Improving agricultural productivity in 
Papua New Guinea 

Strategic and policy considerations 

Dickson Benny, *Todd Benson, Mark Ivekolia, Mekamu Kedir Jemal, and Raywin Ovah 

  

WORKING PAPER #001 FEBRUARY 2022 



ii 
 

CONTENTS 

Executive summary ............................................................................................................................ iv 

Introduction .......................................................................................................................................... 1 

Higher crop yields – why and how? ................................................................................................... 2 

Why focus on improving crop productivity? ...................................................................................... 2 

Raising crop yields – policy priorities ................................................................................................ 5 

Measuring crop productivity ............................................................................................................. 7 

Crop yields ................................................................................................................................. 7 

Other measures of productivity ................................................................................................... 9 

Current crop yield levels in Papua New Guinea .............................................................................. 10 

Staple and cash crop yields in Indonesia – a comparative analysis .............................................. 13 

Critical review of policies and strategies of Papua New Guinea with a bearing on crop 

productivity ........................................................................................................................................ 19 

Master development frameworks – what role for improved agricultural productivity ....................... 19 

Priority given to raising crop productivity in current agricultural sector policies and strategies ........ 20 

Motivating improved agricultural productivity through non-agricultural policies and strategies ........ 25 

Assessment of policy framework for increased agricultural productivity in Papua New Guinea ...... 29 

Prioritized actions to enhance agricultural productivity in Papua New Guinea ............................ 30 

Appendices ........................................................................................................................................ 34 

Appendix 1: Current yield levels of major staple and cash crops in Papua New Guinea ................. 34 

Staple food crops ..................................................................................................................... 34 

Cash crops ............................................................................................................................... 46 

Appendix 2: Synopses of selected policy and strategy statements of the government of Papua New 

Guinea that have a bearing on agricultural productivity .................................................................. 54 

Master development frameworks.............................................................................................. 54 

Agricultural sector policies and strategies ................................................................................. 56 

Non-agricultural policies and strategies with relevance for improving agricultural productivity .. 60 

References ......................................................................................................................................... 62 

 



iii 
 

TABLES 

Table 1: Productivity levels of major staple food and cash crops in Papua New Guinea, actual and 

attainable on farm with optimal management, most recent estimates, mt/ha ........................................ 12 

Table 2: Productivity levels of selected food and cash crops in Indonesia and Papua New Guinea, most 

recent estimates, mt/ha ........................................................................................................................ 15 

 

FIGURES 

Figure 1. Crop yield concepts used by sociologists, agronomists, and economists and plant 

pathologists ........................................................................................................................................... 8 

Figure 2: Provinces of Indonesia ......................................................................................................... 14 

Figure 3: Sweet potato productivity and production areas of Papua New Guinea ............................... 36 

Figure 4: Banana productivity and production areas of Papua New Guinea ........................................ 38 

Figure 5: Taro productivity and production areas of Papua New Guinea ............................................. 39 

Figure 6: Chinese taro productivity and production areas of Papua New Guinea ................................ 41 

Figure 7: Cassava productivity and production areas of Papua New Guinea ...................................... 42 

Figure 8: Yam productivity and production areas of Papua New Guinea ............................................. 44 

Figure 9: Sago productivity and production areas of Papua New Guinea ............................................ 45 

Figure 10: Cocoa productivity and production areas of Papua New Guinea ........................................ 47 

Figure 11: Coffee productivity and production areas of Papua New Guinea ........................................ 49 

Figure 12: Oil palm productivity and production areas of Papua New Guinea ..................................... 50 

Figure 13: Coconut /copra productivity and production areas of Papua New Guinea .......................... 52 

 

 



iv 

EXECUTIVE SUMMARY 

If smallholder farming households in Papua New Guinea achieve higher crop productivity levels, pro-

gress will be made along several dimensions of the development vision for PNG – increasing GDP for 

the agricultural sector and the overall economy; driving growth, diversification, and transformation of 

local rural economies; improving food consumption; and reducing poverty. In this paper, we examine 

recent data on yields for the most important crops grown in PNG, assess what yields might be achieved 

based on productivity data from areas of Indonesia with similar growing conditions, and sketch where 

policy reforms could provide incentives and access to technologies to achieve higher crop yields by all 

farmers across PNG. 

While yield gaps – the difference between the crop yields farmers commonly obtain and what they 

might realize with optimal crop management and the use of high-yielding varieties – could not be esti-

mated for the main cash crops, rough estimates could be made for the main staple food crops. These 

vary by crop. The yield gap for sweet potato is the smallest – actual farmer yields are between one-

quarter to one-third less than attainable yields for the crop. Yield gaps for taro, Chinese taro, cassava, 

and yam are between 40 and 60 percent of what might be attained. Banana shows the largest yield gap 

at only about a quarter of the productivity levels that might be achieved under intensive cultivation. 

Strong agricultural research efforts coupled with effective agricultural extension systems are critical if 

farmers are to maximize the returns that they obtain from their crop production and reduce these yield 

gaps. Moreover, the challenges that farmers face in their crop production today are not necessary the 

same challenges that they will face in the future. Climate change will introduce new production chal-

lenges, whether through changes in rainfall and temperature patterns or through increased incidence of 

pests and diseases that are better able to flourish under the changed climatic conditions. Moreover, in-

creased trade in both cash crops and food crops heightens risks from introduced pests and diseases 

that so far have not affected farming in PNG. To enable farmers in PNG to adapt and surmount to these 

new challenges similarly will require continuous research and effective communication to farmers of the 

improved techniques and technologies that will enable them to manage them sustainably and profitably. 

However, equally as important, if not more so, is strengthening agricultural markets across the 

country so that these markets both efficiently supply farmers with improved agricultural inputs to in-

crease their productivity levels and provide sufficient price incentives to farmers to grow specific crops 

at higher levels of productivity. The performance of the markets into which farmers sell their crops is as 

important as their access to technologies and techniques to achieving higher agricultural productivity 

overall. The comparative case study of current crop productivity levels in Indonesia and Papua New 

Guinea shows that it is uniquely the crops with strong market demand in Indonesia for which productiv-

ity levels there are significantly higher than in PNG. In addition to fostering increased commercial crop 

production through agricultural interventions and strengthening agricultural institutions, such as exten-

sion, government must work to raise market demand for crops by improving the efficiency of crop value 

chains. These efforts include developing techniques and opportunities for food processing and preser-

vation and enhancing domestic markets through improving physical marketplaces, strengthening trans-

portation infrastructure, and ensuring reliable, low-cost communication to enable profitable marketing 

activities. 

To raise crop productivity, specific efforts by crop type will be needed. For broad-based develop-

ment with significant poverty reduction, staple food crops, particularly the traditional root and tuber 

crops, should be the crops that receive most support and investment from the public agriculture sector. 

However, most roots and tubers are difficult to store, are not extensively processed, and are subject to 



v 

spoilage in transport. Building increased incentives for commercial production of the traditional staple 

crops grown in PNG will require research on how best the crops might be processed so that they can 

be readily marketed or stored and still find a significant consumer market, whether domestic or over-

seas.  

However, PNG’s cash crops should not be neglected. Here too some targeting is needed. Particu-

larly those for which smallholders are the main producers – cocoa, coffee, and coconut – deserve more 

effective attention from government and its partners. The apparent stagnation in production and 

productivity levels of these cash crops suggests that government is not as supportive of their develop-

ment as it might be. More effective research, extension, and trade facilitation is required. In contrast, for 

oil palm, now PNG’s largest cash crop, government should primarily facilitate the operations of the firms 

involved, while ensuring that those firms sustainably manage the resources of PNG that they are using 

and that beneficial economic spillover opportunities from the oil palm plantations to surrounding com-

munities are maximized. 

Keywords: agricultural productivity, smallholder agriculture, Papua New Guinea 

 



1 

INTRODUCTION 

In 2018, the agriculture sector was estimated to have contributed about 17 percent of the total gross 

domestic product (GDP) of Papua New Guinea (PNG) (World Bank 2021). This is a substantial drop 

from the sector providing one-third of total economic output 50 years ago. In contrast, the services sec-

tor in 2018 contributed 42 percent of total GDP and the industrial sector, including petroleum and min-

eral production, 37 percent. The secondary importance of agriculture to the national economy is also 

reflected in the share of exports made up by agricultural commodities – statistics from the Bank of Pa-

pua New Guinea show that agricultural commodities made up just over 10 percent of all exports by 

value over the three-year period, 2017 to 2019, while hydrocarbons made up 44 percent and minerals 

36 percent (BPNG 2021).1 

Nonetheless, the livelihoods of many individuals of working age in PNG are centered on agriculture, 

with much of the farming being done by smallholder farming households, relying on family labor princi-

pally to grow their own food. Estimates of the share of the workers in PNG employed in each economic 

sector reflect the central role of agriculture for most PNG households. The International Labour Organi-

zation estimated in 2019 that 56 percent of those employed who are of working-age (15 to 64 years) in 

PNG worked in agriculture, whereas 31 percent worked in the service sector and 13 percent in the in-

dustrial sector (World Bank 2021). Given the share of GDP that is attributed to each sector, the relative 

labor productivity of those engaged in farming is significantly lower than that of those engaged in the 

service sector and especially the industrial sector. The low level of labor productivity in agriculture 

in PNG reflects the development challenges facing the sector and the barriers in PNG to signifi-

cant economic growth overall emerging from agriculture. 

The food systems of PNG historically have been based on self-sufficient production of roots and tu-

bers, in particular, with most farming activities being for own consumption within the farming household. 

