLvely tied to the export prLce with the accornpanying need to maintain sorne flexibLlLty Ln the export market More than anything else a dynarnic cassava sector provides flexibility in Indonesia s food and agrLcultural policy When rice yields start to plateau out at the end of the decade cassava can add flexibility to price and irnport policy for rLce Moreover the starch hLgh fructose sweetner and when necessary the export and/or dornestic feed markets can be a basis IV - 26 - for expanding cassava on the outer islands agricultural areas where a well adapted cash crop for smallholders has been difficult to identify This type of flexibil~ty will be key for balanced agricultural and industr~al development in lndonesia's future IV - 27 - Append~x 4 1 A synthes~s of production and ut~lization This append~x reviews the cons~stency between production and consumption estimates for cassava in Indonesia and develops a supply and utilization table for the year 1978 The table disaggregates the data for Java and the outer islands Two other estimates of cassava supply and d~stribution exist one ~s the food balance sheets for Indonesia put out by the Central Bureau of Statistics and the other is an estimate by Laurian Unnevehr (1982) for Java only These estimates will be used as a point of reference in developing the supply and distribution estimates Food uses are a dominant form of utilizat~on of cassava in Indonesia The most systematic estimates of cassava consumption patterns comes from the periodic National Socioeconomic Expenditure Survey (Susenas) -- see Dixon (1982) for a discussion of the structure of the surveys The 1978 survey (Susenas V) found an average per capita consumption of 20 3 kg of fresh roots and 9 4 kg of gaplek on Java and 20 2 kg of fresh roots and 3 1 kg of gaplek on the outer islands This resulted in an average for Indonesia as a whole of 20 2 kg of fresh roots and 7 3 kg of gaplek or an average of 42 1 kg of cassava on a fresh equivalent basis A standard rate for converting fresh roots to gaplek is more complex in Indonesia than Thailand because roots are peeled and gaplek is not dried to a standard percentage This introduces peel~ng loss moisture content and dry matter content as variables in the determination 9f the conversion rate Field observations suggest a peeling loss of 20% (Unnevehr 1982) which is in accord with standard percentages of peel to root weight of 15 to 20% found at CIAT (Rupert Best private communication) Moisture content of gaplek is apparently highly variable F1.eld observation by Unnevehr suggests levels as high as 25% Studies at CIAT (Rupert Best private communication) have found problems of cont~nuing physiological deterioration and heavy fungal growth on cassava chips with h~gher than 18k moisture even after one week Drying to moisture levels of 20% or above the storage life of cassava is not substantially extended unless there are alternative means of controlling fungal growth Unnevehr did find relatively high losses ~n gaplek storage but only after relatively long periods What average moisture content of gaplek is at the point of consumpt1.on remains somewhat of a question So also does the average dry matter content of cassava roots Dixon (1982) and Unnevehr (1982) both employ a conversion rate of roots to gaplek of 2 S to 1 Assuming a 20/ weight loss due to peeling gaplek at a 2sr moisture content implies a dry matter content of 37 5% wh1.le at 18% moisture a 41% dry matter content is 1.mplied These dry matter percentages are above the normal range at least when compared to different genotypes evaluated at Bogor A more reasonable assumption l.S a 18% moisture content and a 33% dry matter content wh1.ch g1.ves a convers1.on rate of 3 O to 1 for fresh roots to peeled gaplek The 42 1 kg average level of cassava consumption from the expenditure surveys compares to an estimate from the food balance sheets of 74 O kg per capita Food consumpt1.on in the food balance sheets is estimated as a residual after all other uses have been deducted The discrepancy between IV - 28 - the two estimates is significant and provides the first 1ndication of some inconsistencies in either the production estimates or the estimates of other end uses To evaluate such discrepancies the data on the different end uses is first reviewed The est1mates of gaplek and fresh cassava consumption from the SUSENAS surveys are accepted as the best estimate of direct food consumption although if anything these should probably be seen as minimum estimates Gaplek is not only used directly for human consumption but is also exported and Unnevehr (1982) found some gaplek being milled into flour by wholesalers and used in bakery products Gaplek exports from Indonesia are highly variable and in 1978 exports particularly from Java were on the low side Nevertheless export levels for the year 1978 were used Cassava flour on the other hand is assumed to be produced only on Java and Unnevehr s estimate is used Starch is a maJar utilization form in Indonesia and although it princ1pally goes into food uses starch consumption is not included in the human consumption estimates Utilization of cassava as starch comes from starch production estimates The most rigorous evaluation of these estimates is provided by Nelson (1982) for the years 1973 and 1979 His estimates for 1979 are used as the best measure of roots being processed for starch Animal feed provides the only other poss1ble end use of cassava Roche s (1983) survey of cassava production systems suggested no feeding of fresh roots to animals Given the limited importance of swine the dominance of ruminant animals and their ability to utilize lower cost feedstuffs and cassava's role either as a cash or food crop any on-farm feeding of cassava roots would be expected to be limited although there are no reports to conf1rm this assessment Incorporation of gaplek into balanced feeds is also thought to be limited given that market channels for gaplek are directed principally to export Unnevehr in her study of gaplek market1ng channels mentions no movement of gaplek into what is in many respects a very limited feed concentrate industry The assumption will be made then that any use of cassava in animal feed is limited Assessing a waste component is problemat1c Given the intensive nature of production systems the close integration with markets and because of the very limited incomes the tendency for both farmers and middlemen to be very conscious of loss waste on Indonesia would be expected to be lower than in other countries In marketing channels for fresh roots Unnevehr (1982) reports losses of around 8% The more significant losses occur in the storage of gaplek from the main production period for consumpt1on 1n the period of high rice prices Unnevehr reports losses in this context of from 10 to 20% A figure of 8% losses is applied to marketed cassava and 15% to all gaplek for human consumpt1on -- the lower moisture and better storage facilities would mil1tate against such losses in the export trade Utilization figures are compared to production figures in Table 4A 1 For the outer islands there is a reasonable correspondence between production and utilization figures The slight discrepancy is probably due IV - 29 - to estimates for fresh cassava consumpt~on Applying this difference to fresh human consumption yields an annual per cap~ta consumption est~mate of 24 7 kg This is only slightly above the 1978 SUSENAS estimate of 20 2 and well below the 1976 estimate of 34 2 kg On Java however the production estimate is almost 20-' higher than the consumpt~on estima te Unnevehr ( 1982) in her estima te of cassava utilization on Java for 1976 found an even larger difference Rache (1984) suggests a number of problems with the absolute values of the production estimates but cannot deduce any basis for either an upward or downward bias Village level record keeping and crop cutting surveys probably prov~de one of the more accurate estimates of cassava production in Asia Further disaggregation of supply and utilization of cassava on Java reveals that the unexpla~ned production occurs essentially in East and Central Java Rache (1984 Table 2 6) provides some evidence to suggest that yields may be overestimated in Central Java Moreover Mink (1984) found an overestimation of ma~ze yields in official stat~stics in East and Central Java Attributing all the difference to yield overestimation implies a reduction of y~eld of 30% from 9 4 to 7 2 t/ha in Central Java and a reduction of 207 from 9 15 to 7 6 t/ha in East Java For maize in 1978 M~nk found an overestimation of yield of 14% in Central Java and 29-' in East Java Reduction in yield levels are not completely out of the question On the other hand the other maJar area of uncertainty is the size of household starch production The 1976 and 1978 SUSENAS consumer budget surveys show high rates of starch consumption in rural areas of Central and East Java (Dixon 1984) implying consumption from home or nearby production un~ts In other areas direct consumption is low implying purchases of krupuk If the higher figure for rural consumption is assumed and it is also assumed that this comes solely from household product~on then household starch production is at the minimum underestimated by 40 thousand tons in Central Java and 58 thousand tons in East Java Th~s assumes that no household starch production goes into markets for krupuk production This would account for one third and one half of the d~screpancy in Central and East Java ~f a conversion rate of 6 to 1 were assumed Making this adJustment in starch production results in a discrepancy of about 900 thousand tons Attributing th~s to yield overest~mation implies a reduct~on ~n average yields on Java from 9 4 to 8 5 t/ha a not unreal~stic adJustment On the other hand 900 thousand tons represents only a 7% error in the total product~on est~te and could as easily be attributed to underest~mates ~n consumption At this po~nt the choice is arbitrary and Table 4A 1 reflects the adJustment in yield levels • V MALAYSIA Cassava vs Tree Crops in the Compet~t~on for Land The agricultura! economy of Malays~a l~ke that of Thailand has traditionally been export-or~ented Export growth has relied on the fact that Malaysia has always been a land surplus economy and at several points in its history even had to rely on imm~gration of both Chinese and Indians to meet ris~ng labor demand in agr~culture and mining Export orientation within a land surplus economy put a premium on the development of an effective land pol~cy In this aspect Malaysia differed from Tha~land in that the focus of land policy was on promot~ng large-scale plantation agriculture w~th a secondary emphas~s on the development of smallholder agriculture both for the production of rice and export crops A focus on plantat~on agriculture has remained a primary component of agricultura! policy to the present Cassava was the first of the series of export crops that have spread across Malays~an agriculture The establishment of the first tapioca factory in Malacca in the early 1850 s coincided with the rapidly expanding use of commercial steamships The evolution in sea transport together with the opening of the Suez Canal in 1869 opened European markets to agricultura! commodities other than JUSt high valued spices The tapioca ~ndustry expanded rapidly and relied on cassava s part~cular advantages as a frontier crop The forest was cleared to feed the steam engines of the processing plant while cassava was planted in a shifting cultivat~on system character~stic of a land-surplus labor-scarce economy This production system which ostensibly took place w~th~n a plantation-type land concession but where the land was abandoned to lalang when soil fertility declined to unprofitable levels gave cassava the image of a soil-depleting crop especially compared to the rapidly increasing tree crops Although soil depletion was due more to the shifting cultivation system than to the crop ~tself this image has remained upto the present resulting in controls on cassava expansion through restrict~ons on land concessions and leases The oscillations in the export market for tapioca and starch land policy and compet~tion with export-oriented tree crops have remained the key factors ~nfluencing the Malaysian cassava industry to the present Product~on Trends Cassava production ~n Malaysia has never repeated the boom period of 1860-1890 In Malacca cassava area climbed from virtually noth~ng to around a peak of 30 thousand hectares in 1882 In the 1870 s cassava area had also began to expand ~nto neighboring Negr~ Sembilan reaching ~ts peak areas in the 1890 s (Jackson 1968) Area planted to cassava in th~s early period probably did not exceed 45 thousand hectares The cassava industry fluctuated with the prices on the world market through to the turn of the century but then got caught in a squeeze between the rap~dly expanding rubber industry in Malacca and the development of an export oriented cassava ~ndustry on Java These trends were remarkably rapid In 1906 there was 15 thousand hectares planted to rubber ~n the Straits Settlement Provinces (Malacca and Province Wellesley and Penang) versus 43 thousand hectares planted to cassava In the same year Java exported a little over V - 2 - 19 thousand tons of cassava products By 1913 rubber area had expanded to 64 thousand hectares in the Straits Settlements and Javanese exports had increased to over 90 thousand tons Cassava area in the Straits Settlements declined to only 6 thousand hectares (Greenstreet and Lambourne 1933) After this maJar structural shift cassava area oscillated between 10 and 20 thousand hectares over the next 70 years till the present (Table 5 1) The other major element in this stagnation of the cassava industry was the restrictions on land concessions and actual planting of cassava by many of the states Thus Negri Sembilan prohibited planting of cassava in 1912 Perak restricted plantings in 1909 and Selangor did the same in 1925 In Kedah in 1905 cassava was allowed only as a catch crop for tree crop establishment (Greenstreet and Lambourne 1933) Thus in the period between the two world wars the cassava industry shifted to Johore where there were no restrictions on cassava and Kedah where it was grown as a catch crop The shifting nature of the cassava industry continued since following the Second World War and especially after the 1958 Emergency cassava rapidly shifted to Perak which is the locus of the industry today Nevertheless land policy continued to play a dominate role in the organization of production In particular Aw-Yong and Mooi (1973) estimated that in the mid-1960's approximately 75% of the cassava in Perak was planted illegally on unalienated state land or forest railway or mining reserves As a result shifting cultivation remained the dominant production system for cassava Shifting cultivation systems and the uncertainty of access to land for cassava are possibly reflected in recent trends in production (Table 5 2) In cassava area there is significant variation around a relatively stable trend of 16 thousand hectares Yields also are highly variable ranging from 11 to 37 t/ha with no necessary tendency for variat~on in area to compensate variation in yield Product~on as a result is highly variable However this year-to-year variability is not reflected in the output of cassava products Converting starch and chip production to fresh root equivalent shows a consistent rise in root util~zation through the early seventies and a decline from the 1976 peak over the latter part of the decade (Table 5 3) A comparison of the two series suggests much more stability in the utilization series and a consistent underestimation of utilization when using the production series Given the large percentage of illegal plantings the production series probably does not capture all the actual area planted to cassava On balance there is probably much more stability underlying the Malaysian cassava industry than is reflected in production statistics on the other hand over the last half of the decade of the 1970's there has be en a persistent declining trend in cassava production Cassava Production Systems Cassava s principal comparative advantage vis-a-vis other crops is its adaptation to relatively marginal agro-cl~matic conditions and therefore its exploitation of land with a low opportunity cost Because there is no climatic constraints on crop production in Malays~a and tree crops are well TABLE 5 l Malaysia Area Planted to Cassava by Province 1890-1980 Wellesley and Year Mala cea Penang Perak Selangar Johore Kedah Pahang Total (000 ha) (000 ha) (000 ha) (000 ha) (000 ha) (000 ha) (000 ha) (000 ha) 1890 25 5 3 l - - - - - 28 6a 1900 22 5 3 3 - - - - - 25 8a 1905 26 7 4 9 - - - - - 31 6a 1910 7 4 - - - - - 17 o < 1930 b b 4 4 8 9 3 6 8 15 o 1947 o 6 o 2 3 l 2 l 4 o 2 l 2 o 16 9 1965 b l o 8 9 l 2 o 9 o 5 o 5 14 7 w ' 1970 o l o 3 8 8 l 4 2 2 o 5 2 4 17 5 1 1980 neg neg lO 9 o l o 2 o 7 neg 12 5 a Includes only Malacca Wellesley and Penang b Not disaggregated Source V - 4 - TABLE 5 2 Malaysia Area Planted Yield and Cassava Root Production 1960-1984 Year 1960 1965 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 Area Planted (ha) 12 235 16 344 17 667 14 857 13 151 11 820 11 553 15 112 20 908 20 502 17 815 16 635 12 512 9 599 7 654 6 757 5 390 Production Yield (t) (t/ha) n a n a n a na 207 200 11 7 161 768 10 9 279 400 21 1 238 720 20 2 254 326 22 o 218 710 18 6 241 840 11 6 357 345 17 4 197 425 11 1 225 057 13 5 254 309 20 3 211 178 22 o 285 953 37 4 252 442 37 4 201 385 37 4 Source Annual Report Extension Branch Ministry of Agriculture Kuala Lumpur V -S- TABLE S 3 Malaysia Comparison of Root Production Series with Root Equivalent of Starch Pearl and Ch1p Production 1971-83 Year 1971 1972 1973 1974 197S 1976 1977 1978 1979 1980 1981 1982 1983 Starch Pearl Chip Production (t) 161 768 279 400 238 720 2S4 326 281 710 241 840 3S7 34S 197 42S 22S OS7 2S4 309 211 178 28S 9S3 2S2 442 Root Production (t) 220 679 294 S20 314 303 309 824 369 773 444 821 411 240 383 621 393 S88 316 716 310 449 304 347 302 788 Source Appendix S 1 and Annual Reports Extension Branch Ministry of Agriculture Kuala Lumpur V - 6 - adapted to a wide spectrum of tropical soils cassava has no part1cular niche to exploit in the agricultura! economy and must compete w1th tree crops for land Thus of the 2S7 of Malaysian land under cultivation well over 80% is planted to the three principal tree crops rubber oil palm and coconut Paddy land accounts for another 107 leaving under 10% for all other crops Tree crops are by far the most profitable agricultura! activities and in fact cassava is primarily grown in those areas where farmers do not have the option of planting oil palm or rubber Land tenure primarily influences where and the type of product1on system that cassava is grown under in Malaysia The more minor area where cassava is cultivated is as a catch crop in the establishment of oil palm or rubber Th1s is done principally by smallholders although some planting of cassava as a catch crop by tree crop estates has also been reported (Lulofs 1970) The cassava is planted for 2 or 3 seasons as a source of income until the tree crop is established However this is not a widespread practice and is limited to those areas which have access to cassava processing plants The major portian of the cassava is grown in monoculture This is in part due to the fact that a large portian of the crop is planted on land where the grower has no usufruct rights Aw-Yong and Mooi (197 3) in a study of cassava production in Perak in the mid 1960 s found that over 70% of cassava area was planted illegally !llega! planting of cassava is done on a much more extensive basis than legal cult1vation (Table S 4) Area planted is often done on a large-scale sometimes exceed1ng SO hectares Where v1rgin jungle is cleared all work is done by hand However with the rising costs of labor areas covered with lalang which have the possibility of mechanized land preparation are now cultivated more generally than virgin forest This early study reports that most illegal cultivation is done within a system of sh1fting agriculture where the land is planted two or three times to cassava without application of fertilizer and then a new area is opened up and brought under production Whether the rising labor costs of opening new land has caused even 1llegal planting to shift to a more permanent cultivation system 1s only open to hypothesis but certa1nly the incentives are increasingly to shift to more continuous cropping even within an insecure tenure situation Legal production on the other hand is concentrated 1n the hands of smallholders Area planted in casssava averages less than 2 hectares and cassava is usually only one of severa! crops cultivated Even in this situation cassava is often grown on rented land or on state land with temporary occupational licences That is there is suffic1ent uncertainly in tenure not to plant tree crops Also cassava is often a component in the in1tial cropping system in those areas where farmers have recently been settled but have not yet invested 1n tree crops Thus even for the legal planting cassava is only planted in that land where investment 1n tree crops is risky Nevertheless production systems are much more stable Rotat1onal systems with other annual crops are often practiced along with application of fertilizer or manures Over the last couple decades fertil1zation has apparently sh1fted from farmyard manure and woodash (Aw-Young and Mooi 1973) to reliance on chemical fertilizers (Tunku Mahmud 1979) Moreover with the ris1ng cost of labor farmers have as well moved to the applicat1on V - 7 - TABLE 5 4 Malaysia Legal and Illegal Planting of Cassava in Perak 1964-67 Year Legal Planting Illegal Planting Total Area (ha) (ha) (ha) 1964 3846 10 413 14 259 1965 3887 10 324 14 211 1966 3939 10 364 14 303 1967 4502 12 923 17 425 Source Aw-Yong Kong Keong and Mooi Soong Wooi Cult:Lvation and Production of Tapioca :Ln Perak 1973 V - 8- of herbicides in order to control weeds Rising labor costs and the competition with tree crops for land have put a premium on ach1eving low costs of production per ton Land preparation is often mechanized and in Perak ridiging is widely practiced to control root rot under these high rainfall conditions More intensive production methods are now more economic than extensive production methods as the emphasis has shifted to lower labor costs and higher yields In effect shifting production systems have become increasingly uneconomic in Malaysia making cassava s reputation for soil impoverishment more of an historical red herring rather than a point in fact The other major production system for cassava is plantations In the early stages of the cassava industry these systems had the1r impetus in the form of land concessions allocated by the state governments However root production operated on a basis of shifting agriculture and it was not till the advent of rubber at the turn of the century that plantations based on permanent production systems were established At this stage production of cassava on a large scale declined However in the post-war period more permanent cassava plantations have been established usually under government sponsorship The motivation for plantations is usually to assure regular supplies to relatively large-scale starch factories However the operations of large-scale cassava plantations have not met with much success Of four plantations that have been operat1ng in the last decade only one is still operating High labor and overhead costs make plantation production much more costly than smallholder producuon within an industry that is highly competitive both from other domestic producers and international competition from Thailand Yields Cassava is grown purely as a commercial crop in Malaysia and moreover must compete with tree crops for both land and labor Yields are therefore a primary determinant of cassava' s economic viability in the country's agricultura! economy Not surprisingly average yields in Malaysia are high by world standards or even by comparison to other As1an countr1es National product1on statistics suggest an average yield in the range of 11 to 37 t/ha As has been suggested the reliability of these estimates are open to question Nevertheless the few surveys of cassava producers that have been carried out do support the higher end of th1s range of yield estimates Tunku Mahmud (1979) found an average yield of 28 t/ha in the Manong area of Perak Rahman Binti Adam (1974) found an average yield of 18 t/ha in a survey of farmers in Pahang Chan et al ( 1983) report average yields of 12-20 t/ha in Pera k and 20-35 t/ha in Kedah The point where these survey areas reside within the overall yield distribution for the country cannot be specified Aw-Young and Mooi (1973) suggest in Perak a very bread yield variation of from 7 to over 40 t/ha based on differences in soil and production system where the production system as well reflects principally variation 1n soil fertility (Table 5 S) Chan et al (1983) report that in Perak less efficient farmers achieve yields in the 6-10 t/ha range while the better farmers' plots yield 22 5-37 t/ha occasionally reaching levels as high as 45-60 t/ha The fact that cassava is not grown in continuous production sytems as in other parts of Asia contributed to the high yields obtainable in Malaysia V - 9- TABLE S S Malaysia Representative Cassava Yields by Soil Type and Production System Soil Type Virgin Jungle Soil Laterite Soil Clay Loam Sandy Clay Sandy So11 Mine Tailings Production System ShiftLng Cultivation on Jungle Land First Crop Second Crop Third Crop Regenerated Jungle Fust Crop Second Crop Small-Farm RotatLonal System Source Aw-Yong Kong Keong and MooL Soong Wooi Production of TapLoca in Perak 1973 37 26 29 Yield (t/ha) 3 - 44 9 - 29 9 - 32 8 9 9 22 4 - 29 9 14 9 - 17 9 7 5 - 9 o 29 9 - 37 3 29 9 - 32 9 22 4 - 26 9 26 9 - 29 9 22 4 - 2S 4 29 9 - 32 9 Cultivation and V - lO ' Other factors are the favorable rainfall and growing season the existence of relatively high yielding varieties and the apparently wide use of fertilizer on cassava However def~ning the gap between average yields and the potential productivity of the crop remains uncerta~n due to lack of sufficient farm-level data -- see Tan and Chan (1986) for a very good first approx~mation Costs of Production and Labor Utilization Cassava is a highly commercialized crop in Malaysia The crop is fully marketed usually for industrial process~ng Moreover cash costs form a bigh percentage of total costs because most labor is hired land preparation is mechanized and input use is relatively high Cassava farmers are thus responsive to changes in ~nput or output prices and likely to adopt technical innovations Production costs and root prices are therefore princ~pal ~ndicators of economic incentives that cassava producers face Technology development and the evolution of costs have reflected the relative scarc1ty of labor in the agricultural economy Where possible land preparation is mechanized and tractor services are provided by farmers cooperatives Moreover herbicides have assumed increased importance in cassava cultivation in order to reduce labor costs Weeding and harvesting are usually done on a contract basis With th~s tendency to reduce labor use as much as possible labor input is relatively low A survey in Perak (Tunku Mahmud 1979) found an average labor use of 62 mandays/hectare (Table S 6) Any further reductions will require the mechanization of the harvest Labor costs make up just less than half of total production costs for cassava Malaysia provides a counter example to the normal tendency for labor to make up the major portion of total production costs in cassava Moreover weeding is one of the more minor costs items again runn~ng contrary to normal patterns Land preparation fertil~zer costs and harvesting all are usually larger cost items (Table S 7) The tendency toward labor substitution ~s clear in the cost structure however the scarcity of land forced both by government land pol~cy and by high opportun~ty costs has also put a prem~um on yield per hectare as is reflected in the high costs for fertilizer High yields low labor input and moderate input use which is often subsidized by the farmer cooperatives result in a very low variable cost of production per ton of roots comparable to that of Thailand However farm-level prices of roots are normally higher in Malaysia than in Thailand This is princ~pally due to the high opportunity cost of land The annual net ~ncome for rubber was M$36S1 (at a rubber price of M$2 40/kg) and for oil palm was M$S030 (at an oil pnce of M$1200/ton) (Tunku Manour and St Clair-George 1979) This compares to an average net income for cassava in Perak of M$979 (at a root price of M$74/tons) (Tunku Mahmud 1979) High supply prices for cassava in Malaysia reflect the profitabil~ty of alternative crops which has provided some impetus to the search for higher yields and lower product~on costs but is primar~ly reflected in the utilization of land w1th a relat~vely low opportunity cost V - 11 - TABLE 5 6 Malaysia Activity Land Preparation Planting Labor Use in Cassava Production in Perak Labor Use (mandays/ha) 1 2 7 9 Weeding and Herbicide Application 13 3 Fertilizer Application 2 7 Harvesting 27 2 Transport Total Source 9 9 62 2 Tunku Mahmud Bin Tunku Yahya Agronomic Study of Tapioca Small- holders in Manong Perak 1979 V - 12 - TABLE 5 7 Malaysia Costs and Returns for Cassava Root Produc- tion in Perak 1979 Cost Itero Teja Kampar Manong (M$/ha) (M$ha) (M$/ha) Land Preparation 147 7 184 8 222 3 Planting 88 9 86 5 74 1 Stakes 27 2 27 9 19 3 Weed Control 242 6 258 1 146 o Fertilizers 540 9 450 5 168 7 Harvesting 197 3 223 o 222 3 Root Transport 247 o 223 o 271 2 Land Rental 15 1 14 6 14 8 Total Costs 1506 7 1468 4 1138 7 Total Revenue 2124 2 1778 4 1580 8 Net Return 617 5 310 o 442 1 Source Chan Seak Khan et al 'A Special Report on Cassava in Penisular Malaysia 1983 V - 13 - Technology Development Research of a rather sporadic nature has been carried out on cassava since at least the 1920 s The focus of this research was princ~pally oriented a to evaluation and characterizat~on of imported clones and to appropriate fertilization of the crop In the 1970 s a cassava research program was established within the Malays~an Agricultural Research and Development Institute (MARDI) Cassava research broadened in scope at MARDI but continued to maintain traditional lines of emphasis Germplasm evaluation was expanded to include a maJor crossing and selection program The princ~pal breeding object~ves were high y~eld and and high starch content of roots reflect~ng the demands made by the starch and chip markets Agronomic research continued the long tradition of focusing on plant nutr1t~on and ma~ntenance of soil fertility Long-term fertil~ty trials and evaluat~on of nutritional requirements of cassava grown on peat soils became princ~pal lines of ~nvestigat~on The few diseases of any potential significance were incorporated into the program as secondary screening obJectives (Tan and Chan 1986) Little direct impact of this research is yet visible on cassava yields Fert~lizer and herbicide use by farmers has s~gnificantly increased but this is due as much to subs~d~es on these inputs as to the research that has been carried out Breeding on the other hand is a longer term investment and while some lines have been identif~ed wh~ch g~ve superior yields to the dom~nant variety Black Twig none of these as yet has been released as a new variety Emphas~s on increasing yields ~s a well JUStif~ed strategy under Malaysian conditions given the need to achieve higher returns to land A complementary strategy on which there has been some research is to direct technology to low opportunity cost land areas Peat soils have been one area where there has been some research The other area is as a catch crop in the establ~shment of tree crops Little research exists on competitive ~nteractions between these two crops in association and the means to minim~ze them Certa~nly shade tolerance w~ll be a principal ~ssue in such research Markets and Demand Cassava has been cult~vated primarily as an industr~al crop since ~ts introduction The crop is grown as a food source by a few of the hill tribes such as the Seroi Semai (Hohnholz 1980) but in general a food market for cassava has not developed in Malaysia Moreover cassava markets have histor~cally been export oriented as ~nternal demand did not provide a significant base on which to build a cassava industry However with Malays~a s recent industrial growth and rising per cap~ta ~ncomes the 1970 s has seen a sh~ft from dependence on export markets to meeting r~s~ng demand in domestic markets This sh~ft coincides w~th a recent emphas~s ~n Malaysian agr~cultural pol~cy in meeting domestic requ~rements in key sectors pr~ncipally r~ce and to a certain extent sugar Nevertheless such a focus on domestic markets must still recogn~ze the dominance of the export tree sector on factor pr~ces in the Malays~an agricultural sector V - 14 - The Domestic and Export Market for Starch Starch has always dominated the cassava economy of Malaysia Moreover starch production has traditionally been or~ented toward export in line with most of the rest of the agricultura! economy Finally the history of the starch industry in Malays~a has been one of constant movement in search of areas where cassava roots could be produced most cheaply i e where competition with tree crops was least or where illegal land use was not rigidly enforced In the post-war period the starch ~ndustry settled ~n Perak and the following analysis will focus on starch production in that state Only two starch factories existed in Perak prior to 194S By 1968 19 plants were operating in the state with most of the growth coming ~n the 1950's when 10 factories were set up (Table S 8) At this point starch production depended primarily on the sedimentation method as only two plants were using centrifuges Production from these latter plants was higher than for the sedimentation plants (Table S 9) even though the centrifugal plants were only operating at 30% capacity Also the centrifugal plants obtained an extraction rate of between 20 to 23% while the sedimentation plants averaged between 13 to 18% (Onn and Yet 1971) With continuing problems with root supply and increasing competition from Thailand it is not surprizing that a shake-out of the industry would occur in so competitive an environment Thus by 1982 only eight starch factories were operating in Perak (Table S 10) What is clear however is that this shake-out did not occur until the late 1970 s Prior to that -- and contrary to the root production statistics -- the starch industry showed steady growth ~n the post-war period Starch exports increased steadily through the 19SO' s and 1960 s and peaked in 1976 (Table S 11) The shorter series on starch production complementa these export trends and suggests that total starch production also peaked in 1976 at 68 thousand tons Product~on declined from that level and has been stable at about SO thousand tons through the 1980's Exports however declined much more dramatically and Malaysia became a net importer of starch in 1981 (Table 5 12) Two factors were responsible for this reversal rapidly increas~ng domestic consumpt~on and ~ncreased price compet~tion from Thailand Domeptic starch consumption in Malaysia increased very rapidly during the 1970 s rising from less than 20 thousand tons in 1971 -- Onn and Yet (1971) estima te domes tic consumption at 16 3 thousand tons in 1967 -- to about SO thousand tons by the end of the decade Majar users of cassava starch are monosodium glutamate and glucose producers and the textile industry As industrialization proceeds in Malaysia starch demand is certain to continue to increase Particularly any future developments in either the plywood or paper industry should lead to signficant increases in consumption A market with significant potential is the sweetner market This market has expanded rapidly in Japan and Taiwan while Indonesia is currently starting a sweetner industry Malaysia imports about 8S% of its consumption requirements of sugar even though domestic sugar prices are mainta~ned at levels well above world market prices in arder to cover V - 15 - TABLE 5 8 Malaysia Distribution of Starch Factories in Perak Accord1ng to Year of Establishment 1968 Period of Number of Establishment Factories Befare 1945 2 1945-1949 2 1950-1954 6 1955-1959 4 1960-1964 3 1965-1968 2 Total 19 Source Chye Kooi Onn and Loh Wee Yet The Tapioca Processing Industry in Perak 1974 V - 16 - TABLE 5 9 Malaysia Distribution of Starch Factories in Perak According to Output and Processing Method 1967 Monthly Starch Separation Method Production Sedimentation Centrifuge (t) (number) (number) Less than 12 O 1 12 1 - 24 1 2 36 3 - 48 3 4 48 4 - 60 4 1 -60 5 - 72 5 1 84 7 - 97 7 1 96 8 - 108 8 2 133 o - 145 1 2 145 2 - 157 2 1 157 3 - 169 3 1 181 4 - 193 5 1 Total 15 2 Source Chye Kooi Onn and Loh Wee Yet Industry in Perak 1974 The Tapioca Process1ng TABLE 5 10 Malaysia Provine e Peninsular Malaysia Perak Butterworth Kedah Sarawak Total V - 17 - Distribution of Starch and Pearl Factories 1982 Starch Pearl 4 3 4 3 Starch and Pearl 4 4 2 lO Source Federal Agricultural Marketing Authority Kuala Lumpur V - 18 - TABLE 5 11 Malaysia Export and Imports of Cassava Products Exports Imports Year Starch and Pearl Chips Starch and Pearl Chips (t) (t) (t) (t) 1955 7051 3460 1956 6645 883 1957 6455 443 1958 6418 80 1959 13 068 51 1960 16 625 12 1961 21 536 13 1962 18 128 neg 37 1963 22 140 89 1964 24 967 197 207 neg 1965 23 291 11 39 n a 1966 18 443 n a na 1967 16 483 neg n a n a 1968 18 527 na na 1969 20 379 21 281 2 1970 28 176 9 193 1971 17 295 53 727 25 1972 24 982 115 667 6 1973 26 116 800 2033 231 1974 18 289 156 2055 3807 1975 20 979 152 577 1269 1976 27 499 283 273 140 1977 10 831 320 268 8 1978 7 544 44 674 3232 1979 16 912 18 410 59 1980 5 942 5 3965 1981 5 663 n a 5711 n a Note Trade 1s Malaysia only and does not include Singapore Source Import and Export Trade in Food and Agricultura! Products Ministry of Agriculture V - 19 - TABLE 5 12 Malaysia Production Trade and Disappearance of Cassava Starch and Pearl 1971-82 Year Production lmports Exports Disappearance (t) (t) (t) (t) 1971 35 879 727 17 295 19 311 1972 46 872 667 24 982 22 557 1973 50 134 2033 26 116 26 051 1974 50 091 2055 18 289 33 857 1975 52 738 577 20 979 32 336 1976 68 085 273 27 499 40 859 1977 62 400 268 10 831 51 837 1978 57 588 674 7 544 50 718 1979 59 481 410 16 912 42 979 1980 49 828 3965 S 942 47 851 1981 48 929 5711 S 663 48 977 1982 48 517 103 1 331 47 289 Source Monthly Statistical Bulletin Department of Stat~stics Kuala Lumpur V - 20- Malaysian costs of production Sugar imports of 561 thousand tons in 1984 and a protected domestic sugar market offer scope for the development of a high fructuose sweetner industry based on cassava starch Moreover development of this industry requires relatively moderate investment since present starch processing factories can form the basis for an integrated starch-sweetner operation However domestic starch production is the l~iting factor in the development of this industry The other factor influencing recent production and export trends is increasing price competition from Thailand This price competition is amply portrayed in Figure 5 1 Befare 1976 wholesale starch prices in Ipoh Perak were well below Thai wholesale prices This coincided with the period of expanding starch production in Malaysia From 1976 to 1981 Malaysia starch prices in Perak were more or less on a par with Bangkok wholesale prices During this period Malaysia lost export markets even though prices in general were rising In 1981 Malaysian starch became more expensive than Thai starch and Malaysia become a net importer of starch The situation was compounded by a falling price level Thus after two decades of growth the Malaysia starch industry stagnated caught between the high supply price for roots and the prices of imported Thai starch For Malaysia to remain competitive in starch would require further cost reductions in the production of cassava roots The Domestic Animal Feed Market The development of the Malaysian livestock industry is typical of that of Japan Taiwan and South Korea in that to meet rising meat demand Malaysia has developed an intensive pork and poultry industry based on balanced feed rations these feed componente in turn are essentially imported In Malaysia's case the reason for import dependence rests with the export orientation of its agricultura! sector and its comparative advantage in tree crops The agricultura! economy continues to respond princ1pally to international rather than domestic markets and coarse grains are virtually not produced Thus Malaysia has met its growing demand for feed components through rapidly rising imports of maize Malaysia's animal industry never relied on a large production capacity at the village level essent1ally because there were limited grains or grain by-products available to sustain a large village-level animal population Swine production for example was usually associated with larger scale units linked to the by products of processing plants such as cassava starch planta The swine industry was thus the first to develop dependent only on the domestic Chinese market The principal growth occurred dur1ng the 1960 s as the industry switched to low-fat imported breeds and there was a sign1ficant increase of scale in production units (Hertrampf 1985) The majar growth in the poultry industry on the other hand occurred in the 1970's with the rise of intens1ve large-scale production systems Although the domestication of the chicken occurred in Malaysia not until the 1970 s did poultry start to become an important component in the diet The development of the livestock-feed sector over the last decade and a half (Table 5 13) demonstrates the dominance that the poultry sector can achieve even where the swine sector has already undergone significant technical change Part of this difference in growth is due to the larger V - 21 - Fl - - ¡::' .¡ I...JI.J: -:;: _ ¡ -S•:n CT :.,.J1 -- - 1 C•t· --.-t }\.~= ... .... ;:. -, ""r , ... 1 1 - - ....., .¡.,... F t --.::: \.P1 1 r ;_.;:_ ~ ,.. ... -1 '- - ..,, - 40]..,--------------------------~- ?:CI -.p.-, ... -- 50 j ' \ ' 1 J l I,]J. 31.¡ 1 "' 1 1 1 1 1 1 ' J 1 11 ¡' 1 1 : 1 1 ~ 1 1 ' ' 1 11 r' 1 1 1 1 ' .F' ..,( 1 ..,( :::r .., ...... j ...,( ..! 7 t:' ..., ..... .., ..., .., .-. ..., g .-. n ,.. " .- .- .-,.... ... - ¡ _ .~~ r.: o_ :-t :,_,_ :_ =-w. ~'E"'-.­• ¡...