IFPRI Discussion Paper 00800 September 2008 Cotton-Textile-Apparel Sectors of Pakistan Situations and Challenges Faced Caesar B. Cororaton Abdul Salam Zafar Altaf David Orden with Reno Dewina Nicholas Minot Hina Nazli Markets, Trade and Institutions Division INTERNATIONAL FOOD POLICY RESEARCH INSTITUTE The International Food Policy Research Institute (IFPRI) was established in 1975. IFPRI is one of 15 agricultural research centers that receive principal funding from governments, private foundations, and international and regional organizations, most of which are members of the Consultative Group on International Agricultural Research (CGIAR). FINANCIAL CONTRIBUTORS AND PARTNERS IFPRI’s research, capacity strengthening, and communications work is made possible by its financial contributors and partners. IFPRI receives its principal funding from governments, private foundations, and international and regional organizations, most of which are members of the Consultative Group on International Agricultural Research (CGIAR). IFPRI gratefully acknowledges the generous unrestricted funding from Australia, Canada, China, Finland, France, Germany, India, Ireland, Italy, Japan, Netherlands, Norway, South Africa, Sweden, Switzerland, United Kingdom, United States, and World Bank. AUTHORS Caesar B. Cororaton, Virginia Polytechnic Institute and State University Research Fellow, Institute for Society, Culture, and Environment, and Research Collaborator, International Food Policy Research Institute, Washington, D.C. (ccaesar@vt.edu) Abdul Salam, Federal Urdu University of Arts, Science, and Technology Professor and former chairman of the Agricultural Prices Commission of Pakistan Zafar Altaf, Pakistan Agricultural Research Council (PARC) Chairman of PARC and former Federal Secretary, Ministry of Food, Agriculture, and Livestock David Orden, International Food Policy Research Institute Senior Research Fellow, and Director, Global Issues Initiative, Institute for Society, Culture, and Environment, Virginia Polytechnic Institute and State University Reno Dewina, International Food Policy Research Institute Research Analyst, Markets, Trade and Institutions Division Nicholas Minot, International Food Policy Research Institute Senior Research Fellow, Markets, Trade and Institutions Division Hina Nazli, University of Guelph Ph.D. Graduate Research Assistant Notices 1 Effective January 2007, the Discussion Paper series within each division and the Director General’s Office of IFPRI were merged into one IFPRI–wide Discussion Paper series. The new series begins with number 00689, reflecting the prior publication of 688 discussion papers within the dispersed series. The earlier series are available on IFPRI’s website at www.ifpri.org/pubs/otherpubs.htm#dp. 2 IFPRI Discussion Papers contain preliminary material and research results. They have not been subject to formal external reviews managed by IFPRI’s Publications Review Committee but have been reviewed by at least one internal and/or external reviewer. They are circulated in order to stimulate discussion and critical comment. Copyright 2008 International Food Policy Research Institute. All rights reserved. Chapters of this material may be reproduced for personal and not-for-profit use without the express written permission of but with acknowledgment to IFPRI. To reproduce the material contained herein for profit or commercial use requires express written permission. To obtain permission, contact the Communications Division at ifpri-copyright@cgiar.org. iii Contents Acknowledgments vii  Abstract viii  1. Introduction and Overview David Orden and Caesar B. Cororaton 1  References 7  2. Global Cotton and Textile Markets Caesar B. Cororaton 8  References 21  3. Production, Prices, and Emerging Challenges in the Pakistan Cotton Sector Abdul Salam 22  References 51  4. Challenges in the Pakistan Cotton, Yarn, Textile, and Apparel Sectors Zafar Altaf 52  References 96  5. The Impact of Global Cotton Markets on Rural Poverty in Pakistan David Orden, Abdul Salam, Reno Dewina, Hina Nazli, and Nicholas Minot 99  References 117  iv List of Tables 2.1. World cotton supply and use 8  2.2. Major sources of world cotton production (% share) 9  2.3. Harvested area and yield 10  2.4. Major exporters of cotton (% share) 10  2.5. Major users of cotton (% share) 11  2.6. Major importers of cotton (% share) 12  2.7. Direct government assistance to cotton producers, 1997–1998 to 2002–2003 (millions of US$) 14  2.8. Government assistance to U.S. cotton producers, 1995–1996 to 2002–2003 (millions US$) 15  2.9. World prices of cotton, cotton yarn, and cotton fabric 16  2.10. Textile exports of selected economies 17  2.11. Clothing exports of selected economies 18  2.12. Pakistan exports of textiles and clothing to restricted markets, % 20  3.1. Cotton area and production (% distribution) 23  3.2. Staplewise production of cotton (% distribution) 24  3.3. Area, production, and yield of cotton in Pakistan 25  3.4. Major cotton-producing districts in Punjab (1996–2005 average % distribution of area and production) 27  3.5. Major cotton-producing districts in Sindh (1996–2005 average % distribution of area and production) 28  3.6. Cotton yields (kilogram/hectare) 29  3.7. Distribution of cotton-growing farms and cotton area by farm size in Pakistan 31  3.8. Balance sheet of Pakistan cotton (thousand tons) 34  3.9. Support and market prices of seed cotton 35  3.10. International price of cotton, U.S. cents/pound (Index B) 37  3.11. Domestic and international nominal and real prices of seed cotton 38  3.12. Regression analysis of lagged income effect on cotton area 41  3.13. Prices of important inputs and cost of production of seed cotton 42  3.14. Average farmers’ cost of production of cotton in Punjab and Sindh, 2004–2005 crop year 44  4.1. Pakistan export of textile products 53  4.2. Number of ginning factories and machines 60  4.3. Industry losses due to cotton contamination, 2004–2005 61  4.4. Installed and working capacity in the spinning sector, all Pakistan 62  4.5. Installed capacity in the spinning sector, by province (% distribution) 64  4.6. Consumption of cotton and other fibers (% distribution) 65  4.7. Production of yarn, % distribution 66  v 4.8. Production of yarn, by major province (% distribution) 66  4.9. Production and market for Pakistan yarn 67  4.10. Major country destination of exports of cotton yarn from Pakistan 68  4.11. World production of cotton yarn 68  4.12. Production of cotton yarn, countwise (% distribution) 70  4.13. World export of yarn 71  4.14. Equity internal rate of return of different counts 72  4.15. Production and market of Pakistan cloth 73  4.16. Installed and working capacity in the loom sector, all Pakistan 74  4.17. Installed capacity in the loom sector and production of cloth (mill sector), by major province 75  4.18. Production of cloth mill sector (% distribution) 76  4.19. Quality of cloth production, mill sector (% distribution) 76  4.20. Major country destination of exports of cotton cloth from Pakistan 77  4.21. Exports of textile made-ups 78  4.22. Major exporters of towels and cleaning cloths 79  4.23. Pakistan’s exports of towels and cleaning cloths (million dollars) 79  4.24. Major country destination of exports of towels from Pakistan 80  4.25. Major exporters of bed wear 81  4.26. Composition of Pakistan's exports of bed wear 81  4.27. Major country destination of exports of bed wear from Pakistan 82  4.28. World export of clothing 83  4.29. Exports of clothing 84  4.30. Major country destination of exports of garments from Pakistan 85  4.31. Major country destination of exports of knit wear (hosiery) from Pakistan 85  4.32. Recent estimates of losses due to outdated technology 86  4.33. Cross-country evaluation of apparel marketing channels 87  4.34. Cross-country evaluation of apparel human resources 88  4.35. Cross-country evaluation of apparel bureaucracy and infrastructure 88  4.36. Value added in cotton to apparel marketing chain 89  4.37. Financial and rebate cost comparison, Pakistan and India 91  4.38. Cross-country evaluation of ease of doing business 91  4.39. Pakistan’s economic planning process for textiles 92  5.1. Distribution of households by location and agricultural activity 100  5.2. Average per capita (adult-equivalent) annual consumption expenditures by households at the national, provincial, and primary cotton-producing district levels 102  5.3. Distribution of households by per capita (adult-equivalent) consumption expenditure quintiles at the national level 103  vi 5.4. Sources of income of urban and rural households at the national level 104  5.5. Sources of income of landowner cotton-producing households at the national, provincial, and primary cotton-producing district levels 105  5.6. Sources of income of sharecropper cotton-producing households at the national, provincial, and primary cotton-producing district levels 106  5.7. Simulated effects of increased cotton prices on poverty among landowner cotton-producing households at the provincial and national levels 109  5.8. Simulated effects of increased cotton prices on