A mid-2018 rural household survey in communities in Madang, East Sepik, and Sandaun (West Sepik) 

provinces, found that three-quarters of the food the households reported consuming in the previous 

week by value came from the households’ own production (Schmidt et al. 2019). However, this pattern 

is not universal across all rural communities of the country. For households in South Bougainville dis-

trict, the other area in which the 2018 survey was conducted, only 47 percent of the food that survey 

households consumed was reported to come from their own production. Commercial production of co-

coa, as well as artisanal mining, is common in South Bougainville, so households there rely on the mar-

ket for more of their food than is the case in the other survey areas. Moreover, across all of the four sur-

vey areas, it was found that rural households that engaged in some form of non-farm enterprise, in ad-

dition to their farming, were able to increase their consumption levels and improve the quality of their 

diets by expanding the diversity of the foods they consumed through purchases in local markets using 

the income obtained from those enterprises (Schmidt, Mueller, and Rosenbach 2020). However, even 

in those rural areas where purchased foods are a more common part of the household food basket, the 

foods obtained in the market will often be locally produced. Outside of district centers, the diversity of 

 

1 The share of all exports by value made up by agricultural commodities is somewhat higher than the share of exports made 

up by marine (4 percent) and forest (3 percent) products, the other export commodity categories, in addition to hydrocarbons 

and minerals, tracked by the Bank of PNG 

Imports of agriculture-based products, primarily food, accounted for 14 percent of the value of all imports into PNG for the 

same period. 



2 

processed foods available in local shops is quite restricted (Benson et al. 2020). Local agricultural pro-

duction is central to rural food systems in most parts of PNG, whether for own consumption within the 

producing household or through local markets that facilitate the exchange of foodstuffs between neigh-

boring households and communities. 

Based on per capita GNP, PNG is characterized as being a lower-middle income country. 

However, given that the most significant contributions to economic output come from petroleum and 

minerals and not from agriculture, outside of oil palm, the agricultural sector in PNG is better char-

acterized as low income with quite limited development. What might be the agricultural production 

possibilities for PNG farmers if its agricultural sector were developed to the same extent as its petro-

leum and mineral sectors? Such development would involve significant transformations in agricultural 

land use, production techniques and technologies employed, and in the penetration and scope of agri-

cultural markets across the country. 

Higher productivity levels for many of the crops produced in PNG could expand agricultural GDP, 

improve the quality of diets, and, if coupled with improved performance in local markets across the 

country, increase household incomes. The relatively low crop yields realized by farmers in PNG and the 

declining importance of agriculture in the national economy represent important opportunity costs for 

those who farm or might farm and for the economy overall. A central objective in efforts to further de-

velop the agricultural sector must be to enable PNG’s farmers to realize crop yields that are much 

closer to what they might achieve with the use of improved crop germplasm, sustainable soil fertility 

management techniques, effective control of crop pests and diseases, and timely crop management 

operations. Doing so is not only in the best interests of the farmers, but will strengthen local economies 

across PNG. The increased wealth realized by farmers that increase their crop productivity will have 

spillover effects within their communities. Productive farmers will demand more of the labor-intensive 

and generally nonmarketable (outside of the local area) goods and services that their neighbors pro-

duce. This consumption linkage diffuses many of the economic gains from the more productive farming 

of these households to other rural households, deepening local markets, accelerating local economic 

activities, and improving access to food for economically active households in these communities. 

In this paper, we examine the most recent data available on yields for the most important crops 

grown in PNG, assess what yields might be achieved based on productivity data from areas of Indone-

sia with similar growing conditions, and sketch out where policy reforms at national level could provide 

incentives and access to the technologies needed to achieve significantly higher crop yields by all farm-

ers across PNG. However, before considering the data on crop productivity levels in PNG, we first con-

sider from a conceptual standpoint why higher crop yields are important to achieve for the development 

of the agricultural sector and the national economy of PNG. 

HIGHER CROP YIELDS – WHY AND HOW? 

Why focus on improving crop productivity? 

What motivates our focus on crop productivity in PNG is that growth within the agricultural sector funda-

mentally reflects increases in the amount of agricultural output produced, of which crops make up the 

majority, and in the market value of that output. Increased output in the agricultural sector, just as in the 

industrial and services sectors, in simplest terms contributes to overall economic growth at national 

level. Economic growth – that is, increases in the output of the economy – better enables the govern-



3 

ment of PNG to make progress in pursuit of its goal of maximizing social welfare for its citizens. In-

creased output can be used to finance additional production, foster new enterprises in agriculture and 

in other sectors, and to make the investments that will provide jobs offering higher returns to more 

workers. For individuals engaged in agricultural production, higher farm production generally will raise 

the cash income and material resources that they can draw upon to meet the basic needs of their 

households and, with increases in their production levels over time, improve the welfare of their house-

holds and expand opportunities for their children to realize more readily their social, creative, and eco-

nomic potential. These high-level objectives motivate interest in establishing policies and programs that 

will sustainably raise agricultural productivity levels in Papua New Guinea. 

As noted, on a GDP basis, the structure of the economy of PNG is heavily skewed towards petro-

leum and mineral production, with considerable potential for additional expansion in these sub-sectors. 

However, petroleum and mineral production is capital intensive with limited direct spillovers to enhance 

production in other economic sectors. Consequently, the broader impact of petroleum and mineral pro-

duction on the welfare of PNG’s citizens will be dependent to quite a large degree on how revenues 

from petroleum and mineral exports are used to foster growth in other areas of the economy. Given the 

significance of agriculture in the livelihoods of a majority of PNG households, how government 

invests its resources to increase agricultural production and commercial activity in the broader 

agri-food system of PNG will be critical to sustainable poverty reduction and improvements in 

the welfare levels of the many agricultural households across the country. Increased investment in ag-

riculture has repeatedly been shown globally to be superior in reducing poverty levels to investments 

that are directed to other economic sectors (Irz et al. 2001; Diao et al. 2007). 

Beyond strengthening livelihoods for households engaged in some farming, increased productivity 

in the agricultural sector also plays an important role in fostering broad economic growth and lays the 

foundation for a structural transformation of national economies through substantial movement of labor 

and capital out of agriculture and into industries and services and, ultimately, a smaller share of GDP 

being made up by agricultural production. Johnston and Mellor (1961) provided among the first system-

atic assessments of how a vibrant and productive agricultural sector contributes to overall economic 

growth and to accelerating economic transformation. Agricultural growth makes five contributions to 

such processes: 

• Providing increased food supplies – as the population continues to grow and as an economy 

expands and incomes increase, demand for food rises. With greater quantities of food pro-

duced, more agricultural output will be brought into the market for sale. With more reliable 

commercial supplies of food, markets will become more central than traditional subsistence 

production to household food security. As confidence in markets builds, more households 

will choose to pursue non-farm livelihoods. To meet their food needs, these households will 

rely on purchases of foods produced by the increasingly productive farmers remaining in the 

agricultural sector. 

• Expanding agricultural exports – While, given its petroleum and mineral resources, agricul-

tural exports are less pertinent to PNG than to other developing countries, increasing ex-

ports of agricultural products is an important means to increase incomes of households and 

firms and to boost foreign exchange earnings that will enable government to finance and 

step up its development efforts. 



4 

• Enhancing labor markets – Increased and more efficient agricultural production will enable 

labor currently used for farming to work in expanding sub-sectors in the industrial and ser-

vices sectors without adversely affecting agricultural production levels. 

• Generating capital for greater investment - The capital needed for development in the indus-

trial and services sectors can be generated through increased agricultural production. While 

petroleum or mineral export revenues will be the main source of development financing for 

PNG, the agricultural sector can also contribute. The capital requirements for agricultural 

development generally are less than those required for development in the non-agricultural 

sectors, allowing some agriculture-derived surpluses to be diverted to finance expansion in 

the other sectors. However, such taxation of farming can go too far, resulting in agricultural 

stagnation or decline (Schiff & Valdés 1992). Caution is merited. 

• Fostering service and market diversification - The increased rural incomes obtained through 

more productive agriculture expand demand for the output of the growing services and, es-

pecially, industrial sectors, expanding markets for those outputs. 

These contributions of increased agricultural production operate at the economy-wide and eco-

nomic sector levels. However, increased farm productivity can be a central driver in the growth, diversi-

fication, and transformation of local rural economies, bringing about sharp drops in local poverty levels 

(Haggblade, Hazell, and Reardon 2007; Mellor 2014, 2017). The principal mechanism for such rural 

economic growth involves enabling the more commercially oriented smallholder farming households in 

a community to significantly increase their productivity through increased use of modern farming inputs 

and practices over an increasing share of local arable land. As the income of these now more produc-

tive farm households grows, their consumption of locally produced goods and services will increase. 

These goods and services are those that are labor-intensive, require limited capital in their production, 

and typically are not marketed outside of the local community—construction, building repair, and asso-

ciated services; transport and associated services; education, health, and other social services; furni-

ture and handicraft-making; food and beverage processing; and the like. 

As the economic linkages between the more productive farmers and their neighbors who produce 

those goods and services deepen, experience from elsewhere shows that the returns that the neighbor-

ing households obtain from those nonfarm activities will increasingly surpass those that they can obtain 

from continuing their low-productivity, generally subsistence-focused farming, propelling some speciali-

zation in local rural employment patterns (Mellor and Malik 2017). With sustained returns and continued 

investment—including judicious public investment in market support services, transportation infrastruc-

ture, communication services, electricity, and the like—many of the households producing goods and 

services for the local market will expand their activities to serve wider markets, further increasing their 

income. Many will transition from being poor, subsistence-oriented households that engage in some 

farming to become households specialized in livelihoods outside of farming and no longer be poor. But, 

critically, at the core of this mechanism for rural economic development is higher crop productivity 

through the use of improved agricultural technologies and effective crop management, together with 

strengthened markets to provide continued reliable incentives for commercial agricultural production by 

these now more productive farmers. 