¡.. ;:. V -22 - market for poultry in Malaysia since pork consumption is restricted exclusively to the Chinese population The other factor however is the larger effic~ency ga~ns possible with poultry especially for the principal cost component feed These efficiency gains are further reflected in the location of the poultry and feed ration industry The poultry industry is are imported These two coincide in Kuala Lumpur Malacca and Penang where both the feed and poultry industries are concentrated Transport assembly and distribution costs are kept to a minimum The growth in production of balanced feeds over the per~od 1970-83 has been at a rate of 7 9% per annuam which is somewhat below the 10 4/ growth rate in feedgrain imports In fact feedgrain imports are larger than industrial feed production due to the growth in feed mixing by the animal production units More than half of feed production is ~n independently mixed in swine and poultry units Malaysia is already the largest feedgra~n importer in tropical Southeast Asia and with trends in livestock production likely to continue through the end of the decade feedgrain import levels will continue to increase relying on maize imports from Thailand Cassava has been used in the animal feed industry since the m~d-1960 s but its role has always been minor Use of cassava chips in animal feeds reached a peak of 23 thousand tons in the mid-1970 s but has since declined from that point (Table S 14) Although the market for feedstufs has witnessed tremendous growth the cassava chip industry has failed to respond The reason for this was the price squeeze between the price of roots which was determined principally by the starch market and the output price determined by the price of maize As shown in Figure S 2 the price of chips varied significantly in relation to the maize price from as low as 43% of the maize price in 1972 to as high as 86% in 1984 As implied by these statistics cassava chips because less and less competitive in feed rations over this period Cassava enters into least cost broiler rations -- the most exigent for cassava -- at about 68% of the maize price Through most of the 1970's cassava was competitive with maize in poultry rations Cassava use in animal feeds made two b~g jumps in 1972 and 197 S at periods when the cassava-maize price ratio was low Cassava sold at significant discounts to the maize price in these two periods in order to motivate ~n~tial use -- some adjustments in equipment are usually necessary to effectively utilize cassava in feed plants Cassava feed use stabilized from 1976 through 1979 as the cassava price remained at about 6S% of the maize price The year 1980 witnessed the sharp rise in starch prices and a resultant rise in root prices Even though cassava chip prices remained more or less in line with maize prices due to increases as well in the maize price chip production fell due to a lack of cassava roots and compet~t~on with the starch industry In 1983 the maize remained constant chip prices increased and the soybean meal price rose due to the initiation of tar~ff and import licensing to protect a nascent soybean crushing industry (U S D A 1986) Cassava chip prices became uncompetitive and production levels reverted to pre-1972 levels The cassava chip industry much like the starch industry was caught between a relatively high supply price for roots and ~ncreasing price competition from imports as the internacional maize price fell in the mid-1980 s 1 1 1 1 1 1 V -23 - TABLE 5 14 Malaysia Production Trade and Disappearance of Cassava Starch and Pearl 1971-1983 Year Production Imports Exports D~sappearance (t) (t) (t) (t) 1971 3658 25 53 3 630 1972 7145 6 115 7 036 1973 7371 231 800 6 802 1974 5765 3 807 156 9 416 1975 22 629 1 269 152 23 746 1976 16 842 140 283 16 699 1977 16 786 8 320 16 474 1978 17 oso 3 232 44 20 238 1979 16 606 59 18 16 647 1980 8972 5 8 967 1981 8 600 n a n a n a 1982 7 202 2 053 3 9 252 1983 4 039 1 639 5 5 673 Source Monthly Statist~cal Bulletin Department of Statistics Kuala Lumpur 1::" .-, - ..... V - 24 - r '~al J 'i Sl J: R J+•o jf Ccss J J•J Pt ICe ter l'u'lJJIZe Pt ICe o ~ ~------------------------------------------~ • 1 0.7 l " (' ~ " ; ~' ¡ ' ~ ! l ' ' o 6 0.5 YEH¡::·s V - 25- Pricing and Market Eff1ciency The Malaysian agricultural economy is driven by internat1onal commod1ty markets and the small cassava sector is no different Over the post-war per1od the Tha1 export pr1ce for starch has been the dominant influence on domes tic cassava prices (F1gure 5 1) since starch was the concentrated around the maJor population centers and the feed industry around the principal ports since the maJor portien of the feed components principal market and upto 1980 Malaysia was a net starch exporter What is of interest here is the influence of this market structure on formation of root and chip pr1ces The hypothesis is that starch prices -- set in the internacional market upto 1980 and in domestic markets after that point -- set with1n a compet1tive market environment w1ll together with processing costs and conversion rates determine root prices Thus regressing starch pr1ces 1n Perak on root prices in Perak yields the equation Root = 1 406 + O 1448 Starch (O 189) ( 0043) = 9049 The intercept term (in Malaysian Dollars per 100 kg) should measure the normal profits and processing costs and in this spec1f1cation should be negat1ve (see Chapter VII) The pr1ce transmission equation thus reflects low convers1on rates (6 9 to 1) and resultant operating losses This conversion rate is well below the 17 - 18 estimate for sed1mentation plants and 20 - 23 for centrifugal plants given by industry sources This difference in conversion rates would compensate for the losses in the operating margin Thus the price transm1ssion equation captures the nature of the price formation process but does not exactly distinguish the real values of the parameters 1n the profit equation Pr1ce formation 1n the chip market however is much better def1ned The hypothesis in this case in that the chip industry must take the root price as given In this case the est1mated equat1on 1s as follows Chips = 4 28 + 2 34 Roots (O 47) (O 06) 926 The equation reflects a techn1cally very efficient conversion rate -- in line with the high dry matter content of Black Twig -- and a competitive operating marg1n (US$17 per ton of chips) Chip producers thus face a highly compet1tive market situation caught as they are between the starch market and the maize market It is not surpr1zing then that chip production has not expanded g1ven both the low average profitab1lity and the uncerta1nty in the size of the operat1ng margin In fact many chip plants are an extension of a starch operation where the starch operators w1ll move into chip production when margins are adequate The issue currently fac1ng the Malaysian cassava industry 1s the impact of the shift to the domest1c market as the pr1ncipal determinate of cassava prices However this shift does not represent a break with the V -26 - international market but a widening of the band with Thai export prices as Malays1a shifts to being a net importer Thus in 1984 Malays1a imported 10 S thousand tons of starch from Thailand and in 198S S 1 thousand tons (Thai Tap1.oca Trade Assoe1ation 198S) In the 1980's it wl.ll be the starch import price that will be the principal determinant of price formation in the Malaysian cassava sector V - 27 - Conclusions Malaysia much like ThaLland has based its post-war agrLcultural economy on exports and yet in the 1980 s finds domestic markets for agricultural products reaching significant size due to rising incomes industrialization and urbanization Three export crops palm oil rubber and coconut make up 85% of cultivated area moreover Malaysia is by far the largest exporter of both rubber and palm oil and thus has a significant impact on world price levels An interesting policy question for Malaysia is the extent to which growth in the agricultural sector will continue to be based on a few export corps in which the country has a compatative advantage or whether attention should be turned to meeting rising domestic demand for a diversity of agricultural products Malaysia has much more flexibility in its agricultural policy than other Asian countries Export markets are well developed The population pressure on land does not exist since only 25% of the country's land area is cultivated Moreover transport infrastructure is relatively well developed Therefore it is somewhat ironical that in an agricultural economy where labor is the limiting constraint that an estimated 46% (in 1980) of the agricultural population falls below the official absolute poverty line Most of these agricultural households exist in the smallholder rubber and rice sectors Policy has been directed to resolving this poverty problem -- the incidence of poverty fell from 68% in 1970 to 46% in 1980 -- through two principal avenues through resettlement schemes in large land development proj ects and through production subsidies -- fertilizer credit agricultural chemLcals -- through farmer cooperatives Both avenues however focus on increased production of export crops -- and in certain cases rice -- as the means of generating Lncreased incomes Although Malaysia has been very successful with its export strategy this success now brLngs certain uncertainties because of its dominant market share Malaysia accounts for 44% of the world s rubber exports and two-thirds of palm oil exports Palm oil is quite substitutable with other vegetable oils but palm oil is now the major oil that moves in world trade and Malaysian palm oil makes up over 20k of the world vegetable oil market Future expansiona in production and exports must thus consider the impact of world prices and demand In this Malaysia has adopted a two prong strategy diversLfication and increased productivity in existing crops DiversificatLon has continued to focus on tree crops particularly cocoa and to a certain extent bananas and coconut However all of these are crops where there are a large number of competing exporters Malaysia s NatLonal Agricultural Policy Through the Year 2000 focuses on enchancLng its comparative advantage in tree crop exports through increased productivity with a particular focus on mechanization Malaysia's strategy is thus to capture a larger market share in principal exports Malaysia is aided in this by its proximity to the growing markets of Asia For example Malaysia has a transport advantage to the two largest vegetable oil Lmporters India and Pakistan Thus expansion Ln production area will be based on export crops but with an emphasLs on labor and land productivity Land allocation policy WLll continue to play a dominant role Ln the rate of expansion in production and as in the past will provide the government wLth some control over regulating future growth in export supplLes V - 28- The mark of th~s policy committment to export crops is the dropping of the long-held goal of self-sufficiency ~n rice ~n the National Agricultura! Plan Moreover planting crops for animal feed are discouraged in the plan Malaysia will thus rely on imports to service growing domestic markets lt is symptomatic of cassava' s future in Malays~a that it has turned from an export crop to supplying only domestic markets and in do~ng so has lost its ability to compete in international cassava markets Given Malaysia's agricultura! policy this fact would seem to seal the fate of cassava in the future of the country's agricultura! sector However the mere fact that a profitable cassava industry has operated in Malaysia throughout this century is some testimony to cassava's ~nherent productivity since cassava had to compete not with grain crops but w~th much more productive tree crops Cassava was disadvantaged by the production structure which favored centrally processed tree crops Cassava production is not well suited to plantation systems and yet smallholder cassava systems could not compete with smallholder tree crop production Cassava could have potentially competed within a semi-mechanized medium-scale (20 hectares) production system along the lines of that existing in Thailand This type of production scale seems to dominate in the illegal plantings in Perak Cassava was thus relegated to a particular niche in this land surplus agricultura! economy formed by pockets of smallholders with constraints on access to land The growing urban sector -- two-thirds of the growth of the rural labor force in the 1970 s was absorbed in the urban sector -- and the continued expansion of land development projects should continue in the future to reduce this niche In the end the future of cassava in Malaysia will depend on the international market for palm oil and rubber and in this Malaysia's agricultura! policies insure that the country will be a dominant player in these markets to the end of the century V -29 - Appendix S 1 A synthesis of production and utilization Collection of accurate production statistics for cassava in Malaysia is hampered by the illegal nature of a significant percentage of the area planted to the crop In consequence a suspected downward bias exists in estimates of area and production However since basically all the crop is sold for processing and data are collected on production of cassava starch and chips an alternative production series can he constructed (Table SA 1) The utilization series in fact is consistently higher than the root production series Since the downward bias in the production ser1es can be identified there is suffic1ent reason to suggest that the utilization series gives a much more accurate picture of cassava production trends in Malaysia The two series offer quite contrasting views of trends in cassava production The series developed by the extens1on department shows little trend and very substantial variability On the other hand the utilization series displays a steady increase 1n the first half of the 1970's toa peak of almost 450 thousand tons of roots in 1976 Production then declined to about 300 thousand tons in 1980 where it has remained through 1983 The latter series explains very well trends in exports and prices The utilization series is therefore considered as the best estimate of cassava production in Malaysia VI PHILIPPINES Inertia in Market Development Like Indonesia the Philippines is a mult~-island economy yet unlike Indonesia the Philippines has major population concentrations on all the maJor islands although Luzon still figures as the economic center The agricultural economy is dominated by two grains rice and maize and two principal export crops coconut and sugarcane Grain and food production in general are concentrated in the small farm sector while the export crops tend to be dominated by plantation systems although smallholder production of copra is also important The Philippines has an apparent comparative advantage in the production of copra and is by far the dominant exporter of this product This agricultural structure has created something of a dual approach to policy The export crops have attracted increasing government involvement since the early 1970's particularly as a source of tax revenue and as a means of controlling consumer prices at least for sugar and vegetable oil Moreover the government has attempted to stimulate the coconut industry to develop its own crushing capacity often with significant protection The government has generally reduced incent~ves to the export sector In the food sector on the other hand incentives have in general been positive Three themes run through agricultural policy for grains a commitment to self-suffic~ency in grain production apart from wheat very heavy intervention in setting domestic prices and commitment to increasing productivity in the smallholder sector The achievement of self-sufficiency is seen as be~ng dependent on price policy and small farm programs Control over domestic prices is in the hands of the National Food Authority (NFA) which has authority to control imports and exports to huy in the domestic market and to set both support prices and ceil~ng prices Trade in foodgrains and domestic prices as a result are to a large extent administratively determ~ned Policy toward the small farm sector on the other hand has included land reform investment in irrigation infrastructure and specialized credit and extension schemes The stage was thus appropriately set for the advent of the high yielding rice varieties Under the Masagana 99 Program the Philippines went from a consistent net importer to a net exporter of rice in the mid-1970's This success has led to the recent development of the Maisan 99 Program which hopes to achieve self-sufficiency in maize in three years Concern also runs to the large and growing wheat imports and identifying means of either controlling such imports or substituting for wheat flour Cassava fits well into th~s policy context The crop is essentially grown by smallholders although some plantation production does exist Moreover cassava can be a domestically-produced substitute for imported grains This concern for self-sufficiency has even extended to the development of a national alcohol program based on sugarcane and cassava however with the recent fall in world oil prices the program has been scrapped Nevertheless cassava is seen as a crop that can contribute to meeting the increasing demand for carbohydrate sources Since cassava is only a very m~nor crop in the Phil~ppines and since the crop has received VI - 2 - little government support the question to be pursued is what difference government involvement can make in developing cassava as a commercial crop in the Philipp~nes Production Production trends and distribution The official production series for cassava in the Philippines ~s presented in Table 6 1 The series shows relatively stable area production and yields from 1960 to 1974 followed by very dramatic increases in both area and yields Such increases led to more than a tripling in production in three years and to over a quadrupling in five years This remarkable growth immediately begs the questions of what was responsible for this sudden take-off As is discussed in the section on markets and demand there is no corroborating evidence on either consumption or price levels to suggest that such production increases took place On the other hand alternative estimates of area and yield are limited The agricultura! census of 1971 estimated cassava area at 47 061 hectares yields of 5 75 t/ha and production of 270 714 tons Even at this stage there were maJor discrepancies between the census estimate and the Bureau of Agricultura! Economics (BAE) estimate The major difference between the two production estimates is due to the reported area figures the yield estimates are similar at this date This discrepancy with the census figure raises some doubt about the adequacy of the sampling and estimation techniques for cassava estimates This is not surprising given that cassava is such a minar crop in the Philippines The only data which correspond to the BAE' s estima te of increas~ng yields from 1976 to 1979 is the Special Study Division s survey of 901 cassava farmers in the period 1977-79 Average yields for this non-random sample were 4 3 t/ha however this average was biased downward somewhat because the ma]or growing area of Central Mindinao was not included in the survey However even this would not raise yields to the BAE estimate of 11 7 t/ha A regional breakdown of production and area provides insight into the regional locus of this supposed growth in cassava production (Table 6 2) Cassava is produced throughout the Philippines but most is produced in the southern islands There is little production on Luzon apart from the Bicol region lying at the southern tip of the island The major producing areas are the Visayas region and Mindinao The production data suggest that cassava production increased at an annual rate of 20 4% on the island of Mindinao in the period 1970-81 while increasing ~n the rest of Philipp~nes at a 9 6% annual rate Mindinao accounted for 78% of the increase in cassava production ~n the period The years 1975 and 1976 are particularly striking Production in 1975 was 134 thousand tons and in 1976 656 thousand tons This increase almost doubled national production In a single year area increased from 20 to 44 thousand hectares and yields from 6 8 to 14 8 t/ha In JUSt the Central Mindinao region production ~ncreased from 14 thousand tons in 1975 to 1 1 million tons in 1979 These data suggest either Table 6 1 Phlllpplnes Area Productlon and Yleld of Cassava, 1960-1981 Crop Year Area Productlon (ha) (tons) 1960 79,460 442 413 1961 100 310 546 611 1962 92,980 494 805 1963 80 280 457 769 1964 93 540 596,156 1965 93 280 645 720 1966 89,700 614,386 1967 86,520 528 727 1968 83 880 481,928 1969 85 690 482,327 1970 82 620 442 223 1971 81 820 427 055 1972 82,680 439 697 1973 87 420 444 710 1974 96 710 480,015 1975 119,310 684 507 1976 144 650 1 153 958 1977 179 270 1,710 767 1978 181 770 1 781 961 1979 192 360 2 253 824 1980 204 190 2 277 338 1981 211 370 2 255 115 Source Bureau of Agrlcultural Econamlcs publlshed ln Natlonal Econamlc and Developnent Authorl ty Phlllpplne Statlstlcal Yearbook Manlla 1981 Yleld (t/ha) 5 57 5 45 5 32 5 70 6 37 6 92 6 85 611 5 74 5 69 5 35 5 22 5 32 5 09 4 96 5 74 7 98 9 54 9 80 1172 11 15 10 66 Table 6 2 Ph1ll.pp1nes Area ProductJ.on and YJ.eld by RegJ.on 1972-81 Cagayan Central Southern Western Central Eastern Western Northern Southern Central Year Ilooos Valley Luzon Tagalog BJ.=l VJ.sayas VJ.sayas VJ.sayas MJ.ruhnao Ml.nchnao MJ.ndmao MJ.ndJ.nao Area (000 ha) 1972 1 2 1 6 1 o 5 7 15 4 5 3 13 o 10 2 111 4 1 4 2 9 9 1973 1 2 1 7 1 o 6 2 16 4 4 5 14 6 10 7 12 2 5 5 9 5 3 9 1974 1 9 o 7 o 9 5 8 25 8 4 3 25 7 16 9 1 1 6 2 4 o 3 5 1975 1 9 1 2 o 8 7 2 33 4 5 3 25 9 23 8 1 3 7 7 5 9 5 o 1976 1 9 1 o o 9 8 2 27 3 7 8 23 9 29 3 10 2 12 8 6 6 14 6 1977 2 1 1 1 1 1 8 5 27 7 10 7 28 6 31 o 20 4 138 7 5 26 6 1978 2 2 1 o 1 1 8 2 27 8 10 2 28 6 24 4 23 9 16 o 9 3 29 6 1979 2 3 1 o 1 1 7 9 28 8 10 7 29 6 25 9 22 9 23 4 9 o 29 o 1980 2 4 o 9 1 5 8 5 32 4 115 30 7 28 3 23 3 26 4 8 8 29 4 1981 2 3 o 9 1 6 8 4 33 3 12 o 38 o 27 4 25 8 24 1 8 5 29 2 Productwn (000 t) 1972 9 7 14 7 5 o 33 7 63 3 25 4 39 7 57 o 7 8 25 5 37 7 56 2 1973 10 4 14 6 5 6 38 9 61 3 22 2 33 1 53 5 78 o 47 8 60 3 19 2 1974 9 8 6 8 4 2 54 9 139 4 23 9 54 o 52 5 59 56 7 41 5 30 5 1975 111 6 1 4 6 54 2 237 6 30 3 85 2 120 8 8 5 770 34 7 14 2 1976 18 3 3 1 2 9 42 3 220 6 39 2 86 9 84 3 190 9 50 8 40 9 373 8 1977 16 3 3 3 2 2 46 1 230 6 42 2 92 8 98 2 349 9 56 9 40 7 732 5 1978 16 3 2 7 2 3 44 o 269 8 30 8 94 8 114 2 333 8 67 9 42 5 762 8 1979 17 4 5 1 3 5 40 6 308 7 44 6 116 5 1160 297 o 129 5 48 o 1126 9 1980 18 4 3 9 4 5 43 1 293 o 60 8 89 5 126 9 303 6 153 2 53 6 1125 2 1981 16 8 4 4 4 6 44 o 287 o 64 3 75 3 1335 325 o 135 3 47 2 1117 8 YJ.eld ( tjha) 1972 7 91 9 29 5 00 5 93 4 12 4 76 3 06 5 60 o 70 6 15 8 99 5 66 1973 8 36 8 36 5 44 6 31 372 4 93 2 26 5 00 6 40 8 69 6 35 4 89 1974 5 26 9 45 4 73 9 45 5 40 5 61 2 09 311 5 58 9 09 10 42 8 59 1975 5 79 5 15 5 85 7 56 711 5 66 3 29 5 07 6 67 10 04 5 93 2 81 1976 9 29 3 15 3 36 5 16 8 06 5 02 3 63 2 87 18 66 3 96 6 17 25 65 1977 7 65 2 95 2 02 5 40 8 27 3 95 3 24 3 16 17 14 411 5 42 27 50 1978 7 51 2 58 2 03 5 34 971 3 02 3 31 4 67 13 95 4 23 4 58 26 25 1979 7 46 513 3 16 513 10 70 4 17 3 93 4 47 12 96 5 54 5 31 38 07 1980 771 4 32 2 93 5 06 9 03 5 29 2 91 4 47 13 02 5 81 6 08 38 29 1981 7 31 5 10 2 88 5 27 8 61 5 36 1 98 4 87 12 60 5 62 5 54 38 30 So urce Bureau of AgrJ.cultural E=norrucs VI - 3 - explosive structural change in cassava production on MindJ.nao or a maJor revision of the data The starch industry based on plantation systems is concentrated on Mindinao but the data on cassava starch productJ.on suggest no major changes in the industry in 1975-1980 Thus it appears that this major increase in cassava production in the last half of the 1970 s was in maJor part artefact (Independent comparison of production data with the utilization data is found in Appendix 6 !) Cassava production systems Cassava in the Philippines is grown in both plantation and smallholder production systems There are few estimates of the percentage of cassava grown in these two systems However plantation systems are associated only WJ.th starch mills and at least three factories on Mindinao and one in Eastern Visayas operate estates As much as 6 500 hectares may be grown in plantation systems This would imply that the greater portien of cassava is grown by smallholders Smallholder systems will thus be considered J.n most detail Cassava while it is grown throughout the Philippines has never achieved the status of a major commercial crop even on a regional basis Maize is the most prominent upland crop for smallholders The reason for this follows principally from the relatively favorable agro-climatic conditions that exist throughout the Philippines and the relatively universal distribution of paddy lands across the dl.fferent regions A short maturity crop which produces relatively consistent y1elds under upland conditions fits better than a long maturity crop in smallholder systems especially since rice production requires substantial resources during critica! periods of the year In general shortage of rainfall is not a l1mit1ng factor in cassava production nor for the production of other upland crops Because of cassava s better adaptation to poorer soils cassava is often found on the more infertile hillside areas Cassava is planted throughout the year and the only constraint on plantJ.ng time is conflict with rice production activities Such constraints are accentuated because very lJ.ttle hired labor is used in cassava production In the Spec1al Studies Division (SSD) survey about 75"/ of labor use in cassava comes from family labor (Table 6 3) Cassava producers according to the SSD survey operate farms of a little over 3 hectares of which only 6 of a hectare is devoted to cassava Rarely are plots of over 2 hectares planted and of the 916 farmers in this survey only about 40"/ actually owned their land Yet even on cassava producing farms only about JI% of total cash income was derived from cassava Other crop sales accounted for far more income than cassava even though over 80% of the cassava that was produced was sold Cassava was thus grown as a minor cash crop by essentially small-scale producers on land not typ1cally suited for other crops Land is typically prepared by animal traction although some small plots may be prepared by hand Because of the relatively h1gh rainfall the land is either furrowed pr1or to planting or ridging l.S done at the time of the first weeding usually by interrow an1mal cultJ.vation Ridging l.S apparently necessary to control root rot as the crop matures This type of Table 6 3 Ph11l pp1nes Type of Labor Used 1n Cassava Product1on by Reg1on (man days/ha) Reg10n H 1 red 1 ~a 1 d 1 n Cash K1nd O~erator Fam1l~ Exchange Total I 1 ocos 3 7 - 24 4 11 6 o 2 39 9 Central Luzon 4 5 - 28 o 11 5 15 o 59 o Southern Tagalog 15 o - 24 9 25 9 - 65 8 B1col 14 2 - 24 o 25 o o 3 63 5 Western V1sayas 3 5 o 3 14 1 8 o o 3 26 2 Central V1sayas 12 2 21 8 17 5 13 7 - 65 2 Eastern V1sayas 22 8 - 26 6 103 3 2 62 9 Western M1nd1nao 14 9 39 o 16 8 1 3 720 Northern M1nd1nao 8 5 - 29 9 10 2 o 8 49 4 Average 11 1 2 8 24 8 15 6 o 7 54 9 Source E B MeJ1a et al Cassava Soc1o-econom1c and Market1ng Study Ph1l1pp1nes, Spec1al Stud1es D1v1s1on, M1n1stry of Agr1culture No 79-26 Oct 1979 VI - 4 - weeding limHs any type of intercropping and cassava is usually found planted in monoculture Although a substantial range of varieties are found in the Philipp1nes --the SSD survey found 22 d1fferent varieties-- about half the farmers in the survey grew a variety named white while two-thirds of farmers grew either 1 white or 1 yellow 1 (Table 6 4) These variet1es are apparently selected for their good eating quality The one peculiar feature of cassava production systems 1n the Philippines is the very low labor input devoted to weeding (Table 6 5) This partly reflects the use of animal cult1vation but an1mals can be used at most twice for weeding and are often ineffective at controlling weeds with1n the rows Moreover weed control would be expected to be a problem under such relatively high rainfall conditions Low labor input for weeding thus reflects other factors including the reliance on family labor competition with other crops for labor resources and the relatively low commercial status of cassava This same phenomenon applies to other input use In the survey only 18 of 916 farmers or 2 percent used fertilizer on their cassava plots For those farmers who did apply fertilizer the average application rate was about 125 kg/ha of chem1cal fertilizers For smallholder cassava production cash expenses were kept to very low levels which may reflect the risky nature of marketing the crop The riskiness is as well reflected 1n harvesting patterns Cassava 1n general in the Philippines can be harvested anytime after s1x or seven months Farmers in general harvest in small lots partly for home consumption but principallv as a means of insuring d1sposal at a remunerative price in the market Substantial labor is as well expended on trimming cleaning and packing the roots for sale At least one study has shown that there is no loss in yield when harvesting in small lots between 6 and 9 months as compared to a single harvest at n1ne months (Villamajor 1980) -- the border effect may act as a yield compensation mechanism Cassava plantation systems in the Philippines are normally in the range of one to 1 5 thousand hectares in size Planting and harvest are staggered to provide a continuous supply of cassava to the starch factories This production is as well supplemented by purchases from smallholders However in such large estates it has been difficult te achieve any significant economies of scale in cassava production The only significant changes are that land preparation is done by tractor rather than by animal traction and that herbicides are used in weed control The rest of the operations are performed by hand labor usually on a piece rate by farmers contracted in the area A 1978 survey of starch plants suggested that the higher overhead costs resulted 1n substantially higher own production costs as compared to purchase prices from local farmers - 249 pesos/t versus 174 pesos/t (Villanueva and Laguna 1979) Yields Compared to standards elsewhere in Asia cassava yields in the Philippines are low even though agro-climatic condHions are in general more favorable The 1977-79 survey of 916 smallholder found an average Table 6 4 Ph1l1pp1nes Cassava Var1et1es Reportedly Grown on 916 Farms 1976-1979 Van ety Re910n Golden Java 1 Wh1te Yellow Red Nat1ve Yellow Hawa11an Brown Other I 1 oc os Centra 1 Luzon Southern Tagalo9 B1col Wes tern V1 sayas Central V1sayas Eastern V1sayas Northern M1nd1nao Western M1nd1nao Total Farms % Farms 105 36 29 13 27 35 61 48 72 426 44 36 8 45 41 42 172 18 1 Includes 15 other var1et1es 86 86 9 1 14 57 7 79 8 29 9 37 75 8 6 7 13 1 8 3 11 1 3 5 13 27 46 10 5 7 116 11 Source E B MeJla et al Cassava Soc1o-Economlc and Market1ng Study Ph1l1pp1nes Spec1al Stud1es DlVlSlon M1n1stry of Agr1culture No 79-26 October 1979 harvest~ng and marketing cassava crop VI - 6 - Little labor is expended on maintenance of the The impression ~s that resources with a low opportunity cost are principally employed in cassava family labor and animal power in the slack seasons and either marginal land or excess land which cannot be planted to more labor intensive crops given the stock of family labor Scarce resources such as capital are used only when absolutely necessary Cassava is able to yield under such extensive conditions although not at high levels If this is so then the costs of production derived by the SSD may be overestimated since family labor and land were costed at average market prices Just less than 80% of var~able production costs is made up by labor charges (Table 6 6) of the wage bill 70% is imputed to family labor The rest of variable costs are principally delivery and transport charges and for the 19% of farmers who were share tenants the payment in kind to landlords The other principal cost is the interest charged against fixed assets devoted to cassava In the SSD study land was not costed at its rental value but rather as an interest payment (12%) on ~ts value This interest charge to land forma the other major cost component For per hectare production costs there is a certa~n stability in total cost across the different regions What is substantially more variable between regions is yield levels and this resulta in a substantial variability in per ton production costs from 160 pesos/t in Western Mindinao to 338 pesos/t ~n B~col In fact four of the nine region were producing cassava at a higher product~on cost per ton than farmers were receiving as a market price (Table 6 6) However in all cases except region VIII cash income was greater than cash expenses Costing indigenous farm resources at their opportunity cost could make cassava profitable in these other regions as well However what ~s strik~ng is that farm-level prices to a substantial degree natched production costs and that profit or loss depended critically on y~eld level A yield less than 3 S t/ha was just not remunerative at least when costed at market prices Technology development Designing appropriate technology for cassava ~n the Philippines will not be an easy task since the procesa is dependent on answers to several unknowns The basic question is why cassava is grown in such extens~ve production systems when the average farm s~ze of cassava producers is Just over 3 hectares If cultural practices are the principal constraint on yields modifying cultural practices is going to require either providing farmers with further incentives to grow cassava (either higher prices or more assured markets) and/or relieving what may be significant resource constraints within the farm Answers to these questions can only come from a more extensive study of cassava within the complete farm system Moreover although cassava is clearly a commercial crop in these systems what is not clear is the type of market toward which increased production can be directed The two issues of farm~ng systems and markets together define the appropriate design parameters for the development of ~mproved technology Table 6 s Ph1l1pp1nes Labor Use Farm S1ze and Average Cassava Area 1n Cassava Product1on Systems 1977-79 Re 10n I III IV V VI VII VIII IX X Average Labor Ut1l1zat1on (man days/ha) Land Preparat1on 116 20 o 21 9 27 o 10 8 10 8 22 4 16 9 16 3 17 6 Furrow1 ng 2 8 2 2 1 1 3 9 o 2 2 o 3 4 2 6 1 5 2 2 Planting 10 4 6 1 10 5 7 3 5 o 8 5 10 2 8 8 6 8 8 1 Weed1ng 3 6 5 2 11 1 14 9 2 9 5 9 14 o 19 2 6 3 9 5 Harvest1ng 5 9 6 3 15 7 7 8 5 3 27 8 8 7 9 2 7 5 9 8 Pack1ng and Transport 6 7 4 2 4 6 1 9 2 o 1 8 3 9 5 7 10 o 4 4 Peel1ng and Dry1ng - - - - 8 3 - 4 2 1 o 1 3 Total 41 o 44 o 64 9 62 8 26 2 65 1 62 6 66 6 49 4 52 9 Farm S1Ze (ha) 2 25 2 25 2 93 372 4 29 2 82 2 38 3 15 2 50 3 03 Cassava Area (ha) o 65 o 54 o 60 o 79 o 49 o 85 o 47 o 58 o 52 o 61 So urce E B MeJla et al "Cassava Soc1o-econom1C and Market1ng Study Ph1l1pp1nes Spec1al Stud1es Dev1s1on M1n1stry of Agr1culture No 79-26 Oct 1979 VI - S - yield of 4 02 t/ha (Table 6 6) a figure comparable to the pre-1975 BAE estimates of around S t/ha There was some variation in yields between regions but in general yields were uniformly low throughout the Philippines The immediate question is why espec~ally if agro-climatic constraints (except for soils) are not an ~ssue Since the Philippines has had no cassava research program until just recently a potent~al cause of low yields may be the lack of well adapted high yielding varieties The principal evidence that may be brought to bear on this hypothesis is that the first varietal releases by the Institute of Plant Breeding (Lakan 1 and Datu 1) were selections that went by the more common names of golden yellow and Hawaii S These varieties were already being grown by farmers (Table 6 4) and yet the yield trials prior to release of these variet~es gave an average yield of 42 t/ha for Datu 1 and 32 t/ha for Lakan 1 Lack of adequate cultural practices thus appears to be the principal constra~nt on yields Two principal factors appear to be involved lack of appropriate soil fertility management and insufficient weed control As in other parts of Asia (except India) diseases and pests do not appear to be a major problem in cassava apart from the occasional incidence of cassava bacteria! blight One other possible limiting factor is lodging given the frequency of high winds in the Philippines Of these factors the very limited labor input in weed control ~s probably the major constra~nt on higher yields Overcoming this constraint requires a closer study of labor utilization on the farm and the value of the production gain from further labor inputs in weeding of cassava Yields on plantations are considered to be substantially higher although there are practically no published reports of yield levels on estates One estate on Mindinao reports average yields of 18 t/ha (field notes 1982) There is continuous planting of cassava on this estate and apparently there has been problems in maintaining yield levels Yields on newly opened land without fertilizer averaged about 30 t/ha Yields have declined from this level and stabilized around the 18 t/ha average while at the same t~me fertilizer application increased from zero to 400 kg and f~nally to 600 kg/ha On another estate in Eastern Visayas the maximum yield obtained in large fields was 29 t/ha on former r~ce land w~thout fertilizer applicat~on (field notes 1982) On th~s same estate as a whole average yields are in the neighborhood of 20 t/ha with the flat former sugarcane land averaging 25 t/ha and the hilly areas averag~ng 10-15 t/ha Cost of product~on and labor utilization If cultural practices are a pr~ncipal constraint on yields this should be reflected in low rates of labor utilization Labor input in fact is very low (Table 6 S) even by Thai standards where land preparation is performed by tractor At an average of 53 mandays/ha the cassava plots can only be quite extensively managed unless purchased inputs that substitute for labor are used and this ~s not the case The extensive nature of cassava cultivation is particularly reflected in labor expenditure for weeding In more usual labor profiles for cassava weeding usually forms the largest single activity In the Philipp~nes most of the labor is utilized in land preparat~on and planting and secondly in Table 6 6 Ph1l1pp1nes Per hectare Product1on Costs Y1elds and Costs per Ton 1977-79 Cost Item Var1able Costs Labour H1 red Food Fam1ly Land Preparat1on Tractor Amma 1 Plant1ng Mater1al Fert1l1 zer Landlord In k1nd Cash Transport Interest 1 (Work¡ng Cap1tal)l/ Sub- tota 1 F1 xed Cos ts Deprec1at1on Repa1r Interest Y Sub-total Total Costs Y1eld (t/ha) Cost per ton Farm Pr1ce Reg1on I III IV V VI VII VIII IX X - -------------------(pesos/ha)---------------------------- 29 1 26 6 10 4 1 o 288 2 322 6 15 6 1 5 - - o 1 3 4 28 5 8 7 232 2 41 9 73 2 40 9 18 8 688 2 444 2 19 2 5 7 322 1 347 o 28 2 21 3 470 9 520 4 103 5 124 8 28 o 2 1 10 3 280 2 363 4 165 9 o 5 o 9 o 6 o 2 o 9 16 8 17 2 14 9 3 6 14 1 16 8 7 9 414 6 524 8 232 2 24 2 13 9 447 5 485 6 20 4 12 5 2 9 16 5 293 5 344 6 316 8 373 7 181 6 10 1 179 2 32 o 5 6 31 3 19 6 167 o 56 9 267 9 2 7 33 2 2 113 3 51 8 368 8 23 5 13 1 12 3 18 9 75 1 Q 2 266 2 3 4 52 8 4 6 35 9 19 4 28 3 27 7 22 4 479 6 556 1 629 4 469 7 30 2 3 4 386 1 419 7 15 5 3 6 227 3 246 3 110 6 1 217 7 234 8 8 2 21 1 271 7 301 o 1035 1 964 6 900 1 841 5 605 8 899 3 802 4 864 2 770 7 6 19 5 84 3 36 2 49 2 21 5 46 2 16 5 39 4 03 167 2 250 165 2 260 267 9 338 o 274 1 190 230 250 164 7 317 5 160 3 191 2 190 300 240 220 Average 98 8 15 6 282 8 7 o 4 2 o 1 o 1 23 3 30 7 21 1 21 7 505 5 18 9 9 1 325 2 353 1 858 6 4 02 213 6 230 1/ Interest on cash expenses w1th 1nterest rate of 12% 2! Land costs for land owners 1ncluded as 1nterest on land value 1 e ¡mpl1c1t land rent lS 12r of - land value Source E B MeJla et al Cassava Soc1o-economlc and Market1ng Study Ph1l1pp1nes Spec1al Stud1es DlVlSlon M1n1stry of Agr1culture No 79 26 Oct 1979 Table 6 7 Var~ety PR-Cl3 PR-C24 PR-C62 P~l~pp~es cassava Var~et~es Selected for Release by the P~hpp~ne Root Crop Research and Tra~rung Center Months to harvest 10-12 8-10 10-12 Y~eld (t¡ha) 42 43 46 Dry matter (%) 34 39 33 Source The Rachx VolUI!E 2 (1) Jan-June 1980 Table 6 8 Ph1 l1pp1nes Annual Per eap1ta Food eonsumpt1on Patterns by Reg1on 1977-1980 Reg1on llecos eagayan Va lley e entra 1 Luzon Metro Man1la S Luzon B1col W V 1 sayas e V 1 sayas E V1sayas W M1nd1nao N M1nd1nao E M1nd1nao e M1nd1nao Ph 1 l1 pp 1 nes Sweet R1ce Ma1ze Wheat eassava Potatoes {kg/caplta) (kg/caplta) (kg/caplta) (kg/caplta) (kg/caplta) 139 8 101 2 120 1 103 4 118 o 1140 120 7 45 6 104 7 82 o 77 5 101 4 113 4 105 8 3 20 4 1 6 1 6 1 3 3 o 7 5 83 2 19 9 25 o 54 9 28 7 12 7 17 7 7 7 6 9 8 8 17 3 10 8 7 5 6 o 7 1 7 4 6 2 6 9 7 o 8 o 8 5 6 8 o 2 o 4 1 6 4 9 6 o 7 6 5 4 5 2 9 8 9 5 3 5 6 2 5 7 2 o 2 o 2 6 15 6 4 3 6 7 15 9 8 5 6 4 7 1 7 4 6 5 Source Av1guetero et al 1981 VI - 7 - There had been little research on cassava in the Phil~ppines until the formation in 1977 of the Philippines Root Crop Research and Tra~n~ng Center (PRCRTC) The center is located on the campus of the Visayas State College of Agriculture and besides a staff of 15 researchers the center draws on the staff of the College to assist on research proJects Besides cassava the center does research on sweet potatoes yam and taro There ~s no cassava program as such since the different discipl~nes d~vide their t~me between the different root crops except for a breeder whose sole responsibility is cassava breed~ng Research on cassava extends from breeding through crop protection and management to post-harvest util~zat~on The center in its few years of operation has pr~ncipally been involved in defining research strategy and research priorities between root crops Research by each d~scipline is def~ned on a proJect basis wh~ch can be influenced by outs~de funding espec~ally the funding from the Philippine Counc~l for Agriculture and Resources Research (PCARR) Policy development can have a marked influence on research d~rect~on such as was the case w~th the abortive alcohol program The center still is in the process of completing the development of a fully structured breeding selection and varietal test~ng program A germplasm bank has been assembled and evaluated and at least three selections have been suggested as recommended varieties for release (Radix 1980) A crossing and selection program has been started The breeding focus is on higher yield with starch content being a secondary object~ve This program ~s complemented by some cassava breeding wh~ch is done at the Institute of Plant Breeding (IPB) at the University of the Philipp~nes at Los Baños A national varietal testing system has recently been set up with var~etal input from PRCRTC IPB-UPLB and the Bureau of Plant Industry Tr~als are carried out on six different experimental stat~ons