poverty among sharecropper cotton- producing households at the provincial and national levels 110  5.9. Simulated effects of increased cotton prices on poverty among all cotton-producing households at the provincial and national levels 110  5.10. Simulated effects on increased cotton prices on poverty at the primary cotton-producing district, provincial, and national levels 112  5.11. Simulated effects on increased cotton prices on rural poverty at the primary cotton-producing district, provincial, and national levels 113  5.12. Simulated effects at the provincial and national levels of increased wheat prices on poverty among all households producing wheat 115 List of Figures 2.1. Trade ratio—exports/production, % 9 2.2. Nominal cotton price: COTLOOK A and B indices and U.S. price 12 2.3. Cotton versus polyester fibers 14 2.4. World prices of cotton, cotton yarn, and cotton fabric 17 3.1. Arrival of seed cotton (average 1993–2005) 23 3.2.Cotton yield in Punjab and Sindh 30 3.3. Exports/production and imports/consumption of cotton, % 34 3.4. Support and market prices of seed cotton 36 3.5. Nominal and real prices of seed cotton 36 3.6. International price of cotton (Index B) 38 3.7. Nominal prices of seed cotton 39 3.8. Real prices of seed cotton 40 3.9. Seed cotton required to buy selected input 43 4.1. Capacity utilization in the spinning sector, 1958-2005 63 4.2. Average export price of yarn 69 5.1. Harvest and Lahore and Karachi import parity real prices of wheat 114 vii ACKNOWLEDGMENTS This Discussion Paper presents results from one of three main outputs of a research project “Pakistan- India: Cotton Trade Policy and Poverty Study” undertaken by the International Food Policy Research Institute (IFPRI) and its collaborators October 2005 through June 2007. The Agriculture and Rural Development Sector Unit, South Asia Region, the World Bank supported the study. We thank Paul Dorosh, the project manager for the Bank and a keen and experienced analyst of the economies of South Asia, for his encouragement and support throughout the study. We also thank Dr. Sohail Malik, president of Innovative Development Strategies, Ltd., Islamabad, for his support and facilitation of the efforts of the investigators. We thank Dr. Munir Ahmad, Pakistan Agriculture Research Council, and Rizwana Siddiqui, Pakistan Institute for Development Economics, for facilitating computable general equilibrium (CGE) training workshops that were held in Islamabad in March and July 2006 as a related activity of the project. We also gratefully acknowledge the support of the National Center for Applied Economic Research (NCAER), New Delhi, to facilitate the research on the cotton–textile–apparel sectors of India. Finally, we thank Shirley Raymundo, administrative coordinator at IFPRI, for her assistance in preparing this report. The two other main reports from this project are a corresponding Discussion Paper focused on the cotton–textile–apparel industries of India and a research report in which we evaluate the intersectoral linkages among these sectors and their effects on rural and urban poverty in Pakistan in the framework of a CGE model. The results of one or more of the components of this study have been presented at two professional meetings (American Agricultural Economics Association, July 2006; Pakistan Society of Development Economists, December 2006); at several policy outreach/discussion meetings with industry, academic, and government representatives in Pakistan (Islamabad Club, Islamabad, December 2006; Punjab Ministry of Commerce, Lahore, December 2006); at seminars at IFPRI (January and April 2007 in Washington, D.C., and April 2007 in New Delhi, India), NCAER (2007), the World Bank (September 2007), and the University of Guelph (June 2008); at the Conference on Rural Development and Poverty hosted by Pakistan Institute for Development Economics in Islamabad, Pakistan (April 2007); at the World Bank Workshop on Effects of Agricultural Price Distortions on Growth, Income Distribution and Poverty, West Lafayette, Indiana, (June 2007); at a conference of the Poverty Reduction, Equity and Growth Network, Berlin, Germany (September 2007); and at an NCAER-IFPRI conference in New Delhi (July 2008). We thank participants at these presentations and meetings for helpful suggestions and comments. viii ABSTRACT Cotton, textiles, and apparel are critical agricultural and industrial sectors in Pakistan. This study provides descriptions of these sectors and examines the key developments emerging domestically and internationally that affect the challenges and opportunities they face. One-quarter of Pakistani farmers, of whom about 40 percent have household incomes below the poverty line, grow cotton. Export controls and taxes kept cotton prices below international levels until the mid-1990s but have subsequently tracked export parity international levels following reforms to trade and pricing policies and a greater role for the private sector. Pakistani farmers have not formally adopted genetically modified Bt cotton but there is some field evidence of its unregulated use. Despite constraints in its production, storage, and ginning sectors, the production of cotton yarn increased at an annual rate of 4.7 percent during 1990–2005 and Pakistan’s share of world output increased to nearly 10 percent. Cotton-related products account for nearly 60 percent of Pakistan’s export earnings. The textile industry still produces mostly fabrics of relatively low count (low quality) although it has been successful in expanding its exports of some higher-value products. The industry will need further entrepreneurial initiatives to remain competitive in international markets. Among the farm households that produce cotton, about 40 percent of total income comes from its production. The decline in world prices that occurred in the late 1990s adversely affected these households. Household-level simulations suggest that a counterfactual 20 percent increase of cotton prices, which reflects the extent to which real cotton prices declined in Pakistan during this period, would have reduced the percentage of cotton-producing households below the poverty line in 2001 from 40 percent to 28 percent. The estimated effect from declining cotton prices explains about one-sixth of the overall observed increase of rural poverty in the period. Keywords: cotton, textiles, apparel, rural poverty, subsidies, industry policy, world markets 1 1. INTRODUCTION AND OVERVIEW David Orden and Caesar B. Cororaton Cotton, cotton-related products, textiles, and apparel are important commodities and comprise critical agricultural and industrial sectors in Pakistan and India. A number of key developments are emerging domestically and globally that potential will have profound effects on the cotton–textile–apparel sectors of the two economies. The industries face the challenge of remaining competitive in the context of the elimination of the Multi-Fiber Agreement (MFA) quotas on textile and apparel trade under the World Trade Organization (WTO), the emergence of China as a huge textile and apparel exporter, and new and potential intraregional trade agreements. Implementation of the final WTO ruling against U.S. cotton subsidies, a new U.S. farm bill in 2008, and a possible agreement to multilaterally reduce cotton subsidies and tariffs across the related textile and apparel sectors in the Doha Round WTO negotiations may also affect the cotton and related processing industries of Pakistan and India. . This Discussion Paper presents results from one of three main outputs of a research project on the cotton-related sectors of these two countries undertaken by the International Food Policy Research Institute (IFPRI) October 2005 through June 2007.1 In the context of the issues cited above, the overall goal of the study was to assess the intersectoral linkages in production, consumption, and trade from raw cotton through final apparel and to evaluate the effects of changes in domestic policies and world trade opportunities in these products on the related agricultural and industrial sectors and on rural poverty in both countries. There were two principal objectives of the study. • The first research component was to analyze the marketing and producer support policies related to cotton, cotton yarn, textile, and apparel production and trade in Pakistan and India, including assessment of the structure and levels of income of cotton farmers, the cost structure and flows in the cotton and processed cotton product markets; a detailed description of cotton/textile trade, pricing, and marketing policies since 1990; and the calculation of protection coefficients. • Having assessed the responses of domestic farm-level and industry prices in Pakistan and India to changes in world price levels, the second research component was to analyze the effects of changes in world cotton and textile prices and trade opportunities on poverty among farmers, landowners, agricultural and industrial laborers, and other households. Our assessment of the impact of cotton/textile trade policy on poverty rests on two complementary