A final significant contribution that increased agricultural productivity can make to economic devel-

opment and to sustainably improving the welfare of all Papua New Guineans is through mitigating the 

impact of climate change. Higher crop productivity will reduce land cover change, particularly of carbon-

rich forests and wetlands. With higher yields enabling demand for crops to be met from existing 



5 

cropland, forests and wetlands will not need to be cleared and brought into production. This will result in 

reduced greenhouse gas emissions and continued incorporation of carbon dioxide from the atmosphere 

into the forest and wetland biomass through carbon sequestration (Campbell et al. 2014). Farmers also 

can better manage their soils to increase the organic matter in them, resulting in greater productivity, 

increased carbon sequestration in the soil, and reduced greenhouse gas emissions (Lal 2004). Energy 

needs can also be met through farming dedicated energy crops that could replace fossil fuels in a more 

sustainable manner that would reduce the contribution of energy use to global warming (FAO 2013). 

The more efficient use made of natural resources that is inherent to raising crop productivity levels will 

reduce the adverse impact that a growing agricultural sector in PNG would otherwise have on global 

warming processes. 

Raising crop yields – policy priorities 

The previous sub-section provides a condensed answer to the question of why higher crop yields 

should be an important agricultural and economic development objective for stakeholders in PNG’s ag-

ricultural sector and for PNG’s leaders. But how can crop yields be raised? 

In considering what might be done to assist farmers to raise their productivity levels, it has been re-

peatedly shown that farmers in traditional farming systems, far from being ignorant and passive, quite 

effectively and logically choose the production techniques they use to maximize the benefits they obtain 

from their efforts and to reduce the costs they might incur given available technologies, the production 

risks they face, and the incentives to which they respond as to what to produce and at what level of pro-

duction (Schultz 1964). Economic research from around the world has demonstrated that tradi-

tional farming systems generally are efficient in economic terms – farmers allocate their availa-

ble labor, land, and other factors of production up to a level at which the returns to doing so are 

equal to the value of output they obtain from the use of each factor in their farming. Productivity 

levels generally are low in traditional agriculture not because the farmers are bad at farming. Rather, 

productivity is low because the only production technologies that farmers can profitably use given the 

conditions under which they farm generate low yields. That traditional smallholder farmers can be char-

acterized as efficient is conditional on their resource constraints, the incentives motivating their produc-

tion, and the trade-offs that they must weigh in their production decisions, including how best to man-

age adverse production risks and how to balance their efforts in farming with the labor they dedicate to 

non-farm productive pursuits or to reproductive tasks within the household (Ellis 1993). 

The hypothesis of Schultz that smallholder farmers employing traditional agricultural technologies 

are “efficient but poor” within the technological, market, and contextual constraints in which they farm 

suggests several policy initiatives that governments might lead to enable farmers to utilize higher 

productivity farming technologies profitably. If these checks on production levels can be lifted, both indi-

vidual farmers and the agricultural sector of the economy as a whole can be considerably more produc-

tive. The implications of this for both broad economic growth and for poverty reduction, particularly in 

rural communities, are significant. 

Among the areas where government action and investments could result in smallholder farmers re-

alizing increased productivity are (Schultz 1981; Ellis 1993): 

• Invest in research to develop higher productivity technologies or those that enable 

farmers to avoid crop losses to pests and disease. These improved technologies must be 

adapted for local agro-ecological conditions. Technologies developed elsewhere can be 



6 

adapted for PNG conditions by NARI scientists and their partners. However, it should be ex-

pected that farmers will face significant barriers to adoption of many of these technologies 

as innovations within their existing farming systems, since the improved technologies will 

likely be more costly to adopt than the traditional technologies they currently use. But if the 

new technologies perform well under PNG conditions, parallel investments in market sys-

tems will be required so that there is a sufficient demand for the increased crop production 

the new technologies generate. If this is done and if farmers are able to learn about and 

readily access them, relatively rapid adoption of the technologies into traditional farming sys-

tems should be expected. 

• Critical to increasing productivity through adoption of improved agricultural technologies is to 

expand the incentives for farmer to increase their production. Where crop production now is 

primarily for own use by the producing households, the development of competitive agricul-

tural market systems will require investments in physical market infrastructure, reliable trans-

portation networks, good communication services, and financial services. For exportable 

commodities, government should also foster development of trade infrastructure and put in 

place the added regulatory frameworks required to expand trade flows.  

The aim of such efforts should be the development of market systems in which farmers can be cer-

tain they will find buyers who will pay them a profitable price for the crops they bring for sale, even as at 

the same time consumers, at the other end of the value chain, can be certain they will find the food and 

other products they require in the market at a price they can afford. The traders, as the intermediaries 

between the suppliers of the produce and those demanding it, similarly should be able to derive a rea-

sonable return for the necessary services they provide to enable the markets to operate effectively.  

With the objective of significantly and sustainably increasing crop productivity levels across PNG, 

an equally important facet of the development of effective agricultural markets is the develop-

ment of reliable input markets. While it is the prices offered in output markets that provide small-

holder farmers with the incentives to increase their production of marketable crops, to achieve higher 

production on the same land area or with current labor supplies will require that farmers have access to 

the modern inputs – improved planting materials, fertilizer, agro-chemicals, and any farm machinery – 

that will enable them to achieve higher productivity levels. These farm inputs are most sustainably pro-

vided through the market. 

• Invest in both general education and effective agricultural extension services. 

− Building human capital is a foundational component of human development, including for 

economic development. A well-trained (and healthy) workforce is needed so that more 

complex production processes can be used, including in agriculture; more effective ser-

vices can be offered to increase the overall output of the economy of PNG; and all work-

ers can better respond to and benefit from new economic opportunities and incentives. 

Increasing numbers of well-educated citizens expand the human capital stock of PNG 

and, thereby, its economic potential. 

− To improve their agricultural productivity with improved technologies, farmers will need 

to engage in a process of learning and experimentation over several cropping seasons. 

Access to the technologies, whether through the market or through government pro-

grams, will allow them to experiment. However, information on how to make effective 

use of the new technologies is an important additional component in efforts to foster 

farmer uptake of the new techniques so that all adopters realize significantly higher 



7 

yields. This information can be offered by government, nongovernmental organizations, 

or private firms and flow to smallholder farming households using any effective means, 

whether in person or, more remotely, using new communication technologies. 

While not all smallholder farming households in a community will be able to profitably adopt new 

technologies to raise their farm productivity or to significantly increase their participation in agricultural 

markets, those with sufficient land, labor, knowledge, and other productive assets should be able to do 

so. If we accept the hypotheses of Schultz, most smallholder farmers in PNG are as productive as they 

can be given the constraints under which they farm. But several of those constraints can be lifted by 

government and its partners to sustainably increase crop productivity levels. By improving markets and 

changing the structure of incentives that governs the agricultural production decisions of smallholders, 

by improving their access to improved farm technologies, and by improving their abilities to effectively 

use the technologies through improving their skills and decision-making abilities, much higher levels of 

crop output per unit of land or unit of labor used can be attained and sustained. 

Measuring crop productivity 

Crop yields 

Crop productivity is a measure of crop output per unit of input used in producing the crop. In this paper, 

we primarily measure crop productivity as the amount of crop harvested per unit of land farmed in a 

cropping season – that is, crop yields measured in kilograms or metric tons per hectare. This is the 

most common measure of productivity used, because the data required to compute such yields is rela-

tively well understood, even if collecting the data may not be easy. 

Challenges in computing such yields include determining what is the amount harvested, what is the 

land area under production, and what time period constitutes a cropping season. Though the definition 

seems simple, in many circumstances none of these components of the yield measure may be easy to 

determine as both the amount of crop harvested and measures of the land area farmed are prone to 

error and bias and their measurement can be time consuming. In computing yields, Fermont and Ben-

son (2011) highlight the range of different measures of the amount harvested that might be used as the 

numerator in any crop yield computation, depending on the type of analysis being conducted (Figure 1). 



8 

Figure 1. Crop yield concepts used by sociologists, agronomists, and economists and plant 

pathologists 

 
Source: Adapted from Fermont and Benson (2011). 

For this paper, the yield concepts in Figure 1 that are most relevant are those used by economists 

and plant pathologists. Of the five yield levels noted, most of the crop productivity measures reported in 

this paper are those that reflect “actual” yields. However, for most subsistence-oriented farmers in 

PNG, these also will be “traditional” yields, as most do not use modern agricultural inputs on their staple 

food crops. However, the objective to which this paper seeks to contribute is to raise crop productivity 

levels in PNG so that they are much closer to the “economic” yields that can be obtained given availa-

ble farming technologies and prices for both farm inputs and crop outputs. Ideally, as agricultural mar-

kets continue to improve in the country and access to adapted high-yielding agricultural technologies 

improves, the economic yields that farmers obtain will move closer to “attainable” yield levels, even if 

they will not attain those yields, given the cost trade-offs farmers must manage in their crop produc-

tion.2  

However, the yield concepts of sociologists that are presented in Figure 1 are also important in con-

sidering crop yields in PNG, particularly for the main staple food crops. Most of these are roots or tu-

bers that have a high moisture content when harvested. In most parts of the country, due to limited dry 

periods, once harvested, it is difficult to dry or otherwise preserve these crops for long-term storage. 

Consequently, postharvest losses can be high for the dominant staple crops in PNG so that the differ-

ence between the “harvested” and the “economic” yields, as sociologists use the terms, will be signifi-

cant. 