Definition of the potential yield gap that may be exploited remains as yet relat~vely undefined The yield data on the first three select~ons released by PRCRTC (two are already grown by farmers) show the almost traditional yield of prom~sing varieties under experimental conditions of over 40 t/ha (Table 6 7) Defining what potential yield levels are at the farm level is more difficult as well as the even more cr~tical quest~on of how to increase farm-level yields within farmer resource availabil~t~es What probably can be sa~d ~s that a target of 15 t/ha for smallholders is realistic which for the Philippines amounts to a tripling ~n average y~elds Markets and Demand Cassava is grown throughout the Philipp~nes but only ~n Central Mindinao may it be said to be a maJor crop Moreover product~on tends to -be larger in areas where there is access to well developed markets In the Philippines cassava appears to be constrained by what could be termed market ~nertia That is production ~ncent~ves are weak due to poorly developed markets for cassava leading to extensive production systems and low y~elds In turn high per ton production costs prov~de l~ttle ~ncent~ve for further market development Def~n~~g the mechanism for VI - 8 - break1ng this inertia requires an evaluation of the present and potencial markets for cassava in the Philippines Cassava for direct human consumption Where cassava is consumed as a food source in trop1cal Asia it is usually in areas where there is a 'shortfall in rice availabilities e1ther because of limited purchasing power and/or 1nsuff1cient production levels Cassava has not been incorporated as a maJor component 1n the Philippine diet because rice production is in general relatively evenly distr1buted throughout the islands and in regions where rice suppl1es are short carbohydrate requirements are supplemented by maize (Table 6 8) Moreover consumption of wheat products has steadily increased in the post-war period and has reached quite s1gnificant levels in urban areas Root crops are of secondary importance as carbohydrate sources in the diet with cassava and sweet potatoes being of more or less equal rank There is some difference between sources in estimates of actual consumption of cassava Bennagen (1982) reviews these estimates (Table 6 9) and finds an average annual per capita consumption lying somewhere between 4 and 9 kg The locus of cassava consumption is essentially off-Luzon in the southern islands (Table 6 8) and in rural areas (Table 6 10) St1ll even 1n the high consuming areas cassava is still of only secondary importance in the diet Cassava consumption in general coinc1des with the consumpt1on pattern for maize Thus rural households eat twice the amount of less-preferred staples (maize and root crops) than urban households (Bennagen 1982) There is something of a duality 1n consumption forms for cassava In most rural areas cassava is consumed as a caloric staple The roots are either cubed and steamed in the same manner as rice 1s prepared or peeled and boiled Prepared and eaten in this way cassava is a subst1tute for r1ce On the other hand the roots is milled fresh and used to produce a type of cake or other processed snack items The latter 1s probably the principal form 1n which cassava is consumed in urban areas and reflects the fact that the price of cassava is much higher in urban compared to rural areas Demand for cassava should behave more as a caloric staple in rural areas and as a vegetable crop in urban areas The staple nature of cassava demand 1s reflected in the seasonality of prices and consumption In the main rice growing areas on Luzon there is little seasonality to either cassava prices or consumption and consumption levels are relatively low However to the south in Visayas and Mindinao where there are shortfalls in r1ce production there is a more seasonal pattern to both prices and consumption (Table 6 11) On Mindinao cassava consumption sends to be highest in September while on V1sayas it tends to be h1gher in March These are per1ods which l1e outside the rice harvest which occur principally 1n the May-June period and in December Cassava consumption tends to be lowest 1n the main rice harvest in December There thus appears to be substitution between rice and cassava depending on ava1labilities Th1s suostitunon oy rice and the fact that rice is the preferred staple 1s fully reflected in demand parameter estimates for cassava (Table TABLE 6 9 Philipp1nes Comparison of Data for Average Per Capita Consumption of Basic Staples 1978 Food Group Cereals and cereal products Rice Corn grits Wheat and wheat products Starch roots and tubers Sweet potatoes Cassava FNRI (kg) 134 o 109 5 13 9 7 7 13 5 5 1 5 5 SSD (kg) 135 7 107 9 14 7 8 9 18 2 9 9 4 o IAPMP (kg) 148 3 109 5 24 7 11 6 9 9 9 3 Source Eugenia Bennagen 1982 Staple Food Consumption in the Phillippines TABLE 6 10 Philippines Average Per Capita Consumption of Starchy Staples by Urban/Rural Residence and by Island Group 1978 Residence Island Group Philippines Urban Rural Luzon Visa y as Mindanao (kg) (kg) (kg) (kg) (kg) (Kg) Cereals and Cereals Products 134 o 117 9 142 4 131 o 139 10 137 2 Rice and Products 112 4 97 1 120 1 118 2 103 3 102 6 Maize and Products 13 9 6 6 17 9 2 6 31 4 31 4 Other Cereals 7 7 14 2 4 4 10 2 4 4 3 3 Starchy Roots and Tubers 13 5 7 3 16 8 10 2 14 2 26 6 Cassava and Products 5 5 1 5 7 3 2 2 8 o 15 o Sweet Potato 5 1 3 3 6 2 5 5 3 3 6 9 Potato and Products o 7 1 1 o 4 o 7 o 7 1 1 Others 2 2 1 5 2 9 1 8 2 9 4 o Source First Nationwide Nutrition Survey Food and Nutrition Research Institute (FNRI 1978) VI - 9 - 6 12) The elast~city estimates ~n general suggest that cassava is an inferior good i e that consumption actually declines with increasing income and that there is a very strong substitution between cassava and rice and to a more minor degree substitution between cassava and maize These results conform to expectat~on and co~ncide with results for the other less-preferred staples Ma~ze in fact appears to be even more inferior a good than cassava (Bennagen 1982) These demand parameters underlie trends in consumption of basic staples in the Philippines (Figure 6 1) The trend in per capita consumption of rice has been relatively constant with a marked tendency for there to be less year-to-year variability The principal effect of the high-yielding rice varieties has not been on average consumption levels but rather to shift the Philippines from a net importer to a net exporter of rice The constancy in consumption could represent an increase in consumption by the poorer income strata and a decrease by the higher income strata However Bennagen (1982) presents data that does not support this Also there was a shift in relat~ve prices of rice in relation to the non-preferred staples The effect in the 1970's has been to induce a declining trend in per capita consumption of both cassava and maize Maize consumption ~n fact has declined more rapidly than cassava consumpt~on The Philippines food economy appears to be reaching that stage where there is a divers~fication in the diet away from a basic dependence on caloric staples The fresh food market is currently the dominant market for cassava in the Philippines In the best of circumstances it is difficult to build a relatively expansive production base purely dependent on the fresh food market Given the long history of cassava ~n the Philippines it is h~ghly unlikely that cassava will ever develop as a maJor staple In part this was because agroclimatic conditions were not peor enough to favor cassava in any part of the Philippines maize a short cycle crop could always be grown as a secondary staple to r~ce Recent trends in consumption of non-preferred staples including cassava indicate limited future growth in this market Developing cassava as a major commercial crop will thus depend on the development of other alternative markets for the crop The starch market The principal existing alternative market for cassava in the Philippines is for starch production Cassava starch production through the last decade has been relat~vely stagnant (Table 6 13) At the same time net imports of cassava starch while never large have declined to relatively insignificant levels Viewed in isolation these trends would appear to imply a relatively stagnant market for starch yet while cassava starch production has been stationary ma~ze starch production has been increasing (Figure 6 2) indicating quite s~gnificant growth ~n total starch demand At issue then is the compet~t~on between maize and cassava starch for a growing but not expansive market The maJor part of the cassava starch industry is located on M~ndinao together with part of the maize starch industry The industry is by nature large-scale and in 1981. consisted of ten plants with a combined a'l.nual Table 6 11 Ph1l1pp1nes Per Cap1ta Consumpt1on1 of Cassava and Pr1ces2 by Quarter and Reg1on 1973-76 Jan--March A11r1l--June Juli'-se11t · ··· Oct-Dec Reg10n Consumpt1on Pr1ce Consumpt1on Pnce Consumpt1on Pr1ce Consumpt1on Pnce (kg/ca(1lta) (pesos/kg) (kg/caplta) (pesos/kg) (kg/caplta) (pesos/kg) (kg/caplta) (pesos/kg) I 1 4 o 'i3 1 5 o 53 1 8 o 62 1 4 o 51 II 1 9 o 53 1 o o 60 1 7 o 50 1 8 o 55 III 1 9 o 52 1 5 o 61 2 1 o 53 2 4 o 53 IV 2 3 o 41 1 9 o 45 2 3 o 54 2 2 o 54 V 3 9 o 43 2 8 o 44 4 1 o 48 3 2 o 54 VI 2 6 o 47 3 2 o 70 2 1 o 49 2 9 o 48 VII 8 1 o 31 5 2 o 47 3 5 o 41 4 6 o 53 VIII 5 9 o 34 4 8 o 64 5 4 o 38 2 8 o 81 IX 6 1 o 31 4 5 o 66 109 o 29 4 7 o 42 X 4 8 o 40 4 4 077 5 1 o 37 4 7 o 46 XI 5 4 o 38 5 1 o 33 4 o o 36 4 2 o 40 XII 5 5 o 43 5 8 o 41 11 5 o 35 3 9 o 42 1 Per cap1ta consumpt1on expressed on an annual bas1s 2 Constant 1972 pr1ces So urce Calculated from unpubl1shed consumer food consumpt1on surveys carr1ed out by the Spec1al Stud1es Dlv1s1on M1n1stry of Agr1culture TABLE 6 12 Philippines Estimated Demand Elasticities for Cassava Source FNRI IAPMP Binongo Source Income -0 08 o 20 -0 82 Own Price -0 20 -0 68 Cross Price Rice Maize 1 18 o 33 Food and Nutrition Research Institute Integrated Agricultural Production and Marketing ProJect Salome Binongo An Economic Analysis of the Demand for Fresh Cassava and Cassava Products in the Ph~lippines 1985 Table 6 13 Ph1 h pp1 nes Product1on and Trade of Cassava Starch 1968-80 Trade Year Product1on Exports Imports t t) t 1968 22 044 1 201 1969 18 204 350 1970 22 771 193 10 1971 29 277 404 1972 27 867 3 722 1973 15 616 2 211 1974 18,375 4 229 1975 17 425 4 220 1976 17 391 1 2 004 1977 16 576 3 5 1978 17 024 3 3 1979 17 371 1 5 1980 N A 14 4 Source Nat1onal Census and Stat1st1cs Off1ce TABLE 6 14 Philippines Rated Capacity and Production of Cassava and Maize Starch 1976-83 Capacity Capacity Year Capacity Production Utilization Capacity Production Utilization (+) (+) (%) (+) (+) (%) 1976 31 826 10 888 34 2 147 810 58 416 39 5 1977 36 326 14 558 40 1 147 810 65 739 44 5 1978 51 326 16 371 31 9 147 810 74 393 50 3 1979 66 326 16 289 24 6 147 810 72 985 49 4 1980 66 326 13 604 20 5 147 810 55 956 37 9 1981 66 326 18 712 28 2 147 810 65 127 44 1 1982 66 326 19 898 30 o 147 810 68 708 46 5 1983 111 726 38 058 31 4 147 810 72 143 48 8 Source Fortunato Jayme A Report on the Philippine Starch Industry 1 1982 VI - 10 - 1/ capacity of 125 thousand tons of starch - What is impressive is the recent expansion in processing capacity for cassava starch at a time when the cassava starch industry was operating at 39% capacity and the maize starch industry at about 45% (Table 6 14) The maize starch industry went through an expansion phase in the early 1970 s and has maintained 1tself at five plants ever s1nce The cassava starch industry appears to be going through a similar expansion in the early 1980's after having little new investment for over a decade This expansion represente a signif1cant diversification away from Mindinao since two of the new plants are on Luzon and the largest is on Bohol This has come at a time when the overall growth rate in the economy has slowed dramatically and growth 1n the industrial sector has even been negative The need to cover recent capital investments will be constrained by excess capacity in the industry and a certain downturn in aggregate starch demand The profitability of cassava starch production is determined principally by the price of roots the output price and the capacity utilization The output price (and the market share) are largely set by the competition with domestic maize starch prices and not by import prices (Table 6 15) There is a 70% ad valorem duty on cassava starch imports In turn the price of both starches is set by the raw material price In th1s respect cassava root prices have not increased at as fast a pace as maize prices especially since 1980 In 1981 this caused a large different1al to open between maize and cassava starch prices in turn causing cassava starch production to increase and maize starch production to decl1ne What is clear is that the price competition between maize and cassava starch will depend essentially on what happens in raw material prices Even for large-scale plants the costs of producing cassava starch depends principally on the cost of the root Fuel is another large cost component in large-scale plants As can be seen in Table 16 the costs of production are not substantially different from the selling price Small changes in the root purchase price would thus substantially affect the profitab1l1ty of cassava starch production lncreasing capacity utilization depends principally on secur1ng continuity in the supply of roots As is not the case with maize the cassava processing plants must rely on a continuous harvest of roots rather than on stored supplies or imports For the starch industry there appears to be a distinct seasonality to cassava supplies Table 6 17 shows the monthly production of five of the seven starch mills operating in 1978 Only two of the five plants could operate the year round and for these two plants production in the first part of the year was about half of the production 1n the latter part This coincides only to a limited extent with the seasonality in the human consumption of fresh roots but is reflected very clearly in seasonal price variation in Central Mindinao (Figure 6 3) )_/ There are reported cases of household production of cassava starch There are no data to suggest how large such production is but it is assumed to be minor TABLE 6 15 Year 1976 1977 1978 1979 1980 1981 Philippines Trends in the Price of Maize and Cassava and the Respective Starches 1976-81 Maize Cassava Cassava Starch - Grain Starch Root Starch Philippines Bangkok (P /kg) (P /kg) (P /kg) (P /kg) (US$/t) (US$/t) 1 15 2 12 o 28 2 43 326 6 173 4 1 16 2 24 o 30 2 27 306 o 181 o 1 14 2 32 o 32 2 08 282 8 151 6 1 17 2 35 o 37 2 17 293 o 281 3 1 60 2 76 o 44 2 47 329 3 282 1 1 90 3 25 o 47 2 85 361 2 213 5 Source Survey of the Starch Milling Industry in the Philippines Business Research Department Development Bank of the Philippines 1982 Table 6 16 Ph1l1pp1nes Annual Costs of Product1on of Cassava Starch for a Factory w1th a Capac1ty of 20 t/day of Starch 1978 Cost Item Var1able Costs Cassava Roots Labor Fuel Gunny Bags Interest ~n Work1ng Cap1tal Transport (del1vered ex-factory) Total Var1able Costs F1xed Costs Deprec1at1on lnterest on F1xed Cap1tal Total F1xed Cap1tal Total Costs Sell1ng Pnce Total (ooo Pesos) 6300 108 1692 420 96 960 9576 1002 1200 2202 11 778 Per ton of starch (Pesos) 1050 18 282 70 16 160 1596 167 200 367 1963 2100-2400 Source M F Constant1no Cassava Market Study and a General Strategy of Implementat1on for the Cassava Program unpubl1shed M B A Thes1s As1an Inst1tute of Management 1979 Table 6 17 Ph1l1pp1nes Monthly Product1on of Starch by FlVe Starch Faetones 1978 F1rm Month 1 3 4 5 Total (t) (t) (t) (t) (t) (t) January 203 2 1 098 8 656 9 1 954 February 741 o 283 9 1 025 March 42 8 576 4 399 9 1 019 Apnl 123 3 437 7 350 9 912 M ay 173 3 678 5 258 9 1 111 June 180 8 753 2 242 5 69 1 1 246 July 166 1 707 6 412 7 239 8 1 526 Au9ust 195 7 1 028 5 689 1 1136 2 027 September 171 1 1 091 8 644 6 118 9 2 026 October 166 3 81 1 1 110 6 683 7 159 5 2 201 November 161 7 161 3 1 272 o 671 5 165 9 2 432 December 76 7 129 o 1 121 7 704 7 140 4 2,172 Total 1 458 o 574 7 10 612 9 5 999 2 1 007 1 19 652 Source C D V1llanueva and R G Laguna An Intens1ve and Cr1t1cal Survey of Ex1st1ng Industr1al Process1ng of Root Crops and ProJeCtlon for the Next Decade PRCRTC Annual Report 1979 Table 6 18 Ph1l1pp1nes Sources of raw mater1al and un1t costs of cassava roots purchased by f1ve starch factor1es 1978 Own Plantat1on Farmer M1a0leman F1 rm Percent Um t Cost Percent Un1t Cost Percent Un1t Cost (%) (Pesos/kg) (%) (Pesos/kg) (%) (Pesos/kg) 1 - - 60 o o 23 40 o o 23 2 90 9 o 28 9 1 o 18 3 15 o o 18 85 o o 18 4 lOO o 24 90 o o 16 5 88 6 o 37 1 2 o 15 10 2 o 60 !1 Average 18 2 o 25 78 3 o 17 3 5 o 28 !1 Gaplek Source C D V1llanueva and R S Laguna An Intens1ve and Cr1t1cal Survey of Ex1st1ng Industr1al Process1ng of Root Crops and ProJect¡on for the Next Decade PRCRTC Annual Report 1979 VI - 11 - The rat~onale of plantation production ~s to plan supplies in relation to processing needs Ironically the two plants which remained closed for the longest period during the year were exactly those wh~ch relied principally on their own production from their estates The other plants relied to a large extent on purchases of smallholder production (Table 6 18) Moreover according to the companies own estimates it was cheaper to buy cassava from smallholders than to produce the roots in estates Without further efforts at mechanizing cassava production the ev~dence suggests that it is very difficult to achieve economies of scale in cassava production even with such a large yield margin between smallholder and estate production in the Phil~ppines As in most countries the market for starch is not understood in any detail One survey of 64 industrial users showed a relatively broad use in both food and industrial uses (Table 6 19) If the total cassava starch production figures are correct this sample would appear to account for about one-third of total consumption The use of cassava starch in monosodium glutamate production used to be a substantial part of end demand About 1972 m s g producers invested in new equipment which utilized the cheaper molasses as the raw material eliminating most of this demand for cassava starch Constantino (1979) also estimates that about 30 to 35% of cassava starch goes into the manufacture of tap~oca pearl The starch market is currently small relative to processing capacity and growth in that market is uncertain This produces something of a quandary in planning the future direction of cassava development That is the first constraint on the expansion of the cassava starch industry is the limited capacity to produce sufficient cassava roots at a competitive price Indications are that smallholder production is both a more economical as well as socially preferable means of increasing cassava production Yet the nagging question remains that if smallholder productivity and production are increased is starch demand sufficient to absorb major increments in production? The export market will not be an option for surplus starch production unless the world market price is quite high The starch processing capacity that is now in place represents about double the current national product~on of cassava roots Since cassava plants w~ll now be distributed through most reg~ons in the Philippines the starch industry could provide the basis for maJor expansion in cassava production given an increment in farm productivity The starch induszfY thus provides an initial base on which to develop cassava production - However this market does not provide the certainty for major expansion in cassava production nor since large-scale plants are the rule does every farmer have access to this market Analysis of other market alternatives would thus appear warranted 'l:.l Planning is crit~cal to these large-scale plants The farmers in the Bohol region were contracted to supply a new 60 000 ton plant on that island For such a large plant product~on was increased by a major increment over previous levels The plant did not open as proJected and farmers had to chip their product~on and sell at prices which •ere less than half of the previous year s level The plant s ab~lity to contract for the next few year s production was now badly compromised The dried chip market Gaplek-type dried chips are produced in the Philipp~nes but production has never been large enough or sufficiently cont~nuous to allow the development of a broad-based market Chip production is based in the V~sayas and Mindinao areas and principally serves as a means of venting fresh root surpluses where there are constraints on access to fresh markets Prices tend to be cheaper than their fresh root equivalent and chips are absorbed as cheap substitutes in industries such as feed concentrates starch (for making glucose) and flour (for noddles and non-leavened bakery products) In general prices are too low at current yields to provide incentives for increases in chip production Currently, chips are the market of last resort for roots that need to be harvested or once harvested have no ready market High fresh market prices have tended to inhibit the consoladation of a cassava chip market However the question is what would be the potential market for cassava chips if market channels were better developed and root yields were increased? Like a host of other tropical wheat-import~ng countries the Philippines has for a long time had a law which required that wheat flours be substituted with domestically produced flour up to a minimum of 10% Cassava flour was assumed to be the alternative flour with the most promise The law prompted the establishment of at least one cassava flour mill on Luzon The mill never operated at capacity and it was never poss~ble for the wheat flour ~ndustry to meet the requ~rements of the law since sufficient cassava flour at a remunerative price was never available As with similar laws in other countries the market was potentially large but cassava flour could not be produced at a competitive price The composite flour market offers potential if cassava chip prices can be reduced but experience has shown that basing a cassava chip industry on mixed feeds presents far fewer organizational constraints (as well as quality problems) than developing cassava chips for a composite flour industry In the last decade there has been a structural change in the poultry industry as production has shifted from small-scale units to large vertically integrated commercial operations Meat production from these operat~ons has tripled in the last decade (Table 6 20) Such structural change has spawned rapid growth in the feed concentrate industry and the production of mixed feeds has increased at an annual rate of 12 2% over the last decade (Table 6 21) Of total production of the mixed feed industry 70% goes to poultry while the other 30% is sw~ne feed (Table 6 22) A principal feature of the industry however ~s it locus on Luzon where 90% of mixed feeds are produced Since the locus of cassava chip production is in the South ~nter-~sland transport costs will be a majar cost component affecting the farm-level chip price Growth in industrial demand for maize has caused a fundamental change in the structure of the maize market (Table 6 23) Although maize production has increased at the very respectable rate of 4 3% per annum over the last decade increased use of maize for feed and for starch even with decl~ning per capita consumption of maize has entailed a rising level of imports Moreover maize production has stagnated over the past three to four years raising concerns that imports will have to increase even Table 6 l'l. Ph1ll pp1 nes Average Monthly Consumpt1on of Cassava Starch by Type of F1nal Product for a Sample of F1rms 1978 Number of Quant1ty Percent F1nal Product F1rms (t) (%) Kropeck 22 97 19 Noodle 23 41 8 Glucose 2 175 34 Adhes1ve 3 4 1 Cardboard 12 46 9 Monosod1um Glutamate 1 113 22 Detergent 1 38 7 Total 64 512 100 Source C D V1llanueva and R S Laguna An Intens1ve and Cr1t1cal Survey of Ex1st1ng Industr1al Process1ng of Root Crops and ProJeCtlon for the Next Decade PRCRTC Annual Report 1979 Tab 1 e 6 20Ph1l1pp1nes Poultry Stock and Slaughter 1n Commerc1al Operat 1 ons Poultry Year Stock Slaughter (000 head) (000 head) 1970 46 448 34 576 1971 52 526 42 221 1972 52 555 42 276 1973 44 373 32 777 1974 60 609 48 728 1975 69 851 60 928 1976 77 877 64 768 1977 90 315 71 622 1978 103 528 87 813 1979 117 964 101 353 1980 125 362 110 480 So urce Bondad et al 1981 Table 6 21 Ph111pp1nes Product1on of M1xed Feed 1968-1979 Year Total Product1on (mt) 1968 263 744 1969 357 881 1970 314 415 1971 285 143 1972 312 341 1973 387 680 1974 421 266 1975 654 665 1976 625 345 1977 756 877 1978 873 499 1979 935 900 Annual Growth Rate 12 2% Source Lincangeo-Lopez 1979 Table 6 22 Ph1l1pp1nes Volume of m1xed feed product1on by type and reg1on 1978 Locat1 on Type of feed Ph 1 11 pp 1 nes Luzon _. _ Vi_s_a_yas M1nd1nao Poultry Product1on (000 t) 598 4 556 7 41 7 neg % of total by reg1on 100 o 93 o 7 o % of total by feed type 69 o 70 o 75 o Hog Product1on (000 t) 262 5 225 1 13 7 22 6 % of total by reg1on 100 o 86 o 5 o 9 o % of total by feed type 30 o 28 o 25 o 100 o Other Product1on (000 t) 12 6 12 3 o 3 % of total by reg1on 100 o 98 o 2 o % of total by feed type 1 o 2 o Total Product1on (000 t) 873 5 795 1 55 7 22 6 % of total by reg1on 100 o 91 o 6 o 3 o Source L 1 ncageo- Lopez 1979 Table 6 23Phlllpplnes Supply and Ut1 l1zat1on of Ma1ze 1970-1980 Ut111zat10n Crop F-ood · Year Product1on lmports Consumpt1on Feed Starch Seed (000 t) (000 t) (000 t) (000 t) (000 t) (000 t) 1970 2005 31 1248 669 52 39 1971 2013 193 1250 750 73 40 1972 1831 90 1259 680 89 38 1973 2289 94 1337 750 92 45 1974 2568 159 1712 850 96 50 1975 2767 54 1835 900 103 53 1976 2843 160 1669 1150 112 54 1977 2855 134 1647 1230 119 52 1978 3167 56 1600 1338 122 54 1979 3176 94 1657 1580 136 56 1980 3170 351 1604 1699 146 55 SOURCE Bondad et al 1981 TABLE 6 24 Philippines Ingredient Maize Cassava Chip Soybean Me al Fish Meal Coconut Meal Meat Meal Source CIAT Optimal Poultry Rations in Least- Cost Feed Formulation 1981 Price Entry (P /kg) (7) 1 9 40 3 1 3 22 2 2 9 25 8 4 o 7 5 1 2 o 3 9 4 4 VI - 13 - further Stagnating maize production and rising imports thus open the policy question of whether cassava chips can be developed as an alternative carbohydrate source for feed rations The principal question in the potential development of this market is whether cassava can compete with maize in feed rations This is primarily answered in terms of whether cassava enters into a least-cost feed ration Binongo (1982) f~nds that cassava enters into both swine and poultry rations at ruling prices for maize and cassava from 1975 to 1984 However since there are not quoted prices for cassava chips in the Philippines Binongo is forced to use some multiple of fresh root prices Her assumption of 2 5 appears low at first glance However as Janssen (1986) has shown root prices formed in the fresh food market tend to overestimate root costs to processing plants (essentially for quality reasons and the percentage of re)ects for size) Nelson (1986) assumes a factor of 3 O i e a conversion rate of 2 5 and raw material costs being about 80% of total processing costs -- which because of the overestimate of root costs is more like an upper ceiling Unnevehr (1982) found gaplek to fresh root price ratios in Indonesia usually to be below 2 5 although these reflected village market prices and therefore differences in relative marketing costs Assuming a multiple of 2 75 cassava still enters the more exigent poultry feed ration (Table 6 24) ind~cating that cassava can compete with maize even at currently low yield levels There is thus a basis for expanding cassava production and product~vity by developing the market channels to feed manufacturers Private profitability however is not the only basis for a majar policy emphasis on cassava Social profitability offers a more comprehensive basis for assessing crop priorities Gonzales (1984) computes domes tic resource costs (DRC' s) for principal crops produced in the Philippines and finds that cassava offers the highest social profitability of all crops considered However Gonzales used as a border price the export price for high grade cassava starch which is not the market to which increased cassava product~on should be primarily d~rected However the analysis does suggest that the breakeven border prices for cassava is US$101/mt of dried cassava evaluated at an average yield of 2 1 t/ha on a dry basis This pr~ce is quite competit~ve both with the import price of Thai cassava and with the import price for maize (US$157 /t) Given the obvious potent~al for increasing average yield levels and the fact that at current yield levels cassava is already socially profitable to produce further development of dried cassava for the animal feed market would appear to be warranted The Phil~ppines is currently pursuing a self-sufficiency program in maize along the lines of their successful rice program Maize yields at less than one ton per hectare are low and the heart of the Maisagana program is a tropical maize technology in particular a hybrid maize resistant to downy mildew The focus on ma~ze self-sufficiency reflects the growing concern about rising imports Bouis (1983) modeled the rice and maize sector ~n the Ph~lippines and proJected maize imports rising from 244 thousand tons in 1981 to 545 thousand tons in 1990 and to 1 45 million tons in the year 2000 Moreover th~s assumed increases in average maize yields from 97 t/ha in 1981 to 1 41 t/ha h the year 2000 As Bouis concluded only under the most optim~stic assumptions as to technolog~cal change will Development diversified grains VI - 14 - the Philippines be self-sufficient of the cassava chip market strategy in the policy goal of in total cereal production therefore offers a more self-sufficiency in cereal Rowever development of the cassava chip market w1ll not be easy and raising farm level yields will probably be the easiest component 1n the expansion of the chip market A cheap drying technology will be a critica! constraint It is not clear how and whether this can be solved under the generally high rainfall and humidity conditions prevalent in the Ph1lippines Possibly the locus of cassava production could be shifted to the drier areas on Luzon or coconut and rice drying units could be adapted to cassava Second interna! transport costa will play a critica! role in determining cassava's ability to compete Inter-island transport is relatively expensive for a bulky commodity like cassava chips and with most of the cassava production area in the south and the feed 1ndustry on Luzon transport costs will capture a not unsubstantial portion of the output price This however may be counterbalanced by a recent trend to locate new feed mill capacity in Visayas and Mindinao Finally given the Philippines policy focus on improving the welfare of the rural peor development of the cassava crop should take place within the smallholder sector rather than within a plantation system Such a focus would require institutional support to develop production and processing systems and market linkages One such pilot proJect has recently been developed by the Visayas State College of Agriculture A national cassava production program has been formulated by the Ministry of Agriculture The plan focuses on raising cassava yields in all regions in the Phil1ppines Where starch plants are already in operation increased production will be directed at servicing the plant For those cassava production regions that lie outside the effect1ve transport radius of a starch plant, increased production will be chipped and dried Production credit and loans for financing of chipping and drying capacity will be extended through farmers associations The credit will also be extended only on the basis of a marketing contract between the association and an accredited buyer either a starch or feed mill or the National Food Authority The program as currently conceptualized focuses on both production and marketing and foresees the principal market to be for use in feed concentrates Pricing and market efficiency Apart from the supply areas of the starch plants prices for cassava are principally determined by demand in the fresh food market Cassava varies between a vegetable and a staple food in the Philippines Retail prices nevertheless are high and do not consistently follow staple grain prices (Table 6 25) The ratio of retail milled maize prices to retail cassava prices over the per1od 1970-79 varied from 1 4 to 2 4 and varied s1gnificantly from year to year For prices of fresh cassava and milled maize to be equal on a caloric basis the ratio should be around 3 5 Calories derived from cassava are thus expensive compared to maize principally due to the high marketing margin for fresh roots Farm prices make -P as little as 30/ of the eventual retail price (Table 6 26) These marketing marg1ns are broadly typical for cassava Table 6 25. Ph1l1pp1nes Pr1ces of Cassava and Shelled Yellow Ma1ze at the Farm and Reta1l Level 1970-1980 MalZe Cassava MalZe Year (pesos/kg) (pesos/kg) Cassava (% Farm-level 1970 o 33 o 12 275 1971 o 49 o 15 327 1972 o 54 o 15 360 1973 o 56 o 21 267 1974 o 91 o 29 314 1975 o 94 o 29 324 1976 o 94 o 28 336 1977 1 00 o 30 333 1978 o 97 o 32 303 1979 1 01 o 37 273 1980 1 14 o 44 259 Reta 11 1970 o 47 o 32 147 1971 o 80 o 38 211 1972 o 80 o 46 174 1973 o 90 o 53 170 1974 1 24 o 70 177 1975 1 44 071 203 1976 1 43 o 71 201 1977 1 48 o 80 185 1970 1 50 o 74 203 1979 1 60 1 19 134 1980 1 79 1 28 140 So urce Bureau of Agr1cultural Econom1cs Table 6 26 Ph1l1pp1nes Nom1nal and Real Pr1ces of Cassava at Farm Wholesale and Reta1l Level 1970-80 Farm Wholesale Reta1l Year (pesos/kg) (pesos/kg) (pesos/kg) Nom1nal 1970 12 19 32 1971 15 24 38 1972 15 29 46 1973 21 32 53 1974 29 40 70 1975 29 41 71 1976 28 43 71 1977 30 53 80 1978 32 57 74 197 9 37 74 1 19 1980 44 85 1 28 Real (1975 pn ces) 1970 25 40 67 1971 27 43 69 1972 25 48 76 1973 30 46 76 1974 31 43 76 1975 29 41 71 1976 26 40 67 1977 26 46 70 1978 26 46 60 1979 25 51 81 1980 25 49 74 Source Bureau of Agr1cultural Econom1cs Table 6 21 Ph1l1pp1nes Marketwg Marg1n for Fresh Cassava Root for Vanous Types of M1ddlemen 1977-79 Average Buy1ng Average Se ll1 n g Gross ·-M a rket1ri9 Net M1ddleman Pnce Pr1ce Marg1n Cost Return (Pesos/kg) (Pesos/kg) (Pesos/kg) (Pesos/kg) (Pesos/kg) Contract Buyer o 23 o 32 o 09 o 04 o 05 Agent o 23 o 28 o 05 o 02 o 03 Assembler-wholesaler o 16 o 27 011 o 09 o 02 Wholesaler o 28 o 35 o 07 o 04 o 03 Wholesaler-reta1ler o 33 o 42 o 09 o 04 o 05 Reta1ler o 29 o 40 o 11 o 03 o 08 Source E B MeJla Cassava Soclo-economlc and Market1ng Study Ph1l1 pp1 nes Spec1al Stud1es D1vls1on M1n1stry of Agr1culture No 79-26 October 1979 VI - 15 - consumed in urban areas where transport from farm to urban center is relatively expensive However the SSD surveyed 222 cassava middlemen throughout the Philippines and found the gross margins between farmer and wholesaler as well as between wholesaler and retailer to be much smaller than that reflected in the average price data (Table 6 27) Moreover actual marketing costs (without accounting for losses) were low There is thus some doubt as to the extent to which the gross margins as reflected in the BAE price data can be generalized to cassava market channels Nevertheless margins for fresh cassava remain high To evaluate whether cassava is going to compete with grains in alternative markets the relevant price is the farm and not the retail price The price ratio between maize and cassava at this level is much more 37avorable (Table 6 25) Accepting a minimum price equivalent ratio of 3 1 - farm-level prices were in general competitive with maize over the period This would be expected if cassava starch or chips were to be competitive with maize-derived products However what is clear is that there is as yet no consistent market integration between maize and cassava prices This is due to the more fragmented nature of cassava markets and the often specialized nature of these local markets Thus root prices are much lower in the southern regions as compared to Luzon often by as much as half The fresh market can operate at higher price levels than the starch or chip market and has been the principal demand factor in price formation However there is very limited capacity to absorb additional supplies and marketing is risky for farmers There has thus been no incentive to intensify production practices and no effective demand for new technology Pricing in the cassava root market will have to be linked to the coarse grain market creating better price stability and more integrated cassava markets Cassava chip production will be key to such market integration The fresh root market is small enough that making this transition that LS driving prices downward in the fresh market to the maize equLvalent price should be easLly accomplished As a broader based chip market becomes established market efficiency and better market integration between regions should be vastly improved Conclusions The Philippines was the first country in Asia to receive cassava from the New World Cassava was brought by the Spanish from Mexico in the 17th century Yet cassava never established itself as an alternative carbohydrate staple to rice Given the generally favorable rainfall and soil conditions this role was captured by maize Moreover maize while at first being grown as a cheap foodgrain alternative to rice provided the raw material base for the development of both a starch and feed concentrate industry The Philippines is now undergoing a rapid expansion in domestic demand for carbohydrate sources especially the increasing demand for feed components This demand LS principally being met through rising maize imports even though internal maize prices have been kept above their import price and should have acted as a damper on maize demand ]_/ The ratLo assumes a conversion of roots to chips of 2 5 1 and that dried cassava is competitive at 80% of the maize price VI - 16 - ProJections indicate imports levels of almost one and a half m~llion tons by the year 2000 a level which runs counter to a policy obJective of self-sufficiency in cereal grains The future of cassava in the Phil~ppines is clearly dependent on capturing a share of the growing animal feed market Under current maize price policy cassava is already competitive in least cost feed rations although processing capac~ty and marketing channels for cassava chips are as yet not well developed Several factors will influence the development of this market particularly pricing of maize ~mports which is in turn tied to setting of the exchange rate and the relat~ve rate of technical change in cassava product~on versus maize production However the first hurdle is the development of product~on processing and utilization linkages The cassava sector in the Philippines is caught in a market ~nertia induced by the dominance of the fresh food market Price formation depends on local supply and demand conditions local markets are thin and there is little spatial or product price integration Incentives for investment in processing capacity and development of market channels for chips are constrained by the small production base pr~ce variability and uncertain operating margins due to the independence of fresh root and maize prices On the other hand farmers have little incentive to intensify cultural practices and expand area because of uncerta~nty of market access and price variability due to thin markets Expanding the production of cassava chips is the solution to the development of better integrated cassava markets and of a price linkage of maize and cassava markets The potential yield gap in cassava that can be exploited in the Philipp~nes is much larger than in other Asian countries A closer study of cassava within current farm systems is needed to identify the types of technology required to raise yields Increasing productivity however will require appropriate incentives and thus implies simultaneous development of processing capacity and marketing channels In this regard the national cassava program is a step in that direction with its integration of extension of both processing and production technology the opening of credit lines for development of processing capacity and the basing of production credit on marketing contracts Development of a broad-based cassava market will depend on the ability to produce cassava chips Drying technology is potentially the maJor constraint on future development of cassava Various alternatives will have to be tested under various cl~matic cond~t~ons and costs will need to be assessed Given drying constra~nts and relatively high inter-island transport costs consideration of pelleting in southern production areas should be considered at an early stage However processing technology for chips should be maintained small in scale thereby facilitating linkage to small-farm production reducing transport and assembly costs for roots and allowing a more evolutionary growth in relation to the capacity of marketing channels Feed demand in the south will be filled first before there is movement of dr~ed cassava to Luzon The Ph~lippine cassava economy lies between that of Indonesia and Thailand In Indones~a alternative markets developed because of the breadth of the production base and the market integration acheived through VI - 17 - gaplek production for food use In Thailand on the other hand the cassava industry could start from scratch relying on pure cost assessments in evaluating profitabihty In the Philippines the fresh food market makes cassava a non-tradeable and limits market integration Gaplek never developed as a food source because of maize availability and a tradeable market for cassava never emerged Price signals have not provided the relevant information to producers and processors Knowing these constraints and given the potential for yield increases an appropriately designed pilot program where there is an integration of credit for processing investment extension of production technology and development of market channels could provide the base on which dynamic growth in cassava production and utilization could be launched Certainly such a program fits very well into Philippine agricultura! policy with its emphasis on small farmer incomes and self-sufficiency in grains VI - 18 - Appendix 6 1 A Synthesis of Production and Utilization The BAE cassava production ser~es raises several questions about the accuracy of the estimates particularly when they are compared to alternat~ve production or yield estimates Another test of the production series ~s a comparison with data on utilization of cassava Two studies have attempted to reconcile production and consumption data for cassava M E Constantino (1979) compiled known estimates of cassava consumption and found that between 1971 to 1976 these consumption estimates accounted for between only 50 to 80% of est~mated supply (Table 6A 1) The total consumption estimate of 252 thousand tons in 1971 compares favorably with the agricultural census est~te of 271 thousand tons She reconciled the two series by accepting the production series and assuming human consumpt~on as the residual Per capita consumption thus increased dramatically This however is not supported by SSD estimates for human consumption of cassava The Policy Analysis Staff in the Ministry of Agnculture adopted a different tactic Area estimates were assumed reliable and yields were re-estimated based on long-term trends (Table 6A 2) Per capita consumption figures were estimated on the basis of a consumption function The production series human consumption series and starch series were then put together and feed use was estimated as a residual The results shows rapidly rising feed use of cassava in the period 1975-81 There are no other corroborating data that feeding of cassava on-farm has increased dramatically nor that majar increases in the use of dried cassava in concentrates has occurred There is thus no corroborating evidence for the BAE s rap~d rise ~n production since 1975 Real farm level prices ~n the period 1975-80 were very stable and they were only slightly lower than during the first half of the decade All things considered it is probably best to base the production estimate on known consumption data This is attempted by region (Table 6A 3) These regional consumption estimates assume no inter-regional trade in fresh roots Given the bulkiness and perishability of cassava roots this is a reasonable assumption The SSD production and marketing survey in fact found very l~ttle inter-regional trade except on Luzon where there was movement of cassava from regions I III and IV to Manila In the development of the consumption estimates several assumptions were made concerning wastage on-farm feeding of cassava and production of chips Waste was assumed to be a straight 15% of total consumption On-farm animal feeding followed in part from the results of the SSD survey which found that about 57 of production was used in on-farm feeding and that this occurred essentially off-Luzon It was assumed that 10% of small-holder production in M~ndinao and Visayas was fed to swine on farms Production of dried chips was more difficult since there is essentially no data on this consumption form The SSD survey found production of cassava chips in only Central V~sayas and Western and Northern Mindinao These areas were in general areas without access to a starch plant and with ready access to either Cebu City or Cagayan de Oro cit~es where e~ther flour or concentrate m~lls are located Chip product~on ~n these three reg~ons was assumed to be 257 of total small-farm production VI - 19 - The regional utilization estimates more or less follow the regional distribution of production as presented in the 1975 BAE production statistics except for the Bicol region in southern Luzon Up to 1976 the Bicol region was always represented in the production statist1.cs as the major producing reg1.on in the Philippines Yet on the consumption side there is no evidence to suggest what this production is utilized for although there is occasional mention of chip production in Bicol This region remains something of a question mark as far as cassava production and utilization are concerned The utilization est1.mate suggests that cassava is grown throughout the Philippines but that production is larger in the southern islands than on Luzon For most regions there is little alternative to the fresh market for human consumption except in Central Mindinao where the starch industry is concentrated Table 6A 1 Ph1l1pp1nes Supply and Ut1l1zat1on of Cassava as Est1mated by M E Constant1no 1971-77 1971 1972 1973 1974 1975 1976 1977 (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) SuEE.!.l Product1 on 424 7 450 4 444 7 480 o 684 5 794 4 10ll 1 Imports 2 o 18 6 13 8 21 3 21 o 10 o Total 426 8 468 9 458 5 501 3 705 5 804 4 10ll 1 Demand Starch 148 4 157 9 91 9 ll3 1 108 2 97 o 103 6 Ammal Feed 18 3 19 4 19 1 20 6 29 4 34 l 42 5 Ava1lable for Human Consumpt1on1 260 1 291 7 347 5 367 5 567 8 673 3 865 1 Human Consumpt1on2 86 2 125 3 195 2 282 o 237 2 253 o 231 o Tota 1 1 426 8 468 9 458 5 501 3 705 5 804 4 10ll 1 Total 2 252 8 302 6 306 3 415 8 374 9 384 1 377 1 1 Calculated as a res1dual 2 Calculated from SSD food consumpt1on surveys Source M F Constant1no Cassava Market Study anda General Strategy of Implementa- tlon for the Cassava Program unpubl1shed M B A thes1s As1an Inst1tute of Management 1979 Table 6A 2 Ph1llpp1nes Supply and Ut1l1zat1on of Cassava as Est1mated by the Pol1cy Analys1s Staff 1969-1980 Supply Deinand Total Feed and Food Use Year Product10n Imports Supply Waste Starch Total Per Cap1ta (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) (kg) 1969 490 2 492 53 111 328 9 2 1970 448 - 448 41 137 270 7 3 1971 426 2 428 26 173 229 6 1 1972 440 21 461 17 165 279 7 2 1973 489 16 503 34 97 372 9 3 1974 545 24 569 75 112 382 9 3 1975 643 23 666 167 103 396 9 4 1976 750 11 761 247 107 407 9 4 1977 859 - 859 344 102 413 9 3 1978 910 - 910 380 104 426 9 3 1979 928 928 394 110 424 9 o 1980 948 - 948 402 112 434 9 o Source Pol1cy Analys1s Staff M1n1stry of Agr1culture Table 6A 3 Ph 111 ppwes Est1mates of Supply and D1str1but1on of Cassava by Reg1on 1975 Per Cap1ta Total Human Dned Ammal Reg1on Consumpt10n Consum)t1on Starch Ch1ps Feed Waste Total (kg/cap1ta) (t (t) (t) (t) (t) (t) llecos 1 5 4 904 10 370 - - 2 695 17 969 Cagayan Valley 1 9 3 673 - - - 648 4 321 Centra 1 Luzon 1 6 6 736 - - - 1 189 7 925 Southern Tagalog 2 3 11 992 - - - 2 116 14 108 B1col 7 6 24 274 - - - 4 284 28 558 Wes tern V1 sayas 5 5 22 803 18 000 - 4 420 7 981 53 204 Central V1sayas 7 5 25 402 - 12 701 5 080 7 621 50 804 Eastern V1sayas 13 7 35 620 - - 4 749 7 124 47 493 Western M1nd1nao 10 o 20 480 - 10 240 4 096 6 144 40 960 Northern M1nd1nao 8 2 18 975 15 000 13,800 5 520 9 405 62 700 Southern M1nd1nao 4 9 13 304 - - 1 774 2 661 17 739 Central M1nd1nao 11 o 22 770 47 340 - 6 665 13 549 90 324 Mam la 2 5 12 425 - - - - 12 425 Ph1l1pp1nes 5 4 223 358 91 710 36,741 32 304 65 417 449 530 Source CIAT est1mates VII Thailand Rapid Growth Driven by Export Markets Thailand has developed the premier agricultural export economy in the tropics at least in terms of its exporta of carbohydrate sources This export orientation dates to the 1850's when the signing of the Bowing treaty removed a ban by the Thai king on exports of rice The market stimulus to a subsistence economy with surplus land resources was immediate and rice exporta became the driving force in the Thai agr~cultural economy upto the Second World War The beginning of the post-war period marked the diversification of the Thai agricultural economy into upland crops again almost entirely directed to export markets Development of the upland sector has been the principal growth element in the Thai agricultural economy in the post-war period and has been based on expansion in maize kenaf cassava and sugar cane The upland sector in the post-war period has gene through a series of commodity booms These were based on area expansion within a land and labor surplus agricultural economy i e the limited size of domestic markets or the lack of export infrastructure was the most binding constraint on agricultural production The success of these booms resulted in a relative shortage of labor in the 1970's inducing the development of a market for tractor-hire services The motor of this growth process thus was the opening of market channels for export and relative price incentives in these markets However this growth process also reflected the vagaries of world market demand as is epitomized by the rise and collapse of the kenaf industry Cassava is the most recent of Thailand's commodity booms which is not to say that cassava is a recently introduced crop The exact date of introduction to Thailand is not known but cassava was apparently being grown as a food crop in the 18th century However unlike countr~es such as Indonesia and the Philippines Thailand was always able to meet its starchy staple requirements solely through rice Cassava thus never became more than a speciality food ~n the country The genesis for growth ~n the crop has always been non-food markets almost solely directed to export The initial development of such a market was in the 1930 s when cassava pearl wa~/ produced in the South for export through Malays~a (Scheltema 1938) - The Thai cassava industry was based on the starch export market up to about 1960 World War II briefly curtailed this market in Southeast Asia in the late 1940 s but following the war modern processing machinery was ~ntroduced into Chonburi in the eastern region A healthy starch ~ndustry was operating in this region by the mid-1950' s supplant~ng the starch industry in Indonesia and in the south of Thailand However it was starch wastes that became the basis for the real expansion in the crop when a Thai export statistics for cassava do not start until 1950 and the only suggestion of such an industry is Malaysian import statistics VII - 2 - West German importer in 1956 introduced cassava waste as an animal feed to Germany (Phillips 1974 Titapitnatanakun 1979) Low freight rates in this period its lack of alternative uses and high feedgrain prices in Germany made cassava waste particularly price competitive in Europe Since cassava waste was a by-product of starch manufacture shortages resulted and led to the importation of cassava meal starting in 1960 With the introduction of the Common Agricultura! Policy in 1962 and the favorable tariff binding on cassava in the 1968 GATT negotiat1ons the Thai cassava industry shifted to animal feed as its principal market Cassava chips became the dominant export in 1964 native pellets in 1969 and hard pellets in 1983 With this external stimulus Thailand went from a relat1vely minor producer of cassava in the 1950 s to the second largest (1f not the largest) producer of cassava in the world Production Trends Production of cassava has increased from around 400 thousand tons in the m1d-1950 s to almost 20 million tons in 1984/85 (Table 7 1) Th1s represents a sustained growth rate of 16% per annum for over 25 years These sharp increases in production have been based exclusively on expans1on in area planted and have been concentrated in a relat1vely limited number of regions within the country Production has continued to expand in the old starch producing region of Chonburi and Rayong However the bulk of cassava production has shifted from this zone to the Northeast Whereas the Northeast made up less than 10% of the total up to 1969 by 1979 the Northeast was producing over 60% of total cassava Th1s represented a shift to relatively drier production conditions and a movement from the red-yellow podzolic soils to the more acidic latosols Cassava in part displaced kenaf in the Northeast and in part was planted on newly cleared forest areas Cassava has grown from a relatively minor crop in the 1950 s to be the second most important crop after rice in terms of production volume (as measured on a dry weight basis) and in terms of foreign exchange earned As in previous commod1ty booms rapid production increases have been based on area expansion led by demand in international markets Capacity and growth in domestic markets would never have sustained the growth rates that have occured in cassava and the other major agricultura! commodities To understand the cassava industry in Thailand the analysis first reviews the factors on the production side that formed the basis for such high growth rates and then turns to an analysis of the demand s1de which must necessarily consider the changing nature of international cassava markets Cassava Production Systems Agricultura! development in Thailand has been based on exploitation of an agricultura! frontier and reliance on international markets as a surplus vent Unlike Malaysia access to new land has been relatively uncontrolled although a ceil1ng on the s1ze of land holdings fomerly in the public domain was set at 8 ha in 1936 With the expansion in 1nternational markets following World War II planted area expanded rapidly in many cases at the expense of forest lands A satellite census showed that forest land had been reduced from 57% of total land in 1961 to 37% 1n 1974 a loss of 10 million hectares in 13 years (Bertrand 1980) VII - 3 - Whereas the pre-war expansion was based principally on rice for which there was already a large production base diversification into upland crops has been the hallmark of post-war agricultural growth Crops such as maize sugarcane mung bean kenaf and cassava have expanded rapidly from relatively small production bases The final component of this extensive growth pattern was relatively rapid mechanization of the agrLcultural sector based on either animal or mechanical equipment Thus in 1963 68k of farms were usLng animal traction and 147 were using mechanical power or some combination of animals and tractors By 1978 33% of farmers were utilizing tractors Cassava production systems therefore must be understood essentLally in the context of rapid expansion of previously uncultivated land Certainly in the Northeast there was some substitution for kenaf whose area by 1981 had declined by about 330 thousand hectares from its peak in 1967 However cassava area in the Northeast increased by over 780 thousand hectares in the same period at the same time as maize production also expanded quite dramatically Given cassava s adaptation to the drier growing conditions of the Northeast and the profit levels as maintained by EC grain prices the crop expanded rapidly principally by opening up new land The process obviously introduces a dynamic element into characterLzing cassava production systems especially in terms of adaptation of management practices as farmers learn the responsiveness of a new crop and the effects of continuous cassava cultivation on soil fertility Using the agricultural census of 1963 and 1978 as reference points cassava expansion was based on a sizeable increase in the number of cassava growing farms (from 58 to 450 thousand) and in an increase in the average size of cassava plantings per farm from 1 4 to 2 1 ha In 1978 21/ of the farmers Ln the Northeast grew cassava and in most instances probably depended on cassava as their principal source of income By 1978 the modal farm size stratum for cassava farmers was between 3 2 and 6 4 ha (Table 7 2) This is large by overall Asian standards but still relatively small given the agro-climatic potentLal of most growing areas Moreover such a farm size has supported a market for tractor hire servLces but not actual tractor ownership The adoption of tractor hire services has in turn released grazing land formerly needed to support draft animals for cultivation Given the very dynamic nature of the upland sector especially in the Northeast the degree of competition between cassava and other upland crops is diffLcult to define If crop area data are disaggregated by agroeconomic zone (Table 7 3) certain hypotheses at least emerge In the old cassava growing area of Chonburi and Rayong (agroeconomic zone 15) cassava made up 40% of total farm area with the only other upland crop being sugarcane Cassava dominates this zone so thoroughly that Lt appears blanketed by monoculture cassava In the Northeast the SLtuation is more diverse In agroeconomic zones 1 and 5 cassava potentially competes with maize and kenaf In agroeconomic zone 3 cassava competes only with kenaf In none of these latter zones does cassava domLnate the agricultural economy Moreover only in agroeconomic zone 5 do maize and cassava production ares~ really overlap In the two largest maize producing zones only very little cassava LS produced In general in the Northeast there is TABLE 7 1 Thailand Cassava Area Production and Yields 1956-85 Crop Year Are a Production Yield (000 ha) (000 t) (t/ha) 1956-57 39 2 396 o 10 1 1957-58 38 4 418 o 10 9 1958-59 44 1 487 o 11 o 1959-60 62 5 1 083 2 17 3 1960-61 71 S 1 222 3 17 1 1961-62 99 3 1 726 2 17 4 1962-63 122 7 2 076 9 16 9 1963-64 140 o 2 111 1 15 1 1964-65 104 9 1 5S6 7 14 8 196S-66 102 o 1 474 7 14 5 1966-67 130 3 1 891 7 14 5 1967-68 140 9 2 062 8 14 6 1968-69 170 6 2 611 S 15 3 1969-70 189 3 079 16 3 1970-71 224 3 431 15 3 1971-72 220 3 114 14 2 1972-73 328 3 974 12 1 1973-74 41S 5 443 13 1 1974-75 497 6 76S 13 1 1975-76 475 7 094 13 6 1976-77 692 4 10 230 o 14 8 1977-78 846 8 11 839 7 14 o 1978-79 1 165 o 16 3S7 8 14 o 1979-80 845 8 11 101 o 13 1 1980-81 1 1S9 9 16 S40 o 14 3 1981-82a 1 243 1 17 744 o 14 3 1982-83 1 087 2 17 787 9 16 4 1983-84 1 017 8 18 988 5 18 7 1984-85 1 33S 1 19 985 3 15 o a Starting 1981-82 area figures changed from planted to harvested are a this caused an artificial rise in yield figures Source Center for Agricultural Statistics Office of Agricultural Econom1cs Ministry of Agriculture and Cooperatives TABLE 7 2 Thailand Distribution of Area Planted to Cassava by Farm Size 1978 Farm Size Cassava Farmers Cassava Area Strata (ha) Number Less than 32 115 3 1 o 26 213 1 o - 1 6 29 770 1 6 - 3 2 103 824 3 2 - 6 4 167 328 6 4 - 9 6 69 799 9 6 - 22 4 48 523 More than 22 4 4 759 Total 450 331 Source National Statistical Office Thailand Bangkok Percent Rectares Percent o 3 19 5 8 13 429 1 4 6 6 21 721 2 3 23 1 112 212 11 9 37 2 297 336 31 7 15 5 192 920 20 5 10 8 222 699 23 7 1 o 78 732 8 4 lOO O 939 069 lOO O 1978 Agricultura! Census Report TABLE 7 3 Thailand The Relative Importance of Area Planted to Maize and Cassava by Agroeconomic Zone 1974-78 Cassava Maize Percent of Percent of Percent of Agroeconomic Percent of Cassava Total Maize Zone Are a Total Farm Area Are a Are a Farm Area Are a (000 ha) (000 ha) Northeast 1 57 3 3 1 7 7 106 1 S 7 8 4 2 8 2 o 8 1 1 3 8 o 4 o 3 3 107 5 5 7 14 4 3 4 o 2 o 3 4 53 4 3 5 7 1 31 o 2 o 2 4 5 180 6 12 7 24 1 192 o 13 5 15 1 North 6 5 4 o 4 o 7 434 6 34 6 34 2 8 12 2 1 1 1 6 107 2 9 4 8 4 9 1 1 o 2 o 1 62 6 8 4 4 o 10 1 6 o 2 o 2 26 4 4 o 2 1 Central Plain 7 3 8 o 6 o 5 259 S 38 7 20 4 11 12 8 o 8 1 7 10 7 o 7 o 8 12 19 4 2 6 2 6 13 4 1 8 1 o 13 73 4 16 o 9 8 7 o 1 o o 6 14 - - - 15 176 o 39 6 23 6 16 28 2 12 6 3 8 5 8 2 6 o 5 South 17 3 7 o 3 o 5 6 1 o 4 o 5 18 2 6 o 6 o 3 19 1 4 o 5 o 2 Total 748 6 6 1 100 o 1269 6 7 o 100 o Source Pongsrihadulchai Apichart Supply Analysis of Important Crops in Thailand 1981 VII - 4 - still significant scope for expansJ.on of cassava area if not at the expense of other crops then in terms of currently under-utilized land already in farms or in the public domain The rainfall pattern in the Northeast and Central Plain is unimodal with a dry season from November to April and a wet season of varying intensity for the rest of the year as reflected in average annual rainfall for different sites from the Northeast to the South ranging from 900 to 3000 mm Moreover moving to the Northeast rainfall becomes more variable and uncertain Since most of the cassava is solar dried this rainfall pattern creates a trade-off between optimum drying period and opt1.mum planting period The drying season starts in November and farmers rarely leave the cassava in the ground for longer than 12 months though it could be left much longer Where rainfall is more secure that is the Rayong and Chonburi area farmers plant in the dry season as well as the wet season Further to the northeast farmers tend to plant exclusively in the March to June period that is at the beginning of the rainy season (Figure 7 1) Experimental trails have shown that planting at the beginning of the rains gives significantly higher yields (Sinthuprama 1980) Given a eight-to-twelve month growth cycle planting in the November-December period and harvesting in the same perJ.od coincide better with market demand Prices are at their seasonal high in the September-November period befare declining to their seasonal low in March-April Also root starch content is much higher at the beginning of the dry season resultJ.ng in a further price premium There is greater demand for roots at this period because of the significant increase in through-put and thereby lower costs in the chipp1.ng plants due to shorter drying periods There is thus a significant increase in root sales in the dry season (Table 7 4) although harvest occurs throughout the year Cassava production systems in and of themselves are relatively simple The land J.s prepared either by animal traction or by tractor hire services with the latter being increasingly common The cassava is planted either horizontally (sandy soJ.ls) or vertically (loamy soils) depending on the potential drought risk of the soil Planting material comes from recently harvested plants keeping stake storage time to a minimum Cassava is grown in a very strict monoculture system in that no other crop species are interplanted and a single variety tends to dominate throughout Thailand Rayong 1 In weeding hand labor is employed with some animal interrow cultivation Nevertheless in the these activities labor use is kept to the minimum necessary to adequately maintain the crop The most critical issue J.n the rapid expansion of cassava production and the resultant extensJ.ve production systems is the maintenance of soil fertility In general fertilJ.Zer application is low in Thailand when compared to other Asian c2yntries Fertilizer prices are not consJ.stently subsidized in Thailand - and are generally applied to those crops in which margJ.nal returns are highest Of the maJar crops sugarcane has the '!) There are some programs purchase of fertilizer rice which provide a credJ.t subsidy for the These programs are p~imarily orJ.ented to TABLE 7 4 Thailand Percentage Distribution of Monthly Farmer Sales of Cassava Roots during the Crop year 1973 and 1984 North Northeast Central Thailand 1973 1984 1973 1984 1973 1984 1973 1984 Oct - o 4 7 9 12 4 9 o 6 4 8 1 10 2 Nov - - 4 3 8 4 7 4 16 1 5 8 9 6 Dec - - 2 7 8 1 12 9 12 2 7 9 8 5 Jan - 4 6 5 7 15 2 3 9 15 5 4 5 14 5 Feb - 44 1 19 8 24 1 7 9 27 3 12 8 26 2 Mar - 47 o 14 9 17 o 20 4 13 5 17 1 18 4 April - 1 8 14 5 4 2 8 o 6 o 9 2 4 4 May - 2 o 5 5 1 8 5 2 1 5 5 1 1 7 June - - 9 9 o 4 6 7 o 4 7 8 o 4 July - - 7 5 3 6 5 o o 3 8 7 2 6 Aug - - 5 4 4 1 6 1 o 1 6 8 3 o Sept - - 4 8 o 7 7 6 o 9 6 1 o 6 Source Center for Agricultura! Statistics Office of Agricultura! Economcis Ministry of Agriculture and Cooperatives Bangkok VII - 5 - highest application rate followed by rice According to the 1978 census rice consumes fully two-thirds of fertilizer availabilities Sugarcane vegetable and tree crops consume an addit~onal quarter leaving less than 104 or less than 70 thousand tons available for all other major field crops Fertil~zer appl~cation on cassava ~s low In 1973/74 average fertilizer application per cultivated hectare of cassava was only 6 9 kg/ha (Koomsup 1980) On that area where fert~lizer was actually appl~ed (16% of cultivated area) rates were 43 kg/ha Recommended application rates are about 15 times this level By 1980/81 average application rates remained at the same level (Table 7 S) As would be expected fertilizer application is much higher in the old production zones around Chonburi and Rayong while in many areas of the Northeast fertilizer use on cassava is non-ex~stent The very low fertilizer use in cassava raises two critica! issues First has continuous cassava cultivat~on with only minimal levels of fert~lizer use resulted in a declining yield trend? Second what would be the yield gains were fert~lizer application to ~ncrease? To answer partially these issues the analysis turns to an evaluation of cassava yields Yields Average cassava yield levels of 13 to 14 t/ha in Thailand are high even by Asian standards Only India and Malaysia consistently have higher yields than Thailand Moreover Thailand has been able to maintain this level of productivity through the period of rap~d expansion in the crop The nat~onal statist~cs suggest that yields have declined somewhat s~nce 1960 In the early sixties average yields were around 17 t/ha and declined qu~te rapidly to 14 t/ha by the late sixt~es Yields have remained at about this level ever since having fallen below 13 t/ha only once These relatively high yields have been a significant part of Thailand s dominance of the international trade in cassava The difference in agro-climat~c conditions between the Northeast and the Central Plain is only partially reflected in yield differences The older production reg~ons on average maintain a one-to-two ton yield advantage over product~on areas in the Northeast However yields have shown something of a rising trend in the Northeast especially if extended back to 1960 Yield trends ~n the Central Plain on the other hand initially declined in the 1960's and over the past half decade have been remarkably stable at around 15 t/ha Yield levels as expressed in the aggregate production stat~stics thus present a picture of relative stability and give no indication of progressive soil exhaustion The micro-leve! data are only suggest~ve of the factors underlying the dynamics of cassava productivity To start w~th average yields of cassava mask a very w~de yield dispers~on The yield distribution is skewed with the largest segment of farmers producing quite normal yields by world standards of from zero to nine t/ha and with a very extended right-hand side where some farmers produce over 19 t/ha (Table 7 6) The second set of data is long-term fert~lity studies (Figure 7 2) These data show the expected decl~ne in y~elds with continuous cropp~ng after openi"lg up new land However the decline is gradual and in one site yields only declined TABT..E 7 S Agroeconomic Zone Northeast 1 2 3 4 S Thailand Central Plain 7 11 12 13 1S 16 Average Fertilizer Application Rates on Total Cultivated Area 1980-81 b Application Rate (kg/ha) 2 2 1 7 1 9 o 7 o 6 4 ~ 3 7 a a b The survey shows quite high average application rates for organic fertilizers Fertilizer expenditures by farmers were divided by an average fertilizer price of Baht S 1/kg Source Survey of Cassava Production Costs and Returns 1980-81 Office of Agricultura! Economics Ministry of Agriculture and Coooperatives 1982 TABLE 7 6 Thailand Distribution of Cassava Yields 1974-75 Yield Leve! (t/ha) o to 9 4 9 4 to 12 5 12 5 to 15 6 15 6 to 18 8 More than 18 8 Source Phillips Truman 1977 Chonburi Rayong Nakhonrachsima 35 7 20 6 21 4 10 1 12 2 Percent of Farmers Other Changwats 31 1 23 1 14 o 17 8 14 o Thailand 33 2 21 9 17 4 14 3 13 2 A Profile of Thai Cassava Production Practices VII - 6 - from around 30 t/ha to 20 t/ha in a sixteen year period One thorough study found that from an initial yield of 20 to 30 t/ha yields decrease by half within 9 to 20 years (lnterim Committee for Coordination of Investigations in the Lower Mekong Basin 1979) With such rapid opening of new land as has occurred in the case of cassava the yield decline in older plots has been offset by the higher yields of new product~on areas As yield in older plots fall cassava supply becomes more sensitive to price changes particularly since more than half the farmers operate at below average yields Mining of soil fertility has a longer-term social cost of enhanced erosion potential and a permanent decline in the productivity of the land resource This therefore puts prime importance on motivating increased applicat~on of organic and inorganic fertilizers as apparently already is happening in the Chonburi and Rayong area Two factors however complicate increased use of fertilizer on cassava First in most areas cassava must compete with either rice or sugarcane for capital resources for fertilizer Second cassava responsiveness to fertil~zer application is not as certain as in these other two crops There is often no response in the first two to three years after opening up new land (Table 7 7) After that while responses can be shown they cannot be demonstrated consistently (Table 7 8) What remains extraordinary in Thailand is the h~gh yields that farmers achieve in even depleted soils Suttibursaya and Kummarohita (1978) report cassava being grown continuously for 25 years without fert~lization and yet yields have decl~ned to only 16-17 t/ha A fertility restoration experiment selected four farmers fields which had been continously cultivated for 15 years and the average yield of the check plots was 21 t/ha (lnterim Committee for Coordination of Investigations in the Lower Mekong Basin 1979) This suggests that the dominant var~ety Rayong 1 is very efficient in the utilization of limited soil nutr~ents Moreover thirty years of experimental work both on the experiment station and in farmers fields suggest that 30 t/ha is an achievable target with an appropriate fertil~zer regime The results have made fertility management the pr~nc~pal research thrust in cassava ~n Thailand What is the advantage of a large investment in breeding if 30 t/ha is imminently achievable with the current var~ety? However defining a recommendation that gives a consistently profitable response has eluded researchers and inhibited adoption of fertilizer use in cassava Indeed farmers in Thailand utilize fertilizer they however do not apply it to their cassava Until the profitabilHy of fert1lizer response can be signif~cantly increased probably by linking application rates to other environmental variables no effective extension program for fertil~zation of cassava will be successful except possibly in the very badly degraded soils such as now exist in Chonburi and Rayong Thus the relatively high prices for cassava products obtained ~n the European Community was only part of the profit engine that resulted in the rapid expansion in cassava area The other component was the very high initial yields obtained by new adopters of cassava cult~vation Init~al yields in t~e 25 to 30 t/ha range provided a powerful stimulus to expand cassava area and lack of a viable crop alternative kept farmers in cassava However this raises the question of the longer term viability of cassava TABLE 7 7 Thailand Cassava Yields in Long-term Fertilizer Experiments at Rayong 1964-70 Year 1964 1965 1966 1967 1968 1969 1970 a b e Yearly Yearly Yearly First Site Zero Low a Medium b Fertilizer Application Application (t/ha) (t/ha) ( t/ha) 32 5 29 4 29 4 22 5 22 5 21 3 20 o 22 5 18 8 14 4 26 3 28 1 21 3 31 3 28 7 22 5 29 4 28 7 19 o 36 o application of 50-50-25 kg/ha of N P and K application of 75-75-120 kg/ha of N P and K applicaton of 50-50-50 kg/ha of N P and k Zero Fertilizer (t/ha) 25 o 23 8 23 1 22 5 17 5 Second Site a Low Application (t/ha) 25 6 18 8 26 3 26 9 21 3 Source Interim Committee for Coordination of Investigations of the Lower Mekong Basin Agricultura! Research Efficiency in Thailand Volume III Cassava 1979 e Medium Application (t/ha) 25 o 20 o 31 3 31 3 25 6 TABT..E 7 8 Thailand Summary of 121 Fertilizer Trials Across Three Different Soil Types 1968-70 Probabill.ty of Response Soil Series No of Trials N p Huai Pong 14 + Pattaya 25 + Sattahip 82 ++ + a The probabilities are as follows - not probable + probable + fairly probable ++ highly probable ( < 25% of trials showed response) (25-49% of trials showed response) (50-67% of trials showed response) ( ') 6 7% of trials showed response) to Source Sittibusaya Fertilization Chote and K Kurmarohita 1978 Soil Fertility and a K + VII - 7 - as the industry stabilizes as overall yields decline to a low level equilibrium and as output pr1ces come under downward pressure The task is to transform a dynamic industry that has been fueled by private costs being lower than social costs to a sustainable industry where farmers must pay the full cost of soil nutrient extraction Costs of Production and Labor Utilization As yields decline the farmer's in1tial means of maintaining profits are by reducing costs By Asian standards cassava production systems in Thailand are relatively extensive in terms of labor and input use which in turn reflects the relatively high land-labor ratio existent in the country Moreover the existing agricultural frontier and the relatively liberal land policy have further reinforced extens1ve production practices The process has thus favored technologies that substitute for labor rather than those that substitute for land Labor is the major cost component in cassava production systems Estimates of labor input per hectare range from 70 to 100 man days Only maize and broadcast rice have a lower labor input (Table 7 9) Additionally because cassava can be planted almost anytime of the year and can be harvested over a relatively long period labor activities can be scheduled 1n relation to other demands for labor S1nce upland crops must compete with rice for labor this flexib1lity in labor use gives cassava an advantage over other upland crops Finally cassava gives the highest average returns per manday of labor input (Boobst et al ) Cassava thus is very well adapted to the labor economy of Thailand---- The trend is toward further reductions in labor input Land preparation through tractors has rapidly spread through the Northeast With movement to planting in rows interrow cultivation with animals was employed in those areas that still maintained draft animals Increases in sales of herbicides have been reported in the maJor cassava producing area of Chonburi especially since there were no such sales prior to 1973 (Inter1m Committee for Coordination of the Lower Mechong Basin 1979) Thus farmers have been very responsive to technolog1es that have substHuted for labor they have not been responsive in the adoption of land substituting technology Labor or mechanization costs make up over 85/ of total cassava production costs (Table 7 10) Input and fixed costs make up the remainder Moreover normally about half of production costs are paid in cash the rest reflects the opportun1ty costs (evaluated at market prices) of farmer-owned resources The cost structure reflects some flexibility in absorb1ng price declines at least in the short-run since price declines can be absorbed in terms of lower returns on farmer-owned resources MaJor increases in fertilizer costs would sign1ficantly shift this balance again highlighting the importance of a consistent yield response for adoption Supply Response The over the crop was reasons behind the rap1d expansion in last two decades can now be summarized very prof1table During the 1974-1984 cassava area in Thailand First and foremost the per1od average returns to TABLE 7 9 Thailand Crop Rice Cassava Kenaf Peanuts Rainy season Cool season Dry season Vegetables Average Labor Requirements and Returns by Crop Enterprise Northeast 1973-74 Labor Requirements (Man-Days/Hectare) 87 56 lOO 65 161 36 161 78 112 67 155 60 772 os Returns per Man-Day Net of Nonlabor Variable Costs (Do llar /Man-Da y) 1 18 2 02 o 55 1 08 o 93 1 24 o 48 Source Bobst Barry et al in Northeast Thailand Enterprise Selection and Farm Employment 1980 Table 7 10 Thailand Cost Item Variable Costs Labor Costs Land Preparation Man Oxen Tractor Seed Selection Planting Weeding Man Oxen Harvesting Transporting Man Oxen Tractor Input Costs Stakes Agr Equipment Gasoline and Oil Chemical Fertilizer Other Costs Repa1r Agr Equip Working Capital Fixed Costs Land use Average per Rectare Costs of Production of Cassava Roots Northeast 1980-81 Cash (Baht/ha) 2810 6 2590 1 1875 3 58 6 52 9 921 6 8 7 251 5 575 6 1 8 572 1 69 1 2 6 71 o 207 o 134 1 26 1 26 o 20 8 18 3 58 o 58 o Non-Cash (Baht/ha) 2054 3 1290 6 882 9 97 6 93 5 65 8 31 3 154 8 439 1 334 6 72 6 o 5 242 o 242 o 521 8 Total (Baht/ha) 4864 9 3880 7 2758 2 156 1 146 4 987 4 39 9 406 3 1014 6 1 8 906 8 141 6 3 1 71 o 449 o 376 1 26 1 26 o 20 8 18 3 521 8 Depreciation Agr Equip 673 2 647 5 25 7 731 2 705 5 25 7 Total Cost Cost per ton (Baht/t) Price (Baht/t) 2868 6 2726 6 5595 2 406 510 Source Survey of Cassava Product1on Costs and Returns 1980-81 Off1ce of Agricultura! Economics Min1stry of Agriculture and Cooperatives 1982 VII - 8 - cassava never dropped below 25% and were as high as 145% (Table 7 11) Second the kenaf industry was in decline and even further land was available on which to expand Given the high yields on uncultivated land cassava as an income source was unmatched and led to a majar increase in incomes in the relatively depressed area of the Northeast Third farmers did not face a labor constraint as tractor hire services expanded rapidly in the cassava producing areas All of these factors are reflected in cassava supply response Pongsrihandulchai (1981) has estimated supply equations for cassava by agro-economic zone and as might be expected found a very high short-run price elasticity of between O 58 to 2 78 (the median was 1 77) Price responsiveness in cassava was much higher than in rice (O 27) maize (O 70) kenaf (O 87) or sugarcane (O 62) Moreover the supply equations suggested that cassava principally competed for land with kenaf except in the Rayong-Chonburi region where there were no competing crops with cassava These equations were estimated while cassava prices were on the whole increasing The question arises whether farmers would be equally responsive to declining prices and the answer would probably be no There is limited effective competition between cassava and other crops reflecting few other cropping alternatives for land in cassava Farmers would only sign1ficantly reduce area if they were operating at a cash loss Technology Development Research on cassava in Thailand started 1n 1956 with the creation of the Huai Pong Experiment Station in Rayong The station comes under the Field Crop Division of the Department of Agr1culture and since 1956 has beeen the principal locus of cassava research although research on other field crops is also done at the station As research on cassava has increased with the expansion in the crop other field crop research stations in the northeast have also conducted exper1mental work on cassava all of which is coordinated by the Root Crops Branch within the Field Crop Division of the Department of Agriculture For the first two decades cassava research focused on soil management and fertil1zation (see Sittibursaya and Kurmardrita 1978 for a summary of this research) The principal features of this work are well summarized by the Committee for the Lower Mekong Basin (1979) namely high yearly y1eld fluctuat1ons probably related to rainfall conditions rapidly declining yields of unfert1lized plots and variable response to fert1lizers While the research has led to a set of fertilizer recommendations broken down by soil type and while a series of farm level demonstration trials were also carried out only minor adopt1on of fert1lizer has occurred Some research in this area continues to be done even though it follows virtually the same approach The few deviations have been toward evaluat1on of green and organic manures These have shown prom1sing results (Table 7 12) but have not led to any recommendations Lack of progress in the area of fertilization gave impetus to the development of a var1etal improvement program Local clones were collected in 1956 These were evaluated for agronomic characters and yielding ability but were found not to show significant differences One was selected and named Rayong 1 wh1ch was used as a check variety in all TABLE 7 11 Thailand Average Costs of Production and Returns for Cassava 1974-1983 Per Hectare Costs Per Hectare Per Ton Farm Crop Year Cash Non-Cash Total Yield Cost Price a (Baht/ha) (Baht/t) (Baht/ha) (t/ha) (Bath/t) (Bath/t) 1974-75 1593 1558 3151 13 o 242 4 290 1975-76 1854 1674 3528 13 7 256 9 410 1976-77 1701 2390 4091 12 6 325 6 460 1977-78 1696 2116 3812 12 9 294 9 450 1978-79 2059 2089 4148 14 9 282 6 370 1979-80 2217 2227 4444 10 7 415 9 770 1980-81 3114 2757 5871 14 3 411 8 750 1981-82 2820 3221 6041 14 o 432 4 450 1982-83 3399 3018 6417 13 9 446 o 540 a Average price for the crop year Oct-Sept Source Production Economic Section Office of Agricultural Economics Ministry of Agricultura and Cooperativas Bangkok VII - 9 - succeeding experimental work While some selection from collected open-pollinated seed started in 1971 a controlled hybridization program did not begin till 1974 (Sinthruprama 1978) Initial crosses were between Rayong 1 and other local cultivara In 1977 varieties from CIAT were introduced as well as seed from controlled hybridization This served to s1gnificantly expand the germplasm on which the crossing program was based Initial selection is based on high root yield and high starch content In later evaluations earliness and appropriate plant type for intercropping are introduced as selection characteristics Promising materials are evaluated for drought tolerance resistance to the few cassava diseases and pests that occur in Thailand and in some cases for edible quality characteristics A testing program of regional and on-farm trials resulted in the release in 1983 of the first promising variety Rayong 3 Ita principal advantages over Rayong 1 are a higher starch content and a higher response to chemical fertilizer As yet it is too early to evaluate the adoption of this variety New production technology has not been necessay to the rapid expansion in cassava cultivation The high yields obtained with the local variety as new land was cultivated and the high prices set by the European Community were sufficient to maintain high profits in cassava cultivation These profit levels are now coming under pressure from two sources the decreasing yields as soil fertility declines and uncertain access to the European Community as the EC attempts to reduce cassava imports The latter will require lower price levels as Thailand looks to alternative international markets which in turn will result in a cost-price squeeze at the farm level effectively increasing the demand for improved technology The research program is in a position where a new variety in and of itself will not have a high probability of markedly 1mproving yields This will occur only if the variety is combined with a viable soil fertility management strategy The first signs of farmer adoption of fertihzer are occurring in the old production areas of Chomburi and Rayong Motivating this trend will provide the base for yield ga1ns though new varieties Markets and Demand The development of the Thai cassava economy (together with that of Malaysia) has followed the reverse of the normal pattern That is growth in production was initially driven by export market development Only after export market channels were well in place did domestic markets of any size begin to develop Price formation was always based on cassava as a tradeable good in international markets and Thai farmers and cassava processors based their dec1sions on price 1ncent1ves set in these markets An analysis of the Thai cassava economy in thus dependent on an evaluation of cassava demand in international markets (see Chapter VIII) and