approaches. First, using available household data for each country, we characterize different types of rural households and their dependence on cotton production and cotton-related employment. We then evaluate the impact of lower cotton prices on rural poverty among cotton- producing households using partial equilibrium (single equation) simulations for Pakistan and India. This provides an analysis of both short-run (supply-fixed) and long-run (supply price–responsive) direct effects of changes in cotton prices. Second, for Pakistan, a more comprehensive CGE analysis, which explicitly models the economic responses of producers to the price incentives they face and the consequent intersectoral effects on production and household incomes and consumption, complements the partial equilibrium poverty assessment. The CGE model captures interindustry linkages, particularly vertical product linkages, in cotton production and procurement, yarn, and textile and clothing production, building on a recently completed social accounting matrix (SAM) constructed by Paul Dorosh, M.K. Niazi, and Hina Nazli (2004). There has been substantial progress recently in the integration of household information with 1 The project was “Pakistan-India: Cotton Trade Policy and Poverty Study” (EW-P091261-ESW-TF055329), supported by the Agriculture and Rural Development Sector Unit, South Asia Region, World Bank. 2 CGE model simulations, and we incorporate these innovations into our analysis to assess disaggregated effects on poverty from the policy simulations. This Discussion Paper addresses the first project objective by presenting a description of the characteristics of the cotton–textile–apparel sectors of Pakistan and the challenges that these sectors face. We address the second objective by presenting the partial equilibrium analysis of the effects of price changes on rural poverty in Pakistan. A companion Discussion Paper provides similar analysis for India (Bedi and Cororaton, 2008). A third report presents the CGE analysis of the project (Cororaton and Orden, 2008). 1.1. Overview of the Findings Since 1990, Pakistan and India have undertaken substantial reforms in their cotton/textile industries, increasing the role of the private sector. This Discussion Paper provides a careful review of the effectiveness of these reforms in Pakistan. Senior research staff at IFPRI and collaborators in Pakistan drafted the individual chapters. We examine the industry structure at various stages of production, processing, and marketing by reviewing recent industry literature and analyzing industry trends using secondary data. Focused interviews of major players in the industry—including farmer organizations and progressive farmers, selected cotton ginners, textile association representatives, traders, and the manufacturers and exporters of cotton and various textiles products, as well as with policymakers— provide a key dimension. These discussions and interviews focused on sector-specific issues in the factor markets, the product and export markets, the policy environment, existing constraints and prospects facing the industries, and likely challenges and opportunities in the near future. We present original simulations assessing the effects of cotton prices on poverty among cotton-producing households. The remainder of this introduction and overview summarizes the analysis from each chapter. Chapter 2: Global Cotton and Textile Markets This brief introductory chapter provides an overview of world markets in cotton, textiles, and apparel as context for the country-level analysis that follow. Global cotton production has doubled since the early 1980s and increased by about 20 percent since 1990 due primarily to yield growth. Acreage, although varying annually, shows little trend growth. The United States, China, and India are the dominant cotton- producing countries, accounting for nearly 65 percent of world production. Cotton production has increased at a faster than world average pace in India and Pakistan since 1970; and as a result their shares of total cotton output have increased over the past 35 years, with Pakistan now providing about 9 percent of world output and India about 20 percent. In India, the implementation of the Bt cotton program in 2002 increased cotton production by 106 percent from 2002 to 2006. The United States, Brazil, Africa, and Australia dominate exports of cotton. Like China, which now imports about one-fifth of the world’s total cotton traded, both Pakistan and India have declined as cotton exporters, and in some years they are net cotton importers, as their domestic spinning and textile industries have expanded. Cotton prices, and specifically the effects on world prices of the subsidy and trade policies of developed countries, have been controversial in the Doha Round of WTO negotiations. Chapter 2 traces the movement of world cotton prices, noting their decline over the past 10 years from relatively high levels in the mid-1990s. We review a set of studies estimating the impact of subsidies in driving prices lower than they would otherwise be. We put these effects in the context of other short- and long-run supply and demand forces affecting the cotton market. Cotton has lost market share to man-made fibers since the early 1990s, but relative prices do not appear to be the main driving force behind this shift. To complete the overview, Chapter 2 briefly examines trends in world textile and clothing markets. The value of textile trade has doubled between 1990 and 2005 to over $200 billion with an average annual growth rate of 3.9 percent. The European Union, United States, and China are both large importers and large exporters of textiles, with China a large net exporter, the United States a net importer, and the European Union having nearly balanced trade. Pakistan and India are large net exporters of textiles with very limited imports. The European Union, United States, and Japan are the three largest 3 clothing importers and the European Union, China, and Turkey are the largest exporters. Pakistan exports about $3.5 billion of clothing (about half the value of its textile exports) and India over $8 billion (about equal to its textile exports). The cotton and related processed goods sectors account for over 60 percent of Pakistan’s foreign exchange merchandise earnings, whereas they account for about 15 percent of those of India. Among other important exporters of textiles or clothing are Korea, Indonesia, Mexico, Bangladesh, Romania, Thailand, Sri Lanka, Malaysia, and the Philippines. Chapter 3: Production, Prices, and Emerging Challenges in the Pakistan Cotton Sector In Chapter 3, Dr. Abdul Salam, former chair of the Agricultural Prices Commission of Pakistan, examines the history of production and the price support and trade policies related to cotton in Pakistan. Cotton is the principal cash crop and is second only to wheat in total acreage. About 80 percent of the production comes from Punjab Province and about 20 percent from Sindh. Cotton-planted area shows an upward trend but is also affected by lagged income from cotton production and substantial unexplained annual variability. Yields show much more annual variability than does area, and has had no significant upward trend since 1990. About 25 percent of the farmers in Pakistan, including many small farmers, grow cotton. Almost half of the cotton-producing households own less than 5 hectares, and these households account for nearly 20 percent of cotton production. Thus, both yield and price variability affect small farm incomes and rural poverty levels. We analyze this in depth in Chapter 5. Until the mid-1990s, government price and trade policies heavily influenced cotton prices and trade. The state-controlled Cotton Export Corporation monopolized the cotton trade from 1974 through the late 1980s. Subsequently, the private sector has been allowed to purchase cotton from ginners and market it domestically and internationally. During the heavy intervention period, export taxes and domestic price policies kept cotton prices in Pakistan below world levels by as much as one third. This policy intervention was to ensure availability of low-cost primary inputs to the domestic processing industries. It had the effect of taxing farmers by depressing farm-level prices of raw cotton, which reduced incentives for production and investment. The price management system finally broke down in the mid-1990s. Subsequently, domestic market- and farm-level prices have tracked quite closely with their export parity equivalents for Index B cottons in world markets. These prices are about 20–25 percent lower than import parity; so domestic cotton production remains an important source of competitiveness for the domestic spinning, weaving, and apparel industries, even without the explicit subsidies that earlier policies gave the processing sectors. Cotton prices in Pakistan fell by about 20 percent in real terms from a three-year average around the peak price year of 1994–1995 to a three-year average around the lowest price year of 2000–2001. This was less than the decline