Among the challenges in measuring the land area dedicated to a crop to compute yields are: 

 

2 “Potential” and “attainable” yields will be of most interest to agricultural researchers as they work to develop crop varieties and 

other technologies that best control both the biotic and abiotic stresses on a crop that reduce output. However, farmers are 

unlikely to be able to profitably attain such experimental yield levels. Given diminishing returns in outputs to increased use of 

inputs, trade-offs among production objectives, and variable input to output price ratios, highest financial returns are obtained 

at lower productivity levels – the economic yield level. 

 

Sociologists Agronomists Economists and plant pathologists 

   

 

Biological 

or gross 

yield 

Harvested 

yield 

Economic 

yield 

Harvest 

losses 

Post-harvest 

losses 

Potential or 

theoretical 

yield 

Attainable 

yield 

Actual or 

farmer yield 

Water and 

nutrient 

limitations 

Losses due to 

weeds, pests, 

diseases, and 

pollutants 

Potential  

yield 

Attainable 

yield 

Traditional 

yield 

Losses due to imperfect 

management 

Actual 

yield 

Economic 

yield 

Losses due to pests 

and disease 

Losses due to biotic 

and abiotic constraints 

Losses due to not using 

improved agricultural 

inputs 



9 

• How a measurement of the land area is determined. Various technical approaches can be 

used to compute plot areas ranging from traditional chain and compass land survey meth-

ods to the use of Global Positioning System (GPS) devices in the field to remote analysis of 

aerial photographs or, increasingly, satellite imagery. Field methods are expensive given the 

trained staff required, while the costs of obtaining recent remote imagery and accurately an-

alyzing it pose a high barrier. A common alternative is to rely on farmer estimates of the size 

of their plots, because obtaining such estimates can be done as part of a survey of farm 

households and does not necessarily involve data collectors going to the actual farm plot, 

reducing costs. However, smallholder farmers often do not have a good understanding of 

the size of standard land area units, such as hectares or acres, and most have never had 

their land measured by anyone with such expertise. Recent rigorous analyses demonstrate 

that such farmer-reported estimates are not accurate and often are biased (e.g., Carletto, 

Gourlay, and Winters 2015). Making use of estimates from farmers of their cropped area to 

compute yields to guide policy and to design programs is not recommended. Nevertheless, 

continued use of such self-reports of land area remains common in computing crop yields, 

likely including for many of the crop yield values reported in this paper. 

• Heterogeneous crop performance is high in smallholder farming and can be seen at plot 

level, within a community, and between regions. Average yield levels computed from small 

samples of farmers are likely to not be as representative of yields for all farmer as we would 

wish. 

• Intercropping or mixed species cropping is a technique smallholder farmers use to spread 

risks, diversify their production, and increase total output of individual fields, including in 

PNG (Bourke and Allen 2009, 107). The final impact of intercropping on individual crop 

yields is a result of complex interactions of many factors that include relative time of plant-

ing, plant density, rainfall, soil fertility, and the optimal timing among the intercrops of the 

farmer’s crop management activities, among others. Estimating yields of individual crops on 

a farm plot into which two or more crops are grown at the same time presents difficulties for 

any crop yield estimation method, with significant overestimates or underestimates common 

for the individual crops in intercropped plots (Fermont and Benson 2011). 

The final important challenge in computing crop yields, particularly relevant to PNG, is the definition 

of a cropping season or other time period over which crop production is computed. In farming systems 

which show significant seasonality in cropping patterns every year, this problem typically does not arise 

since annual crops dominate in those farming systems. However, in PNG, due to the relatively even 

distribution of rainfall in large parts of the country, farmers may plant and harvest crops throughout the 

year. Many important crops are perennials or require more than a year of growth before harvest, includ-

ing cassava and banana. Other crops have extended harvest periods, including cassava, sweet potato, 

banana, and coffee (Fermont and Benson 2011; Kilic et al. 2021). This uncertainty about the period of 

time over which production takes place complicates the estimation of both total production and produc-

tivity levels for such crops. 

Other measures of productivity 

Crop yield – the physical amount of crop harvested per unit of land farmed in a cropping season – is not 

necessarily the most useful measure of crop productivity to use to assess the economic performance of 

the agriculture sector at farm level. For example, in assessing productivity at the entire farm or broader 



10 

levels of aggregation in which multiple crops are involved, land-based measures of productivity will gen-

erally use the market value of output produced, rather than the physical quantity, enabling the inclusion 

in the productivity measure information from all crops that a farmer might produce, even those for which 

the harvests have very different physical characteristics. However, in this paper, our interest is crop-

specific, so we will use physical quantity-based measures of yield. 

Yields are not the only measure of crop productivity, however, because land is only one factor in 

agricultural production. Labor and capital are also critical factors.3 Productivity measures can be com-

puted over a specific time period for these other factors using either the physical amount or the value of 

output from a plot, farm, or broader agricultural area as the numerator and for the denominator a meas-

ure of the input of labor (person-days; total labor costs) or some partial or complete measure of capital 

employed (e.g., value of purchased inputs, value or number of farm machines or other tools, value of 

capital improvements to plot or farm, etc.). However, the challenges to collecting complete and reason-

ably accurate data on the use of labor or the value of capital invested in production in smallholder farm-

ing systems are daunting. Moreover, in smallholder farming systems, most of the labor is supplied by 

members of the farming household, which is unpaid, often difficult to measure given its heterogeneity 

(children, adults, or elderly; male or female; trained or untrained), and usually ignored by the household 

members in estimating production costs for the crops they produce. Consequently, measures of 

productivity based on these other factors are not commonly produced by statisticians for monitoring 

crop productivity trends or farm and farming system performance. Researchers are more likely to calcu-

late them for analytical purposes. 

These factor-specific productivity measures are partial productivity measures. A measure of total 

factor productivity can also be computed. Such a measure for a farm, farming system, or agricultural 

sector can be computed by calculating the market value of all farm outputs for the unit of interest as the 

numerator and the market value of all factors – land, labor, and capital – used to produce that output as 

the denominator. While we will not consider total factor productivity in this paper, for studies of eco-

nomic growth, both broad and agriculture-specific, and for studies of economic structural transfor-

mation, changes in the measure over time or over space provides important insights into these eco-

nomic development processes (e.g., Benin 2016; McMillan, Rodrik, and Sepúlveda 2017). 

CURRENT CROP YIELD LEVELS IN PAPUA NEW GUINEA 

The overall aim of this paper is to consider how higher agricultural productivity levels for many of the 

crops produced in PNG could be increased to foster significant growth in the agricultural sector of the 

economy and to improve the welfare of the many households across the country that rely at least in 

part on farming to meet their basic needs. We focus on a set of six basic staple food crops – sweet po-

tato, banana, taro (Colocasia esculenta and related species), Chinese taro (Xanthosoma sagittifolium), 

cassava, and yam (Dioscorea species) – and the principal cash crops produced in PNG – cocoa, cof-

fee, palm oil, and coconut / copra. We also discuss the productivity of sago obtained from naturally oc-

 

3 Land, labor, and capital are the classic factors of production in the economics literature. But over recent decades entrepre-

neurship has increasingly been seen as an additional equally important factor (e.g., Naudé 2008). However, standard 

measures of entrepreneurship in smallholder agriculture remain to be developed. In consequence, we are unaware of any 

productivity measures for developing country agricultural systems based on entrepreneurship inputs. 

 



11 

curring perennial stands of sago palm. Sago is more a managed source of staple food, rather than culti-

vated. The most recent statistics on agricultural production from across the country for these eleven 

crops are summarized in Table 1 to provide a situation analysis on crop productivity.4 More detailed in-

formation by crop is presented in Appendix 1, including maps of where the crops are significant compo-

nents of local farming and food systems across PNG. 

Yield gaps – the gap between the yields farmers commonly obtain and what they might obtain with 

optimal crop management and the use of high-yielding crop varieties –could not be estimated for the 

main cash crops, except coffee, due to no data on attainable yields for those cash crops in PNG being 

found. However, for the main staple food crops, yield gaps could be roughly estimated from the statis-

tics presented in Table 1. Yield gaps vary by crop. The yield gap for sweet potato is the smallest – ac-

tual farmer yields are between one-quarter to one-third less than the attainable yields for the crop. Yield 

gaps for taro, Chinese taro, cassava, and yam are similar with actual yields obtained by farmers being 

between 40 and 60 percent of what they might attain with improved planting materials and excellent 

crop management practices. The data on actual versus attainable yields for banana shows the largest 

yield gap – PNG banana growers are obtaining only about a quarter of the productivity levels they might 

achieve with more intensive cultivation of the crop.  

The only cash crop for which yield gaps could be computed from the available productivity data for 

PNG is for coffee. Coffee producers are generally harvesting about half of what they might harvest if 

they used improved coffee production practices. 

Given the challenges that farmers in PNG, particularly smallholder farming households, face in 

seeking to obtaining good information on how best to produce these crops under local conditions and to 

obtain improved inputs to do so, that there are significant yield gaps for these major crops is not sur-

prising. Some of the information constraints that farmers face can be addressed through adaptive agri-

cultural research coupled with sound systems for supplying farmers with advice on how best to manage 

their crops to maximize the returns that they obtain from their crop production. 

However, equally as important is strengthening agricultural markets across the country so that 

these markets both supply farmers with improved agricultural inputs at prices which allow them to earn 

a reasonably financial return on the use of those inputs and to provide sufficient price incentives to 

farmers to grow specific crops at higher levels of productivity. Without strong markets, crop productivity 

will remain low because, as was highlighted in the previous section of this paper, the only production 

technologies that farmers can profitably use given both the agro-ecological and market conditions un-

der which they currently farm generate low yields. With access to improved technologies and crop pro-

duction techniques and appropriate market incentives in place, PNG farmers can achieve yield levels 

higher than what they now realize and much closer to the ‘attainable’ yields noted in Table 1. 