of price format1on in these markets The Cassava Pellet Export Market The export market for cassava chips and pellets dominates the Thai cassava economy High grain prices in Europe first in West Germany and TABLE 7 12 Thailand Yield Effect of Various Green Manure Crops on Succeeding Crop of Cassava 1970 Treatment Crotalaria JUncea Dolichos biflorus Vigna sinensis Phaseolus mungo Phaseolus calcaratus N-P-K (50-50-25) No green manure Yie1d (t/ha) 26 8 29 6 32 2 27 3 25 5 27 3 20 4 Source Inter1ma Committee for Coordina- tion of Investigation of the Lower Mekong Basin Agricul- tura! Research Effic1ency in Thailand Cassava 1979 VII - 10 - later w~thin the larger EEC have provided the genes~s for Tha~ chip and pellet exports These markets have been able to absorb the rapid expansion in export volumes to the extent that Thailand has not had to diversify its markets that is uptil 1983 Thai success however has given rise to European discontent and in 1982 a agreement for voluntary export restraint was negotiated and signed between the two parties (a lengthy discussion of the structure of the European market of the history of cassava imports into Europe and of the details of the quota is found in Chapter VIII) The quota while slowing growth in Thai exports nevertheless has not stopped it completely (Table 7 13) The pattern of growth in the Thai cassava industry is relatively unique when compared to cases of rapid expansion in other agricultural commodities especially the gra~ns The difference comes in the fact that cassava has to be processed very close to the production po~nt because of its bulkiness and rapid perishability Sugar cane and palm o~l have similar characteristics and in their case relatively large scale processing units have usually been linked to core plantations though if properly planned smallholders can provide a certa~n percentage of the raw material production However in the case of cassava the expansion in root production and processing has been based on link~ng small-scale producers to relatively small-scale processing capacity Decentralized small-scale processing is thus a solution to the problem of minim~zing transport costs where in the case of sugar cane or palm oil the solution is plantat~ons Moreover growth ~n production can be more easily syncronized with needed investment in processing capacity This is typical of cassava development other examples are gari in West Africa and farinha de mandioca ~n Braz~l This development pattern allows cassava both to maintain a small-farm focus to maximize the employment generation in production and processing and to distribute more equitably income growth as the industry expands The development of investment in processing capacity is portrayed in Table 7 14 The data suggest a pattern that first depends on concentration of investment in a few limited areas About 78% of all chipping plants in 1973 were located in only four changwats 60% were located in only two Rayong in the Central Plain and Nakhon Ratchas~ma in the Northeast By 1978 these same four changwats accounted for JUSt 417 of all chipping plants Root product~on followed much the same organic growth process That is development of the industry was based in~tially on the establ~shment of growth nodes where increasing dens~ty of production made for a more efficient cassava root market This concentration in turn allowed the orderly evolution of market channels to the export points By 1978 the next phase in th~s growth process ~s apparent i e rapid expansion of processing capacity into other changwats especially in the Northeast and expansion in processing scale in those original areas where product~on density had reached a certain critical point such that transport costs were not a constraint on scale expansion A certain production density is necessary to support efficient large-scale cassava processing This organic development of the Tha~ cassava ~ndustry has induced a continual search for cost reductions especially in processing storage and transport In the 1960 s this was policy induced as the EEC varied its tariff rates en meal versus chips (see Chapter VIII) The binding of the duty in 1968 provided the market security to JUStify investments leading to other cost reductions The first large ~nvestments came in the form of TABLE 7 13 Thailand Exports of Cassava Products Dest~ned for Animal Feed Use 1960-83 Year Chips Me al Pellets Waste Total (000 t) (000 t) (000 t) (000 t) (000 t) 1960 3 o 64 6 25 ü 93 6 1961 8 4 188 4 18 6 215 4 1962 12 7 267 7 9 6 290 o 1963 93 4 189 8 22 4 305 6 1964 339 4 202 3 45 5 587 2 1965 400 5 79 o 97 8 577 3 1966 359 8 65 8 107 9 533 5 1967 337 4 174 8 70 2 582 4 1968 323 2 388 8 33 1 853 7 1969 56 4 27 7 752 7 16 9 1 181 9 1970 8 1 4 o 1 163 9 5 9 972 1 1971 2 5 1 5 963 9 4 2 1 181 6 1972 2 4 o 6 1 177 4 1 2 1 659 o 1973 18 2 o 6 1 638 7 1 5 2 139 6 1974 105 3 1 o 2 031 5 1 8 2 240 5 1975 70 6 2 168 7 1 2 3 484 9 1976 43 4 o 2 3 441 3 3 752 9 1977 65 6 o 5 3 686 7 o 1 6 052 3 1978 255 6 o 2 5 796 1 o 4 6 052 3 1979 142 o o 4 3 695 8 o 3 3 838 5 1980 159 2 2 7 4 811 2 4 973 1 1981 334 4 o 6 5 620 2 o 6 5 955 8 1982 523 1 9 7 6 892 8 o 5 7 426 1 1983 280 o 4 8 4 545 1 o 3 4 830 2 Source Center for Agricultural Statist~cs Office of Agricultural Economics Ministry of Agr~culture and Cooperatives Bangkok 'OO!LE 7 14 Tha.i1m1d Evolution of Processing Capacity for Cassava Chlps and Pellets by ~t 1973-85 Chlp Pellet ~t 1973 1978 1985 1973 1978 1985 (rumber) (rumber) ( OOJ t capacity) (rumber) (number) ( 00) t cap<1 North 88 95 900 lO 24 2312 4 JCarrvbaeng Phet 80 35 24 3 6 5 360 o Nakhon Sawan 5 34 18 4 1 10 943 2 Ch:!ang Rai 10 7 1 1 Phitsarrulok 6 35 5 2 4 345 6 Uthai Thani 2 4 o 1 1 2 532 8 Northeast 421 1 777 7 860 7 24 305 20 736 o Kalasin 36 159 625 o 2 5 381 6 Klnn Kaen 252 7750 58 4 406 4 Chaiyaphum 2 41 632 5 17 10440 Nakhon Phancm 6 28 172 3 1 7 871 2 Nakhon Ratchasima 356 617 3 934 2 10 114 7 855 2 Buri Ram 4 108 543 7 4 21 1 036 8 Maha Sarakharm 1 60 284 3 23 396 o Roi Et 3 97 221 1 7 475 2 Nong Khai 1 45 203 4 2 9 410 4 Udon Thani 4 18 234 1 3 235 1 540 8 Surin 24 222 10 1 483 2 Central Plain 641 1 375 1 812 3 141 287 19 843 5 Kanchanaburi 25 58 63 9 4 5 158 4 Suphan Buri 29 62 47 9 4 8 828 o 01achengsao 40 134 315 8 29 3 420 o Chon Buri 113 348 991 2 115 126 8 553 6 Trat 27 58 21 8 15 6 Prachin Buri 32 230 120 4 33 1 785 6 Rayong 345 328 176 6 11 62 2 368 8 Total Tha.i1m1d 1 152 3 254 13 698 175 618 42 892 Source Division of Factory Control and Industrial Economics Ministry of Industrv Bangkok VII - 11 - pelleting capacity The obJective here was to reduce transport costs by increasing the density (Table 7 15) These were first based on the importation of European pelleters but this was shortly followed by the manufacture of pelleting machines in Tha~land This gave r~se to a quality distinction of brand versus native pellets with the latter having a lower density being softer and not having a pure composition (Mathot 197 4 explores in detail the techn~cal and economic factors determining pellet quality in Thailand) Accord~ng to export statistics Thailand converted from exporting mea! and ch~ps in 1968 to exporting virtually all pellets ~n 1969 that ~s 750 thousand tons Reports suggest the first pelleters were established in 1967 Investment in pelleting capacity was thus rapid and was independent of chip processing Investment in pellet~ng rel~ed on a signif~cant chip production capacity and a marg~n defined by transport cost advantages both internally and in the export trade Nevertheless pelleting plants were not large A 1974/75 survey identified three types of plants a small-scale plant with an annual capacity of 1260 tons a medium-scale plant producing 3310 tons and large-scale plants with a capac~ty of 7280 tons (Titapiwatanakun 1979) Interest~ngly these were not much larger than the average production capacity of chip plants and thus suggest no economies of scale in pelleting That is since chipping and drying gets over the perishability and transport constra~nt and since chip product~on was relatively concentrated any economies of scale in pelleting would have suggested ~nvestment in larger centralized plants There were no econom~es of scale in native pellets however for hard pellets produced with steam and/or a vegetable oil binder scale economies did seem to exist The cost savings on the ut~lization side in hard pellets are three First density is greater so there is a transport savings Second for feed concentrate manufacturera hard pellets do not require as much modification ~n factory transport systems i e essentially adapted for grains Third hard pellets can be stored longer allowing fewer storage losses Also there was a significant decline in dust pollution which previously had remained an externality and was dealt with by public funds in ports such as Rotterdam The price d~fferential result~ng from these savings however was through the 1970 s never suff~cient to motivate a larger production of hard or brand pellets Most majar cassava users ~n Europe especially ~n the Netherlands made the necessary investments to handle the higher meal content of native pellets in the feed plants and the ports Investment in hard pelleting capacity started to ~ncrease in 1982 at the start of the quota and by 1985 over 804 of pellet exports were in the form of hard pellets What is ~ronical ~s that ~nvestment came at a time when prospects ~n the EEC market were very uncertain Two factors prompted th~s conversion First the quota resulted in a large stock build-up initially due to the quota restriction and beginning ~n 1983 as a means for the Thai government to allocate the quota (see Chapter VIII) Storage costs (pellet dens~ty) and storage time thus become key constra~nts leading to an ~nternal demand for hard pellets Second the quota allocat~on procedure forced the big shippers [transnational corporations in the international grain trade (see T~tapiwatanakun 1982) who managed the European end of the market] to secure more certain control over supplies in order to guarantee their forward contracting in Europe They did th~s by backward integration into large-scale hard pellet~ng plants usually of European manufacture Thai manufacturera did follow with their own cheaper models to upgrade native pelleting plants These produced a quasi-hard-pellet an intermediate product between nat~ve and hard pellets As the ~ndustry developed large investments were also made in storage and loading facilit~es at export points A reflection of this investment is the change in size of sh~p that carried cassava Table 7 16 charts the progressive change to larger bulk-cassava carriers which in turn impl~ed investment in loading facilities in Thailand In 1980 the average cargo size for a ship hauling cassava was 87 thousand tons This compares to an average size of 41 thousand tons for ships hauling grains of North American origin The Thai cassava trade was able to capture significant economies of scale in ocean transport with Rotterdam being the only port that could take advantage of these scale economies Prices of cassava pellets in Hamburg for example are as much as 50 deutsche marks more expensive per ton than in Rotterdam Moreover cassava shipments to the United K~ngdom are usually unloaded in Rotterdam and sent on lighter to U K ports As in biology so in econom~cs growth is a far more complex process than surface -- or macro -- appearances would suggest Thailand in many ways offers an idealized growth pattern for cassava Early growth based on small-scale production and processing insures syncronization between the two in the growth process Economies of scale are possible then when critica! market size and production densit~es are reached It is important to v~sualize cassava in this more dynamic sense when the comparative advantage of cassava versus grains is discussed later in the chapter Also what is important about the Thai cassava case is the rapid growth in investment in a industry characterized by relatively small-scale plants and the forward linkages that were made to domest~c manufacturing capacity Investment in small-scale rural based industries is a particular characteristic of Asian agriculture -- one is tempted to attribute this to the constra~ned land resource base and the need for alternative employment in the rural sector the history of investment in the rural sector particularly irrigation and generally low incomes which makes even margins in small-scale processing attractive Cassava is in more ways than one well adapted to Asian conditions (see Chapter IX) Pr~ce Formation Price is the trottle that has controlled growth in the Tha~ cassava ~ndustry Understanding how prices for cassava pellets are formed will thus provide a basis for assessing both future prospects and an appropr~ate response to the EEC quota Because the maJor portion of Thai pellets are exported of which almost all go to the EEC the price of pellets in Thailand and the price of pellets in Europe are interdependent The policy history of cassava in the EEC is discussed in Chapter VIII but suffice it here to say that since the binding in GATT of cassava at a 6/ ad valorem duty in 1968 cassava has had a competitive edge over gra~n imports which must enter under the EEC s variable levy system Since domestic grain prices ~n the EEC are normally well above world grain prices and through the Common Agricultura! Policy insulated from international market cond~t~ons the cassava pr~ce is formed within the relative confines of the EEC market The implications for the cassava price is shown ~n Figure 7 3 where the Rotterdam cassava pr~ce and the maize threshold price TABLE 7 15 Thailand Weight per Unit Volume for Differ- ent Cassava Products Product Chips Native Pellets Hard Pellets (Steam) Weight/Volume 3 (g/ cm ) 412 569 808 Percentage Increase in Density (%) 38 96 - - -- f ~ ,_,1_1 .., i"""" --... u_.•- _,. - ~ J - - 1 -- TABLE 7 16 Year 1967 1970 1975 1980 Thailand Twin Deck Vessel (%) lOO lOO 43 2 Size of Ship Unloading Cassava in the Rotterdam Port 1967-80 Percent of Cassava Trade Carried by Bulkcarrier Bulkcarrier Less than 60 000 tons More then 60 000 tons (%) (i') - o o o o 57 o 8 90 Source Graan Elevator Maatschappij (g e m ) b v Rotterdam VII - 13 - are compared to the cif price of maize in Rotterdam World market maize prices and interna! EEC maize prices have significantly diverged over the last decade and a half However although cassava prices have remained above world market maize prices (at least on a feed equivalent basis) cassava has gotten relatively cheaper compared to EEC priced grains Export demand for Thai cassava and therefore the export price is determined by the prices for feed componente ~n the EEC -- import demand for cassava in Europe is analyzed in Chapter VIII -- however supply side factors may as well be affecting price formation in cassava The structure of the pellet market argues for the formation of cassava prices in the EEC feed component market with European prices being transmitted back to Thailand The carriers or shippers are key agents in price formation and transmission They are the interface between the European and Thai markets Moreover cassava is sold on an fob basis in Rotterdam That is the sh~ppers assume ownership of the cassava until its unloading in Europe Grains on the other hand are sold on a cif basis where the feed compounder has assumed ownership in say the Chicago market As well the maJor portion of cassava is sold on a forward basis That is a compounder contracts a certa~n quantity of cassava at a specified price for delivery some months forward and the shipper in turn buys in Tha~land in order to lock in the margin on his sale The shipper obviously must be in a position to monitor market conditions in both Tha~land and Europe and companies such as Krohn & Co Peter Cremer and Alfred C Toepfer are European-based companies with significant investments in Thailand To demonstrate the price linkage between the two markets and to evaluate the locus of price formation European and Thai cassava prices are analyzed in a framework which evaluates "causality between the two price series The concept of Granger causality is used in the sense that European prices cause Thai prices if the European prices lead the Tha~ prices in a sense defined by correlation between lags in the two series (see Bessler and Brandt 1982 Spriggs Kaylen and Bessler 1982 and Adamowicz Baah and Hawkins 1984) The methodology rests on prefiltering any autocorrelation in each series using an ARIMA estimation In this case the series of residuals could be reduced to a white noise series using the same prefilter -- this allows a valid test of Granger causality (Sims 1972) The residuals were then cross-correlated with varying lags The correlations then suggest the degree to which European prices lead (cause) Thai cassava prices Four European price series are utilized representing two markets Rotterdam and Hamburg and representing spot market prices and the two-month forward contract price All European prices are from the German agr~cultural market intelligence paper Ernahrungsdienst These series are analyzed in relationship to the Bangkok wholesale price for cassava pellets published by the Thai Tapioca Trade Associat~on in their Tapioca Products Market Review Pr~ces were available on a bi-weekly and a monthly basis and a ser~es of both time periods are analyzed from 1974 through 1985 The period is divided into two pre-quota and post-quota ~n order to assess the impact of import restrictions on pr~ce relationships between the two markets The cross-correlations between the Tha~ and European price series are presented in Table 7 17 First considering only the bi-weekly series two TABLE 7 17 Tha:iJBnd Cross-correlations betloeen Prefiltered Prlce Series for Thailand and Europe 197~5 Thailand Tho funth Forward Price Spot Price Leads(+) or lags(-) over Rotterdam Hamburg Rotterdam Hsmburg Furope Jan 1974 Oct 1982 Jan 1974 Oct 1982 Jan 1974 Oct 1982 Jan 1974 Oct 1982 Sept 1982 Dec 1985 Sept 1982 Dec 198S Sept 1982 Dec 198S Sept 1982 Dec 1985 Bi"""kl~ +3 periods 010 006 o 03 004 ...{) 03 o 02 -006 o os +2 periods o 07 o 01 009 o 03 o 07 o 01 009 000 +l period o 21** ...{) 07 o 44** o 12 o 19** 02()k o 18** o 25* sinultanerus o 52** o 29** o 32** o 21* o 44** o 26* o 44** 026* -1 period 006 029** 011 02()k o 07 013 ...{) 01 -007 -2 periods 009 o os o 01 006 004 ...{) 02 006 o 02 -3 periods 008 011 o 03 -010 o 03 -009 ...{) 05 008 M:mthl~ +3 periods o 05 -O lO o 06 ...{) 17 o 15 ...{) 20 o 06 ...{) 19 +2 periods o 19* 011 o 03 o 33* o 07 000 o 05 006 +1 period o 15 013 o 14 o 29* -006 011 -009 o 01 s:urultanerus o 51** o 23 o 62** o 27 o 54** o 30** o 48** o 43** -1 period o 22** o 38* 022** -008 o 25** o 27 023** o 03 -2 periods o 07 o 12 o 07 o 22 008 ...{) 02 ...{) 02 o 14 -3 periods -011 o 23 ...{) 23** o 39* ...{) 23 040k ...{) 23 o 24 Note ** int>lies s:lgnificance at 1% level and * int>lies s:lgnificance at 10% level Source CIAT structural features of the market are confirmed that is the forward price generally gives a higher correlation between markets than the spot price and in the case of the forward price the Rotterdam market is more closely linked to the Thai market then is the Hamburg market (for the spot price the correlations are virtually the same comparing Rotterdam and Hamburg) Considering then only the case of the forward price Bangkok and Rotterdam prices in the 1974-82 period are significantly instantaneously correlated i e within the two-week time frame This representa relatively efficient flows of information between the two markets and therefore relatively close price integration Somewhat contrary to expectation there is also some residual tendency for the Bangkok price to lead (cause) the Rotterdam price In the very short-run this indicates that the short-term supply situation in Thailand i e the ability of the shipper to fill his forward contracts influences the price negotiated in Europe This situation is even more marked in the case of Hamburg and again indicates that Hamburg is not as rapidly integrated with the Bangkok market as is Rotterdam The quota has radically changed this situation The strength of integration between the two markets has declined as reflected in the lower correlation coefficients As will be shown later this has resulted in a widening in the margin between the two price ser1es Moreover although instantaneous causality between the two series is still apparent European prices under the quota lead Bangkok prices Under the quota short term supply needs are adequately met by stocks while in Europe cassava supplies are constrained by the quota Cassava does not have to sell at much of a discount to grains in order to move available supplies Therefore short-term price formation shifted over to demand side factors but with a decline in the strength of the direct pr1ce transmission back to Thailand Price transmission between Europe and Thailand in the past has run in both directions but for monthly data at least the analys1s suggests that Europe leads the Thai price The price transmission process is then analyzed by making Thai cassava prices a function of European prices at varying lags the transport costs and a dummy variable for the quota period The results in Table 7 18 suggest that only 49% of price changes in Europe is passed back to Thailand in the first month and another 29% in the second month The transport cost variable was negative as expected but not significant This was due to the inability to construct a series that reflected the change in scale of shipping during the period the variable as specified assumes the same size ship Finally the dummy variable for the quota period is negative implying that the marg1n between Europe and Thailand has widened under the quota Th1s is to be expected with upward pressure on cassava prices in Europe due to a constrained supply and downward pressure on prices in Thailand due to rising stock levels As is explained in Chapter VIII Thai quota management pol1cy has utilized this larger margin to finance third-country exports rather than allowing a w1defall profit to accrue to cassava export companies The previous analysis argued that the locus of price formation in this cassava market occurs either at the level of negotiations between the shipping company and European feed manufacturer or between the shipping company and Thai suppliers the type of supplier depending on how far back into the market the shipping company is integrated This impl1es that root and chip pr1ces are determined by pellet prices whether set in Europe or TABLE 7 18 Thailand Estimates of Price Transm~ssion Equations between Europe and Thailand 1974-8 4 Dependent Variable European Price Thai Price Intercept 8 36 -1 66 (2 05) (2 31) Price (no lag) o 64 o 48 (O 08) (O 06) Price (one month lag) o 11 o 28 (O 09) (O 06) Price (two month lag) o 14 o 02 (O 08) (O 06) Transport Cost Index o 07 -0 03 (O 02) (O 02) Quota Dummy 4 30 -1 73 (O 98) (O 99) R2 o 62 o 55 Note European pr~ces were monthly two month forward cassava pellet prices in Rotterdam Thai prices were monthly wholesale Bangkok prices for cassava pellets Extimates were corrected for second-order autocorrelation Numbers in parentheses are standard deviations Source CIAT in Thailand This pattern is distinct from grains were normally processing is a mark-up on grain prices set in bulk wholesaling markets In the cassava situation the standard accounting for the chip and pelleting processing are P = e P + e + R and e e r e e P = e P + e + R p p e p p where p representa price e is conversion rate e is operating cost and R is operating profit and the subscripts refer to roots(r} chips(c) and pelleta (p} However given the assumptions on price formation price transmission equations for cassava chips and roots are as follows p 1 p - (e + R ) and = r e e e e e p 1 p (e + R ) = -e p p p e p Making the variables stochastic and assuming an error term the above equations were estimated and the results are presented in Table 7 19 The pellet equations follow expectations with the estimated conversion rates being within a reasonable range of but somewhat below the figure of 976 cited by industrial sources The estimated operating margin (per 100 kg ) however is s~gnificantly below the actual budgeted costs of pelleting (see below) Nevertheless what the price transmiss~on equations for pellets do suggest is quite restricted margins and therefore a very competit~ve industry The chip equations on the other hand only partially confirm expectations The convers~on ratea in ehonburi and Rayong are very close to the 372 figure used by industrial sources while the estimated conversion rate in Korat is unreasonably high suggesting a far higher leve! of efficiency than can be expected to be the case On the other hand the operating margin estimates cover a wide range from being reasonable in Korat to being significantly positive in ehonbur~ i e reflecting operating losses The equations suggest a delicate balance between operating margins and conversion rates a binding charactistic in the profitable operation of a chipping plant The equations again demonstrate the limited margins with~n which the chipping plants have to operate to turn a profit Given the chip price competition within the industry has generated relatively high root prices and limited operating margins Price formation in summary in the Tha~-European pellet market is efficient reflecting the very competitive nature of the Thai cassava industry Any excess profits when they occur either accrue to cassava farmers or result in inflated margins for the sh~pping companies (Figure 7 4) The later has occurred as a result of the imposition of the quota but Thai policy has issured that these windfall profits are directed towards opening up new markets for cassava pelleta 1 .-1 ·- Tl-¡IJ'' ......... ""1 1 j 1 """' 1- f •1Qt 91Íl [1? lf >?'C ¡: lJ!"?r,"' ,r l_ :::s:c 'r!] ,.!E' E-'""¡ Eut op'? ord F o n 1 -l'?., e ill'~j¡lfll/111 !lllll 1 1111 '" li ¡1'!111¡ 11 h 1 1111 !illhl 111 1' ,~, li,llhljllll 1 ll¡¡lf Jhllll'¡ll' 111" 11 1111111 11 hlll11111' ~ ......... -~ ~~ ........... ~ ~:--" .... ~ :r-~~~ ..,qr-c¡rt;:et' d :11 ,¡ Praduc'i:n::n Cc.rí: ~ j F .. rmitr Pr-cf¡t,. 11 Profitability of the Cassava Pellet Industry The very marked rate of growth in the Thai cassava industry was driven by the relative profitability of the industry especially since prices set in Europe were efficiently transmHted to cassava root producers The profitability of cassava at the farm level is shown in Figure 7 4 wh~ch presents a graphic picture of margin development in the cassava industry Farm-level profits were highly variable but even in years with low prices profits were sigm.ficant Not surprisingly root production showed continuous growth even with quite significant variability in prices Another major characteristic of the cassava industry is that the farm-level root price makes up only between 40 to 504 of the eventual f o b pr~ce By comparison farm level production costs make up 83% of f o b costs of maize in the U S A (Ortmann Stulip and Rask 1986) The ability of cassava to compete with grains thus lies in ~ts relatively low production costs and an efficient processing industry As seen in F~gure 7 4 the processing margin did not vary significantly over the 197 5-84 period Cassava is very profitable for Thailand A complete cost accounting for 1981 is summarized in Table 7 20 (see Appendix 7 2 for details) The costs are disaggregated by domestic factor costs foreign import costs and government taxes including tariffs All costs are at 1981 market prices with interest rates being at the commercial loan rate of 19% Íhere are no ind~cations of any market imperfections that would cause market prices of factors to deviate from their opportunity cost (see Bertrand 1980 and Lokaphadhana 1981) Nor until the quota was there any intervention by the government in the cassava export trade The Thai cassava industry was one of the few examples of an industry that functioned without government intervention Deducting taxes and tariffs thus closely approximates social costs of producing cassava The cost breakdown suggests that root production costs are two-thrids of total f o b costs of cassava pellets Chipping pelleting and export costs relatively equally divide the other th~rd Labor is by far the largest cost component making up 4 7% of total costs Import costs are relatively low making up only 11% of production costa Comparing costs to 1981 prices implies that almost 30% of the f o b price was garnered by the economy as social profit with almost two-thirds of that going to the cassava farmer From a social point of view cassava was very profitable to the Thai economy and especially for the incomes of the population in the poorest sector of the economy the rural Northeast The quota has made apparent the polit~cal underpinnings of the internat~onal market for cassava pellets Uncertainty about long-term access to the European market has raised the question about the ability of the Thai cassava industry to compete in the larger international feedgrain market The first point to emphasize is that because Thailand did not sell cassava in the international feedgra~n market up till the quota does not necessarily ~mply that cassava could not compete in that market The analysis to date and that presented in Chapter VIII clearly shows that Thailand could sell all its production ~n Europe at prices above what could have been obtained on the world feedgrain market obviously it was more profitable for Thailand to sell all ~ts production in the European market This situation has changed with the quota and the issue of cassava' s TABLE 7 19 Thailand Estimated Equations for MargLn Determination for Chips and Pellets 1974-84 Roots to Chips Chips to Pellets Chonburi Rayong Korat Chomburi Korat Margin 8 63 o 53 -18 09 -6 39 -8 41 (Baht/100kg) (2 19) (2 05) (3 35) (1 81) (2 12) ConversLon Rate o 35 o 37 o 52 o 94 o 91 (O 01) (O 01) (O 02) (O 01) (O 01) R2 077 o 82 o 79 o 98 o 97 Note Numbers Ln parentheses are standard deviations Source CIAT TABLE 7 20 Thailand Social Cost Accounting of Cassava Pellet Exports 1980-81 Total Farm Chipper Pelle ter Exporter Costs (Baht/t) (Baht/t) (Baht/t) (Baht/t) (Baht/t) Purchase Price 1480 1792 1958 Sales Price 1480 1792 1958 2471 2471 Factor Costs Land 140 4 140 4 Labor 655 1 45 4 51 1 43 7 795 3 Capital 251 8 74 9 119 1 131 4 577 2 Foreign Exchange Costs 76 4 48 o 59 2 183 6 Total Costs 1123 7 1648 3 2021 4 2133 o 1696 5 Government Tax 22 7 23 6 27 9 18 4 92 6 Rent 333 6 120 1 -91 3 319 5 681 9 Source Appendix 7 2 TABLE 7 21 Comparison of Costs of Maize from ~~jor Exporters and Cassava (on a maize equivalent basis) from Thailand cif Japan Maize Cassava U S A Argentina Brazil Thailand ($/t) ($/t) ($/t) ($/t) Production Costs Variable Costs 60 o 37 9 66 6 52 6 Fixed Costs 59 8 32 9 68 2 7 7 Total Costs 119 8 70 8 134 8 60 3 Marketing and Processing 24 7 25 3 33 9 33 8 F O B Costs 144 5 96 1 168 7 94 1 Freight to Japan 26 o 32 4 34 2 10 o e r F Costs 170 5 128 5 202 9 104 1 Yield (t/ha) 6 25 3 36 2 22 5 22 Note All costs are at 1985 prices and exchange rates Thai cassava costs represent 1981 costs multiplied by wholesale price index and divided by 1985 exchange rate Costs are then put on a maize equivalent basis by dividing by O 7 Source Maize Ortmann G U J Stulp and N Rask International Trade and Economic Development Examples of Comparative Costs in Inter- national Commodities 1986 and Cassava CIAT ability to compete (In Chapter VIII wider market while allotment) in the wider feedgrain market is now a policy concern the issue is addressed of how Thailand develops this continuing to garner the social profits from the quota International comparative advantage has commonly been analyzed within a domestic resource cost framework (Pearson Akrasanee and Nelson 1976) This methodology takes border prices (f o b prices for exporters and e i f prices for importers) as the measure against which comparative advantage is assessed A good summary statistic is the resource cost ratio (Page and Stryker 1981) where any country with a ratio less than one has a comparative advantage in the production of that commodity For cassava in 1981 us1ng Thai f o b prices the RCR was 71 indicating significant comparative advantage in supplying cassava to the European market To evaluate social profitability of selling on the international grain market the break-even price (the f o b price at which the RCR is one) is calculated This price is $77/t Assuming that under normal circumstances cassava competes with maize at about 7 of the maize price (see Chapter VIII) then the maize equivalent price is $110/t This compares very favorably to the f o b price of maize in Thailand and in the U S in the 1980 1 S The issue can be taken one step further and f o b costs compared to f o b costs of major maize exporters (Table 7 21) Comparing Thai cassava costs on a maize equivalent basis with those developed by Ortmann Stulip and Rask (1986) shows that cassava is very competitive with maJor ma1ze exporters How much cassava Thailand will produce at currently declin1ng world market maize prices is another issue but the same could be asked of countries such as the United States and France if price and income support policies were eliminated In summary the Thai cassava industry has shown itself to be very responsive to export opportunities and to the vagaries of policy changes in import markets The EEC became virtually the sole market for Thai pellets essentially because it was the most profitable outlet Moreover because of efficient price transmission between the two markets Thailand could respond very quickly to the changing needs of the European market The imposition of the quota in 1982 has forced Thailand to begin to restructure its export markets a subject discussed in Chapter VIII What that analysis shows is that Thailand has adjusted to the quota by opening new markets in East Asia thereby allowing domestic production to continue to grow The growth of the Thai pellet industry also offers a more general lesson about the development of comparative advantage in the crop Comparative advantage of cassava versus grain substitutes is based on certain physical characteristics particularly the availability of land with low opportun1ty cost and an agricultura! sector with a relatively small farm-size structure However there is also a time and scale dimension to comparative advantage because of the critica! importance of the processing component since it makes up from a third to a half of the total costs In cassava economies of scale 1n processing develop over time in relation to the concentration of production on the one hand and the size of the output market on the other Malaysia and Indonesia have VII - 18 - attempted to force the issue through plantation development but in cassava these have not been notably successful The soc~al equity benefits from cassava development (marginal agr~cultural areas small-scale producers and rural employment in small-scale agro-industry) prov~de strong support in certain circumstances for an infant industry argument to support cassava in the initial development of its process~ng capacity In Thailand this ~nitial protection was provided by the EEC market The Thai case suggests that cassava can compete with grains but in the evaluation of the comparat~ve advantage of cassava in the feedgrain market a time perspective should be incorporated for processing costs The Cassava Starch Market The cassava industry in Thailand developed initially on the bas~s of the market for starch Starch production and exports have continued to grow throughout the post-war period but the industry has declined in relative importance having been eclipsed by the cassava pellet market Nevertheless the cassava starch industry ~n Thailand v~es with Indonesia as being the largest in the world It continues to be dynam~c suppling starch to both an expanding export market and an increasing domestic market Constructing a supply and util~zation series for cassava starch must rely on data from different sources and this produces some inconsistencies The series in Table 7 22 is developed from independent export product~on and util~zation estimates and representa the author s efforts at achieving consistency between the estimates What the data suggests is quite significant growth in starch production driven through the 1970 s by rising domestic consumption and in the 1980 s by a sudden spurt in the export market Cassava starch has a w~de number of end markets in Thailand The principal use is as a raw mater~al in the production of monosodium glutamate In th~s industry starch competes directly with molasses which is interchangeable with cassava starch Starch is also important in the expanding pulp and paper industry in textile production and in food industries All of these are grow~ng industties and cassava starch will continue to enj oy an increasing domes tic market throughout th~s century However unlike other starch markets in East Asia one market which cassava starch has not entered is the glucose and sweetner market This is principally because Thailand is a producer and net exporter of sugar High fructose sweetners derived from cassava have been advocated as another possible market since 52% of industr~al sugar consumption ~s for beverage production (Frankel 1981) Moreover the Tha~ government has a pol~cy of subsidizing sugar exporta when world prices are low and taxing exports when prices are high (Lokaphadhana 1981) Nevertheless the price variability in cassava starch prices has made the investments needed in large-scale plant and capacity too risky and there has been no development in th~s market Tha~land is virtually the sole exporter of cassava starch and the largest exporter ~n the world of starch in general The export market was relat~vely stable through the 1960 s and 1970 s but increased dramatically in the 1980 s as new non-traditional importers came into the market (see Chapter VIII) Thailand between 1980 and 1985 was able to expand exports by 50% in two years and virtually to double export volumes in four years without too much affect on domestic consumption levels Th~s suggests the investment in significant excess production capacity for starch on the one hand and the ability of the starch industry to compete effectively with the pellet industry for roots in 1984 and 1985 root prices were relatively low due to the quota The starch industry needs to be very competitive in the sense that its margins are defined by root prices principally set by the pellet export market in the EEC and starch export prices set principally by international maize prices i e the dominant cost in maize starch product~on (see Chapter VIII) The starch industry very early began a search for scale economies in processing essentially based on large-scale plants but with equipment manufactured in Thailand -- in Indonesia on the other hand these scale economies in starch production do not exist (Nelson 1984) Based on the development of this market Thailand is a now net exporter of cassava starch equipment including complete plants However with this competition to invest in order to lower processing costs excess processing capacity was created allowing the industry to respond so quickly to new export markets Price Formation and Profitability Like other cassava processing industries profHabilHy in starch production is primarily dependent on the conversion rate and the margin between the root buying price and the starch selling price Unlike the pellet industry where the price of the processed product leads the price of roots the starch industry must take the root price as a given The starch industry rarely has been able to underbid the chipping plants The root price thus sets the price of starch Competition for limited markets in turn insures both downward pressure on margins and the search for reductions in processing costs The above scenario for price formation is adequately captured in the pr~ce transmission equations in Table 7 23 and the processing cost analysis in Table 7 24 Note that contrary to the chip industry starch price is the dependent variable in the regression equat~on The estimated conversion rates are only slightly higher than the estimate of 4 34 tons of roots for every ton of starch given by industrial sources Even the estimated rates suggest very high technical efficiency in starch extraction The estimated operating margin compares favorably with the budgeting analysis in Table 7 24 Again the evidence suggests a very competitive industry where there is no indication of excess profits Moreover a domestic resource calculation would be redundant in the case of Thai starch since Thailand sets the world price for cassava starch and apart from import duties on starch processing equipment there is no government intervention in the starch market Continued growth in the starch industry is dependent principally on the supply price of starch which in turn is dependent on the root price and the changing dynam~cs of the pellet market The tendency in the medium term is for cassava starch prices to come in line with maize starch making cassava starch more competitive The other major factor of course is growth in export markets Prospects in the international starch market are TABLE 7 22 Thailand Cassava Starch Production and Dlsappearance 197o-83 llcm!stic Consumption fu1osodium Paper Textile Food Total Year Glutamate Industry Industry Industry Other Export Di.sappearance Production (OCOt) (OCOt) (OCOt) (OCOt) (OCOt) (OCOt) (OCOt) (OCOt) 1970 234 6 8 6 8 360 7 1 144 7 224 8 173 6 1971 290 7 9 8 4 37 1 8 1 149 8 240 3 157 6 1972 33 3 104 9 o 382 107 129 2 230 8 201 1 1973 346 103 10 1 39 3 139 176 7 284 9 286 8 1974 346 133 10 o 404 17 4 2525 368 2 315 7 1975 366 112 108 41 S 20 S -M!U 265 3 409 9 1976 33 S 15 4 131 42 S 24 6 236 3 365 4 513 o 1977 37 2 18 9 135 43 6 288 200 8 342 8 538 S 1978 408 20 1 14 3 447 33 2 235 9 389 o 411 o 1979 382 24 7 14 S 45 7 387 122 S 284 3 3050 1980 37 2 26 2 15 8 460 43 1 243 6 411 9 432 9 1981 57 7 31 3 14 3 469 361 ~3081 494 4 504 1 1982 547 37 3 14 8 47 8 429~~ 3870 584 S 590 1 1983 608 444 15 3 488 47 2 tfV 363 S 5800 573 9 Note Di.sappearance and production data are derlved from different sources llireover change in stocks are not included There is a definite discrepancy in the 1970-72 perlad Soorce Production Industrial Econonú.cs and Planning Division Ministry of Wustry Bangkok Danestl.c Consumption Titapiwatanakun Boon]it 'IX:rnestic Tapioca Starch Consumption in Thailand 1982 Exporta Center for Agricultura! Statl.stics Office of Agricultura! &omm:l.cs Ministry of Agrlculture and Cooperatives Bangkok TABLE 7 23 Margin Thailand Estimated Equations for Margin Determination in Starch Processing 1974-84 Chonburi Rayong 108 7 116 4 (25 6) (20 3) Conversion Rate 4 73 4 91 (O 35) (O 29) R2 o 61 o 70 Note Numbers in parentheses are standard deviations Source CIAT