of nearly 50 percent for equivalent world prices because of real depreciation of the Pakistan rupee. The final subsection of Chapter 3 identifies challenges faced by the cotton sector and makes some recommendations in the context of the past experiences in production and price policy. We identify yield variability as an ongoing concern, with producers subject to boom and bust differences on this account. We examine the costs of production, including significant expenditures for pest control. We recommend steps to raise production efficiency and, particularly, to improve control of insects and diseases through integrated pest management (IPM), strengthened variety research, farmer education, and investments in testing facilities. We recommend strengthened regulatory oversight to improve the quality and consistency of the chemical products for pest, disease, and weed control available to farmers. We describe the growing but unregulated use of Bt cotton based on field observations and articulate the need for clear- cut policy and strengthened institutional capacity to evaluate biotechnology and regulate seed markets. We recognize the need for improvements in marketing practices and ginning capacity to enhance cotton quality. We also cite price variability as a concern. We advocate some fine-tuning of the price support program for cotton to mitigate adverse effects from unscrupulous traders. Yet, we recognize the limited scope for price interventions and the gains from the elimination of past state control of cotton exporting and the setting of domestic prices below international parity levels. 4 Chapter 4: Challenges in the Pakistan Cotton, Yarn, Textile, and Apparel Sectors In Chapter 4, Dr. Zafar Altaf, former Pakistan federal secretary at the Ministry of Food, Agriculture, and Livestock, presents an overview of the economics and political economy of the entire value chain, including growing raw cotton, ginning into lint, spinning into yarn, weaving into fabric, producing cotton “made-ups,” such as towels and other nonapparel goods, producing apparel, and marketing. These combined sectors contributed 11 percent to Pakistan’s gross national product in 2004–2005, nearly 50 percent of manufacturing output, and more than 60 percent of the country’s foreign exchange earnings. Dr. Altaf’s analysis has an optimistic theme. He addresses the successes of Pakistan’s industries in terms of its rapid growth of yarn and textile production levels and highlights cases of highly competitive and successful entrepreneurs. But, he is also critical of the industry overall for failing to have sufficient entrepreneurial spirit, which he argues is necessary in the globalized fibers-to-apparel economy that has emerged. He lays the roots of weakness in the protected market environment in which Pakistan’s industry developed—not just the multilateral quotas of the MFA but also its own protected market, including its historically captive market in Bangladesh when that country was East Pakistan. His assessment recounts incidents of the distortions this protected market created. He raises many challenges to the industry; chief among these are the upgrade of the work force, the development of modern entrepreneurs, and greater attention to product differentiation and value, which requires marketing expertise and initiative. In Chapter 4, Dr Altaf also includes additional insights to complement the discussion of the cotton-producing sector. The historic tilt of pricing policies toward the domestic industry led to complacency and inefficiencies. Not only were incentives diminished for cotton production and quality, but also the spinning industry concentrated on yarn of low quality (low count yarn). The industry resisted efforts to raise quality standards and even as late as 1999 suppressed an attempt to increase competition to the benefit of farmers by introducing e-commerce bidding. At the same time, the cotton price support system, at price levels suppressed compared to international markets, kept farmers from developing mature marketing outlooks and strategies. These problems spilled over into the cotton ginning sector, which failed to make investments in modern machinery even as it expanded dramatically in the number of ginneries after deregulation in 1986–1987. The cotton ginning sector continues to use outdated storage (on open ground) and processing techniques (lack of cleaning before ginning, outdated ginning machinery) that yield a relatively low quality of cotton. Despite these difficulties, over the period 1990–2005 production of yarn increased at an impressive annual rate of 4.7 percent. Exports peaked at nearly 50 percent of output in 1991–1992 but have subsequently declined as the domestic textile industry has grown. Pakistan’s share of world production of cotton yarn has increased to nearly 10 percent. It has increased its utilization of man-made fibers to nearly 20 percent of all fiber content but still lags behind other major yarn producers in the production and diversity of its synthetic and mixed-fiber yarns. Altaf argues that Pakistani producers have continued to concentrate on low-value yarns despite higher potential returns from producing yarns of higher value. For this, he faults the manner the industry was first established by fiat when Pakistan gained independence as well as the highly protected markets in which it operated. Similarly, the weaving sector has grown at a high annual rate but remains concentrated on unprocessed greige fabrics. Again, Altaf attributes this outcome to relatively poor quality from the weaving sector (too many faults in the cloth and too much dependence on cotton fiber) and the lack of entrepreneurship and marketing knowledge for more specialized products by the Pakistani entrepreneurs. In terms of final products, Pakistan has been highly successful in several cotton made-ups, particularly towels and cotton bed wear. Altaf discusses how one firm that exported towels to the European Union succeeded by not compromising on quality, cataloguing its designs and products to guide the purchasers’ selections, and making heavy investment in high-quality imported machinery. Yet beyond such microsuccesses, Altaf notes the concentration of Pakistan exports in a few markets. He raises the concern that Pakistan has received some preferential market access in the aftermath of the 9/11 terrorist attacks in the United States and that its ability to gain entrance into new nonquota markets remains largely untested. 5 The notion that the better the textile product quality, the more marketing needed applies especially in the apparel sector. Pakistan’s share of world apparel trade has been stagnant and is concentrated on a few narrow categories. The apparel industry is dispersed into many cottage-based small units. Losses arise from a lack of a trained workforce, and Pakistani exporters are argued to lack flexibility in the apparel production process. Altaf recommends that public institutions, such as a fashion school, could help develop a more competitive and market-focused industry. The concluding sections of the chapter address the challenging market environment in which the Pakistan textile and apparel industries have operated since 2005 due to both increased competition facing its exports and the opening of its domestic market to increased import competition. The industry has responded with a multipart call for assistance from the government. Some of the assistance the industry needs is direct subsidies for investment or production cost offsets, which it argues are warranted because its competitors receive similar support. The industry also suffers from missing public institutions and a weak infrastructure, the improvement of which would generally enhance the competitiveness of the industry. Altaf fears the industry may fall back on its past lackluster ways of operating if the government bails it out again. But he recognizes that for these critical industries to prosper, in a cricket phrase, the Pakistan wicket must make some major changes. Chapter 5: The Impact of Global Cotton Markets on Rural Poverty in Pakistan Rural poverty increased in Pakistan during the late 1990s after declining during the 1980s and early 1990s, raising concern that agricultural growth was not being translated into poverty reduction. In Chapter 5, David Orden and co-authors examine the effects of cotton prices on rural poverty, particularly poverty among households producing cotton, using data from the 2001–2002 Household Integrated Economic Survey (HIES). Identifying the effects of prices on cotton incomes and poverty is an empirical issue of importance to policymakers, who need to understand the causes of the rise in rural poverty levels. At the national level, about 70.6 percent of households in Pakistan are classified as rural and about 40.7 percent are engaged in farming. Among farmers, about one quarter produce cotton and almost all these farmers also produce wheat. Nearly 70 percent of the cotton farmers are landowners, whereas 30 percent are sharecroppers or have other tenancy arrangements. Cotton production is