 

4 Most of the crop data used is also over a decade old. While it is still useful for the purpose of the analysis here, readers 

should recognize that current production and productivity levels for the crops examined and where they are grown in PNG may 

differ somewhat from what is presented. Stronger data systems and greater applied analytical capacity within the agricultural 

sector in PNG are areas of rising urgency for identifying realistic and achievable strategies for increasing the productivity of 

PNG’s farmers and developing the agricultural sector. 



12 

Table 1: Productivity levels of major staple food and cash crops in Papua New Guinea, actual 

and attainable on farm with optimal management, most recent estimates, mt/ha 

Crop 
Elevation or other production 
conditions Actual Attainable 

Staple food crops    

Sweet potato lowlands (<1,200 masl): 13 1 15 to 20 2 

 highlands (1,200 to 1,800 masl): 15 1 20 to 30 2 

Banana lowlands: 12 1 -- 

 highlands: 9 1 -- 

 on-farm with good husbandry: -- 38 to 50 3 

Taro lowlands: 6 to 8 1 -- 

 highlands: 10 1 -- 

 on-farm with good husbandry: -- 15 1 

Chinese taro lowlands: 14 1 -- 

 highlands: 11 1 -- 

 on-farm with good husbandry: -- 25 1 

Cassava lowlands: 22 1 -- 

 highlands: 16 1 -- 

 on-farm with good husbandry: -- 45 1 

Yam lowlands: lesser yam: 15 1 
greater yam: 13 1 

-- 

 highlands: greater yam: 11 1 -- 

 on-farm with good husbandry: -- lesser yam: 40 1 
greater yam: 30 1 

Sago average, per trunk: 180 kg 1 500 kg 1 

Major cash crops    

Cocoa smallholder:  0.2-0.4 4 -- 

 plantation: 0.4-0.6 4 -- 

Coffee    

Arabica smallholder:  0.75-1.25 4 -- 

 plantation: 1.00-2.40 4 -- 

 on-farm with good husbandry: -- 1.25-2.50 3 

Robusta average: 0.90 4 -- 

 on-farm with good husbandry: -- 1.90 3 

Oil palm 
(fresh fruit bunches) 

smallholder:  6.0 to 14.5 4 -- 

plantation: 19.0 to 21.5 4 -- 

Coconut 
(in form of copra) 

smallholder: 0.40-0.70 4 -- 

plantation: 0.70-1.00 4 -- 

Sources: 1: Bourke et al. (2009); 2: Allen and Bourke (2009); 3 Acland (1971); 4 Allen, Bourke, and McGregor (2009). 

Note: -- = “not available”; masl = “Meters above sea level” 

Yam – Lesser yam = Dioscorea esculenta; ‘mami’ or ‘taitu’. Greater yam = Dioscorea alata; ‘yam tru’. 

However, we should not minimize the challenges involved in improving the markets for the main 

staple foods produced in PNG. The principal staple crops of PNG are roots and tubers. While the 

crops are well adapted for the agro-ecological conditions in which they are produced and often have 

long harvest periods that contribute to a consistent supply of food for the producing household and for 

neighboring households and communities, these staples are difficult to store after harvest for any 

extended periods of time. Their high perishability also makes them difficult to transport and market 

without incurring significant losses to wastage. The principal exception to this rule is yam, which can be 



13 

stored for quite long periods without spoilage or other losses. The other staple crops require extensive 

processing and drying or cold storage to avoid significant losses after harvest. 

It is common in many subsistence-oriented economies globally to dry staple crops in the open air 

using solar radiation during the regular dry seasons. However, few areas of PNG have dry seasons 

of sufficient length with sufficient sunshine to adequately dry the harvested roots and tubers. 

These staples could be dried using heat, primarily using firewood as the source. However, the adverse 

environmental ramifications of doing so are considerable. With significantly expanded rural electrifica-

tion, both drying and cold storage of these staple crops become possible. However, whether doing so 

will be cost-effective is unlikely. 

In consequence, these staple crops are primarily marketed in unprocessed forms either locally or, 

where there is reasonably good transportation infrastructure to connect producing areas to PNG’s main 

cities, in the wider domestic market. The tradability of PNG’s principal food crops poses a challenge for 

intensive commercialization to raise producer incomes. It is difficult to see where demand-side incen-

tives for PNG’s main staple crops will come from outside of local areas, given that they are difficult to 

transport if not processed into more storable forms. There are likely only to be niche export markets, 

at most, for PNG’s principal staple foods. Building increased incentives for commercial production of 

the traditional staple crops grown in PNG is a significant agricultural development challenge. 

STAPLE AND CASH CROP YIELDS IN INDONESIA – A 
COMPARATIVE ANALYSIS 

The broader aim of this paper is to identify what sort of policy reforms and public investments could be 

made, particularly at national level, to provide improved incentives and better access to the technolo-

gies needed to achieve significantly higher crop yields by all farmers across PNG. Virtually all staple 

crop production and much of the cash crop production in PNG is carried out using traditional production 

methods and technologies. This in part reflects the relatively limited development of PNG’s agricultural 

sector.  

Indonesia is a middle-income country with a growing economy that is the largest in Southeast Asia 

and is among the 20 largest globally. Neighboring PNG to the west and sharing half of the island of 

New Guinea, the main islands in the eastern half of the Indonesian archipelago have the same agro-

ecological conditions as are found in PNG. From a biophysical perspective, agricultural production pos-

sibilities in PNG should be relatively similar to those in eastern Indonesia. 

However, from an economic perspective, much of the agricultural production in Indonesia is done 

within the context of a strongly commercially-oriented farming sector. While smallholder agricultural 

systems remain a significant part of farming systems in Indonesia, in contrast to PNG, the in-

centives many Indonesian smallholders respond to in choosing what and how much to produce 

are more driven by considerations of the market-based income they might derive from their 

farming than by the subsistence consumption needs of their household. Smallholders in Indone-

sia also are able to better access modern farm inputs, primarily through the market, and information on 

how best to employ those inputs to raise their crop productivity levels than is the case for farm house-

holds in PNG. Indonesian smallholder farmers have access to larger and deeper markets than those 

into which smallholders in PNG might sell their crops. Theoretically, based on the stronger agricultural 

market systems in which they are emmeshed, we would expect that smallholders in Indonesia attain on 

average higher crop yields than do smallholders in PNG. 



14 

To determine whether the crop productivity patterns in Indonesia relative to PNG are higher due po-

tentially to the economically more developed and greater commercial orientation of agriculture in Indo-

nesia, we compare in Table 2 average productivity levels reported in Indonesia and PNG for the food 

and cash crops considered in this paper for which data is available. Where possible in the data, we fo-

cus on productivity levels in the eastern provinces of Indonesia that agro-ecologically are most compa-

rable to PNG – the islands of Kalimantan and Sulawesi, the Maluku Islands, and the provinces of Indo-

nesia (Papua and West Papua) on New Guinea itself (Figure 2). 

Figure 2: Provinces of Indonesia 

 

Source: https://commons.wikimedia.org/wiki/File:Indonesia_provinces_english.png. Copyright remains with the author(s). This map is licensed 

for use under a Creative Commons Attribution 4.0 International License (CC BY 4.0). To view this license, visit https://creativecommons.org/li-

censes/by/4.0. 

The crop yields reported in Table 2 for PNG are the actual farmer yield estimates presented in Ta-

ble 1, which are discussed in more detail in Appendix 1. For banana, cassava, cocoa, and coffee yields 

in Indonesia, as these crops have sufficiently detailed data, we further disaggregate productivity levels 

across the 15 provinces of the eastern part of the country for the period 2015 to 2017. We also further 

disaggregate productivity estimates for these crops by three groups of provinces in eastern Indonesia– 

average yields for the two Indonesian provinces in western New Guinea, which likely best reflect the 

conditions under which crops are produced in PNG just to the east of those two provinces; average 

yields for the six provinces on Sulawesi island, which has significant areas of highland; and those for 

the five provinces on Kalimantan and the two provinces in the Maluku Islands, which generally have 

lowland and coastal agro-ecologies. For oil palm and coconut (copra), sub-national production data for 

Indonesia are not readily available. For these crops, only average national yields are presented. No re-

cent national or sub-national productivity estimates for taro, Chinese taro, and yam in Indonesia were 

found, despite extensive online searching. 

https://commons.wikimedia.org/wiki/File:Indonesia_provinces_english.png
https://creativecommons.org/licenses/by/4.0
https://creativecommons.org/licenses/by/4.0


15 

Table 2: Productivity levels of selected food and cash crops in Indonesia and Papua New 

Guinea, most recent estimates, mt/ha 

Crop 
Papua New 

Guinea Indonesia 
Eastern 

Indonesia 

Papua 
(Indonesia New 
Guinea – mainly 

lowland 
production) 

Kalimantan 
and Maluku 

(lowland 
production) 

Sulawesi 
(some highland 

production) 

Sweet potato lowlands: 13 
highlands: 15 

17.1 13.2 13.1 13.9 13.0 

Cassava lowlands: 22 
highlands: 16 

24.0 21.8 12.1 22.5 22.1 

Banana lowlands: 12 
highlands: 9 

55.4 42.1 38.2 36.5 50.1 

Cocoa smallholder: 
0.2-0.4 plan-
tation: 0.4-

0.6 

0.36 0.37 0.24 0.28 0.39 

Coffee,  
Arabica 

smallholder: 
0.75-1.25 
plantation: 
1.00-2.40 

0.55 0.38 limited produc-
tion 

no production 0.42 

Copra (coconut)       

smallholders 0.40-0.70 1.1 na na na na 
plantations 0.70-1.00 2.8 na na na na 

Oil palm       

smallholders 6.0-14.5 15.3 
potential: 44.6 

na na na na 

plantations 19.0-21.5 19.5 
potential: 48.8 

na na na na 

Source: PNG yields – Bourke et al. (2009); Indonesia banana, cassava, cocoa, and coffee – KPRI (2021); Indonesia oil palm – GYGA (2021); 

Indonesia copra – Alouw & Wulandari (2020). 