TABLE 7 24 Thailand Costs of Production of Starch in Large-Scale Processing Plant 1981 Cost Item Variable Costs Roots Labor Electricity Fuel for drier Fuel for vehicles Repair and maintenance Transport to Bangkok Working capital Sub-total Fixed Cost Admim.stration Capital depreciation Fixed capital costs Sub-total Total Costs Costs no including roots Starch Price Value of Cassava Waste Cost (Baht/t of starch) 2608 7 142 o 366 7 235 o 16 o 264 8 120 o 30 6 3783 8 41 8 116 3 251 7 409 8 4193 6 1584 9 3750 365 Note The capacity of the plant is 100t of starch per day and produced 15 S thousand tons in average year The conversion rate is 4 35 tons of roots for 1 ton of starch Source CIAT survey VII - 20 - analyzed in Chapter VIII and suggest that markets open only where the country loses the ability to meet its own domestic needs The Animal Feed Market There is no better illustration of the lack of integration between world market maize and cassava prices than the comparative role that these two export crops have played in the development of Tha~land s domestic feed concentrate industry Maize has formed the carbohydrate base for this rapidly growing industry basically because it has been more profitable to export the cassava On those relatively rare occassions when the prices of the two commodities have come into line cassava has been used domestically in the manufacture of animal feeds This has happened more often since the imposition of the quota and given the current size of the domestic market the animal feed market could start to play a larger role in putting an absolute floor under cassava prices Starting in the late 1960's basic structural changes in the production of both swine and poultry have formed the basis for the rapid expansion in the feed concentrate industry Prior to this time both swine and poultry were raised in small-scale integrated crop-livestock systems Swine continues to be raised principally in the central plain This region is relatively close to the Bangkok market and forms the main rice growing area where rice bran and other by-products provide a plentiful feed source Commercial operations of over 50 hogs have increased their production share from approximately 12% in 1974 to 14-' in 1978 to around 15% in 1983 (Chesley 1985) Development of commercial swine operations however has been constrained by the Animal Slaughtering and Meat Control Act of 1959 which allows only local authorities to establish slaughterhouses and prohibHs shipment of carcasses outside the legally defined market area of each slaughterhouse This has resulted in local monopsonies in slaughter facilities resulting in h~gh costs and inefficient wholesaling of carcasses (see Chesley 1985 for further discussion) A high percentage of the slaughter is done illegally but this is difficult for large commercial growers Nevertheless swine numbers have continued to increase especially since the mid-1970 s (Table 7 25) Structural change in the poultry industry has been even more rapid (Table 7 25) often motivated through vert~cal integration of feed companies backwards to commercial poultry production units The broiler industry has been by far the most dynamic animal sector in Thaüand increasing nine-fold in the 1974-82 period Partly this arises from the restrictions on the pork sector and partly from the very rapid technical change in the poultry sector The later is reflected in the declining relative price of chicken compared to other meats (Figure 7 5) and a virtual doubling of per capita consumpt~on of chicken over the course of the 1970 s The only limita on growth in this industry a technically efficient industry with access to cheap feed sources is the size of the domestic market With total per cap~ta meat consumption still at relatively low levels and population and income still pro)ected to grow there ~s no hint yet of a downturn in growth Moreover Thailand is developing as a major exporter of poultry in the East Asian and Middle Eastern market TABLE 7 25 Thailand Swine and Poultry Population 1970-82 Year 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 Swine (thousand) 3215 3348 3335 3004 3256 3866 5201 5420 6713 7343 6589 6448 n a Poultry Cornmercial Village Chickens (million) Layers (million) -------- 154 2 --------- -------- 156 9 --------- -------- 148 2 --------- 126 2 7 4 105 9 7 o 83 1 8 9 92 9 9 o 76 9 9 6 61 1 10 4 Broilers (million) 36 4 41 6 58 2 78 o 104 o 130 o 200 o 234 o 286 o Total (million) 136 3 150 7 166 8 182 2 190 6 198 5 206 4 211 6 216 9 222 o 301 9 320 5 357 5 Source Derived from Chesley Merritt The Demand for Livestock Feed in Thailand 1985 VII - 21 - The dynamism in the meat sector has been integrally linked to a dynamic industrial feed sector Production of balanced feeds have ~ncreased from a mere 64 thousand tons in 1968 to 2 1 mill~on tons in 1984 Although initially based on swine feeds the real growth in production has come in broiler feeds This expansion in the feed sector has induced rapid ~ncreases in the derived demand for carbohydrate sources This demand has been met almost exclusively by domestically produced maize The maize sector has also been very dynam~c in the last two decades (Table 7 26) increasing from a production level of just over half a million tons in 1960 to well over 4 million tons in 1984 Production growth in the 1960's went almost exclusively into exports However since about 1970 a growing share has gene to meet the needs of the domestic feed sector and s~nce that point exports have been relatively stable at around 2 million tons Cassava's potential as a carbohydrate source in the animal feed market is defined in Table 7 27 and Figure 7 6 Cassava comes into the least cost feed ration when its price is about 67 of the price of maize This rat~o is somewhat low because the prices of soybean meal which is principally imported are maintained relatively high through import taxes These taxes have risen from 5 to 6 percent in the late 1970 s to 8 5 percent in 1983 (Chesley 1985) Thus cassava came into the ration in 1981 and again in 1984 Over the period 1971-85 cassava was never competit~vely priced with maize for any extended period of time (Figure 7 6) Thus cassava has never been a feature of the domestic feed market Nevertheless in 1985 feed manufacturers for the first time began to use sign~ficant volumes of cassava in their feed mixtures An estimated 625 thousand tons was used in feeds in 1985 However these competitive price relationships did not last through the end of 1985 and cassava again moved out of the ration This situation is in fact quite favorable for cassava producers The animal feed industry has a solid raw material supply in maize but when substitutes are cheaper manufactures can profitably mix them in their rations Price is the determining factor for these feed components not continu~ty of supply Since cassava is readily available feed manufacturers can easily move into cassava when price relatives are favorable As domestic feed manufacturers gain experience in us~ng cassava initially in swine feeds the domestic feed market could put an absolute price floor under the cassava market At these times cassava will essentially be competitive with world market feedgrain prices but the logical market on which to sell is the domest~c rather then the export market When cassava prices are above maize prices the cassava producer is much the better off The domestic animal feed market is now large enough that it can play such a role in supporting cassava prices Conclusions Cassava led the rapid post-war expansion in upland agriculture in Thailand While maize and sugarcane expanded principally in the Central Plain provinces cassava area increased first in the East and then expanded rapidly in the poorest area of Thailand the Northeast Thailand was able to base exploitation of an agricultura! front~er aided by labor-substituting technologies in the 1970 s on development of export markets This was as true for maize as it was for cassava The expansion in cassava started in the 1950's and continued through the early 1960's TABLE 7 26 Thailand Maize Production and Utilization 1960-61 1982-83 Total Feed Use Domestic Use Domes tic As % of b Use as Total Total Feed % of Total Domes tic Cropyear a Production Exports Total Use Production Use (000 t) (000 t) (000 t) (000 t) (%) (%) ----------------- (1000 tons) ----------------- 1960-61 544 519 10 2 2 20 1961-62 598 589 15 4 3 27 1962-63 665 722 15 4 2 27 1963-64 858 923 20 6 2 30 1964-65 935 896 25 10 3 40 1965-66 1021 1132 29 10 3 34 1966-67 1122 liBO 35 13 3 37 1967-68 1315 1214 55 25 4 45 1968-69 1507 1289 104 75 7 72 1969-70 1700 1502 176 140 10 80 1970-71 1938 1663 220 180 11 82 1971-72 2300 2111 280 235 12 84 1972-73 1315 1039 295 270 22 92 1973-74 2339 2ll2 348 300 15 86 1974-75 2500 1872 608 560 24 92 1975-76 2863 2442 313 250 11 80 1976-77 2675 1982 787 730 29 93 1977-78 1677 1297 397 365 24 92 1978-79 2791 2155 614 560 22 91 1979-80 2863 1825 652 590 23 90 1980-81 2998 2418 797 749 25 94 1981-82 3449 3079 846 821 24 97 1982-83 3002 2244 971 942 31 97 a All data are for July-June cropyears b Does not include beginning or ending stocks therefore exports and domestic consumption do not add up to production Source Ches ley Merritt The Demand for Livestock Feed in Thailand 1985 1 1 1 1 .J 1 p .:S J - .... 1 L ·:r :; -; l ! ' il 1 i o . .., r 1 O.t 11 J' - 1:' u.-' 11 1 ' 1 ' ~ ~ 1 11 11 1¡ Ra i:~o c.-= l='rbces ~"1 1 1 ti " i d ~ • 11 1'11 1 ' 1 ' ! 1 1 ' [1 1 ,.ll r • TABLE 7 27 Thailand Optimal Composition of Poultry Rations Derived in Least Cost Feed Formulation 1981-84 1981 1982 1983 1984 Ingredient Price Entry Price Entry Price Entry Price Entry (Baht/kg) O'l (Baht/kg) (%) (Baht/kg) (%) (Baht/kg) ( .C:l Cassava 1 91 9 6 2 11 o 2 51 o 1 70 25 o Maize 2 91 45 8 2 87 56 7 3 15 56 7 3 08 25 3 Soybean Meal 7 74 21 4 7 46 14 4 7 46 14 4 7 so 24 9 Fish Meal 11 09 7 S 10 54 7 S 10 99 7 S 11 00 7 S Note All ingredients are not shown here Kapok meal entered at a significant leve! in 1982 and 1983 Source Prices are wholesale Bangkok and are from the Office of Agricultura! Economics the model was developed by CIAT VII - 22 - being based princ~pally on the starch export market It is a mark of Thailand s ability to take optimum advantage of changes in international market conditions that with the GATT binding of the cassava tariff in 1968 creating a hole in the EEC s variable levy system Thai cassava exports could respond so rapidly Thus the Thai cassava boom should not be seen as un~quely determined by a favorable tariff rate in the EEC but equally ~mportant was the dynamism of upland agriculture and the addit~onal land and labor resources that could be brought into production in response to profitable export markets Thai success in cassava however has been at the expense of the EEC s political objectives The resulting voluntary export quota has created an air of uncertainty as Thailand has had to rapidly develop its own policy response and control procedures It is ironical indeed that Thailand' s only policy intervention in the cassava sector is a negative one even though forced by the EEC The uncertainty however should not be interpreted as portending eminent decline in the cassava industry Rather a period of structural adjustment has been forced on the industry which in the end will lay the basis for more d~versity in end markets and even more efficient production The short-run policy problem for Thailand has been to develop a policy that allows the country to capture the social prof~ts earned in the EEC and to the extent possible to transfer these benefits to cassava producers especially in the Northeast The solution requires an analysis of alternative export markets and this is left till Chapter VIII Suffice it to say that Thailand has managed to make the adjustment and expand its export markets principally in East Asia Moreover root production has even increased during the quota period Future growth w~ll be based on continued penetration of these new export markets Nevertheless there has been downward pressure on farm-level pr~ces under the quota and the more the need to export to third-country markets the more the downward pressure on root prices Over the past twenty years Thailand has significantly reduced cassava processing costs Farmers have also adjusted to rising labor costs by adopting labor-saving technolog~es What has not happened and what is becoming critical as root prices come down is the adoption of yield-~ncreasing technology Yields have remained relatively constant over the past twenty years even though area has expanded into more marginal areas and fertilizer has not been used in trad1tional growing areas Under current monocropping conditions y~elds will eventually decline catching farmers in a cost-price squeeze A fertilization and soil management strategy that quarantees a profitable return is needed to complement improved varieties This will insure the ability of Thai cassava to compete in the wider feedgrain market allow~ng Thailand the required flexibility in restructuring its export markets Most important of all cassava will then have achieved parity with grains in international markets establishing a new claim for carbohydrate exports from the tropics a role palm oil has recently carved out in the world vegetable oil market VII - 23 - Appendl.X 7 1 A Synthes~s of Production and Utilization Cassava production has grown rapidly in the last two and a half decades w~th most of the root production being processed for export Domestic consumption of cassava is limited te starch and the occas~onal use of chips ~n animal feed concentrates Thailand should be a country therefore where cassava utilization and production data are relatively consistent A production series is produced both by the Division of Agricultura! Economics (DAE) and the Department of Agr~cultural Extens~on (AEX) both of which form part of the Ministry of Agriculture and Cooperat~ves Both the DAE and AEX maintained the same series through the 1968/69 crop year but diverged then when the DAE changed procedures In general the DAE series is most ut~lized ~n the literature and is the ene reported by FAO Both ser~es show the same basic upward trend but in any particular year can diverge by as much as 25% Converting exports te a fresh weight basis and comparing this export series te the production series (Table 7A 1) shows that the production data tended te be consistently underestimated in the case of the AEX before 1973/74 and in the case of the DAE before 1982/83 Titapiwdtanakun (1979) reviews this discrepancy in some detail and attributes the difference te a failure te accurately monitor the rapid expansion in area especially where cassava was being planted in more frontier-like conditions in the Northeast The DAE production ser~es thus provides a relatively consistent underestimate of actual production and the export series probably provides a more accurate minimum estimate of actual product~on The Ministry of Commerce has developed supply and utilizat~on estimates for cassava (Table 7A 2) These clearly highlight the dominance of the export market but also identify a not un~mportant domest~c market for both starch and animal feed The other dominant component is the very high stock levels being held in this period The production estimate constructed from utilization data is about 11% larger than the DAE estimate of production Thus Thailand prov~des ene of the few cases (Malaysia is the other) where cassava production tends te be underestimated TABLE 7 A 1 Thailand Comparison of Root Production Series with Implied Production from Export Series 1960-85 Agricultura! Export Year Economics Extension Series (000 t) (000 t) (000 t) 1960 1083 1083 1109 1961 1222 1222 1706 1962 1726 1726 1298 1963 2077 2077 1341 1964 2111 2111 2089 1965 1557 1557 1864 1966 1475 1475 1850 1967 1892 1892 2265 1968 2063 2063 2487 1969 2611 2611 2684 1970 3079 2474 3645 1971 3431 2432 3169 1972 3114 3673 3575 1973 3974 4436 4995 1974 5443 7770 6554 1975 6765 9503 6238 1976 7094 11 638 9778 1977 10 230 13 554 10 242 1978 11 840 13 024 15 953 1979 16 358 12 877 10 023 1980 11 101 13 864 13 442 1981 16 540 17 204 16 160 1982 17 744 n a 20 147 1983 17 788 n a 13 718 1984 18 989 n a 17 014 1985 19 985 n a 18 812 Source Office of Agricultura! Economics Ministry of Agriculture and Cooperatives and Department of Agricultura! Extension M1.nistry of Agriculture and Cooperatives TABLE 7 A 2 Thailand Disappearance Domest~c Consumption Starch Animal Feed Export Starch Pellets and Chips Change in Stocks Total Production Harvested Unharvested Total Supply and Disappearance of Cassava (fresh weight basis) 1984-85 Fresh Root Equivalent (000 t) 1 100 625 2 435 15 365 1 731 21 256 21 256 1 000 22 256 Source Ministry of Commerce Bangkok VIII World and Asian Markets for Cassava Products World trade in cassava products has ~ncreased rapidly over the last three decades rising from about 200 thousand tons c~n product weight) in the early 1950's to a peak of 8 4 milhon tons in 1982 The latter represents a l~ttle less than 207 of total world product~on of cassava a very sign~ficant f~gure when compared to a commodity like rice where only 4% of production moves in world trade While the volume traded is sizeable by world commod~ty standards eg world r~ce trade amounts to a little over 8 million tons the number of countries involved is relat~vely small In fact over 904 of trade is accounted for by exports of Thailand to the European Community For a commodity trade of such volume this is a particularly narrow base Trade dominates the cassava economy only of Thailand and in the 1980-82 period China Trade ach~eves a more limited importance although rarely exceeding 10% of domestic production -- in Indonesia and ~ In all other cassava producing countries internatio~l trade has rarely been an option and is currently of only marg~nal importance This relatively un~que trade structure ra~ses a number of issues which will be explored in this chapter Most importantly the reasons surrounding the relatively narrow participation in world cassava trade will be examined This analysis will then lead to an evaluation of the potential for broadening the import markets for cassava followed by some prognosis for increasing the number of exporting countries The discussion will be rooted in an historical evaluation of the changing determ~nants of comparative advantage an approach which will allow some speculation on the future role of cassava in world trade in carbohydrate sources Protectionism and Substitution Decl~ne in the World Starch Trade World trade in cassava started with starch exports from the Malayan peninsula in the mid-1800 s Early trade relied on cassava s advantage as a starch source the higher value-added of starch compared to other processed cassava products and the proportionately lower freight costs for starch compared to dry cassava Starch was the maJor cassava product in value terms moved in world cassava trade throughout the present century up till the 1960 s The market for starch is relat~vely small in compar~son to trade in wheat or feed grains Moreover while this market exhibited moderate growth from the turn of the century to the Second World War there has been little growth ~n the post-war period while the grain trade has grown at histor~cally h~gh rates However underlying these trends ~n starch trade ~s a market structure undergoing s~gnificant change influenced by shifting comparative advantage dynamic technical change rapidly shifting end markets and trade barriers It is in these terms that the world market for cassava starch will be analyzed Demand for starch is marked by the product s versat~lity Almost ever"l ma]or industry has found a use for starch and as a result the process of industrial~zation normally coincides with a significant ~ncrease in the demand for starch This industrialization affect is partially reflected in the h~storical ser~es on ~mports of cassava starch over the present century At the turn of the century the Un~ted Kingdom was the largest ~mporter of cassava and other starches By the 1920 s the Un~ted States although a major producer of starch itself became the largest importer In the late 1970's the U S was overtaken by Japan and in the early 1980's Japan was superceded by Taiwan This pattern closely tracks the industrialization process characten.zing the world economy over the present century / However a possibly more important phenomenon is the eventual decline of imports of cassava starch into principal markets This decline in imports is not due to any falling off 1.n overall starch consumption ~t 1 rather the substitution of im orted starch by domestically produced starch 1 Over t1.me this substitution process has een accelerate on t e one hand 1' by advances in starch chemistry and the ability to chemically modify -1 <,_ \ starches thereby making starches more substitutable and on the other 1¡ hand by techn1.cal change in both maize product1.on and the maize wet 1 milling process reducing the unit costs for this starch and making it over ¡ the post-war per1.od the predominate starch produced 1.n the world Events l 1.n the U S played a dominant role in the declining market share of cassava and the rising share of maize in world starch consumption The analysis thus turns briefly to a consideration of the starch industry 1.n the Un1.ted States and the effect this industry has on the world starch market By the turn of the century following on the development of a successful processing technique in 1842 (Radley 1968) maize was the dominant starch produced and consumed in the U S Production of maize starch increased from 141 thousand tons in 1904 to 2 27 million tons in 1982 a sustained annual growth rate of 3 6% over the course of almost 80 years (Figure 8 1) This growth in product1.on sped up in the post-second-world-war period rising to an annual rate of 4 87 between 1954 and 1977 In this same post-war period exports of maize starch fell while imports of cassava starch first increased through to the mid-1960's and then fell dramatically to levels not reached since the turn of the century (Figure 8 2) A convergence of factors influenced these trends in production and trade in maize starch but the driving force was the declining real pr1.ce of maize in the U S during the post-war per1.od -- except for a small hiccup in the years from 1972 to 1976 (Figure 8 3) The declining price was due to rap1.d technical change 1.n maize product1.on in the U S as per hectare yields increased from 2 4 tons 1.n 1950 to 7 6 tons in 1986 ~ The consequences of th1.s were far reaching 1.n its effect on world starch production and trade In the U S the declining price to the maize starch industry for its raw material allowed the industry to expand its markets resist the invasion of tradit1.onal markets by synthetic resins and to subst1.tute for imported cassava starch The two dom1.nant trends in the U S starch market was the expansion of starch use in the eaper and cardboard 1.ndustry (Table 8 1) and the technical advances in the modification of starch The expanding starch use in the paper products industry caused the increas1.ng demand for unmod1.fied starches while advances in starch modification and the advent of waxy maize allowed import substitutio~and __ con~l.nued competitiveness in the other end uses Thus over the post-war period unmodified starch maintained its market share while the number of different types of modified starch expanded signif1.cantly (Table 8 2) Finally the wet-m1.lling industry was able to achieve increasing returns to scale 1.n processing as output per plant has expanded rap1.dly over the period - 7 ·¡giJre : -' Unrted S t•Jtes: t; e'at.c r sr1,¡::: Expc• +s Gnd ¡:;e !JI :: t 1ces 1 l"'l.:u:::;: Y u: lo! '¡ 1 1 Y1eld \ \ 111 ~ 1"\:un Prn::z 1 !:-.¡-~ ton 'h:: • ' l --l 1 1 11 1 ' 1 1 - HU: S - ·="l .. - .. ~~ .,., • /\ ¡ t • )~~ --.._ f 1 :-X- m· t::; 1 \ -- n 1 l1 1, - 1-- - ,• ''· ( ' 1 r·plJ.':I, L ....... ' " ...,1 \ ..... , .. \ f \, \ 1 1 ,.. ;' \ --- - 1"" • .:11 :~t E rts of Cassava Pellets Chips and Meal 1960-1985 Urúted Year Nether 1ands Genrany France Belgium Italy Dernmrk K:lngdan Ireland Total (CXXlt) (OOOt) (CXXlt) (CXXlt) (CXXlt) (CXXlt) (CXXlt) (CXXlt) (OOOt) 1960 4 1 322 8 27 1 444 5 - 7 - 399 6 1961 6 6 357 1 26 4 868 8 - 1 2 - 479 o 1962 1 2 366 1 23 6 229 neg - 2 - 414 1 1963 4 8 387 3 20 o 721 - - a - 484 2 1964 16 9 461 5 18 5 105 4 - - a - 602 3 1965 76 5 519 6 18 o 100 5 6 - a - 715 2 1966 95 7 701 7 16 6 70 7 20 - a - 8868 1967 158 8 532 7 19 6 113 3 1 2 - a - 825 7 1968 234 3 4800 14 4 123 4 1 5 - a - 853 7 1969 424 9 548 1 14 8 209 5 3 9 - a - 1 201 1 1970 475 8 587 4 11 1 267 3 1 4 neg o 2 - 1 343 o 1971 510 9 522 o 39 o 273 2 2 o neg o 1 - 1 347 2 1972 670 4 429 2 140 o 290 8 1 3 neg o 1 - 1 531 9 1973 756 6 331 3 159 o 188 9 o 2 os neg neg 1 436 6 1974 1 067 8 429 4 164 3 3814 o 7 3 6 23 7 neg 2 070 9 1975 1 200 4 483 5 146 5 4418 - - o 3 neg 2 272 S 1976 1 541 1 660 2 175 1 552 8 12 9 7 9 7 1 1 9 2 959 1 1977 1 823 8 920 4 2010 672 9 neg 53 2 6 6 15 o 3 693 o 1978 2 293 1 1 409 7 713 4 863 1 219 2 127 3 134 804 5 719 4 1979 2 001 8 1 463 1 567 6 714 2 189 8 82 2 22 2 42 8 5 083 7 1980 2 158 5 1 336 5 364 9 620 9 98 9 545 28 1 39 9 4 702 2 1981 2 401 5 1 547 6 680 4 841 2 237 o 91 2 4018 307 6 231 2 1982 2 827 4 1 993 9 786 6 1 029 9 212 2 57 6 798 6 804 7 786 6 1983 1 121 5 1 796 7 239 8 906 3 997 o 9 314 3 47 5 4 526 7 1984 2 432 1 1 830 8 263 6 799 5 108 o 5 o 126 3 18 8 S 584 1 1985 2 982 o 1 674 6 307 o 801 6 108 7 o 4 770 50 8 6 002 1 a Cassava not broken out as separate item in these years Source ElJROSTAT Foreign Trade Analytical Tables {NIMEXF) and foreign trade statistics of individual countries VIII - 10 - Unification of grain prices however became difficult to mainta1n with realignment of exchange rates of member countries As grain prices were specif1ed for a 12 month period 1n un1ts of account (UA) any exchange rate adJustment vis-a-vis the UA would cause grain prices to d1verge Price unification became particularly difficult with the floating exchange rates adopted in the early 1970 s Thus with the realignment of the franc and mark in 1969 green exchange rates -- that exchange rate at which common prices are established -- and border taxes (MCA's) were instituted in order to manage CAP administrative prices The result of these pol1cies was that member countries grain prices began to diverge again that is when evaluated in dollar terms at market exchange rates This differentially affects demand for cassava in the individual countr1es since each country faces a single market price for cassava but in relation to different grain prices (see Nelson 1983 for a discussion of this po1nt) The CAP completely changed the dynamics of animal production in Western Europe Growth in animal populations occurred in those areas with the cheapest feed sources and these are precisely the areas which have transport advantages in the import of those feedgrain substitutes that do not come under the variable levy The process was extraordinarily rap1d and was especially pronounced in the swine industry Between 1965 and 1970 swine populations increased 59% in the Netherlands and 103% 1n Belgium compared to only 16% in Germany and 21% in France (Table 8 10) In the period 1970 to 1985 the swine population increased 103/ in the Netherlands and only 19% in Germany and actually declined in France These trends are correlated with the use of grains in compound feeds Overall the proportional use of cereals in balanced feeds has declined 1n the EEC but especially in the Netherlands Cereal use in compound feeds in that country has dropped below 20% (Table 8 11) whereas worldwide the f1gure is closer to 60% Cereal substitutes are essent1ally imported and the princ1pal one is cassava Cassava imports into the EEC over the past two decades and a half have shown dramatic growth increasing from 400 thousand tons in 1960 to a high of 7 8 mill1on tons in 1982 (Table 8 9) Every country 1n the EEC imports cassava but the Netherlands is by far the largest 1mporter Cassava importa by West Germany remained relatively stagnant until 1976 at which point 1mports more than doubled in two years In 1975 nat1onal grain prices 1n West Germany finally recovered to their pre-1967 level From that point national prices continued to rise The mark in 1976 also started to appreciate rapidly against the dollar and the 1nternational price (in marks) of cassava declined significantly in 1977 and 1978 This made cassava very attractive in Germany again and imports increased markedly The basic rationale beh1nd the Common Agricultura! Policy was that the European consumer would bear the principal costs of the h1gher prices paid to farmers Moreover EEC consumers as well paid the cost of the h1gher prices of cereal substitutes even though they were not subJect to the variable levy cereal substitutes garnered h1gher prices in the EEC grain market and these higher prices were transferred to exporting countries as social profits above what could have been earned on the world market 'levertheless cereal subst1tutes did not add to the EEC s tax revenue account and budgetary outlays by the EEC government for the costs of 1ts TABLE 8 lO Year 1960 1965 1970 1975 1980 1985 European Community Evolution of Growth in the Swine Population 1960-1985 Germany France Netherlands Belgium (000) (000) (000) (000) 15 787 8 603 2 934 l 579 17 723 9 238 3 987 l 885 20 532 ll 215 6 340 3 966 19 805 ll 890 7 016 4 679 22 553 ll 963 lO 186 S 011 24 360 lO 956 12 908 5 521 Source EUROSTAT AnLmal Production EEC-9 (000) 69 584 68 663 77 293 80 983 ThBLE 8 11 furopean Camunity Raw MaterlBl Used in CarqJound Feeds 1978 Oilseed cakes Com gluten Cereals Cassava and ueals feed Other Total Country (OOOt) (%) (OOOt) (%) (OOOt) (%) (OOOt) (%) (OOOt) (%) (OOOt) (%) West Gennany 4 506 303 900 6 1 4900 33 o 670 4 5 3 876 26 1 14 852 100 Nether l1mds 2 470 18 3 1 904 14 1 2 349 17 4 1 152 8 6 5 597 41 5 13 472 lOO Belgl.um 1 724 35 1 618 12 6 1 055 21 5 o o 1 518 309 4 915 100 United K1ngdan 5 578 49 4 o o 1 377 12 2 o o 4 336 384 11 287 lOO France 5 862 44 1 710 5 3 2 500 18 8 200 1 5 4 028 303 13300 lOO Camunitytotal 27 643 380 4 557 6 3 15 793 21 7 1 717 2 4 22 961 31 6 72 671 lOO Scurce Falcon et al 'The Cassava Econany of Java 1984 VIII - 11 - grain policy started to increase s~gnificantly in the early 1980 s In that period the EEC became a net exporter of grains the dallar started to apprec~ate against European currencies making the domestic costs of export subsidies high and cassava importa reached record high levels in 1981 and 1982 The budgetary costs of the CAP grain policy started to reach levels that were putting strains on the capacity of the EEC to generate tax revenue Cassava started to play a significant role in the ability of the CAP to sustain its objectives In an econometric model of the EEC feedgrain market Rastegari (1982) found that cassava imports and consumption had a positive impact on livestock product~on -- thereby confirming the previous analysis -- and had a negative impact on feedgrain imports The latter effect is expected and resulta in the loss of tariff revenues to the EEC treasury The more significant f~nding was that cassava imports had a negative effect on the setting of threshold prices Cassava imports were reducing the flexibility of the EC to set domestic farm pr~ces especially when the EC moved into a net export position in grains where export subsidies were large and dumping developed political repercussions with traditional grain exporters especially the U S The EEC was under significant pressure to reduce the growth in budgetary costs of the CAP without the political flexibility of legislating majar structural reform ~n agricultura! policy The EEC sought to resolve the situation by reduc~ng the growth in imports of cassava Because the 6% ad valorem import duty on cassava was bound in the GATT the EEC sought to--negotiate voluntary export restraints w~th princ~pal supplying countries especially Thaüand The EEC found this to be the politically most tractable solution since unbinding of the tariff would have required agreement of compensation with exporting countries with which the binding had been negotiated and with the country (if different) which is the majar suppl~er Moreover all the EEC countries would as well have had to agree to the unbind~ng In November 1980 Thailand agreed in principle to the voluntary limitation of cassava exports to the EEC however it is not t~ll September 1982 that the voluntary export restraint agreement was ratified by both part~es Thailand felt that she had little bargaining power at th~s stage She had already negotiated a quota agreement for textile exports to the EEC an industry in which investments had been large and which was a principal component of her industrial~zation strategy Moreover Tha~land d~d not want to put a polit~cally sensitive industry such as cassava (because of its importance as a source of farm income ~n the Northeast) at r~sk by relying only on the difficulty of EEC members reach~ng agreement among themselves on an unbinding of the duty In addition Thailand was promised a s~gnificant increase in agricultura! development aid to be spent on cassava diversification in the Northeast Finally as Blyth (1984) has shown in another context from the exporters viewpoint voluntary export restraints are the least harmful form of provid~ng protection against imports into the EEC Weighing the options Thailand chose the less risky course However as Britain s Overseas Development Institute observed The story combines all those elements which so often bring the CAP into disrepute m~sdirected public expend1ture (in this case of aid money) insensitive protectionism and uncr~tical acceptance of the views of VIII - 12 - European farming intereses at the expense of consumers (in this case other farmers) and overseas suppliers (House of Lords 1981) As a concession to Thailand the EEC also comm1tted 1tself to mainta1ning Thailand' s position in the European cassava market The EEC thus sought voluntary export restraints from other principal exporting countries In 1982 an agreement also was reached with Indonesia and Brazil who were then GATT members which unbound the tariff and replaced it with a tariff-quota The agreement for all parties concerned was limited to a five-year period (Table 8 12) Thailand was part1cularly disadvantaged in the agreement by being the only country whose export quota would decline over time Also in the initial understanding the EEC would also bear in mind the importance of imports of carbohydrate products which would compete directly with manioc (House of Lords 1982) Significantly the other cereal substitutes of importance were maize-gluten feed and citrus pulp pellets the principal supplier of which was the United States The EEC has not found it possible politically to restrain the imports of these produces and during the quota period imports of maize gluten feed rose dramatically This situation underscores a basic point about the political economy of cassava which is that cassava's vested intereses have always lain with the economically powerless Before the end of 1986 the EEC and the principal cassava exporters i e Thailand had to come to terms on a new agreement or return to the situation prevailing before 1982 By late 1986 Thailand and the EEC had both ratified a new agreement on export controls of cassava The agreement covers four years from 1986 through 1989 and specifies a maximum export volume of 21 million tons over the period This amounts to S 2S million tons a year some improvement on the 4 S million ton quota of 198S-86 However exports to Portugal and Spain as well would now come under the agreement Some minor flexibihty was allowed in distributing the quota from year to year as Thailand could export up to S S million tons in any single year This pattern of periodic deliberation and renewal of a new agreement on export restraint will most l1kely continue to be the pattern of EEC-Thailand trade in cassava Demand for Cassava in the EEC With the voluntary export restra1nts 1n place since 1982 estimation of import demand for cassava is something of a moot point at least as far as total quantity imported by the EEC is concerned However price and the distr1bution of those imports w1thin the EEC does have an effect on the profits to be earned by the Tha1 cassava industry and the comparative cost of animal feed across EEC countries How prices for cassava are determined thus is of key importance to Tha1land especially in its management of the restra1nts on exports to the EEC The feed industry in Europe is highly competitive and factories base their purchas1ng decis1ons on least-cost feed formulation models In general cassava will enter into swine rations first that is at h1gher cassava prices than its entry into poultry rations A large feed manufacturer in the Netherlands in 198S maintained a 40% maximum 1ncorporation level for swine rations and a 2S/ inclus1on maximum for poultry rations McK1nzie et al (1986) cite max1mum inclusion levels of 3Sk in swine rations and 20k for poultry rations for Dutch feed manufacturera in 1980 With1n any individual country cassava demand is a TABLE 8 12 Year 1982 1983 1984 1985 1986 Source EEC Export Restraint Agreement on Cassava Negotiated between the EEC and Principal Trading Countries 1982-86 Other GATT Other Third Thailand Indonesia Members Contr1es (OOOt) (OOOt) (OOOt) (OOOt) 5000 500 90 370 5000 750 132 370 5000 750 132 370 4500 825 146 370 4500 825 146 J70 Council Regulation No 2646/82 30 September 1982 TABl..E 8 13 European Cumminity Estimates of Price Dependent Cassava Demand Equations Netherlands Germany Standard Standard Variable Coefficient Deviation Coefficient Deviation Intercept o 74 1 88 -2 99 3 os Cereal Price o 85 o 23 o 31 o 10 Soybean Meal Price -0 03 o 06 o 03 o 04 Swine Population -0 03 o 27 o 54 o 31 Net Imports -0 03 o 01 -0 02 o 008 Quota Dummy o os o 07 o 07 o 47 R2 o 21 o 14 Note The dependent variable was the spot price for cassava in Rotterdam and Hamburg The cereal price was maize in the Netherlands and barley in Germany Data were monthly observations 1973-1984 Equa- tions were estimated in double-log form corrected for second order autocorrelation SOURCE CIAT VIII - 13 - step function operating between the price when ~t first enters the swine ration to that price at which cassava reaches maximum incorporation levels in all rations Because internal grain prices vary between countries cassava will be utilized f~rst in those countr~es with relatively high grain prices As Nelson (1983) points out cassava demand will be relatively elastic in these countries between the price at which it first enters the ration and the maximum incorporation rate For additional imports demand becomes less elast~c as the cost of transporting cassava from the port increases and it must compete in reg~ons where feed-grain prices have been lowered by green rates The import demand function for cassava is fraught w~th difficulties in specification Given a short enough time period so that supply cannot respond demand theory would suggest a price dependent funct~on Moreover s~nce grain pr~ces vary between countries a market clearing price for cassava will be defined ~n each of the major importing countries with some potential for arbitrage between neighboring countries Using monthly data price dependent import demand funct~ons were estimated for the Netherlands and Western Germany with the internal cassava price be~ng a function of the market price for the dominant feedgrain net imports of cassava the soybean meal price and the swine population The results of this estimation (Table 8 13) show that cassava prices respond to changes in feedgrain prices As would be expected cassava prices are more responsive to changes in maize prices in the Netherlands the ma~n importer than to barley prices in Germany However although cassava imports have a significant and negative effect on cassava prices in both countries the sue of the coefficient is remarkably close to zero suggesting very little elasticity in the market This result is counterintuitive given the rapid rate of growth in cassava imports and the ease of substitution in feed componente McKinz~e et al (1986) estimate a demand elast~city for cassava in the Netherlands of -2 4 using transformad solutions of least-cost feed models It is therefore worthwh~le to analyze more closely the mechanisms surrounding price formation of cassava Cassava prices are quoted in Europe in Deutsch marks on an fob Rotterdam basis which ~s d~stinct from the cif Rotterdam quotes for other commodit~es such as soybean meal The difference is the point at which the buyer takes ownership of the commodity In the case of soybean meal it is purchased on the Chicago Board of Trade and the feed manufacturer pays the freight and insurance at the unloading point in Rotterdam In the case of cassava he buys on a customs cleared basis from the shipper in Rotterdam The shipper pays the freight and ~nsurance discharge costs and customs duties The sh~pper has ownersh~p of the cassava till d~scharge ~n Rotterdam while in the case of soybean meal he does not providing only freight services The reason feed manufacturera have gone to th~s system was essentially the uncertainty of quality and customs clearance At one stage Thai pelleters were ~ntroduc~ng rice hulls which under EC tar~ff rules would be classified as a compound feed dutiable at a very high tariff Under the current system the shipper guarantees the quality and the price and the buyer assumes no risks However this system potentially reduces the efficiency of pr1ce transmission between the two markets This last point is reflected in the determination of a market price for cassava in Europe Most buyers purchase cassava on forward contracts so that continuity of supplies is guaranteed and storage costs are kept to a minimum In general cassava 1s contracted between 2 to 6 months forward Thus approximately 90% of each shipment from Thailand has already been contracted Only a small percentage is sold on a spot market or at the so-called afloat price the price normally quoted from trade sources Moreover the afloat pr1ce generally reflects speculators in the market who have not yet covered their contracta and is therefore more variable than the forward price The market price for cassava is therefore a negotiated forward pr1ce between shipper and feed concentrate manufacturer and this price is often not quoted The shippers can negotiate on the basis of known production costs for pellets in Thailand known handling and freight cost -- in 1985 $4/t for loading $9/t for freight and insurance and $5/t for discharge -- and the tariff while the buyers will negotiate on the basis of the shadow price of cassava in their feed cost models and their sense of the cassava price in Thailand and Europe The analysis of price transmission between Thaiiand and Europe (see Chapter VII) suggested that forward prices in Europe were much better correlated with Thai prices than afloat prices and that prices were transmitted instantaneously with some residual tendency for prices in Thailand to lead those in Europe before the quota and those in Europe to lead Thailand after the quota The forward contracting and the nature of price transmission suggests that the cassava price is given exogenously --in the context of a monthly import demand equation--and thus the endogenous variable in the demand function should be cassava imports An import demand equation was thus estimated using net cassava imports as the dependent variable Since th1s is an amount which is forward contracted traders have suggested that an average period is about three months and so imports were lagged three months Lagged imports were then made a function of the forward price for delivery in three months current swine stocks current soybean meal pr1ces and the grain threshold pr1ce three months forward Since grain prices are fixed on a monthly basis before the crop year the threshold price is the best estimate of the future gra1n price Because a fixed amount of cassava must be allocated among the various countries the equations were estimated us1ng Zellner's seemingly unrelated regression technique The results (Table 8 14) are significantly better than the previous specif1cation The direct import elasticity is relatively elastic although lower for the Netherlands than for Germany This is expected in a country where cassava imports already are 30% of the combined product1on of pig and poultry feeds and moving add1tional amounts involves more radical price changes Conversely cassava importa in the Netherlands respond much more strongly to changes in grain prices than in Germany In Germany a large part of the concentrate and animal industry is in the South and cassava use 1n rations in th1s part of the country is moderated by the TABLE 8 14 European Commun1ty Estimates of Import Dependent Cassava Demand Equations Netherlands Germany Standard Standard Variable Coefficient Deviation Coefficient Deviation Intercept 3 08 3 74 -65 1 8 35 Cassava Price -1 49 o 32 -0 90 o 31 Cereal Price 1 87 o 64 2 77 o 58 Soybean Meal Price o 26 o 29 o 54 o 26 Swine Population o 61 o 61 6 69 o 87 Quota Dummy -o 60 o 16 -0 06 o 12 R2 o 33 o 55 Own Price Elasticity o 71 1 15 Cross Price Elasticity with Cereals o 65 o 36 Note The cassava and cereal pr1ces were three month forward prices and imports were lagged three months Zellner s Seemingly Unrelated Regression procedure was used to estimate the coeffcients Source CIAT VIII - 15 - transport costs from port areas Grain prices have to move more radically to get the same response in demand for cassava imports Finally the quota is principally affecting cassava use in the Netherlands where cassava imports have declined other things being equal to what they were prior to the quota Why Netherlands should be worse affected than Germany by the quota is not clear and in the end is counterintuitive However this result may be short term in nature since in 1985 the Netherlands recovered in import volume what it lost ~n 1983 and 1984 This result may therefore reflect forward contract committments at the time of ~mplementation of the quota The soybean meal coefficient remains something of an anomally since it suggests that cassava and oilseed meals are substitutes particularly ~n Germany where the coefficient is sign~ficant Misspecif~cation is possible s~nce the current price rather than the future price was used a future price in Europe was not available Nevertheless Nelson (1983) in bis model of EEC import demand did not get a significant coeff~cient for soybean meal either though the sign suggested complementarity McKinzie et al (1986) working with least cost feed models in the Netherlands find a complementary relationship between cassava and oilseed meals Nevertheless even using such a robust technique the cross-price elasticity estimated is only -0 3 i e there is a response of cassava use to changes in oilseed meal prices but ~t is not large In Germany o~lseed meals make up 30 to 40% of feed concentrates Because oilseed meals are often similarly priced to grains they enter as a calorie as well as a protein source Changes in oilseed meal prices would thus have little influence on cassava use since the protein restrictions in the least cost models are already more than met The effects of the quota thus have been (1) to reduce the efficiency of price transmission between Europe and Thailand while shifting cassava price formation essentially to demand-side factors in Europe (2) to widen the margins between Europe and Thailand a factor which Thailand is using to open third-country markets and {3) to reallocate cassava imports between countries On the latter point Spal.n and Portugal s entry into the EEC the suggested elimination of green rates and MCA s and the environmental constraints being placed on expansion of livestock enterprises l.n northern Europe all suggest potentJ.al for shiftJ.ng the locus of growth J.n an1mal production to these two countries ~f based on the abil~ty to efficiently import feed componente whJ.ch do not come under the variable levy Given grain shortfalls in both these countries rJ.sing grain prices as the grain sector comes under CAP prices some experJ.ence with importl.ng cassava in 1984 and 1985 and the prOJected improvement in port facill.tl.es condJ.tions seem appropriate for such a restructurl.ng Moreover the quota on cassava imports will probably have little impact on increased grain use Hillberg (1986) developed a simulation model of the West German feed sector and found only gradual substitution of grains for cassava in swine and poultry ratJ.ons in northern Germany However the quota also led to higher feed prices a decreased demand for feed concentrates and in consequence the impact of changes l.n ration composJ.tion favoring grains was dampened by the accompanying hJ.gher finished rat~on costs (Hillberg 1986) Moreover as McKinzie et al (1986) find the high cross-price elasticities suggest that a specific commodity ~mport restriction would substantially reduce that commodity' s TABLE 8 1S Asia Per Capita Chicken Meat and Pork Consumption Trends in Selected Countries 196S-1982 Country 196S 1970 197S 1980 1982 (kg) (kg) (kg) (kg) (kg) Non-Cassava Producing Japan Chicken 1 6 3 8 S 2 7 7 7 7 Por k 3 1 4 7 6 S 9 6 9 6 Taiwan Chicken 2 o S 6 8 4 12 3 13 3 Por k 16 8 18 o 17 S 26 2 2S 4 South Korea Chicken O S 1 4 1 6 2 3 2 S Por k 2 o 2 6 2 8 6 3 6 o Cassava Producing Countries Tha~land Chicken n a n a 3 3 S 9 7 2 Por k n a 4 9 S 1 7 8 n a Philippines Chicken n a 2 S 3 1 3 2 3 4 Por k n a 8 1 9 o 8 S 9 1 Malaysia Chicken n a 6 8 9 2 10 3 10 3 Por k n a S 9 4 9 S O S 4 Indonesia Chicken n a o 3 o S 1 1 1 3 Por k n a o 3 o 4 o 4 o 4 useage but that use of other non-grain importa could be expected to rise greatly Such appears to be the case with corn gluten feed imports from the Un1ted States (Siamwalla 1986) The world market for cassava feedstuffs is something of the reverse of that for cassava starch In the case of feedstuffs tariff and price policies in Europe have created a large market insulated from world trade conditions in feedgrains Since the market is politically defined (though almost every agricultura! market has its political dimension) cassava' s impingement on other EEC objectives has resulted in restraints on future growth of EEC importa The European market is nevertheless providing the base for the restructuring of trade in cassava pellets and to understand this process requires some analysis of the feed and livestock sector in East Asia The Asian Regional Market for Cassava Feedstuffs Do cassava feedstuffs have a w1der international market than just the European Community ? Trade and price policies as in all trade matters dealing with cassava hold the key to the answer The issue is being forced by the EC itself through its imposition of import quotas which in turn has caused Thailand to devise mechanisms to open third country markets The solution mimics the EEC' s export subsidies with one big difference the European consumer rather than the EEC budget is in effect subsidizing Thai exports to non-EC countries Th1s is irony of a high order that the EEC should be subsidizing Thai cassava exports to third countries This outcome is to the international grain trade what epicycles were to Ptolemaic astronomy a further complication to produce a workeable system but where the central thesis of that system is faulty For cassava what it achieves is time to develop a more rational system and the bulwark of such a system will inevitably be the Asian market for feedstuffs which is currently dominated by imports of U S coarse grains Food consumption patterns in East and Southeast Asia are changing rapidly The causes for these changes arise as much from the supply side -- technical change in food production and processing improved foreign exchange availabilities allowing an increase in and diversification of food 1mports and improvements in marketing -- as from the demand side -- increasing per cap1ta incomes urbanization declining influence of religious prohibitions on certain foods and changing relative pr1ces Changing food consumption patterns are thus set within an evolving economic system which reflects fundamental structural change and basic shifts in food processing marketing home preparation methods and purchasing patterns as the population shifts from rural to urban residence The most fundamental shift in food consumption patterns in Asia has been the rapid increase in the consumption of l1vestock products especially meat (Table 8 15) For example in Japan 1n the two decades spanning the period 1960 to 1980 per capita consumption of beef grew at an annual rate of S 6% pork at a rate of 11 14 and chicken at a susta1ned rate of 16 7% Even after such high rates of growth per capita meat consumption in Japan is still only about a quarter of levels in the United States This 'nghlights the first salient feature of meat consumption patterns in Asia that growth in consumpt1on has started from a very small VIII - 17 - base since for most countries no more than 5 O kg of meat per person was consumed Ln the early 1960's Only the PhilLppLnes and Taiwan would appear to have had a higher consumpt1on base due essentially to the larger role of swine in farming systems and rural consumpt1on patterns Pigs also were important Ln large parts of China Swine have played a differential role across Asian countries in definLng meat consumpt1on patterns partly because of relig1ous restrict1ons such as Moslem taboos in Malays1a and Indonesia and Buddh1st prejudices in Thailand and Japan and partly because of feed availability on farms in swine producing countries usually the root crops sweet potatoes or cassava and rice millings In the two decades encompassing 1960 to 1980 annual growth in per capita GNP was over 47 in all countries under study here except for the Philippines which grew at 2 8% per year Meat demand is very income elastic in As1a (Table 8 16) and yet income elasticities and income growth do not explain all the growth in per capita meat consumpt1on In Asia income growth has also precipitated diversification of the diet as reflected in the very low per capita consumption f1gures for meat in the early 1960's Also income growth 1s closely related to other bas1c changes 1n the economy that affect food consumption patterns particularly urbanization and the growth of food reta1ling networks Implic1t in migration from a rural to urban setting is a shift in food sources from one based primar1ly on production to one based on purchases Also conven1ence becomes an important factor in food choice in preparat1on methods and 1n food storage in the home Finally food preferences become more susceptible to advertising and to the diversity found in eat1ng out of the home Therefore implicit in income growth are the basic changes in lifestyle that imp1nge on food consumpt1on patterns these have had a large impact on the rising demand for meat in Asian countries Income elasticit1es do not vary sign1f1cantly across the different meats except for the lower estimates for pork in the high consuming countries Income growth does not account for the very signif1cant d1fferences in growth rates between the d1fferent meats Thus wh1le income expla1ns much of the growth in total meat consumption price is the more relevant variable 1n analyzing growth rates in indiv1dual meats In all meats the own-price elastLcity is very high and whüe cross-price elastLc1ties are normally s1gnificant (Table 8 17) substitution has not yet played a dom1nant role in meat consumption patterns in Asia as it has for example 1n Latin America Differences in growth rates in consumption of the various meats is due to the differentLal trends in real prices of the meats especLally the declLne in chicken and to a certa1n extent pork prices vis-a-vLs stabilLty or increases Ln the price level of beef It is the fundamental effect of prices on meat consumpt1on that makes bas1c cost changes on the supply side so Lmportant Japan has the longest hLstory in the modernizat1on of 1ts feed and livestock industry and thus in many respects will presage the future developments 1n the lLvestock industry of many Asian countrLes The dominant factor in the expansion of the livestock sector 1n Japan was technical change This is shown in Table 8 18 which shows rap1d expansion 1n meat production of ch1cken and pork even though product prices were declining relat1ve to feed prices This relationship 1s the more impress1ve considering that feed makes up 35/ of pork product1on costs and about two th1rds of chicken productLon costs (Coyle 1983) Three TABLE 8 16 Asia Income Elasticities for Meats Country Por k Chicken Beef Non-Cassava Producing Taiwan 39 1 10 97 Japan 1 02 1 64 1 09 South Korea 1 19 1 54 1 38 Cassava Producing Philippines 85 1 00 80 Thailand 58 44 41 Indonesia 1 4 2 2 n a Source Wu Cho Sawada ASEAN Prusarn Monteverde TABLE 8 17 Asia Own Price Elasticities of Meats Country Por k Chicken Beef Non-Cassava Pr-oducing Taiwan 44 55 -1 99 Japan -2 os -1 25 -1 53 South Korea -1 53 -1 64 -1 34 Cassava Producing Phihppines Urban -2 00 -1 30 -1 30 Rural -1 so -1 00 Source Wu Cho Kester ASEAN ~mportant efficiency VIII - 18 - changes account for these rapid increases in production changes that are now occurring in other Asian countries F~rst structural change in livestock product~on has been rapid Production has moved from small units on farms to specialized large-scale enterpr~ses In Japan this process has been particularly ~mpressive in both swine and broiler production (Table 8 19) Structural change in livestock production has not implied a gradual increase in animal populations on farms but a rapid shift away from farm units to specialized production units In the process the number of producers declined rapidly In Japan the number of swine producers declined from 800 thousand in 1960 to 156 thousand in 1979 (Coyle 1983) Statist~cs on total animal populations usually mask quite marked shifts in sources of production Thus in disaggregating the statistics for Thailand for poultry (Table 7 25) while growth in the total population has been moderate the increase in large-scale commercial operations has been very rapid and on-farm populations have declined This search for scale economics through structural change has characterized the pork and poultry sectors of all the countries under study here except Indonesia and China In China the very rapid rise in pork production and consumpt~on since the political changes of the late 1970's has been due to shifts of production from collectives to individual households and intens~fication of product~on through the improved availability of grains (Sicular 1985) In Indonesia on the other hand income distributional obJ ectives have been translated into a 1983 poli e y which limits the size of poultry operations to a thousand layers and 750 broilers (see World Bank 1984 for a more extensive discussion of the policy) This policy may limit the price declines in poultry that have come in other countries and therefore the expansion in consumption On the other hand s~nce the population is still overwhelmingly rural the policy may in fact lead to decentralization of production away from urban areas and increased rural consumption as is occurring with pork in China The feed companies appear willing to respond by developing rural feed d~stribution channels Indonesia and China may offer an alternative livestock development strategy oriented towards rural consumption However eventually when the policy turns toward urban consumption the development of large-scale poultry and swine units will be essential to cost and price reductions for urban consumers The second important change in livestock systems in Asia is the shift to balanced feeds as the principal source of animal nutrition The impact of this on production efficiency has come through improved animal nutrition which has allowed quicker weight gains usually higher slaughter weight and improved reproductive capacity Whether balanced feed is cheaper than on-farm feed sources is questionable especially for swJ.ne since feedstuffs with relatively low opportunity costs are used Mixed feeds however allow balanced nutrition especially for protein requirements and expand the availability of feed sources which are usually constrained at the farm-level Development of a mixed feed industry has been especially critica! in the growth of the poultry industry Development of a mixed feed industry usually leads the structural change in livestock production with the initial linkages generally being TABTE 8 18 Japan Trends in Meat Production and Meat-Feed Price Ratios 1960-79 Beef and Vea! Chicken Por k Annual Annual Annual Annual Change in Annual Change in Annual Change in Production Meat-Feed Production Meat-Feed Production Meat Feed Period Growth Ratio Growth Ratio Growth Ratio (%) (7) (%) (%) (7) (%) 1960-6S 6 1 n a 36 o n a 20 9 3 6 196S-70 6 9 S 4 19 8 - 2 6 12 4 - o 7 1970-7S 4 3 2 9 8 4 - 3 o S 2 2 6 197S-79 S 2 S 8 9 6 - 2 9 13 2 - 4 9 1960-79 S 6 4 6 18 4 - 1 o 12 8 o 2 Source Coy le William Japan s Feed-Livestock Economy 1983 VIII - 19 - made with the poultry sector Growth in compound feed manufacture has been very rapid in East and Southeast Asia in the last one to two decades Most countries have managed annual growth rates of well over 107 with Japan maJ.ntaining a 9 9% annual rate of growth over a period of 22 years from 1960 to 1982 (Table 8 20) Growth can be remarkably rapid in the early stages in the establl.shment of the industry Thus in the 1960's Japan's compound feed industry grew at annual rate of 17h comparable to the growth of South Korea's industry in the 1970 s of 18% but well below the remarkable growth in Thailand of 30/ per annum through the course of the 1970's There is a chicken or egg questl.on in the gestation of a compound feed industry In most cases the establishment of the industry is based on the development of commercial poultry enterprl.ses Wl.th the two often vertically linked in the initial phases The feed industry often assumes the inJ.tiative in the development of its market If developments in the J.ndustry follow the example of Japan then eventually divestment of the poultry enterprises takes place and diversificatJ.on occurs with a significant rise in swine feed and dairy feed production However significant differences will be expected to occur across countries in the development of the latter two industries because of Moslem prohibl.tions of pork consumption l.n Malaysia and Indonesia and lactose indigestability in many Asian populations In Asia more so than any other continent the development of the livestock industry is and will be based on either the purchase of mixed feeds by livestock producers or the purchase of the feed ingredients by the livestock producers to mix their own feeds Expansion of the livestock industry in Asia will not be based on an l.ntegrated farm system l.n which own production of feed components is ll.nked to livestock production The third element responsJ.ble for rapid technical change in the livestock sector is the improved feed conversion rates in the animal population This is due to both more efficient animal breeds and improvements in management especJ.ally in animal health A particular trend in swl.ne production is the movement away from breeds with a hJ.gh fat carcass to those with a much higher percentage of lean meat However aggregate feed conversion rates only partially reflect this improvement since they as well incorporare the movement away from on-farm feed resources -- that is those feed components which do not usually figure J.n data on feed avaJ.lability -- to compound feeds (Table 8 21) Aggregate feed conversJ.on rates thus first J.ncrease and then decline when the conversion by livestock producers to compound feed has stabilized Comparison of these aggregate rates across countries will not differentiate between improvements in the efficiency of feed conversion and the degree of penetratJ.on of compound feeds in the livestock sector What the limited data in Table 8 21 indicate is that aggregate feed conversion rates are still rising in all countries but Japan that l.S the changes in the production structure of animal production is stJ.ll the dominate J.nfluence Rising demand for livestock products and the structural change in livestock production have created a very rapl.d increase in the derived demand for feedstuffs especJ.ally carbohydrate sources The response to this situation l.n all cases but Thailand has been to increase l.mports of feed grains In the non-cassava and non-maize producing countries the growth in feed grain imports has been very rapid J.ndeed In 1960 Japan TABLE 8 19 Japan Structural Change in Average Herd or Flock Size 1960-79 Period Swine Broilers Layers --------- animals per farro --------- 1960-65 4 o n a 25 9 1965-70 9 7 1 852 8 62 2 1970-76 23 3 S 101 o 186 6 1975-79 46 4 10 081 o 492 1 1979 60 7 12 684 o 670 3 Source Coy ele 1983 TABLE 8 20 Asia Production of Compound Animal Feeds in Selected Countries 1970-83 Non-Cassava Producing Cassava Producing Year Japan South Korea Thailand Philippines Malaysia (000 t) (000 t) (000 t) (000 t) (000 t) 1960 2 884 n a n a na n a 1970 15 097 508 109 4 314 4 236 7 1975 16 897 901 486 5 654 7 315 6 1976 18 671 1382 666 4 625 3 389 8 1977 19 948 1899 725 5 830 o 386 2 1978 21 210 2693 922 8 960 o 444 8 1979 22 796 3880 1 173 9 994 o 457 3 1980 22 292 3462 1 350 o 1 061 o 548 6 1981 22 173 3491 1 560 o 1 147 o 564 6 1982 22 896 4420 1 710 o 1 161 o 569 2 1983 n a 5852 1 962 o 1 061 o 636 2 Source Statistics of feed associations and government agencies TABLE 8 21 Asia Feed Convers~on Rates (kg of feed per ene kg of meat) for Seleeted Countries 1970-80 Meat and Year Japan (kg) Swine 1970 5 36 1975 4 36 1980 4 34 Poultry 1970 2 07 1975 3 13 1980 2 90 Beef 1970 4 18 1975 5 61 1980 8 08 a b e Poultry meat and eggs Grain only Commereial produetion only South Korea Tha~land e China b (kg) (kg) (kg) n a 3 85 n a 2 40 n a n a 3 27 n a 4 o a 2 55 2 55 n a 3 79 a n a n a 5 51 a 2 o n a n a n a o 43 n a 2 41 6 o Souree Coyle 1983 Dyek and Sillers 1986 Chesley 1985 Sieular 1985 TABLE 8 22 Southeast Asia Trends in Production and Trade of Maize 1960-84 Thailand Philippines Indonesia Malaysia Year Production Net Exports Production Net Exports Production Net Exports Production Net Exports (000 t) (000 t) (000 t) (000 t) (000 t) (000 t) (000 t) (000 t) 1960 544 515 1210 14 2460 n a 4 -120 1965 1021 804 1380 - 6 2283 5 9 - 53 1970 1938 1371 2005 - 1 2606 282 16 -212 1975 2863 2072 2767 -121 2903 50 14 -275 1976 2675 2388 2843 - 96 2572 - 51 26 -269 1977 1677 1518 2855 -148 3143 1 18 -288 1978 2791 1955 3167 -105 4029 - 5 12 -310 1979 2863 1988 3123 - 35 3605 - 63 8 -436 1980 2988 2175 3110 -250 3994 - 19 8 -430 1981 3449 2549 3290 -253 4509 4 8 -400 1982 3002 2800 3126 -341 3234 -193 9 -683 1983 3552 2630 3134 -528 5087 - 33 20 -775 1984 4226 3117 3439 -182 5288 100 22 -953 Source National production and trade statistics VIII - 20 - Taiwan and South Korea together imported less than 2 million tons of coarse grains By 1984 the import level for these three countries stood at 27 6 million tons Domestic production of feedstuffs in these countries declined during the period especially barley 1n Japan sweet potatoes and barley in South Korea and cassava and sweet potatoes in Taiwan which thereby reinforced the linkage between domest1c livestock production and feed grain 1mports Decline in domestic production of feedstuffs in these countries was due to the demise of integrated livestock-crop farms and the rising costs of farm labor as a result of industrialization and rural-urban migration In ma1ze-producing countries however development of the livestock sector has been one of the factors stimulating increases in grain production Thus in the Philippines Indonesia Thailand and China feedgrain product1on has increased significantly (Table 8 22) but this has not been sufficient to keep up with rising demand except 1n the case of Thailand The Philippines moved from the position of net exporter or m1nor net importer of maize to a major net 1mporter in 1971 Indonesia did the some in 1976 and China has significantly increased its imports in the last five years Finally Thailand has not been able to increase significantly its maize exports even through domestic production has increased from 2 3 m1llion tons in 1973 to well over 4 million tons in 1984 In all countries feed demand has increased at a much more rapid pace than domestic production of feedstuffs Significant scope therefore exists in the tropical countries in Southeast Asia to link increasing interna! demand to production growth in feedstuffs thereby improving farmer income in principally upland areas The rapidly rising demand for carbohydrate sources for the growing animal feedstuff industry in East and Southeast Asia thus raises a dual potential for cassava that is exports from Thailand to the large import markets in Japan South Korea and Taiwan and increased domestic utilization in the cassava producing countries As regards the former the quota placed by the EEC on cassava imports has had the secondary affect of shifting Thai surpluses into principally East Asian markets The mechanism by which this has been accomplished has to do with Thailand s interna! management of the quota on the one hand and liberalization of tariff barriers on cassava for animal feed by the princ1pal importing countries in East Asia Since the agreement between Thailand and the EEC restricting cassava flows to Europe 1s a voluntary export restraint Thailand had to accept the responsibility for managing the quota (as Blyth 1984 has shown voluntary export restraints are the least harmful form of protection from the exporter's view point) Since the agreement which covers the period 1982 to 1986 was not s1gned t1ll September of 1982 only 1n 1983 did Tha1land begin to effectively lim1t cassava exports to the EEC During 1983 the Min1stry of Commerce in Thailand adopted an export licensing system and attempted several forms of allocating the licenses F1rst the quota was allocated on a quarterly basis to exporters based on historical shares in the export business Then the quota allocation was sh1fted to a first-come-first-serve system where licenses were granted for the quarter upto the point that the quota for the period was exhausted Finally by the end of 1983 Tha1land had arrived at a workeable system for allocation of the export quota Start1ng 1n 1984 the year was d1vided into seven periods Export allocations in a period were based on the stocks held by exporters such that those holding higher stocks would be given a higher percentage share of the export quota In addition a bonus system was instHuted in which any exports to th1rd countries in the previous period would allow first priority to export allocation in the next period depending on the Sl.ze of the third country exports The bonus system was established on a 1 1 basis and the ratio was changed to 1 25 1 at the end of 1985 that is a one ton quota allocatl.on for every 1 25 tons exported to third countries However due to the declining stock levels in mid-1986 the bonus ratio was changed back to 1 1 in June of that year The reversal indicates that the Ministry of Commerce recognizes the pol1cy role of the bonus ratl.o whereby market surpluses can be managed by adjustment in this ratio The result of this quota allocation system has been the development of a two-tiered price structure at the export point The system has allowed Thailand to appropriate the rents to be accrued in the European market while maintaining a unified domestic price structure The divergence in prices at the export point is due to the situation where cassava prices in Europe are determined by the grain price set under the Common Agricultura! Policy and those in th1rd countries are set by the world price for feedgrains As one of the results of the quota has been an increased price spread between Tha1land and Europe the Ministry of Commerce has developed l.ts export allocation policy to divert these exporter rents in arder to finance exports to third countries As export allocations have been as low as 11% of total stock holdings (Figure 8 6) there is sign1f1cant l.ncentive for exporters to guarantee their access to the European market by utiliz1ng some of these profits to sell in third countries Thailand has thus taken the logical step of stratl.fying its market On the import market side there has been a progressive liberall.zation of tariff and quota restrictions on cassava in most markets With the recogn1zed shift to dependence on imports to meet their animal feed requirements East Asian countries have progressively liberalizad import restrictions on feed components In general liberalization of feed grains especially maize and sorghum precedes that of cassava In Japan and South Korea this has been due to a vestl.gial desire to protect domestic sweet patato producers and in Taiwan to protect both sweet patato and cassava producers Nevertheless in 1968 Japan reduced its tar1ffs on cassava l.mports for feed use to zero In South Korea the liberalization has been much more recent Upto 1984 the general tariff for cassava was 407 compared to 5% for maize -- cassava chips for alcohol manufacture were imported at a lower duty under a quota system In 1984 cassava tariff rates were reduced to 207 and in 1985 to 77 which was then equal to the rate on feedgrain imports Taiwan on the other hand has continued to maintain a low tariff rate on maize of 3% Wl.th a significantly higher rate for cassava Taiwan has been reluctant to liberalize the duty because of its own cassava producers even though domestic cassava does not go into animal feed concentrates East Asl.an markets have easily absorbed the surpluses from Thailand Thai exports to East and Southeast Asian markets increased from 48 thousand tons l.n 1982 (this was all chip exports to South Korea for alcohol production) to 129 thousand tons in 1983 225 thousand tons 1n 1984 and TABLE 8 23 Year 1980 1981 1982 1983 1984 1985 Thailand Size and Distribution of Cassava Pellet Exports 1980-85 Destinat~on Total Exports EEC East Asian Countries (000 t) (000 t) (000 t) 4973 4811 o 5954 5883 4 7426 7331 49 5094 4964 129 6201 5867 225 6616 4708 954 Source Department of Customs Bangkok Note The voluntary export restra~nt carne into effect October 1982 VIII - 22 - finally to 954 thousand tons in 1985 (Table 8 23) In 1985 Japan took over 400 thousand tons and South Korea and Taiwan over 200 thousand tons each The potential market for cassava in East As~a is more than even current cassava export levels as long as it is competitively priced with maize East Asia will develop as the secondary or residual market for Thai cassava with Europe having first call on Thai cassava exports upto the quota l~m~t On the other hand for the cassava-producing countries in Southeast Asia increased cassava production is one of the means for meeting the rapidly rising domestic demand for carbohydrate sources in feed ratwns (Table 8 20) Feed concentrate production has been increasing rapidly in most countr~es in Southeast Asia as demand for animal products have increased and technical change has taken place in animal production systems In Malaysia and the Philippines feed component demand has been met to a significant extent by increased maize imports In Thailand increasingly maize production has been diverted to meeting domestic demand while exports have largely stagnated Finally in Indonesia structural change in an~mal and feed production is JUSt beginning and if Indonesia follows trends in the other countries Indonesia will also become a net feedgrain importer Therefore the potential exists to link increasing domest~c demand for feed energy sources to increased cassava production Realization of this potential depends on cassava being price competitive with other carbohydrate sources in animal feed diets In Asia this is maize supplemented by broken rice when available Cassava ~s competitive if it enters into the solution of a least cost feed formulation model For the period 1982 to 1984 cassava enters into the least cost diet in Indonesia and the Philippines Cassava comes in and out of the diet in Thailand and does not enter at all in Malaysia To enter the diet cassava in general has to be priced at about 65 to 70i' of the price of maize depending on the price of soybean meal Viewed in the longer term this maize-cassava-price ratio has been very variable in Indones~a and Tha~land reflecting the disarticulation between the two international markets In Malaysia the trends in this pr~ce ratio have been consistently rising In Malaysia cassava has progressively gotten more expensive in relation to maize Starting in 1980 cassava began to be period~cally uncompetitive and in m~d-1982 this trend became relatively permanent In Indonesia on the other hand cassava has become relatively cheaper compared to ma~ze although with significant variability This analysis reinforces conclusions from the previous chapters In Malaysia in the 1980's cassava has failed to remain compet~tive with maize imports In Thailand cassava will come in and out of the ration depending on price relationships for maize and cassava defined in two ~ndependent but nevertheless internacional markets In Indones~a cassava could form a more important component of the as yet nascent feed industry Cassava ~n some years is extremely competitive with maize and yet cassava has not been ut~lized ~n this industry Use in this industry could put a more effect~ve price floor under cassava on Java However since the feed industry has so far rel~ed on imported ma~ze through BULOG the marketing channels there have yet to develop In the Philippines cassava is competit~ve but an even further step is required of developing cassava processing capacity In general there is sufficient demand in existing domest~c markets to absorb VIII - 23 - cassava production in these countries Cassava s entry into the growing animal feed market will apart from Tha1land depend on increased domestic production Conclus1ons The previous analysis suggests a rather basic question what is a world market for cassava? The world cassava market is something of an odd animal only because it presents the reverse image of the dominant world market for grains The distinctions here are many but a few will suff1ce in order to characterize the world cassava market First cassava moves as a semi-processed product whereas grains are essentially bulked and shipped being processed in the importing country Processing makes cassava a tradeable go~ and unlike other root crops links cassava producing areas to international markets However the processing defines the end market where it will be util1zed i e starch human food or an1mal feed End use in cassava is defined at or near the product1on point whereas in gra1ns end use is defined near the consumption po1nt The issue is critical in international trade because rocessed roducts e or lour, n general ha ve higher tariff protection than raw materials Thus a world cassava trade is not defined in the same sense as a world maize trade Rather there is a cassava starch trade and a cassava pellet trade each with their respective world prices Second government policy plays a very d1rect role in price formation for cassava in world markets just as in the case of grains However for grains world prices are principally determined by policies in maJor exporting countries which support the price or incomes of their grain producers In cassava on the other hand pr1ces are principally set by the pol1cies of importing countries There are virtually no policies wh1ch directly intervene to support e1ther farm prices for cassava or cassava producer incomes The distinction is important in regards to the standard by which cassava is judged to be price competitive with grains in international markets Cassava competes essentially with grains but the current organ1zation of international trade in cassava and grains results in a situat1on where they do not compete directly at internationally determ1ned prices Thus the common assessment that cassava is not competitive in international grain markets in something of a red herr:mg because prices are formed w1thin two very distinct policy structures and prices in both cases are not an adequate measure of actual production and transfer costs Finally the degree of substitut1on between cassava and grains has measurably increased over the post-war period and much of the growth 1n world trade 1n cassava has been based on cassava s d1rect subst1tut1on for grains in the different end markets Cassava s future in world markets does in fact depend on its ability to compete with grains To date this competition has been determ1ned by grain price pol1c1es and tariff structures of importing countr1es and because of this cassava trade is more vulnerable to policy changes than the international grain trade where pr1ces and volumes are principally set by the gra1n policies of the exporting countr1es Thus while cassava competes on a cost bas1s in the wider international grain market (Table 7 21) it cannot compete on a pr1ce basis The political economy of international trade in carbohydrate VIII - 24 - sources is such that cassava which comes closest to being produced and traded under laissez-faire market principies and perfect competition economic princ1ples and furthermore which is produced solely in developing countries cannot compete in an international grain market where income support policies (and to a lesser extent export subsidies and government-to-government sales) of developed countries are necessary for producing at 1nternational prices The future of a world market for cassava is principally a matter of political economy and not of pure economics and the policy structure within which cassava must compete will be set outside the influence of cassava producers themselves Does cassava have a comparat1ve advantage vis-a-vis grains in international markets? The dominant world market for both grains and cassava in the near future is the animal feed market Cassava would move as pellets competing against maize and sorghum What is striking about current world trade in coarse grains is that trop1cal countries are net importers with the volume growing over time In the tropics only Thailand has remained a large and consistent exporter of coarse grains in the last decade Sudan Burma and Zimbabwe have exported smaller amounts These exporters essentially trade in their own regional market anad their comparative advantage over the large temperate exporters often rests on transport costs quality (white maize in Africa) and demand for bagged grain The temperate zone appears to have a significant comparative advantage over the tropics in the production and export of maize and sorghum Part of th1s is due to edapho-climatic conditions -- longer day length longer growing season better soils and reduced d1sease and pest pressure -- but the primary factors are agricultura! research and efficient transport and marketing systems For example the large investments in maize research in the United States since the early 1900's was responsible for a signif1cant rate of growth in maize yields over the post-war period This increased production was princ1pally directed to export markets at declining real prices (Figure 8 3) The issue then is whether tropical cassava has a comparative advantage against temperate grains and whether this comparative advantage can be further shifted towards cassava through investments in agricultura! research processing and marketing Cassava is perfectly adapted to tropical cond1tions it grows well 1n acid soild of low nutrient status can w1thstand periodic drought is relatively resistant to disease and pest attack and is very flexible in its planting and harvesting dates Its productivity under such cond1tions 1s unequalled by grain crops in the tropics Moreover cassava has a very limited research h1story with almost no basic research on the crop Compared to temperature grains research on cassava 1s in its infancy and to date there has been little impact on cassava productiv1ty from 1mproved technologies Average yields of cassava in exporting countries are far below their potent1al indicating significant scope to shift relative comparative advantage to cassava 1n the same way that tropical palm oil has gained an 1ncreased market share over temperate soybean oil in the last two decade Comparative advantage between grains and cassava (and also between cassava producers) will also depend on processing and marketing costs The development of the cassava sector in Thailand offers something of a model in the development of scale economies in cassava process1ng assembly and VIII - 25 - transport If growth in cassava exports are to be based on small farm production there is an argument for basing ~nit~al growth on small-scale processing units and achieving scale economies only at critica! production volumes Large-scale processing units without the production base are a non-starter or usually result in plantation production Something of an infant industry argument exists for developing an export capacity in cassava that is competit~ve with the Thai industry where scale economies have already been developed Thailand because of the efficiency of its processing and marketing sector is fully competitive on a cost basis with U S coarse grains Sustaining the infant industry argument would call for developing a critica! production volume based on domestic markets In this lies the real future of a world cassava market since as has been stated tropical countries are maJor net importers of coarse grains and increased cassava product~on will be directed to meeting domestic requirements f~rst Any export surpluses will depend on the growth in domestic demand vis-a-vis the growth in production As has been the case in Asian cassava producing countries apart from Thailand production has not been able to meet rising demand for cassava products In this regard then improved production technology would provide the increases in volumes necessary to meet domest~c demand and should surpluses develop would result in the cost reductions that allow the country to compete in international markets The international market for cassava products will continue to be ruled by trade policies technical change and shifting comparative advantage IX A COMPARATIVE ANALYSIS OF CASSAVA PRODUCTION AND UTILIZATION IN TROPICAL ASIA Cassava was probably first introduced into Asia dur1ng the Spanish occupat1on of the Philippines Accord1ng to Rumphius cassava was being grown on Ambon one of the outer islands of Indonesia by 1653 (Nelson 1982) Cassava was 1ntroduced from Java to Maurit1us in 1740 and from Mauritius to Sri Lanka in 1796 (Greenstreet and Lambourne 1933) Certainly by the beginning of the 19th century cassava had been effectively distributed throughout tropical Asia Expansion of cassava production in the 19th century was hastened by colonial admin1strations first by the initiation of a cassava processing and export industry in Malaya in the 1850's followed by the Dutch in Java and second by the promotion of cassava as a famine reserve particularly by the Dutch in Java and the British 1n Southern India Of the new world food crops introduced into tropical Asia cassava has become the most important on a production basis Characteristic of the crop the development of cassava has responded to different forces in each country as is particularly reflected in the utilization patterns for the different countries in Table 1 Cassava is an important food source only in India and Indonesia an important export crop in Thailand and an important source of starch in all countries Just as cassava has filled a particular market niche in each country the crop also occup1es a different productioa niche in each country that 1s in terms of the type of land resource which has been exploited and the type of cropp1ng system which has evolved The crop' s pecul1ar adaptability to upland cond1tions particularly where there are either soil or moisture constra1nts and its multiple end-market uses g1ve cassava a certain malleability in adapting to quite different demand and production conditions By utilizing a comparative approach this paper propases to bring out the diversity and s1milarities in systems of cassava production and ut1lization in tropical Asian countries From this conclusions will be drawn about potential for and constraints on further development of the crop in the region An issue dominating this discuss1on will be whether princ1pal constra1nts have their orig1n on the production or the demand s1de or vice versa whether growth has been production or demand led This view departs substantially from the more orthodox perspective in Asia - which is dominated by the case of rice - which suggests that the restriction on increased food supplies 1s lack of sufficient factors of production especially land and the solution is therefore improved production technology and land productivity The question for cassava on the other hand 1s whether improved technology is a suff1cient st1mulus for the expansion of production or whether this as well needs to be integrated with market development A Comparative Analysis of Production Cassava 1s essentially an upland crop in tropical Asia Only in rare cases when water 1s l1miting such as occurs w1th well-fed systems in Tamil Nadu in Ind1a or during the secondary season on sawah soils of Java is cassava planted in irrigated areas The agro-cl1matic cond1tions under Table 1 Product1on and Ut1l1zat1on of Cassava 1n Pr1nc1pal Produc1ng Countr1es Domest1c Ut1l1zat1on Human Consum2t1on An1mal Country Product1on Export Fresh Dr1ed Starch Feed (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) (OOOt) Ind1a (1977) 5688 22 2610 619 ,(]~ Kerala 4189 22 2437 619 499 Tamü Nadu 1310 - 126 - 1162 Indones1a (1976) 9686 801 3444 2212 \!