concentrated in a number of primary cotton-producing districts of Punjab and Sindh. Household incomes are lower for rural than for urban households; and farmers who produce cotton have incomes slightly below the average. Sharecroppers are a particularly disadvantaged group, with over 65 percent of those producing cotton falling in the lowest two quintiles (40 percent) of the national income distribution. Income from cotton is quite important to the cotton-producing households in Pakistan and accounts for nearly 40 percent of total household income among landowners and nearly 45 percent among sharecroppers. We evaluate the effects on poverty among cotton-producing households of cotton prices rising by 10 percent to 40 percent from their low levels at the time of the 2001–2002 HIES by simulating the increase in net income this would generate for each household. Assuming that the additional income would be used for consumption expenditures, we evaluate these direct effects using the recognized national poverty line and the Foster-Greer-Thorbecke measures of poverty. Our analysis focuses on the effects of a 20 percent increase in prices, which reflects the extent to which real cotton prices declined in Pakistan during the late 1990s. We determine that at the national level 40 percent of cotton-producing households are in poverty. With a 20 percent increase in prices, this percentage declines to 28 percent and the depth (poverty gap) and intensity (poverty gap squared) also decline. In Punjab, the decrease due to a 20 percent rise in cotton prices is from 36 percent to 27 percent, and in Sindh it is from 50 percent to 32 percent. Among sharecropper cotton-producing households, poverty declined from 57 percent to 42 percent nationally. Two final parts of the analysis provide further context. First, at the regional level within the primary cotton-producing districts, the direct effects on poverty among cotton-producing households lower overall rural poverty by 3 percent in Punjab and 6 percent in Sindh. This concentrated effect 6 explains a 2 percent decline in rural poverty nationally compared to a reported increase in rural poverty of 12 percent in the late 1990s. Thus, low cotton prices need to be taken into account in designing antipoverty strategies and are important on a regional basis but are only one among several explanations for the overall observed increase in rural poverty. Second, more households produce wheat than cotton in Pakistan; but net incomes of these households are less dependent on wheat production, and wheat prices affect net household income only for that portion of the crop that is sold commercially. In addition, global wheat prices did not decline as much as global cotton prices did in the late 1990s. For these reasons, the effects of cotton prices on those households producing cotton are sharper than the effects of changes in wheat prices on households producing wheat during this period. Because more households produce wheat, the overall effects of a given percentage change in wheat prices on poverty levels among all farmers, and on poverty measured at the provincial or national levels (also taking into account the effects on non–wheat-producing households), are similar to the deeper but more concentrated effects of an equal percentage increase in cotton prices. 7 REFERENCES Bedi. J., and C.B. Cororaton. 2008. Cotton-Textile-Apparel sectors of India: situation and challenges faced. Discussion Paper, Washington, D.C.: International Food Policy Research Institute. Cororaton, C., and D. Orden. 2008. Pakistan’s cotton and textile economy: intersectoral linkages and effects on rural and urban poverty. (Research Report in press) Washington, D.C.: International Food Policy Research Institute. Dorosh, P., M.K. Niazi, and Hina Nazli. 2004. A social accounting matrix for Pakistan, 2001-02: methodology and results. (A Background Research Paper for the Pakistan Rural Factor Markets Study) Washington, D.C.: South Asia Rural Development Unit of the World Bank. 8 2. GLOBAL COTTON AND TEXTILE MARKETS Caesar B. Cororaton 2.1. Introduction To provide a basis for the sections that follow, this section provides a review of world cotton, textile, and apparel markets with some focus on Pakistan. In Section 2.2, we describe broad trends in production, consumption, trade, and prices in the international market for cotton and highlight some factors as determinants of the movements in the international price of cotton. Section 2.3 examines trends in textile and clothing trade since 1990. 2.2. Global Cotton Markets 2.2.1. Trends in Production, Consumption, and Trade The total global area devoted to cotton production changed little over the period 1965–2004. Its average growth is 0.1 percent (Table 2.1). However, productivity in terms of yield (kilogram per hectare) improved by an average of 1.8 percent. Thus, the average output growth of 1.9 percent was largely due to the improvement in yield. International trade is a major component of the cotton market. However, whereas exports and imports of cotton grew relatively faster (average rate of 2.5 and 2.4 percent, respectively) than production and consumption (average rate of 1.9 and 2 percent, respectively) over the period 1965–2006, the export- to-production ratio exhibits a declining trend after the mid-1970s, when it reached a peak of nearly 50 percent (Figure 2.1). Table 2.1. World cotton supply and use Year Beginning 1 Aug Harvested Area (millions of hectares) Yield (kilogram/ hectare) Supply Use Beginning Stocks Production Imports Consumption Exports Ending Stocks (million 480-pound bales) 1965 33.3 372.5 29.0 56.9 17.4 53.8 17.0 32.6 1970 31.8 380.5 22.4 55.6 24.6 57.1 23.6 21.8 1975 29.9 393.4 33.4 54.0 26.1 61.6 26.0 25.9 1980 32.4 426.3 21.2 63.4 27.3 65.0 26.3 20.6 1985 31.6 552.5 42.1 80.2 28.7 75.3 28.1 47.6 1990 33.2 572.2 25.0 87.1 30.4 85.5 29.6 27.4 1995 36.0 567.2 31.9 93.7 27.4 85.8 27.4 39.9 2000 32.0 604.0 49.2 88.9 27.3 92.2 26.4 46.8 2001 33.7 637.4 46.8 98.8 29.9 94.3 29.0 52.1 2002 30.4 631.0 52.1 88.3 30.6 98.3 30.3 42.3 2003 32.1 646.0 45.4 95.3 34.8 98.1 33.2 44.3 2004 35.8 742.9 44.3 122.1 34.6 108.7 35.0 57.4 2005 34.9 734.5 57.4 117.7 45.9 116.0 44.5 60.4 2006 34.7 765.1 60.4 121.9 123.3 Average growth* 0.1 1.8 1.8 1.9 2.5 2.0 2.4 1.6 Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, United States Department of Agriculture (USDA). Note: *1965–2006 geometric growth, %; 1965–2005 for imports, exports, and ending stocks. 9 Figure 2.1. Trade ratio—exports/production, % 0 10 20 30 40 50 60 Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA The largest producer of cotton is China, which captures about a quarter of world production (Table 2.2). Historically, the United States has long been the second major producer of cotton, but in the past two years, India has surpassed it. Over the past 35 years, cotton production in India has grown 4.6 percent on average. Since 2000, cotton production in India has grown rapidly at 11.6 percent. The surge in cotton production in India is mainly due to the introduction of Bt (Bacillus thuringiensis) cotton in 2002.2 On the other hand, over the same period, the average cotton production growth in Pakistan was 3.7 percent. This relatively high growth enables Pakistan to double its share in the overall world production of cotton. At present, it is the fourth major producer. Table 2.2. Major sources of world cotton production (% share) Period Average China United States India Pakistan Brazil Former Soviet Union Turkey Others 1970–1974 17.3 19.4 8.5 4.8 4.6 18.4 3.9 23.1 1975–1979 16.8 19.4 9.3 4.1 4.0 20.4 3.8 22.2 1980–1984 25.7 16.9 9.6 4.9 4.5 16.0 3.4 18.9 1985–1989 23.1 16.5 10.7 8.0 4.3 15.6 3.3 18.7 1990–1994 24.3 19.9 11.8 8.6 3.0 11.7 3.3 17.4 1995–1999 22.4 19.2 14.4 8.4 2.4 8.0 4.2 21.1 2000–2003 24.1 19.6 13.4 8.8 4.8 7.2 4.1 17.9 2004 25.4 19.0 15.6 9.1 4.8 6.6 3.4 16.1 2005 25.1 20.3 16.2 8.6 4.0 7.1 3.0 15.7 2006* 29.1 17.7 17.9 8.1 5.7 6.7 3.2 11.5 2007† 29.7 15.8 19.7 8.2 5.9 6.9 2.8 11.0 Average growth‡ 3.3 1.7 4.6 3.7 2.6 −0.7 1.6 0.1 Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA. Note: *Estimates; †forecast; ‡1970–2007 geometric growth of volume production. 2 Bt cotton contains a gene, derived from soil bacteria (Bacillus thuringiensis), that protects cotton crop against bollworm by producing a special protein. The bollworms feeding on Bt cotton leaves become sleepy and lethargic and thus cause less damage to the crop plants. 