No recent national or subnational yield estimates for Indonesia for taro (Colocasia spp.), Chinese taro (Xanthosoma spp.), and yam (Di-

oscorea spp.) could be found through standard online searching. Sago productivity estimates were not sought. 

Note: na = “not available”. 

Our comparison of crop yields from Indonesia with those from PNG do not provide consistent re-

sults across the crops examined. Sweet potato and cassava show comparable productivity levels in In-

donesia with those recorded for PNG, particularly for the eastern provinces of Indonesia. It is unclear 

the degree to which production of these two staples is commercialized in Indonesia. However, given the 

technical challenges in processing sweet potato and cassava for long-term storage, likely the crops in 

Indonesia are principally sold in fresh form, similar to how they are marketed in PNG, or grown for own 

consumption by the producing household. As a consequence of their restricted commercial poten-

tial, as is the case for these crops in PNG, low levels of investments in crop breeding or in crop 

value chains are likely seen for sweet potato and cassava in Indonesia. Given this neglect in both 



16 

countries in developing varieties of sweet potato and cassava with higher productivity potential or in de-

veloping deeper marketing systems that offer farmers good incentives to raise their levels of productiv-

ity of the two food crops, we find similar productivity levels for the crops in the two countries.5 

In contrast, for banana, the third food crop listed in Table 2, productivity levels are far higher in In-

donesia than in PNG. While Indonesia is among the top ten banana producing countries globally, little 

of the production is exported. With over half of all Indonesians living in urban areas – over 150 million 

urban dwellers, there is a large domestic market for banana. The high productivity levels for banana in 

Indonesia almost certainly reflect research investments to develop banana varieties with higher produc-

tivity potential and to identify optimal agronomic practices for banana production. These investments 

make sense in the context of significant and growing demand for banana in Indonesia’s cities. As a re-

sult, much of the banana produced in Indonesia is under intensive, specialize farming systems, in con-

trast to PNG where much of the banana is produced out of plots in which banana is one of several in-

tercrops. The low productivity levels for banana in PNG reflect the low commercial potential for the 

crop. The country does not have the large domestic market for banana seen in Indonesia, with less 

than one in five of PNG’s citizens living in urban areas. In consequence, most banana production in 

PNG remains for own consumption or for local trade. Efforts to raise banana production and productiv-

ity levels in PNG would likely require close attention to developing export markets for PNG’s bananas. 

For the cash crops examined, we also find that productivity levels in Indonesia are not consistently 

higher than in PNG. Cocoa yields estimated for Indonesia are at the lower end of yield ranges for 

the crop in PNG. Indonesia’s cocoa sector, which is centered on the island of Sulawesi and is domi-

nated by smallholder production, is at least ten times larger in terms of total annual production than that 

of PNG – Indonesia is the third largest producer of cocoa globally. Given the size of the sector, we 

might assume that there is more effective outreach to producers with improved planting materials and 

advice on optimal cocoa production techniques in Indonesia than in PNG. However, productivity levels 

shown in Table 2 do not support such an assumption. Analyses of constraints facing the cocoa sec-

tor in Indonesia highlight many of the same challenges facing cocoa producers in PNG – sub-

optimal crop management, old trees that should be replaced, limited access to credit for cocoa 

farmers, and volatile prices for both cocoa beans and inputs, among others (Fahmid et al. 2018). 

A similar pattern is seen with Arabica coffee. Like cocoa, coffee also is a smallholder crop in Indo-

nesia – 90 percent of production comes from smallholders. Indonesia is the fourth largest coffee 

producer globally. However, despite the size of the sector, the yield levels obtained by Indone-

sian coffee farmers are not necessarily any higher than those of coffee producers in PNG. The 

constraints noted above as being faced by smallholder cocoa producers in both Indonesia and PNG are 

also many of the same constraints that coffee producers in both countries face. Indonesia has not 

solved the challenge posed by low coffee yield levels constraining growth of the sector any 

more than PNG has. 

 

5 Although we have no comparative data on taro, Chinese taro, and yam, we have found no evidence to indicate that yield lev-

els for the three would differ between PNG and Indonesia. Moreover, in explaining why we see on-farm yield levels in PNG for 

the three crop being considerably lower than what is potentially attainable (Figure 5, Figure 6, and Figure 8), the arguments 

for why yields of sweet potato and cassava in Indonesia are low related to restricted market potential also apply to these other 

three staple crops. Moreover, that it is difficult to obtain information on their production or productivity levels in either PNG or 

Indonesia is in part due to the crops being viewed as purely for subsistence, with any commercial potential they have being 

primarily restricted to local markets and not for wider domestic markets or for international trade. Agricultural researchers simi-

larly give them much less attention than more commercial crops, further exacerbating data gaps on the productivity of these 

crops. 



17 

However, productivity estimates in Table 2 show that this pattern is not replicated for the other two 

cash crops examined, coconut and oil palm. Both show higher productivity in Indonesia. The Food and 

Agriculture Organization of the United Nations (FAO) reports that Indonesia is the leading producer of 

coconut globally. While PNG is among the top ten global producers, FAO coconut production es-

timates for PNG for 2019 are less than 10 percent those of Indonesia (FAO 2021). While both Indo-

nesia and PNG are among the largest exporters of copra globally, demand also is high in the Indone-

sian domestic market for a wide range of coconut-based food products (Alouw & Wulandari 2020). A far 

larger share of the coconut produced in Indonesia is consumed either directly or after commercial pro-

cessing than is the case in PNG. The size of the domestic market in Indonesia supports considerable 

specialized processing of the crop into a range of food products. Because of this greater commercial 

potential, coconut producers in Indonesia, most of whom are smallholders, demand more productive 

varieties of coconut and the information that will allow them to produce the crop most profitably. The 

higher productivity levels reported for copra in Indonesia reflect, at least in part, effective re-

sponse to these demands of Indonesian coconut farmers on the part of agricultural researchers 

and extension service providers, whether public agencies or private firms within the coconut 

sector. 

The Indonesia case shows that PNG coconut producers can be considerably more productive. 

However, to replicate the outcomes of patterns of investment in Indonesia in the coconut sector in PNG 

will be much more challenging, simply because the domestic market for coconut products in PNG is 

much smaller and unlikely to provide the returns on public and private investments to improve produc-

tivity that is apparent in Indonesia. However, that coconut in PNG (as in Indonesia) is a smallholder 

crop is an important consideration in prioritizing which crops should see higher investments to 

improve their productivity. The direct welfare and poverty reduction benefits for coconut producing 

households of targeted investments to raise coconut productivity levels would certainly be much higher 

than they would be for investments in crops that are less commonly produced by smallholders. 

Oil palm is much less widely grown by smallholders. Table 2 shows that oil palm producers in PNG, 

both smallholders and plantations, are keeping pace with the productivity levels of Indonesian produc-

ers. This should not be so surprising given that oil palm plantations in both countries are vertically inte-

grated and often part of multinational corporations. These features facilitate any innovations in produc-

tion of oil palm to be quite quickly disseminated and scaled up. Moreover, smallholder outgrowers work 

closely with the plantations, so have good access to improved planting materials, other inputs, and ad-

vice. However, we see in Table 2 that there is considerable scope for further productivity improvement 

for oil palm. Based on crop modeling which assumes optimal management of improved planting materi-

als that are perfectly adapted for local growing conditions and have the highest production potential, 

researchers estimate that productivity levels for oil palm in Indonesia could more than double current 

levels. While those levels are unlikely to be realized by most farmers, the estimates demonstrate that 

there is room for more efficient and more profitable production of oil palm in Indonesia. There are no 

biophysical barriers to similar productivity yields being realized by oil palm producers in PNG. 

Consequently, we can expect that there is considerable potential for growth in output from the oil 

palm sector in PNG simply through greater oil palm productivity. Raising productivity in the sector will 

increase the amount of physical product produced and, as most oil palm in PNG is exported, expand 

agricultural export revenues, and increase both agricultural and the overall GDP of PNG’s economy. 

Indirectly, this economic growth should provide significant benefits to the welfare of poor households 

across PNG. However, in determining whether public investments, in particular, should go to increasing 

oil palm productivity, the structure of the oil palm sector in PNG should factor in such decisions. The 



18 

organization of oil palm production in PNG, like petroleum and mineral production, is in enclaves with 

rather limited linkages to the broader economy in the provinces in which those oil palm plantations and 

their associated smallholder outgrower schemes are located. In consequence, the direct benefits of 

increasing productivity of oil palm are likely to be significantly less than would be equivalent 

improvements in the relative productivity levels of other crops, particularly staple food crops, 

but also cocoa, coffee, and coconut, since smallholders mainly produce them. This has implica-

tions for the poverty reduction objectives that guide government’s decision making as to how public re-

sources should be invested. More so than for the other crops considered, if public investments are 

made to enhance oil palm productivity, policy makers will need to identify how the benefits from those 

investments will flow through the economy to result in improved welfare for PNG households who are 

not involved in oil palm production. 