}_)7 Java 6317 253 1815 1760 2134 Off-Java 3369 548 1629 452 613 Malaysu (1977) 432 66 - - 302 43 Ph1l1pp1nes (1975) 450 - 223 37 92 32 Thaüand (1977) 13 554 9 996 - - ~ 16 ' - Source Unneveh1 1982 T1tap1watanakun 1979 CIAT data hles Waste (OOOt) 653 503 131 482 355 127 21 65 2797 IX - 2 - which cassava is grown ~n the upland areas of Asia vary enormously but the defining factor in major cassava producing zones is the existence of a constraint on plant growth In areas such as Kerala India the off-islands of Indonesia or the eroded slopes of eastern and central Java the lim~ting factor is soils In the northeast of Thailand Tamil Nadu in India or Madura ~sland in Indonesia the problem is moisture stress Cassava produces high carbohydrate yields under such cond~tions compared to other crop alternatives Cassava has thus tended to be concentrated in those areas where compet~tion with other crops is relatively insignificant This however is too broad a generalization for cassava competes quite effectively at both the extensive and intensive margin (Table 2) Cassava is grown in upland areas where farro size is a major constraint on farmers' crop production such as Kerala and Java Cassava is selected because of its high yields and yield responsiveness even where there are agro-climatic constraints Exploitation of the yield potential of cassava is clearest in the irrigated area of Tamil Nadu Here farm-level yields commonly exceed 50 t/ha On the other hand cassava is well adapted to more land extensive production systems such as occur in frontier areas Cassava has been a maJor crop component in the transm~grat~on schemes in Indonesia and where infrastructure has developed cassava has expanded rap~dly such as the Lampung area an Sumatra The same applies in the Mindinao area of the Philippines where cassava has become a major crop In such areas infrastructure development is a principal st~ulus in moving cassava from essentially subsistence status to a major cash crop In Malays~a as compared to other Asian countries cassava s role in the agricultura! economy is defined more by access to land than by land quality Malaysia is by Asian standards a land surplus country and much of the unexploited land remains under control of the federal government Cassava is the crop of first choice for squatters on federal land and apparently much of the cassava grown in Malaysia is grown by squatters In the major producing state of Perak a 1976 estimate ind~cates that 3 892 ha of cassava were planted legally while 10 240 ha were planted ~llegally (Hohnholz 1980) Given cassava s demonstrated ability to exploit the heterogenity of the land resource in Asia a ma)or factor determin~ng the production potential of cassava is its ability to compete with other crops for land in the upland areas An important point emerges on the product~on side cassava rarely competes for land with the same crops with which it competes on the demand side That is cassava rarely competes with food or feed grains There is some competition with ma~ze in the central plain of Thailand and to a more limited extent ~n Mindinao ~n the Ph~l~ppines but the one area where maize and upland rice overlap with cassava is on Java and Lampung and here the three are often found in an ~ntercropping system In areas where rainfall ~s limiting such as the northeast of Thailand or the unirrigated areas of Tam~l Nadu cassava has no effective competing crop In most of principally with rubber in Kerala the other cassava producing areas cassava competes tree crops coconuts in the Phil~ppines coconuts and o~l palm and rubber in Malaysia and the off-islands of Country Ch1na Ind1a 1 Indones1a Malaya1a Ph1hpp1nes Thaüand Table 2 Type of Land Constra1nt 1n the Pr1nc1pal Cassava Product1on Zones L1m1ted Farm S1ze Guangdong Kerala Tamü Nadu (ungated) Java (leve! sawah) V1sayas Central Pla1n Type of Land Constra1nt Marg1nal Agro-Cl1mat1c Cond1t1ons Guangx1 Tamü Nadu (non-1rr1gated) Java (eroded h1lls1de) Peat so1ls Northeast Front1er Area Transm1grat1on schemes Land development zones M1nd1nao Northern reg1on IX - 3 - Indonesia and rubber ~n the southern part of Thailand Southeast Asia has an internat~onal comparat~ve advantage in these crops over 80% 85/ and 90% of world exports of rubber coconut oil and palm oil respect~vely originate from the region Expansion possibilities in these crops are limited by the growth potential of world markets and moreover these are markets in which close substitutes exist Cassava' s ability to compete with tree crops for land labor and cap~tal ~n these areas is an open question but it will essentially depend on the relative importance given to expanding export markets versus meeting domestic demand for carbohydrate sources While it is the land issue that largely determ~nes where cassava is grown it is relative endowments of land to labor that determines how cassava is grown that is in what type of cropping system Cassava-based cropping systems vary substantially across As~a (Table 3) and the labour intensity of these systems is fairly consistent with the land/labor rat~o in each country (Table 4) In the countries with the highest land/labor ratios Malaysia and Thailand tractor services for land preparation are widely used in cassava production systems In the Philippines animal traction is common while in Indonesia and Kerala land is princ~pally prepared by hand A similar trend is found ~n weeding ~ntensity and the propensity to achieve a higher land productivity through intercropping and fertilizer applicat~on One common theme that does run across cassava cropping systems in Asia is the low use of chemical fertilizers (Table 3) Even in Kerala and Java chemical fertilizer application to cassava is low despite the fact that application levels on other crops particularly rice is very high To a signif~cant extent in Indonesia and India farmers compensate for this by applying organic manures and wood ash In India what green manure that remains in the field is incorporated into the soil below the planted stake Although many publ~shed fertilizer experiments have shown a y~eld response of cassava to fertilizer application the fact rema~ns that few farmers utilize chemical fertilizer in significant quantities A better understanding of the fert~l~zer response issue at the farm-level is needed but it does appear to offer one potential avenue for significant yield gains These differences ~n cropp~ng systems lead to significant d~fferences in labor input per hectare production costs and yields across Asian cassava production zones (Table 5) The largest cost component in cassava production is cons~stently labor D~fferences between countr~es ~n total per hectare labor costs are substant~al However once differences in yields are taken into account there is a signif~cantly reduced range of yyriable production costs per ton Expressed on a dr~ed equ~valent basis - these production costs must be seem as low compared to per ton production costs of grains l/ As a gross approximation 2 S t of fresh roots produce 1 t of dr~ed cassava expressed on a 14% moisture bas~s This will obv~ously vary depending on the dry matter content of the roots Table 3 Character1st1cs of Cassva Cropp1ng Systems 1n MaJor Product1on Zones Thaüand MalayS1a Indones1a Ph1l1pp1nes Ind1a Character1stJ.c Northeast Pera k Java M1nd1nao Kerala PrJ.ncl.pal Power Source Tractor Tractor Manual Bullock Manual Intercroppl.ng Monoculture Monoculture MaJ.ze and upland Monoculture Peanut rJ.ce pr1nc1pal recent 1ntercrops 1ntercrop Labor Input for WeedJ.ng (man days/ha) 37 6 13 3 hJ.gh 12 8 h1gh Fert1l1zer Use - OrganJ.c (t/ha) - - O to 8 6 non e hJ.gh - InorganJ.c (kg/ha) 9 6 198 21 7 non e 19 SeasonalJ.ty J.n Plant1ng 50/ planted Apnl-June shght 75% planted Modera te 60-65,. Nov-Jan planted Apnl-June - Average YJ.elds (t/ha) 138 27 2 9 7 4 7 13 6 - 1 Subs1stence Consumptl.on non e non e 277 17/ 6or Source Tha1land M1n1stry of Agr1culture and Cooperat1ves 1982 Tunku Yahya 1979 Roche 1982 HeJ1a et al Uthamal1ngam 1980 Tam1l Nadt Bullock Monocultt 96 7 18 5 200 MaJar po: plante1 Jan-Ma· 24 5 neg 1979 IX - 4 - However 1t is probably yield rather than per hectare production costs that is the principal variable in the determination of costs per ton Cassava as compared to the gra1n crops has a potent1ally high yield variance Y1elds as low a 2 t/ha are not uncommon in many parts of the Philippines wh1le farm yields reaching as high as 80 t/ha have been recorded in Tamil Nadu Ind1a This very large yield potential has always been the hallmark of the crop and it is in Asia that this yield potential has been most exploited Compared to Africa or Latin America yields in Asia are high Part of this is due to the significantly lower disease and insect pressure since Asia is outside cassava' s center of origin The other factor is the more intensive cassava cropping systems found in Asia The other basic characteristics of the crop however is it adaptation to marginal growing condit1ons Yield potential must therefore be defined in terms of agro-climatic condit1ons Because of the differences in agro-climatic conditions of the majar production regions and 1n cropping systems between these regions there is a large variation in yield levels within tropical Asia (Table 6) While general causes for the differences 1n yield between regions can be postulated there has been no systematic work which has specifically related differences in agro-cl1mat1c conditions input 2fevels variet1es and management practices to variation in yield levels - Without th1s information it is very difficult to assess the principal constraints on cassava yields and in turn the potential far increasing cassava productivity The potent1al yield gains from new technology and 1n large measure the definition of that technology still remain rather amorphous Nevertheless the range of yields suggested in Table 6 are at least suggestive of substantial scope for yield improvement in many countries A Comparative Analys1s of Consumption The food economies of tropical Asia are dominated by rice any other starchy staple is only of secondary importance in the regional diet Within this context cassava has achieved a significant role in the food econom1es of Indonesia and Kerala and only maize is as signif1cant a calorie source in tropical Asia Tne impetus for the early expansion of the cassava crop 1n Kerala the Phil1ppines and Indonesia was to supplement inadequate supplies of rice and it was in land-scarce Kerala and Java tbat cassava production expanded most significantly In Thailand and Malays1a on the other hand the incentive for production expansion came from non-food markets The locus of cassava consumption in Indonesia and Kerala is in the rural sector and among the lower 1ncome strata Moreover because cassava The research by Rache is the one exception labor input the other system dummies (1982) on cassava cropp1ng Apart from age at harvest explanatory variables were systems on fertil1zer regional or Java and land Table 4 Land-labor Rat1os and Average Farm S1ze for Var1ous As1an Countr1es 17 Land-Labor Rat1o - Country (ha/person) Indu (Kerala) o 12 Indones1a o 22 Java N A Malays1a o 65 Ph1l1pp1nes o 44 Thalland o 51 Average Farm S1ze (ha/farm) o 49 (1971) 1 os (1963) o 4 (1973) 2 19 Jj ( 1970) 3 59 (1960) 372 (1978) !/ Arable land and land 1n permanent crops d1v1ded by rural populat1on 1980 11 Does not 1nclude estates wh1ch make up 317 of cult1vated area Source FAO 1981 agr1cultural censuses of d1fferent countr1es IX - S - is very much a secondary staple in the food economy of these countries it is sign1ficantly less preferred than rice in the d1et These characterist1cs to a large extent def1ne cassava s role in these food econom1es as a cheap calor1e source wh1ch supplements shortfalls in the availability of r1ce whether due to insuff1cient supplies or restricted purchasing power Cassava has thus come to play a signif1cant role in the calorie nutrit1on of that population most at risk in the reg1on (Figure 1) While food policy in these countries will st1ll have rice as 1ts central component cassava can add a certa1n flexibility to these rice-based policies Unfortunately it is rare that pol1c1es on secondary staples are 1ntegrated with those on rice 1n developing an overall food and nutrit1on policy The role of cassava in nutrition plann1ng has been analyzed most rigorously in Indonesia (Dixon 1982 Timmer and Alderman 1979 Timmer 1980) Cassava s low cost relative to r1ce the very skewed d1str1bution of consumpt1on toward the low income strata the existence among the peor of calorie intake well below recommended standards and among the lowest income strata the significantly pos1tive income elasticity for cassava (D1xon 1982) create a situation where increased cassava production and lower prices will impact exclusively on the peor consumer Overall inelast1city 1n food markets wh1le providing substantial benef1ts to consumers when improved technology is introduced does not provide much scope for increasing farm incomes Cassava is a cash crop 1n Asia Even in Indonesia and Ind1a where there is some subs1stence food consumption the major portien of the cassava moves into market channels Where cassava production has expanded rap1dly in the reg1on this expansion has been associated with dynamic markets Thus 1f cassava 1s to play a role in food pol1cy there must be a means of maintaining 1ncent1ves to producers Cassava s role in generating increases 1n farmer 1ncomes is therefore associated w1th markets other than traditional food markets Where trad1tional food markets are important development of these alternative markets provides something of a price floor to susta1n farmer incomes The economies of Southeast Asia have been chang1ng rapidly in the last two decades (Table 7) Industrialization rapidly ris1ng income and significant rates of urbanization have created signif1cant changes 1n domestic demand for food Food demand within the reg1on 1s being driven principally by changes occurring outsides the agricultural sector yet it is this sector wh1ch must cont1nue to generate both the bulk of employment in the economy and continued 1ncreases in marketable surpluses Increas1ng demand in the quantity and var1ety of food products can be a st1mulus to the agricultural sector or can put unwanted pressure on internal food prices-- and thus affect the nutr1t1on levels of the peor-- and/or food 1mports Th1s situation is potent1ally aggravated by the winding down of the product1on ga1ns achieved by the dwarf r1ce variet1es and by the s1gn1f1cant portien of resources devoted to export tree crops One of the dominant trends in As1an food economies is the r1sing demand for l1vestock products and the der1ved demand for carbohydrate and prote1n sources for concentrate feeds (Table 8) Th1s growth 1n demand for livestock products has been most striking 1n the poultry sector that is Table 5 1/ Labor Use and Cost Structure 1n Cassava Product1on Systems - Country Locat1on Per1od Labor Input (m d /ha) Land Costs (US$/ha) Var1able Costs (US$/ha) Labor Land Preparat'lon Fert1l1zer Pest1c1des Seed Total held Var1able Costs (US$/ton) Indones1a Gunung K1dul 1979/80 345 8 o 97 8 o o o 2 6 lOO 4 2 6 38 6 Indones1a Ked1r1 1979/80 237 2 233 7 227 o 106 7 114 9 o 4 8 453 4 17 5 25 9 Thaüand Cholbur1 1977/78 74 8 28 9 76 2 59 2 16 6 2 7 16 6 171 3 10 9 15 7 Tha1land NakornraJ s1ma 1977/78 67 2 74 8 64 o 33 5 o o 1 9 99 4 13 7 7 3 l/ Domest1c currency converted to US dollars at ex1st1ng exchange rate 2/ Share tenancy - 33/ of gross value '}_/ Herb1c1des Ind1a Salem 1978/79 138 5 121 3 90 9 13 4 59 8 o o 164 1 10 7 15 3 Ph1hpp1nes Central V1sayas 1976/77 65 o 46 4 Jj 50 1 5 1 o o o 55 2 5 5 10 o SOURCE Rache 1982 T1nprapha 1979 Uthamal1ngam 1981 MeJ1a et al 1979 Tunku Yahaya 1979 Malays1a Perak 1977/78 62 2 17 3 116 4 38 9 25 9 12 1 '}_/ 3 5 196 8 27 2 7 2 IX - 6 - for meat and eggs The poultry and feed concentrate sector has developed rapidly over the last decade in the cassava producing countries of Thailand Philippines and Malays~a and in the non-producing countries of Taiwan Japan and the Republic of Korea The sector is only in a very formative stage ~n Indonesia However per capita consumpt~on levels remain low and FAO (1983) antic~pates annual growth rates to the year 2000 on the arder of 8 8 and 6 3% for poultry meat and eggs ~n the Far East Maize is universally the princ~pal feedgra~n used in the feed concentrate industry in the region and only Thailand Philipp~nes and Indonesia are signif~cant producers of which only Tha~land is in a net export position Without a doubt Southeast Asia w~ll have a continuing deficit in production versus consumption of feedgrains However at present only very insign~ficant amounts of cassava enter into an~mal feed rations in the region At around 15 thousand tons Malaysia is apparently the largest ut~lizer of cassava for feed concentrates A large and growing domestic market thus remains unexploited in most countries After direct food use starch is by far the largest form of domestic utilization of cassava in the region As in the case of livestock products consumption levels of starch have increased rap~dly in most countries in the last decade (Table 9) In countries such as Indonesia and Malayaia and regions such as Tamil Nadu India and Mindinao Ph~lippines starch process~ng dominates the market for roots These similarit~es contrast with significant heterogenity across countr~es ~n the end market for cassava starch competition with other starch sources principally maize and the scale of processing technology within the starch industry These latter factors determine to a large extent the future growth potential for cassava starch in each of the countr~es The other maJar cassava market is the export market exports are dominated by chips/pellets although there ~s a s~gn~ficant volume of cassava starch that is exported as well While all of the majar cassava produc~ng countries in the reg~on have exported cassava products in the recent past only in Thailand is production principally directed to export markets In all other countries the export market ~s minar when compared to the domestic market Ind~a and China have been interm~ttent exporters while Indones~a has been a consistent exporter but with large fluctuations in quant~t~es Malaysia has been a consistent but decl~ning exporter For these latter countries the export market serves as someth~ng of a surplus vent wh~ch usually is operational only at relatively h~gh world market pr~ces Th~s was part~cularly the case in 1979-80 and demonstrates the role that the export market can play ~n setting a price floor under domest~c markets even though at historically low to moderate world price levels domestic prices in most countr~es make cassava exports uncompet~tive A multiple market structure has developed for cassava in most countr~es in the region with each country having developed its own part~cular ut~lization patterns Yet as has been noted sign~ficant untapped potent~al ex~sts for cassava in undeveloped markets such as the domestic feed concentrate markets Other markets wh~ch have been unment~oned are the composite flour market especially where the wheat flour is used pr~ncipally in noodles and in sugar-import~ng countr~es Table 6 Comparat1ve Y1elds Der1ved from Nat1onal Stat1st1cs and Product1on Surveys Nat1onal Stat1st1cs Country/Regwn Year Y1eld (t/ha) Ind1a 1978-79 16 7 Kerala 1978-79 14 6 Tam11 Nadu 1978-79 31 2 Malays1a 1978 17 4 Perak N A Indones1a 1977-79 12 9 West Java 1977-79 10-12 Central Java 1977-79 9-11 South-Central Java 1977-79 7-9 East Java 1977-79 10-11 1 Ph1l1pp1nes 1977-79 10 3 Central Luzon 1977-79 2 4 B1col 1977-79 9 6 Central V1sayas 1977-79 3 5 Eastern V1sayas 1977-79 4 2 Western M1nd1nao 1977-79 14 7 Northern M1nd1nao 1977-79 4 6 Tha1land 1980-81 13 1 North 1980-81 17 o Central 1980-81 15 5 Northeast 1980-81 13 3 Source Uthamal1ngam 1980 Tunku Yahaya 1979 et al 1979 M1n1stry of Agr1culture and nat1onal stat1st1cal sources 1 Non-1rr1gated and 1rr1gated cond1t1ons Product1on Surve:¡: Year Y1eld ( t/ha) N A 1978-79 13 6 and 1978 27 2 1979-80 6-20 1979-80 5-12 1979-80 2-10 1979-80 10-40 1977-79 5 8 1977-79 2 5 1977-79 5 5 1977-79 2 2 1977-79 5 4 1977-79 4 o 1980-81 14 2 1980-81 15 1 1980-81 13 8 Rache 1982 MeJ1a and Cooperat1ves 1982 23 o1 IX - 7 - such as Indonesia high fructose syrups A natural question is what has been constraining the development of these alternat1ves markets and in turn whether improved production technology could be a motivating factor in their development At the heart of this issue is the original question of whether it is production or demand that is constraining or generating further development of the crop and to answer this question the issue of price formation must first be analyzed Marketing and Price Formation In a multi-market situat10n it is essentially price which allocates the cassava roots between the different end uses It is axiomatic that the price must be able on the one hand to cover the farmer s costs of production and on the other hand to compete with substitutes in the various markets Forces on the supply side such as increasing input or factor costs or the advent of more profitable crops may drive the production cost of cassava out of line with the market price of substitutes Vice versa forces on the demand side such as inelastic output markets or falling price of substitutes may drive the market price out of line with production costs at least for more high cost producers At issue in this section then is delineation of the principal factors determining cassava price in the different countries and of the mechan1sm influencing the allocation of cassava between different end uses The cassava products in the different cassava markets tend to compete with different substitutes This sets up someth1ng of a h1.erarchy of markets in which cassava in some markets can be competitive at higher prices than in others Thus in Kerala India the fresh food market is the principal demand-s1de factor in price formation Since there are severe supply-side constraints on expanding cassava production cassava prices set in the food market tend to be higher than are profitable for the operat1on of the starch industry which absorbs seasonal surpluses and roots of inferior quality In the Phil1ppines on the other hand the fresh food market usually sets a higher root price than the starch market but because the size of the food market is so limited the starch factories tend to be the ma]or market force in their supply area However expansion in th1s starch market has been apparently constrained by competition with maize starch There is potent1al for expanding cassava area and production for the animal feed market but y1elds need to be higher than their current average of around S t/ha and therefore costs of production lower Factors determining cassava prices are very different between countrfes (Table 10) and the constraints on further development of the crop also vary markedly In Thailand and the Phil1pp1nes the constraint is on the demand side wh1le in Ind1a Malays1a and Java the constraint 1s very much a production constraint Where cassava production has expanded rapidly in Asia such as Thailand and the Lampung area of Indones1a there has been the convergence of access to a very expansive market and underut1lized land to support area expansion In the other areas apart from the possible case of Malays1a growth in production w1ll depend on increasing yields whether to make cassava competit1ve in alternative markets or as a means of substituting for land where land availability is very l1mited Table 7 Selected Econom~c Ind~cators of Pr~nc~pal Cassava Produc~ng Countr~es Percent of GNP of 1980 GNP Per CaE1ta Industr~al Or~g~n 1 of Populat10n Growth ~n Urban PoEulat~on Country 1980 Leve! Growth 1960-80 1960 1980 ~n Urban Sector 1960 70 1970-80 ($US) 0') (/) (k) (%) (%) (k) Ind~a 240 1 4 20 26 22 3 3 3 3 Indones~a 430 4 o 14 42 20 3 6 4 o Malays~a 1620 4 3 18 37 29 3 5 3 3 Phü~pp~nes 690 2 8 28 37 36 3 8 3 6 Thaüand 670 4 7 19 29 14 3 5 3 4 Source World Bank 1981 IX - 8 - For a crop where in most countries prices are so dependent on forces within domestic markets and where there l.S such a diversJ.ty l.n market structure the expectation would be that cassava prices would very markedly across countries Evaluated at current exchange rates farm-level prices are consistently the lowest in Thailand and are the highest either in Indl.a or Indonesia (Table 11) although the latter are probably inflated because the series is based on village-level prices Clearly however the competitive position of Thailand in the world market is firmly established while the other countries remain either minor or intermittent exporters Moreover it is only in Thailand that there has been any clear trend in real farm-level prices over the last decade and this has been a downward trend which is consistent with the very rapid expansion in production In the other countries farm prices have been relatively stable which would appear to imply a relatively stable supply-demand situation The case in Indonesia is more complex than that but certainly for the other countries there has been little incentive to develop lower-priced markets Different end markets and different forms of marketing cassava raise the second J.ssue of how price allocates the cassava roots and dried products between the different markets As it has been noted only a relatively small part of cassava production remains on the farm for subsistence consumption and this occurs only in Indonesl.a and Kerala the greater portl.on moves into marketl.ng channels Farmers market the major part of their production as fresh roots and it is generally the assembly agent who decides on the end market to whl.ch the cassava will go However farmers also have the option of producing gaplek-- by peeling quartering and drying the root This practice predominates in Indonesia and is utl.lized to a much more limited extent in Kerala and the southern region of the Philippines Gaplek plays a fundamental role in Indonesia in integrating cassava markets across different forms space and time Various demands are made on a cassava marketing system due to the bulkiness and extreme perishabl.lity of the roots the different end uses and forms and in most countries the seasonality of production Seasonality is a problem in only the maJor cassava producing countries of Thailand Indonesia and Indl.a In Thailand about 50k of cassava area is planted in the April-June period in Kerala 60-657 is planted in the same three month period and in Java 75% of area is planted in the November-January period In Thailand the seasonality problem is overcome by processing all the cassava roots and by the availability of a large storage capacity In India and Indonesia where consumpt1on of fresh roots as food is important there is a definite seasonality l.n consumption as can be seen for the case of Indonesia in Table 12 In Indonesia and to a much lesser in India gaplek although a less preferred food serves to extend the consumption period thus resolving the seasonality problem not by adJustments l.n the production system but through adjustments in marketl.ng processl.ng and consumption form Gaplek provJ.des the storage capabill.ty in cassava markets and thus tends to l.ntegrate them through time Gaplek also permits economical transport of cassava and thus tends to integrate cassava markets across space as well That is consumption points for fresh roots normally draw on only a very small supply area due to the high transport costs and the perishability constral.nt This situation would tend to create relatively Table 8 Product1on of Feed Concentrates 1n Relat1on to Coarse Gra1n lmports Feed Concentrate Growth 1n Concentrate Coarse Gra1n Growth 1n Coarse Country Productwn-1980 Product1on 1970-80 Imports 1980 Gra1n Imports 1970-80 (OOOt) (k) (OOOt) (/) Cassava Producers Tha1land 1350 28 6 - 2 175 Ph1hpp1nes 936 1 12 92 351 27 5 Malays1a 549 12 23 431 7 4 Indones1a 410 N A 34 3 5 Non-Cassava Producers Republ1c of Korea 47754 5 25 2 364 27 2 Ta1wan N A N A 3 618 N A Hong Kong N A N A 270 4 4 Japan 19 8766 N A 17 165 5 7 S1ngapore N A N A 55? 14 o 1 1979 2 1970-79 3 1972-80 4 1981 5 1972-81 6 1977 Source FAO 1975 and 1982 CIAT data f1les IX - 9 - independent markets in which prices vary s1gnificantly between areas These would tend to occur in countries in which food markets for fresh cassava dominate that 1s the Phil1ppines and Kerala (Table 13) Widely traded commodities such as starch and gaplek where arbitraging is poss1ble have more of a national market where prices are determined more by aggregate rather than local supply and demand situations Because farmers and/or assembly agents have the option of supplying roots to these markets gaplek and starch prices will tend to integrate fresh root markets with1n the economy as occurs in Thailand and Indonesia (Unnevehr 1982) Price integration across markets space and time is critical in fostering growth in cassava production and utilization Integrat1on provides incentives for cassava to be grown in areas where production is most efficient it maintains competitive price formation and it provides the necessary information implicit in nationally determined market prices to motivate investment 1n processing capacity for which there is greatest market potential Fragmented markets in a crop such as cassava can sign1ficantly 1nh1bit wide-spread investment in processing plants by making cassava appear too costly in price terms in relat1on to its actual production cost This 1s certainly one factor in explaining the lack of growth in Philippine cassava production compared to that in Thailand and Indonesia Finally an observation arises on the role that gaplek can play in price integrat1on between different and markets Gaplek is in many ways a cassava grain If properly dried it can be stored which provides food supplies out of the harvest season Because it is peeled it can be ground for composite flour production or go into domest1c or export animal feed markets Starch plants in India and the Phil1pp1nes occasionally use gaplek for starch processing especially for glucose production when fresh root supphes are limited Apart from kokonte in Ghana and farinha de raspa in Brazil dried cassava chips of this quality are only produced in Asia almost solely in Indonesia Interestingly Indones1a has the most diverse end markets for cassava and is probably the most fully integrated cassava market where the bulk of production is for domestic use Motivating a gaplek market of a certa1n minimum critical size would appear to give the cassava economy a large degree of flexibility 1n responding to changing economic and market conditions Cassava s Future Role in Asia Beyond the central role that rice plays in the food economies of tropical Asian countries the agricultural sectors of these countries are very diverse Cassava product1on and utilization has adapted itself to this divers1ty As l.S apparent in the previous analys1s it is the differences rather than the similarities that are most striking in comparing cassava sectors across countries Cassava has developed withl.n d1fferent types of land constraints and multiple markets have evolved around the crop w1th the particular market structure reflecting the overall development of the economy The rate of development of most of these economies has accelerated over the past two decades creating a potential demand for further broadening of cassava production and utilization Table 9 Character1st1cs of the Cassava Starch Iodustry 10 the Pr1oc1pal Produc1og Couotr1es untry 1a ones1.a ays1a l1pp1nes üand 1974-79 Cassava Starch Product1oo 1980 (000 t) 415 662 50 17 3 416 2 1972-80 Growth 1n Cassava Starch D1sappearaoce 1970-80 (~) 3 1979 N A 8 9 1 9 9 2 2 9 4 7 7 4 1970-79 ~urce Nelson 1982 CIAT data f1les Growth 10 Total Starch D1sappearaoce 1970-80 (/) N A 8 9 1 9 9 2 7 9 4 7 7 Two Largest F1oal Eod-Uses Tap1oca Pearl Cloth S1z1og Krupuk Other food Iodus- tr1es N A Glucose Monosod1um Glutamate Food Industry Monosod1um Glutamate Modal Scale ' Process1ng Med1um Med1um to L Large Large Large Large IX - 10 - Rapid development of the crop in most cases will depend on ~ncreases in yields either to relieve land constraints or to be compet~t~ve in these emerging markets It is natural ~n an As~an context where expans~on of crop area ~s frequently constrained that there should be a bias toward crops with very h~gh yield potent~al more so when this is high y~elding ability under upland conditions Very high productivity is already being achieved ~n certain areas but in general average y~elds remain below the known potent~al of the crop What still rema~ns largely undefined is the means to achieving this high yield capability across trop~cal Asia Obviously the type of technology necessary will vary requiring a cont~nued commitment of research resources to mainta~n the cassava research capac~ty in As~a that has emerged over the last two decades since the found~ng of the Indian program ~n 1963 Governments however require some JUstification for research investment wh~ch follows from the role cassava could play in the policy arena Cassava' s adaptation to a wide range of upland conditions and its multiple-use character~stics give cassava a substantial flexibility in agricultura! policy As has been stressed cassava's role in each country's agricultura! economy w~ll be different (Table 14) but ~n each case cassava can be a basis for meeting multiple policy object~ves In India and Indonesia cassava can play a clear role in nutrition policy In all countries even ~n India and Indones~a cassava because of its multiple-market potent~al can play a maJar role as a source of income generation for small-scale farmers in upland areas A further advantage in satisfying growing domestic markets by increased domestic production is the positive ~mpact on balance of payments Further market d~versification of cassava however will require both improved production technology and appropriate processing technology together with in some countr~es better integrated markets The Green Revolution that swept the continent ~n the late-sixties and the seventies was limited to the irrigated areas The next majar challenge ~s to raise crop productivity and farmer incomes in the upland areas With probably l~mited prospects for further maJar growth in world demand for rubber palm oil and coconut oil with grow~ng domestic markets that could absorb cassava products and w~th a growing regional market for carbohydrate sources for l~vestock cassava is a maJar if not the maJar crop in a pos~tion to foster income growth in the upland areas of'trop~cal Asia Table lO Co try Ind a Ind1 K l Tam.1l N du Tha1l nd M laysu Ph 1 ppln s M nd nao R t of Cou try Pr1n ipal Factors D t rm n1ng Cassava P 1 F rmati and C n tra1n1ng Expans on of Cas ava P oduct1on and Ut1l1 at1on p DClp 1 MaJor Mark t S co dary Market Sta h and F d gapl k F d-F h Ro t F d F h R t St h Starch Food F b Root Exp t Pell ts Exp t St h Sta eh An1mal F d Star h Food-F sh R ot Fo d F h Ro t St h p n p 1 C nst a nt of Alt rnauv n O v lopment Hark ts S pply síd Java Farm S1 Con t nt Off Java Comp t1t on w1th Tr cr p Farm S1 Con tra1nt FrmS Const nt Pn o stort1ons R 1 t1v8 D mand 1d Java Ex st ng G owth Ha k t Off Java-Infra t uctu e H gh Pr1c s 1n Food Ma ket EX1st ng Growth Mark t to Gra1ns Creat d by EC Exp rt Ha k t Land Us Poll.cy Compet1t1on w1th lmport d Ma1z Lack of Integr t1on of Appr pr ate P duct1on and PToce 1ng Technology Lack of Int grat1on of Appr pr1ate Produ tion and Pr T chnol gy SS ng D m n nt nstra nt 1n E pan 100 of Produ tlon nd Ut1li t n J va-S pply S d Off-J va O mand S1d Supply 1d Supply-s d D mand- 1d s pply ud Demand s d Demand s1de T bl 11 Farm 1 1 p e s of Ca a R t R al (1975 lOO) D m t e Cu y Pr e s and US Dalla P l.C S 1970 81 I d 1 Indo s a 2 Malaisla 3 Ph 1 E:E:l 4 Th 1 d 5 Y ea R al -Prlce Doll p R 1 Pr e D llar P ce R al Pr1 Dolla Pr Real P Dll p e Real P Dll p (R 2 /t) (US$ / t) (R 2 /kg) (US$/t) (M$/t) (US$/t) (P s/kg) (US$/t) (B ht/kg) (US$/t) 1970 N A N A 19 7 22 N A N A 25 20 79 24 1971 391 29 17 7 19 83 20 27 23 82 25 1972 406 31 21 5 23 56 15 25 22 72 23 1973 446 40 28 3 40 65 22 30 31 38 14 1974 423 47 16 1 32 79 32 31 42 30 14 1975 400 48 17 6 42 78 30 29 40 40 19 1976 449 44 23 4 67 73 29 26 37 44 22 1977 376 37 21 9 70 76 33 26 40 43 23 1978 353 39 19 9 64 58 28 26 43 29 18 1979 411 49 19 4 53 67 36 25 50 56 36 1980 N A N A 20 3 67 89 51 25 58 47 37 1981 N A N A 19 7 73 72 43 N A N A 30 25 1 K r 1 F rm-1 1 2 J d Mad R r 1 V llag leve! 3 p k F 4 Ph1l1pp1n Fart~rl 1 a t ry Buy g p e Av rag 1 N g t d d lg t d d t n S CIAT D t F 1 Table 12 Indones~a Seasonal~ty ~n Consumpt~on and Pr~ces of Fresh Cassava and Gaplek 1976 Consumpt1on (kg/cap~ta) Java-Rural Fresh Cassava Gaplek Indones~a Fresh Cassava Gaplek Pr1ces (Rup1ah/1000 calor1es) Indones~a Fresh Cassava Gaplek Source Dnon 1979 January- May- September- Annual Apr~l August December Average 33 7 24 7 33 3 19 7 21 14 25 1 31 6 27 o 25 3 24 13 15 8 33 9 17 o 23 o 26 20 24 9 30 1 25 7 22 6 23 16 Table l3 Reta1l Pr1ces of Cassava Fresh Roots 1n D1fferent Market Areas Kerala and the Ph1l1pp1nes 1979 Kerala Reta1l Pr1ce Phü1pp1nes Reta1l Pr1ce (D1stnct) (Rupee/kg) (Repon) (Pesos/kg) Tr1vandrum o 50 Ilocos 1 29 Quüon o 48 Cagayan Valley 1 34 Alleppey o 59 Central Luzon 1 11 Kottayam o 63 Southern Tagalog 1 01 Idukh o 70 B1col l 07 Ernakulum o 60 Western V1sayas 1 53 Tr1chur o 51 Central V1sayas l 15 Palghat o 47 Eastern V1sayas o 95 Malappuram o 56 Western M1nd1nao 1 18 Kozh1kode o 62 Northern M1nd1nao 1 os Cannanore o 87 Southern M1nd1nao 1 30 Central M1nd1nao 1 00 So urce CIAT data hles Table 14 Potent1al Role of Cassava 1n Agr1cultural Pol1c1es of Selected As1an Countr1es Contr1but~on accord~ng to country Agr~cultural pol~cy obJect~ves Indones~a lnd~a Tha~land Ph~l~pp1nes Malays~a Food and nutr~t~on pol~c~es 1 a Flex~b~l~ty 1n r~ce pol~c~es b Nutr~t~on of the poor Farm ~ncome and land use a H~gher small-farm ~ncome ~n upland areas b Explo~tat~on of front~er areas Balance of payments a lncreased export earn~ng b lmport subst~tut~on X X (gaplek) X X (except Java) X (sugar) X X (fresh) X X X X X X X (~n the NE) (~n M~nd1nao) (peat so~ls) X X X (feed gra~ns) (feed gra~ns) 1 In Indones1a there ex1sts a pr1ce pol1cy on r1ce and 1n Ind1a r1ce comes under a food rat1on1ng system -- -r F1gure 1 D1str1.but1on of staple food consumpt1on Java 1976 Percent of Total Stapl e Food Calor1es lOO 90 80 70 60 50 40 30 20 R1ce ~ .... ', Cassava ,_ lOl~~~-=--:-~~~5:~ -------Sweet Patato .._ ________ _ Rp per <1 000 1 000 2 000 3 000 4 000 5 000 6 000 8 000 10 000 >15 000 cap1ta 1 999 2 999 3 999 4 999 5 999 7 099 9 999 14 999 per month Percent of Populat1on l 2 17 6 27 3 19 4 12 2 7 2 7 3 3 2 3 o l 6 So urce D1.xon 1982