10 Table 2.3 shows the data on harvested area and yield for the four major cotton producers. Except for the variability around a flat trend, there is not much change in area in either China or the United States. Nevertheless, there are some noticeable increases in India and Pakistan. The yield in China and the United States is higher than the world average and lower in India and Pakistan. However, some catching up has occurred. Over the period 1970–2006, whereas the improvement in world yield was 76 percent, the improvement in China was 149 percent, in India 193 percent, and in Pakistan 101 percent. The improvement in yield for the United States over the period was 66 percent. Table 2.3. Harvested area and yield Period Average World China United States India Pakistan Harvested Area (millions hectare) Yield (kilogram/ hectare) Harvested Area (millions hectare) Yield (kilogram/ hectare) Harvested Area (millions hectare) Yield (kilogram/ hectare) Harvested Area (millions hectare) Yield (kilogram/ hectare) Harvested Area (millions hectare) Yield (kilogram/ hectare) 1970–74 32.9 400.2 5.0 458.6 4.9 526.8 7.6 147.1 1.9 330.5 1975–79 31.8 409.4 4.8 450.7 4.7 540.3 7.7 158.2 1.9 280.6 1980–84 32.3 476.1 5.8 680.4 4.4 594.0 7.8 190.5 2.2 342.7 1985–89 31.4 548.4 5.0 797.0 4.1 701.1 7.1 257.0 2.5 548.3 1990–94 32.7 570.3 5.9 773.5 5.0 741.3 7.6 287.6 2.8 594.2 1995–99 33.7 580.1 4.6 966.3 5.4 706.9 9.0 311.2 3.0 568.9 2000–01 32.9 621.7 4.4 1095.7 5.4 750.9 8.7 292.1 3.0 601.0 2002–06 33.6 704.1 5.4 1141.4 5.2 875.0 8.4 431.3 3.1 665.9 Average 1970– 2006 532.1 771.1 673.7 256.7 479.9 Average growth* 76.0 148.9 66.1 193.1 101.5 Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA. *Between two subperiods: 1970–1974 and 2002–2006, % Table 2.4. Major exporters of cotton (% share) Period Average China United States India Pakistan Brazil Former Soviet Union Africa* Australia Others 1970–74 0.5 17.8 0.6 2.9 3.7 37.3 2.4 0.1 34.7 1975–79 0.4 21.1 0.7 1.7 0.6 41.3 2.9 0.4 30.9 1980–84 1.4 23.6 1.4 4.2 1.3 38.4 3.5 1.8 24.5 1985–89 7.0 18.4 1.6 8.7 1.5 34.5 5.7 3.7 18.9 1990–94 2.3 25.9 1.8 3.6 0.8 32.6 8.0 6.0 19.0 1995–99 1.9 25.0 1.7 1.7 0.1 22.9 13.0 9.8 23.9 2000–03 1.5 36.0 0.7 1.0 2.0 17.6 12.6 10.2 18.3 2004 0.1 41.2 1.9 1.6 4.4 17.0 11.8 5.7 16.3 2005 0.1 39.4 7.8 0.6 4.4 16.3 10.0 6.5 14.9 2006† 0.2 34.6 13.5 0.7 3.5 18.3 10.1 5.7 13.5 2007‡ 0.1 39.4 12.2 0.6 6.8 16.8 7.4 3.5 13.1 Note: *Includes Benin, Burkina Faso, Cameroon, Chad, Ivory Coast, Mali, Niger, Senegal, Togo, and Central African Republic; †Estimates ‡Forecast. 11 The major source of world cotton exports is the United States (Table 2.4). From the average of 17.8 percent in 1970–1974, its share increased to 36 percent in 2000–2003. In 2004, the share increased to 41.2 percent but declined slightly to 39.4 percent in 2007. The former Soviet Union used to capture a large part of cotton exports in the 1970s, but its share has dropped significantly, especially in the first half of the 2000s. Exports from the African region have improved through the years, as have those from Australia, except in some recent years. Cotton exports from China, India, and Pakistan are relatively limited although there is substantial annual variability in their exports. The size of the textile industries largely determines consumption of cotton. China, being the world’s leading producer of textiles is also the major cotton user. At present, it consumes more than a third of world production (Table 2.5). India and Pakistan have increasingly become major cotton users as well, due to their relatively larger textile industries. Table 2.5. Major users of cotton (% share) Period Average China United States India Pakistan Brazil Former Soviet Union Turkey Others 1970–74 19 13 9 4 3 15 2 37 1975–79 20 11 9 3 4 14 2 37 1980–84 24 8 9 3 4 12 2 36 1985–89 24 9 10 4 4 11 3 35 1990–94 24 12 11 8 4 7 4 31 1995–99 23 12 15 8 4 3 6 29 2000–04 30 8 14 9 4 4 6 25 2000–03 29 19 14 9 4 4 6 14 2004 35 19 14 10 4 3 7 8 2005 39 20 14 10 4 3 6 4 2006* 41 15 15 10 4 3 6 8 2007† 43 16 15 10 3 3 6 5 Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA. Note: *Estimates; †Forecast. Some years, the cotton production in China does not meet domestic consumption, thus, it relies on importation. Cotton imports to China were significant in the middle of the 1990s and in the first half of the present decade (Table 2.6). Cotton imports in the former Soviet Union, EU-25, and Japan dropped steadily over time, whereas they increased in Indonesia and Thailand. Cotton imports into both India and Pakistan have increased in the past 10 years. 12 Table 2.6. Major importers of cotton (% share) Period Average China United States India Pakistan Brazil Former Soviet Union Russia EU-25 Japan Indonesia South Korea Thailand Taiwan Others 1970–74 4.4 0.2 1.6 0.0 0.0 28.2 0.0 28.6 14.2 0.9 2.4 1.1 2.8 15.7 1975–79 6.7 0.1 0.8 0.0 0.0 27.9 0.0 25.2 11.9 1.4 3.8 1.5 3.7 17.1 1980–84 5.7 0.1 0.0 0.2 0.1 25.6 0.0 25.7 12.4 2.0 3.8 1.7 4.2 18.5 1985–89 2.1 0.0 0.2 0.0 1.1 25.0 10.8 25.1 10.7 3.2 3.2 3.4 5.5 9.8 1990–94 6.0 0.0 0.7 0.7 4.5 15.7 11.7 21.2 8.0 6.6 3.5 5.4 4.6 11.3 1995–99 6.2 1.0 1.7 1.4 6.5 6.0 4.2 19.8 5.0 7.8 3.7 5.2 4.9 26.5 2000–02 4.2 0.1 5.9 2.6 1.6 7.0 5.8 15.0 3.7 8.3 3.5 6.1 4.3 31.9 2003 25.3 0.1 2.3 5.2 1.6 5.0 4.2 9.5 2.2 6.2 3.7 4.8 2.9 27.0 2004 18.5 0.1 3.0 5.1 0.6 4.9 4.2 9.3 2.4 6.4 3.9 6.6 3.9 31.4 2005 42.0 0.1 0.9 3.5 0.7 4.0 3.1 5.3 1.4 4.8 2.2 4.1 2.5 25.5 2006 26.8 0.0 1.0 5.8 1.3 4.8 3.6 5.4 1.5 5.6 2.7 4.9 2.9 33.4 Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA. 2.2.2. Trends in International Cotton Prices Figure 2.2 shows three indicators of international cotton prices: the COTLOOK A and COTLOOK B Indices,3 and U.S. price. Together, these indices move generally in the same direction. The COTLOOK A index is generally higher than the COTLOOK B Index, and the U.S. price index is either below or above the two indices. Cotton from Pakistan is grouped within the COTLOOK B Index. Figure 2.2. Nominal cotton price: COTLOOK A and B indices and U.S. price - 20 40 60 80 100 120 US ce nt s p er p ou nd COTLOOK A Index US price COTLOOK B Index Source: International Cotton Advisory Committee Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA. Note: Number converted from 480-pound bale to metric tons. 3 COTLOOK A Index is the average of the five lowest quotations of 16 styles of cotton (middling 1-3/32 inches) traded in North European ports from the following origins: Australia, Brazil, China, Francophone Africa, Greece, India, Mexico, Pakistan, Paraguay, Spain, Syria, Tanzania, Turkey, the United States, and Uzbekistan. COTLOOK B Index is the average of the three lowest quotations of eight styles of coarser grades of cotton from Argentina, Brazil, China, India, Pakistan, Turkey, the United States, and Uzbekistan. 13 There is a high degree of variability in the international price of cotton. Although an increasing trend in nominal prices is observed from the second half of the 1960s through the 1970s, there was no clear direction in the 1980s. The early 1990s saw a sharp hike in cotton prices until 1994, then a significant drop is observed in the second half of the 1990s until 2001. During these years, international cotton prices (COTLOOK A and B indices) fell nearly 60 percent, whereas U.S. cotton prices fell by 40 percent. Wide swings in cotton prices continued from 2002. After a recovery in 2002 and 2003, prices dropped in 2004. However, the past three years saw improvement in cotton prices. 2.2.3. Some Factors Influencing Movements in International Cotton Prices Various factors, such as expectations, production, and inventories affect short-term fluctuations in the international price of cotton. For example, natural calamities coupled with a significant drop in stocks in China resulted in a sharp increase in prices in 2003. Lower than expected consumption and the expected bumper crop resulted in a decline in domestic price in 2004 (Food and Agriculture Organization, 2006). Over the long term, improvements in yield affect international prices of cotton due to improved inputs, such as expanded use of irrigation, fertilizers, and chemicals. Other technological developments that reduce the cost of production, such as the introduction of genetically modified (GM) varieties, also affect prices. Competition from substitute fibers and trade-distorting policy shifts in major cotton- producing and exporting countries also affect international prices. One recent development in cotton production is the focus on cost reduction through the less intensive use of chemicals (Baffes, 2004). Contributing to this development has been the introduction of GM seed technology. The technological developments of the 1990s that resulted in the introduction of Bt cotton present potential for reducing cost and thereby for increasing profitability. The leading cotton- producing countries that have introduced this technology include China, India, and Mexico in the Northern Hemisphere, and Argentina, Australia, and South Africa in the Southern. Brazil, Indonesia, Israel, Pakistan, and Turkey are presently in the trial stage.4 However, the largest user of Bt cotton is the United States, where it is estimated that 70 percent of its cotton area was sown with GM varieties in the 2003–2004 season. In Australia, 44 percent of its cotton area was sown to such varieties in the 2002–2003 season. In China, more than 20 million hectares were planted with such varieties in 2002. Indeed, the introduction of this technology is significant. At present, it is estimated that 22 percent of the world’s cotton planting is now in GM varieties, up from 2 percent in 1996–1997 (Baffes, 2004). Synthetic fibers such as rayon and polyester are substitutes for cotton fibers. Since the early 1990s, there have been major structural shifts in the share of cotton and polyester fibers (Figure 2.3). In the 1980s, cotton and polyester shares were each around 50 percent. However, from 1992 onward, the share of polyester improved to about 60 percent, whereas that of cotton dropped to about 40 percent. The synthetic/cotton price ratio does not appear to be the main factor behind the shift in consumption. Over the past two decades, their prices generally move in the same direction. One of the most likely reasons behind the shift is the durability of polyester-based (or polyester mixed with cotton) clothing compared to pure cotton based. 