This comparative case study of food and cash crop productivity levels in PNG and Indonesia, alt-

hough somewhat constrained by limits on the data available, demonstrates some of the enabling condi-

tions that need to be in place to raise crop productivity levels in a country. Demand for the crop is the 

most important driver of sustainably increasing crop productivity. In this, the differences be-

tween Indonesia and PNG are stark – Indonesia has a domestic market of over 275 million per-

sons, while that of PNG is under 9 million.6 A general pattern is apparent of Indonesian farmer’s re-

sponding to higher (particularly domestic) market demand with greater production achieved in part 

through higher productivity. In both countries, however, we find that several of the main staple crops 

have not been commercialized and have only quite local value chains or only supply the crop to rela-

tively thin domestic markets in perishable form. Consequently, there is no indication that sweet potato 

or cassava producers in Indonesia have any greater incentive to improve their productivity than do pro-

ducers in PNG. Likely the same conditions constrain improving productivity for taro, Chinese taro, and 

yam producers, as well, as these staple foods also are characterized by limited market demand. Both 

banana and coconut are good examples of crops with good markets in Indonesia, but relatively 

small ones in PNG, resulting in sharply contrasting productivity levels in the two countries. 

While growing market demand is the principal driver, investing in agricultural research and putting in 

place effective agricultural extension services also make important contributions to increased productiv-

ity. It is evident that for both banana and coconut in Indonesia and for oil palm in both countries, doing 

so has resulted in quite high average productivity levels. The somewhat stagnant cocoa and coffee sec-

tors in both Indonesia and PNG suggest that more effective research and extension efforts on those 

crops may enable producers to increase their productivity and realize higher incomes. 

Policy reforms are also important to raising productivity of specific crops. That productivity levels of 

the traditional roots and tubers staple crops in both countries remains much lower than what might be 

realized with better planting materials and crop management and effective market demand for greater 

production quantities in part reflects policy decisions that prioritized investments in other crops. How-

ever, that the traditional staples have generally been neglected in PNG, as well as Indonesia, 

works against development visions in both countries to realize sharp reductions in poverty. In-

creased productivity of these staples with parallel development of markets and diversification in the 

value chains for each will stabilize production and prices. The benefits of reliable production levels and 

 

6 A possibly more useful detailed comparison would be between the two provinces of Indonesia on New Guinea, Papua and 

West Papua, which have a total population of about 5.5 million people, and PNG. However, doing so would require overcom-

ing additional gaps in the data to what was faced with the comparison with Indonesia as a whole. 

 



19 

stable and, ideally, lower prices for staples will result in improved consumption levels for poor house-

holds, for which staples are the major component of their consumption basket. If the development vi-

sion for PNG, as well as for Indonesia, goes beyond economic development alone to consider higher 

level development objectives like poverty reduction, increased attention to the productivity of staple 

crops should receive greater policy attention. 

Our discussion now considers the strengths and gaps in the policies and strategies of PNG that 

have a bearing on raising crop productivity levels, both within and outside of the agricultural sector. 

CRITICAL REVIEW OF POLICIES AND STRATEGIES OF 
PAPUA NEW GUINEA WITH A BEARING ON CROP 
PRODUCTIVITY 

The policies and strategies of the government of PNG that might contribute to raising agricultural 

productivity levels were reviewed based on the discussion in section 2 on why improvements in agricul-

tural productivity levels are important for economic development and how such improvements might be 

achieved. The policy architecture around improved agricultural productivity as a development objective 

in PNG is anchored to the development vision for the country and to both 20-year and 5-year develop-

ment plans. These three documents are examined from the perspective of what priority they give or, if 

updated, might potentially give to improving productivity levels in the agricultural sector as part of the 

pathways to achieving the development vision they lay out for PNG. Thereafter, we turn to examine ag-

ricultural sector policies and strategies, before concluding our discussion by examining non-agricultural 

policies and strategies that could play a role in enabling farmers across PNG to adopt higher productiv-

ity methods in their farming by providing them with both the knowledge and the incentives they will re-

quire to do so. 

More detailed synopses on some of the policies or strategies reviewed, particularly the major ones, 

are presented in Appendix 2 to this paper. In those synopses is additional discussion to what is pre-

sented in this section on how the content of each might contribute to improving agricultural productivity 

levels in PNG. 

Master development frameworks – what role for improved agricultural 
productivity 

Three documents make up the current master development framework for PNG – the Papua New 

Guinea Vision 2050 (NPC 2009), which lays out the development goals for the country to 2050, the 20-

year Papua New Guinea Development Strategic Plan 2010-2030 (PNGDSP) (DNPM 2010); and the 

current 5-year Medium Term Development Plan III (MTDP III) (DNPM 2018). The two plans are at the 

apex of development planning within government, providing a cross-sectoral assessment of what gov-

ernment needs to invest in, undertake, or facilitate to achieve the development vision laid out in the Vi-

sion 2050 document.  

Published in 2009, the vision seeks that by 2050 PNG be a “Smart, Wise, Fair, Healthy and Happy 

Society”. From a strategic standpoint, the key economic development challenge identified in Vision 

2050 is to move the economy from being dominated by the mining and energy sectors, which 

typically operate in enclaves with somewhat limited employment opportunities, to an economy 

in which there has been a significant expansion in the economic contributions of more broad-



20 

based sectors including agriculture, as well as forestry, fisheries, tourism, and manufacturing. Vision 

2050 is supported by seven development pillars, two of which are particularly relevant to raising agricul-

tural productivity: (i) human capital development; and (ii) wealth creation through building a strong, dy-

namic, competitive, and productive economy. However, no detail is provided in the document on the 

role that a growing agricultural sector might play in achieving this development vision or on how to fos-

ter increased growth in the sector through improved productivity. The vision document states that how 

this is to be done will be detailed in forthcoming sectoral policies and strategies, particularly a National 

Agricultural Development Plan. This agricultural sector plan is discussed in the next sub-section of this 

paper. 

More detailed all-of-government guidance for achieving the goals set in the 2050 Vision was formu-

lated in the PNGDSP and in a sequence of 5-year Medium Term Development Plans, of which 

MTDP III is the current one. PNGDSP provides cross-sectoral integrated development planning and in-

cludes more detail on agricultural sector priorities than do either the Vision 2050 document or the 

MTDP III. A set of 20-year production targets for cash crops relative to current levels are estab-

lished in PNGDSP. Staple food production is to be increased to fully meet demand from domestic mar-

kets and downstream industries. These increases in agricultural production are to be achieved 

through both increasing the productivity of PNG’s farmers by 60 percent and by expanding the 

area of land farmed by 180 percent. An overriding aim of PNGDSP is to convert most subsistence 

producing farm households into small and medium agricultural enterprises. The plan also highlights the 

need to establish the research basis for higher productivity across PNG’s economy, including agricul-

ture. Equally important, the PNGDSP devotes considerable attention to rural development efforts criti-

cal to providing farmers with the skills and incentives they require to raise their productivity and, 

thereby, the income and consumption levels of their households. 

The MTDP III has less content on agricultural development than does the PNGDSP. However, the 

principal course correction for the economic development of PNG laid out in the five-year plan is an in-

creased focus on investing in renewable sectors of the economy, particularly agriculture, and away 

from petroleum and mineral exploitation. Three of the eight Key Result Areas of MTDP III, if significant 

progress is achieved under them, will either directly contribute to improved agricultural productivity or 

provide an enabling context, the skills, or the incentives to farmers to increase their productivity – (i) in-

clusive economic growth; (ii) continuing infrastructure development; and (iii) higher quality education 

and skills development. 

However, the financing plan of the MTDP III has no detail on public support to the agricultural sec-

tor. Consequently, it is not clear whether we should expect improvement in agricultural productivity dur-

ing the five-year implementation period of MTDP III. This general critique of a lack of specificity in any 

strategic approaches to agricultural development in PNG can be applied to the other master develop-

ment framework documents, too. Although the PNGDSP is the strongest of the three in this regard, 

even in that plan, what defined steps will be taken to transform agriculture in PNG to contribute to im-

proved welfare and lower poverty are not provided. 

Priority given to raising crop productivity in current agricultural sector poli-
cies and strategies 

We now critically examine agricultural sector policies and strategies that are relevant to increasing agri-

cultural productivity. The National Agriculture Development Plan 2007-2016 (NADP), although now 

dated, remains the overarching statement of the priorities and strategies of the government of PNG for 



21 

the development of the agricultural sector to achieve Vision 2050 (DAL 2007). Its objectives are rural 

economic development, food security, and improved livelihoods for all. In addition to a 200-page narra-

tive document, a 300-page implementation framework for NADP was developed. There are eight prior-

ity areas under the NADP. Three are pertinent to improving farmer productivity – (i) Food and horticul-

tural crops development, which concerns the production, processing, and marketing of traditional sta-

ples, fresh vegetables, and domestically-produced wheat and rice, with a focus on fostering commercial 

food production; (ii) Tree and industrial crops development, with a strong focus on reversing the declin-

ing productivity seen in recent years in all cash crops, except oil palm; and (iii) Agriculture research and 

extension, particularly through substantially improving the access of farm households to relevant agri-

cultural extension services. However, the implementation framework budget privileges cash crops, with 

the industrial crops priority area receiving just over half of the proposed ten-year budget, while food 

crops receive 12 percent and agricultural research and extension activities receive 22 percent. These 

relative budget allocations call into question how broadly the benefits of NADP implementation might 

have been spread – cash crop producing farm households are likely to have received the most direct 

benefits from implementation of NADP. 