4 In Pakistan the Ministry of Food, Agriculture, and Livestock announced plans on January 5, 2007 to release the first home- grown insect-resistant BT cotton variety during the next sowing season “to maximize the production of cotton crop for domestic needs and exports.” (http://www.checkbiotech.org/blocks/dsp_document.cfm?doc_id=14159) 14 Figure 2.3. Cotton versus polyester fibers 0 5 10 15 20 25 30 35 0 10 20 30 40 50 60 70 80 Pr ic e ra tio : P ol ye st er /C ot to n Sh ar e of C ot to n an d Sy nt he tic Share of cotton Share of polyester Price ratio: polyester/cottonSource: International Cotton Advisory Committee Source: Cotton and Wool Situation and Outlook Yearbook, Economic Research Service, USDA. Note: Number converted from 480-pound bale to metric tons. In the early 1990s, Townsend and Guitchounts (1994) estimated that countries that implement some form of trade-distorting government policies such as taxes and subsidies produced about two-thirds of cotton. Recently, the International Cotton Advisory Committee (ICAC) found that eight countries provided direct support to cotton production: Brazil, China, Egypt, Greece, Mexico, Spain, Turkey, and the United States (Table 2.7). By far the largest direct government assistance to cotton producers is in the United States, which reached nearly $4 billion in 2001–2002. The government support in the United States comes in various policy instruments (Table 2.8). Table 2.7. Direct government assistance to cotton producers, 1997–1998 to 2002–2003 (millions of US$) Country 1997–98 1998–99 1999–2000 2000–01 2001–02 2002–03 United States 1,163 1,946 3,432 2,148 3,964 2,620 China 2,013 2,648 1,534 1,900 1,196 750 Greece 659 660 596 537 735 718 Spain 211 204 199 179 245 239 Turkey n.a. 220 199 106 59 57 Brazil 29 52 44 44 10 0 Mexico 13 15 28 23 18 7 Egypt 290 n.a. 20 14 23 33 Source: Quoted from Baffes (2004). Original sources are ICAC (2003) and USDA. Note: n.a. = not available. 15 Table 2.8. Government assistance to U.S. cotton producers, 1995–1996 to 2002–2003 (millions US$) Policy Instruments 1995–96 1996–97 1997–98 1998–99 1999– 2000 2000–01 2001–02 2002–03 Coupled payments 3 n.a. 28 535 1,613 563 2,507 248 PFC/DP n.a. 599 597 637 614 575 474 914 Emergency/CCP n.a. n.a. n.a. 316 613 613 524 1,264 Insurance 180 157 148 151 170 162 236 194 Step-2 34 3 390 308 422 236 196 n.a. Total 217 759 1,163 1,947 3,432 2,149 3,937 2,620 Source: Quoted from Baffes (2004). Original sources are USDA (assistance), ICAC (production). Note: n.a. = not available; PFC = production flexibility contracts; DP = direct payments; CCP = countercyclical payments. A number of studies have attempted to quantity the impact of government support on world prices and production, particularly focusing on the 1994–2002 period in which prices dropped sharply. Orden et al. (2006) and the Food and Agriculture Organization (FAO, 2004) surveyed those studies and found that, generally, the elimination of the subsidies would likely improve international prices of cotton. However, the magnitude of the impact depends on the method used, such as the CGE model, the partial equilibrium model, and econometric estimates of supply response. To cite some conclusions from individual studies, the estimates of the Overseas Development Institute (Gillson et al., 2004) indicate that if the cotton market were to be liberalized, production in the United States and the European Union would fall and world prices of cotton would increase between 18 and 28 percent. This would increase export earnings of all developing countries by $610 million. West and Central African countries could gain between $94 million and $355 million in earnings from cotton production. ICAC (2003) finds that the removal of subsidies will result in lower production in countries concerned and will, therefore, increase world prices of cotton by 21 percent in 2000–2001 and 73 percent in 2001–2002. The study of Goreaux (2003) indicates that export earnings of West and Central Africa were reduced by $250 million owing to cotton support policies. The removal of subsidies is estimated to increase world prices of cotton by 18 percent. The study of Reeves, Vincent, and Quirke (2001) finds that the removal of production and export subsidies by the United States and the European Union could lead to a 20 percent reduction in U.S. cotton production and 50 percent fall in U.S. cotton exports. This in turn could increase prices by 10.7 percent from the observed benchmark. The study carried out by the Australian Center for International Economics (Townsend, 2002) indicates that the removal of subsidies would increase world cotton prices by 10.7 percent. Sumner (2003) finds that without U.S. cotton subsidies during 1999–2002, world cotton prices would have been higher by 13 percent. At the lower end of estimates, Tokarick (2003) finds that multilateral trade liberalization across cotton and other agricultural markets will improve cotton prices only by 2.8 percent, whereas Poonyth et al. (2004) find the improvement in cotton prices would range between 3.1 and 4.8 percent. From these studies, the impact of trade-distorting policies in major producing and exporting countries on world cotton prices is significant, with many estimates in the range of 10–20 percent. This would have far-reaching effects on rural farm households, especially in cotton-producing developing countries. Estimates from the Food and Agriculture Organization (FAO, 2001) indicate that as many as 100 million rural households may have been directly or indirectly involved in cotton production. 2.3. Prices of Cotton Yarn and Cotton Fabric Cotton is processed into yarn and then fabric, both of which are also heavily traded internationally. Unlike the COTLOOK A and B indices, no similar price indices for cotton yarn and cotton fabric are readily available. To provide an idea of how world prices of cotton yarn and fabric move with the world prices of cotton, we derived the traded price indices of these cotton products using data from the United Nations 16 Commodity Trade Statistics. We selected major world exporters of cotton yarn and tracked their data on value and quantity traded from 1990 to 2005. Similarly, we tracked the data on value and quantity traded of cotton fabric of major exporters. We computed price series for these products and express them, including the COTLOOK B, in index form with index 2000 = 100 in Table 2.9. For the period 1990– 2005, the coefficient of variation of COTLOOK B is 22.9 percent, whereas cotton yarn is 13 percent and cotton fabric 7.7 percent. Figure 2.4 also shows that COTLOOK B is more volatile than cotton yarn prices and cotton fabric prices. The period 1994–2001 saw a drop in COTLOOK B of 57.8 percent. Over this period, the price of cotton yarn dropped by 27.4 percent and by 38.8 percent from lagged peak-to-trough yarn prices in 1995 to 2002. The drop of the price of cotton fabric is not as dramatic at 6.4 percent over the 1994–2001 period and by 19.4 percent from the peak textile prices in 1996. Using these reduced 1994–2001 form relationships, the “elasticity” between COTLOOK B and the price of cotton yarn is 0.47 and the that between the price of cotton yarn and the price of cotton fabric is 0.23 during the 1994–2001 period. Table 2.9. World prices of cotton, cotton yarn, and cotton fabric COTLOOK B Cotton Yarna Cotton Fabricb 1990 144.9 100.8 125.8 1991 108.9 104.3 124.3 1992 100.0 116.6 111.7 1993 125.3 106.4 99.8 1994 171.9 123.4 107.0 1995 150.9 136.8 121.7 1996 139.4 125.8 124.2 1997 132.2 116.9 115.0 1998 101.1 111.7 113.3 1999 92.3 105.1 106.9 2000 100.0 100.0 100.0 2001 72.5 89.5 100.2 2002 97.6 83.8 116.0 2003 124.1 97.5 111.1 2004 95.3 101.9 118.4 2005 95.3 94.9 116.9 Mean 115.7 107.2 113.3 SD 26.5 14.0 8.7 C.V. % 22.9 13.0 7.7 1994–2001 Change, % −57.8 −27.4 −6.4 Ratioc 0.47 0.23 Source: United Nations Commodity Trade Statistics and International Cotton Advisory Committee (ICAC). Note: C.V. = coefficient of variation; SD = standard deviation. aCotton yarn: Standard International Trade Classification Revision 3-6513 (Countries: China-Hong Kong-Special Administrative Region (SAR), China, India; Pakistan, United States, and Italy). bCotton fabric, woven: Standard International Trade Classification Revision 3-652 (Countries: China-Hong Kong-SAR, China, India, Pakistan, United States, Italy, Germany, Japan, France, Republic of Korea, Belgium, Netherlands, and United Kingdom). cFor cotton yarn: change in the price of cotton