The Agriculture Medium Term Development Plan 2020-2022 (AMTDP) is the current plan for the 

agriculture sector (DAL 2020). It aligns with the MTDP III, setting a target for the sector of increasing 

overall production by 30 percent. Organized around eleven sectoral growth drivers, several of these are 

pertinent to improving productivity at farm level. The discussion on the driver on food security, quality, 

and nutrition highlights the importance of increased domestic food production, as well as downstream 

processing and preservation, while the driver on quality provision of agricultural services is centered on 

how those services will increase agricultural productivity. Less directly, the drivers on agricultural trade 

and investment, on agricultural marketing, and on downstream processing are critical to providing en-

hanced incentives to farmers to raise their agricultural production, in part through greater productivity. 

The agricultural land mobilization driver is to bring into production land that can be farmed in a sustain-

able manner but which is not now being used for agriculture, raising crop output overall. Notably, 

AMTDP presents a broad set of specific measurable targets to achieve by 2022. Most have to do 

with improving cash crop exports, reducing food imports, or undertaking policy or administrative reforms 

in the sector. There are only a few targets related to improving the productivity or increasing pro-

duction levels of staple foods and most of these targets will be difficult to measure. 

The NADP and AMTDP are the agriculture sector-wide development plans of the government of 

PNG. However, we also reviewed sub-sectoral strategies within agriculture, if they existed – agricultural 

research, agricultural extension, and agricultural marketing and trade. While here we discuss all three, 

only for agricultural research has much strategic thinking found its way into policy documents. No prior-

ities for facilitating increased supply of extension services and for expanding marketing and 

trade have yet been established for the agricultural sector in PNG. 

For agricultural research, NARI has a mandate to conduct research on all aspects of agricul-

tural development in PNG, with particular attention to meeting the information and technology needs 

of smallholder farmers and rural communities. The overall strategic objective for NARI under its Strat-

egy and Results Framework (SRF) for the period 2011 to 2020 is the “enhanced productivity, efficiency, 

stability, and sustainability of the smallholder agriculture sector … [through] … enhanced food and nu-

tritional security, increased cash income generation, increased gainful rural employment, and a sustain-

able resource base (NARI 2011, p. vii)”. NARI’s efforts are to contribute to the higher-level impact of 

improved welfare for rural families and communities. 



22 

Four programs are described under the NARI SRF for 2011 to 2020. It is the first on improving 

the productivity, efficiency, and stability of agricultural systems that is most focused on raising farm 

productivity levels. Seven sub-programs under this program address many of the limitations that 

smallholder farmers face in raising their productivity levels – access to improved planting materi-

als; strengthened market systems; effective management of weather and climate-related shocks; sus-

tainable soil fertility management; effective and sustainable control of pests and diseases; farm mecha-

nization where it can be used efficiently; and profitably integrating crop, livestock, and aquaculture pro-

duction. Recalling the conceptual discussion in section 2 on why improved agricultural productivity mat-

ters and how it might be achieved, these agricultural systems sub-programs are well justified and pro-

vide a useful guide to how NARI should strategically allocate both its human and financial resources. 

However, priority setting across different crops will still be required. 

The second program on creating an enabling environment for smallholders to increase both their 

production and productivity levels is equally well thought out and necessary. The sub-programs here 

include building marketing opportunities from the outputs of smallholders through infrastructure devel-

opment, domestic market and international trade policy reform, and strengthened market information 

systems; policy reforms to foster increased investment in agriculture production and agriculture-based 

enterprises; appropriate and predictable government regulation of both production and marketing; and 

land reform. If implemented, these sub-programs will result in farmers being more willing to increase 

their production to supply markets, increasing household income in doing so. 

This now outdated SRF for 2011 to 2020 for NARI is conceptually strong. The strategic priorities 

laid out in it are well justified. These are as important for agricultural development in PNG ten years af-

ter the SRF was written as they were in 2011. Whether much progress has been achieved in attaining 

the objectives of the SRF programs will to a large degree have been dependent on whether NARI was 

able to amass sufficient qualified and dedicated staff and material and financial resources to implement 

the strategic programs effectively – that is, whether the fourth program under the SRF on NARI institu-

tional management and development was successfully implemented. Unfortunately, success under this 

fourth program has been patchy, at best. That this is the case has been a continuing challenge for both 

NARI specifically and for the public agricultural sector in PNG more broadly. 

With the end of the period of application of the 2011-2020 NARI SRF, a new SRF is now under 

preparation for the period 2021 to 2030 (NARI 2021). While still in draft form, this document was shared 

for review. Its content is built from the earlier SRF and retains the same objective of “enhanced produc-

tivity, efficiency, stability, and sustainability of the smallholder agriculture sector”. However, some major 

changes are made to its content. Setting aside a focus on increased productivity to foster growth in the 

agricultural sector, the 2021 to 2030 NARI SRF draft proposes to shift the focus of NARI’s efforts 

to commercialization of production, to building resilient farming systems in the face of shocks, 

and to improving nutrition. An immediate question that arises with this movement away from NARI’s 

traditional priority of sustainably improving agricultural productivity is whether NARI has the capacity 

and expertise and is best placed among public institutions in PNG to address the three new priorities 

laid out in the draft SRF. The expertise needed to improve market systems and specific agricultural 

value chains or to improve the nutritional status of the population of PNG are quite different from the 

classic areas of expertise in plant breeding, agronomy, plant pathology, soil science, agricultural pro-

duction economics, and so on that enabled NARI to develop and adapt new more productive and sus-

tainable agricultural production technologies. Can NARI effectively take on these new priorities 

which have somewhat higher-level objectives than what NARI scientists are trained to address? 



23 

However, it should be noted that the outcome matrix for the draft of the new NARI SRF does not 

neglect improving agricultural productivity as a central measure of the performance of the Institute. In-

creases in agricultural GDP attributed to smallholder food production, increases in factor productivity 

and resource use efficiency, and increased productivity per unit of limiting resource in specific farming 

systems are all listed as indicators of success for the SRF. However, there remains a conceptual gap 

between actions to address the three priorities under the draft SRF and actions to achieve these 

production and productivity-focused outcomes. 

On agricultural extension, neither the Department of Agriculture and Livestock (DAL) nor the overall 

government of PNG have developed any sub-sectoral strategies. Most efforts to increase the 

productivity levels of smallholder farmers across PNG will be hampered without effective means 

to ensure that farmers are able to access context-adapted information on how they might be more 

productive and efficient and realize greater profits in their farming. This is the role of agricultural ex-

tension agencies. Although there have been calls for a comprehensive agricultural extension policy or 

strategy for PNG for several decades, such a document has not yet been formulated. The description 

by Sitapai (2012) of how agricultural extension services were provided in PNG ten years ago remains 

largely accurate. The provision of such services is done by all three tiers of government – national, pro-

vincial, and local level governments – as well as by crop-specific commodity boards, most of which are 

parastatal, so reliant, in part, on government funding.7 While extension services focused on horticul-

tural crop and fruit production are provided by the Fresh Produce Development Agency, there is 

no dedicated agency focused on staple crops. Most of the other extension service providers focus 

on increasing the production of cash crops. 

While there is no agricultural extension policy document, both NADP and AMTDP have detailed 

sections laying out a rationale for increased investment in more effective extension services. NADP de-

votes a full chapter to agricultural extension and research efforts and what must be done to improve the 

delivery of the services they might offer farmers. In a critique of agricultural extension efforts nationally, 

NADP highlights poor coordination between the agencies with agricultural extension responsibilities; 

poor linkages between levels of government; limited public budget allocations; human capacity and 

management deficiencies; and the low priority given to agricultural extension relative to other develop-

ment priorities of the various level of government. To address these deficiencies, NADP calls for the 

formulation of a formal National Agricultural Research and Extension System (NARES) that will 

“provide timely and equitable access by farm households to relevant, quality agricultural extension, ed-

ucation and information services (p. 61)”. 

AMTDP also bemoans the poor state of agricultural extension service provision in PNG, highlighting 

that public-sector agricultural extension had been in decline. Responsibility for most extension services 

was devolved from the national Department of Agriculture and Livestock to the provinces and districts. 

In many provinces, extension is virtually non-existent due to lack of funding and capacity is-

sues. However, little detail is provided in AMTDP for strengthening the provision to farmers of agricul-

tural extension services, beyond cash crop-specific efforts. In discussion in AMTDP of the Digital Agri-

 

7 The Organic Law on Provincial Government and Local-Level Government 1995, which was a reform of the earlier law of 1977 

on the decentralization of government administration and public service provision, stipulated that the provision of agricultural 

extension services would primarily be the responsibility of provincial governments. 



24 

culture growth driver for the agricultural sector, the provision of extension services through digital Infor-

mation and Communication Technologies (ICT) platforms is noted as an early area of investment under 

this driver. An E-agriculture Strategy has since been developed and is discussed in Appendix 2. 

Effective agricultural extension service provision will be necessary if rapid and sustained increases 

in the productivity of smallholder farmers across PNG are to be realized. However, strategies for ensur-

ing that farmers have access to all of the information and improved technologies that they require to 

profitably realize increased levels of productivity have yet to be developed. How agricultural exten-

sion services will be provided is an important gap in efforts seeking to increase productivity lev-

els to expand PNG’s agriculture sector. 

On agricultural marketing and trade, in section 2 it was repeatedly noted that increased commercial-

ization of agricultural production will be necessary to provide incentives to smallholder farmers to ex-

pand their production beyond levels that solely met the subsistence needs of their households. This in-

creased agricultural production for the market – whether domestic markets for food crops or 

overseas for cash crops – is fundamental to sustained expansion of the agricultural sector 

within PN