yarn over change in COTLOOK B; for cotton fabric: change in the price of cotton fabric over change in the price of cotton yarn. 17 Figure 2.4. World prices of cotton, cotton yarn, and cotton fabric 0.0 20.0 40.0 60.0 80.0 100.0 120.0 140.0 160.0 180.0 200.0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 COTLOOK-B Cotton Yarn - 6513 Cotton Fabric - 652 2.4. Global Trends in Markets for Textile and Clothing 2.4.1. World Markets This subsection presents trends in the world markets for textiles and clothing, the position of Pakistan in these markets, and some information on Pakistan’s world exports of textiles and sources of its imports. In 2005, the size of the world market for textiles was $203 billion (Table 2.10). It has grown strongly in the past 15 years. In the 1990s, the average annual growth of the market was about 5 percent. In 2003 and 2004, its annual growth was more than 10 percent, slowing in 2005 to 3.9 percent. The European Union (EU-25) captures a third of the total world export of textiles. This is mainly intra-EU trade. Its textile trade with the rest of the world accounts for less than 12 percent of the total. China has a rapidly growing share in the world textile market. In 1990, China accounted for 6.9 percent of the world export of textiles. Its exports surged after 2000. By 2005, China had a share of 20.2 percent of the world market. The shares of the other major producers of textile are generally stable, implying falling shares for several other countries. Hong Kong’s share, which is mostly due to reexporting, is about 7 percent, and the United States has about the same level. The share of India was about 4 percent in 2005 and Pakistan’s was 3.5 percent. Table 2.10. Textile exports of selected economies 1990 2000 2003 2004 2005 World (billion US$) 104.4 157.1 173.7 195.4 203.0 Average annual growth, % 5.1 10.6 12.5 3.9 % World European Union (EU-25) n.a. 35.9 37.4 37.0 33.5 Intraexports n.a. 24.9 25.2 24.5 21.9 Extraexports n.a. 14.7 9.7 7.4 11.6 China 6.9 10.3 15.5 17.1 20.2 Hong Kong 7.9 8.6 7.5 7.3 6.8 Reexports 5.8 7.8 7.1 7.0 6.5 United States 4.8 7.0 6.3 6.1 6.1 18 Table 2.10. Continued 1990 2000 2003 2004 2005 Republic of Korea 5.8 8.1 6.2 5.5 5.1 Taipei, China 5.9 7.6 5.4 5.1 4.8 India 2.1 3.8 3.9 3.6 3.9 Pakistan 2.6 2.9 3.5 3.1 3.5 Turkey 1.4 2.3 3.0 3.3 3.5 Japan 5.6 4.5 3.7 3.7 3.4 Indonesia 1.2 2.2 1.7 1.6 1.7 Source: International Trade Statistics, 2006. Note: n.a. = not available. Textile: Standard International Trade Classification, Revision 3-65. Table 2.11 presents the structure of the world market for clothing. In 2005, the total world exports of clothing amounted to $275.6 billion, somewhat larger than the world market for textiles. It is also growing strongly, with an average growth of 8.3 percent in the 1990s, rising to 17.6 percent in 2003, 11.4 percent in 2004, and then slowing to 6.4 percent in 2005. Similar to the world market structure for textiles, the European Union has the largest share in the world market for clothing, and, again, this is mostly intra-EU trade. There is remarkable growth in China’s exports of clothing with its share of the world market increasing from 8.9 percent in 1990 to 26.9 percent in 2005. India’s share is stable at about 3 percent. The share of Pakistan is also stable at about 1 percent. Table 2.11. Clothing exports of selected economies 1990 2000 2003 2004 2005 World (billion US$) 108.1 197.8 232.6 259.1 275.6 Average annual growth, % 8.3 17.6 11.4 6.4 % of World European Union (EU-25) 26.9 29.4 29.7 29.2 Intraexports 20.1 22.0 2.2 20.9 Extraexports 6.8 7.4 7.4 8.2 China 8.9 18.2 22.4 23.9 26.9 Hong Kong, China 14.2 12.2 10.1 9.7 9.9 Reexport 5.7 7.2 6.4 6.5 7.3 Turkey 3.1 3.3 4.3 4.3 4.3 India 2.3 3.1 2.8 2.6 3.0 Mexico 0.5 4.4 3.2 2.9 2.6 Bangladesh 0.6 2.0 2.1 2.2 2.3 Indonesia 1.5 2.4 1.8 1.7 1.9 United States 2.4 4.4 2.4 2.0 1.8 Romania 0.3 1.2 1.7 1.8 1.7 Thailand 2.6 1.9 1.6 1.5 1.5 Pakistan 0.9 1.1 1.2 1.2 1.3 Sri Lanka 0.6 1.4 1.1 1.1 1.0 Republic of Korea 7.3 2.5 1.6 1.3 0.9 Malaysia 1.2 1.1 0.9 0.9 0.9 Philippines 1.6 1.3 1.0 0.8 0.8 Source: International Trade Statistics, 2006. Note: Clothing: Standard International Trade Classification Revision 3-84. 19 2.4.2. Liberalization of International Trade in Textiles and Clothing Three major shifts in the rules have governed the international trade of textiles and clothing during the past 30 years. From 1974 to 1994, the rules set in the MFA provided the parameters for bilateral negotiations of how quotas on textile and clothing trade were determined. Under the MFA, discriminatory quotas were allowed in areas where the increase in imports had the potential to cause domestic market disruptions. The European Union, Austria, Canada, Finland, Norway, and the United States applied quotas exclusively to developing country exports. With the advent of the WTO in 1995, the WTO Agreement on Textiles and Clothing (ATC) was designed to provide a transitional phase between the MFA and the full integration of the textile and clothing industry into the multilateral trading system. Under the ATC, Canada, the European Union, Norway, and the United States retained some quota restrictions until January 1, 2005, when the quotas on textile and clothing trade were lifted and replaced by tariffs only. Before the lifting of the quotas, a number of studies estimated the potential effects of liberalized international trade of textiles and clothing. Nordias (2004), for example, argued that China and India would come to dominate world trade. The share of China alone was predicted to reach more than 50 percent during the post-ATC period. Tables 2.10 and 2.12 indicate the rapid increase in the world share of China in both textiles and clothing. Although the world share of India has not shown significant enlargement thus far, India’s share in the world market will likely improve in the near future with the surge in cotton production because of the implementation of the Bt cotton program and the ongoing policy reforms in the textiles and apparel sectors in India (Bedi and Cororaton, 2008). Martin (2004) examined the possible effects of quota elimination on Pakistan and argued that improvement in productivity is the key issue if Pakistan is to gain shares in the world markets. This is because the international markets will be more price responsive after the abolition of the quota. This will present opportunities for suppliers with high productivity, whereas suppliers that lose competitiveness can expect to suffer losses in market shares. Thus, for Pakistan, Martin concludes that “raising productivity— either by improving the efficiency of the production process or the range and the quality of the products produced—is key to reaping the benefit from the abolition of the MFA.” The same implication may hold for India as well. Even with the abolition of the MFA, Pakistan’s exports of textile yarn, fabric, etc. that goes to the restricted markets have not declined relative to its overall exports of these items. In the data presented in Table 2.12, the share of Pakistan’s exports of textile fibers that go to markets of the European Union, United States, Canada, and Norway has declined from 34.4 percent in 2002 to 20.7 percent in 2006. This is due to Pakistan’s efforts to increase value added by processing fibers into yarn, fabric, garments, and textile made-ups as discussed in the next section. However, the shares of textile yarn, fabric, etc. and clothing and accessories remain high. The combined ratio increased from 52.9 percent in 1990 to 70 percent in 2005 and 68.6 percent in 2006. This indicates that Pakistan remains particularly competitive in some specific textile product lines. 20 Table 2.12. Pakistan exports of textiles and clothing to restricted markets, % Textile fibersa Textile yarn, fabric, etc.b Clothing and accessoriesc Combinedd 1990 28.9 43.9 88.4 52.9 1995 22.4 37.6 89.9 50.8 2000 24.9 49.6 90.1 61.4 2002 34.4 54.5 84.5 63.6 2004 16.1 55.9 86.7 65.1 2005 19.8 63.7 85.1 70.1 2006 20.7 58.7 88.7 68.6 Source: United Nations Commodity Trade Statistics. Note: Restricted markets include European Union, United States, Canada, and Norway. aStandard International Trade Classification Revision 3-26. This is the ratio of Pakistan exports of textile fibers to these markets and the overall Pakistan exports of textile fibers. bStandard International Trade Classification Revision 3-65. This is the ratio of Pakistan exports of textile yarn, fabric, etc. to these markets and the overall Pakistan exports of textile yarn, fabric, etc. cStandard International Trade Classification Revision 3-84. This is the ratio of Pakistan exports of clothing and accessories to these markets and the overall Pakistan exports of clothing and accessories. dThis is the ratio of Pakistan exports of fibers, textile, and clothing to these markets and the overall Pakistan exports of fibers, textile, and clothing. 2.5. Conclusion There are major developments in the world markets for cotton, textiles, and apparel. The increase in world production of cotton was largely due to the improvement in yield arising from improved inputs such as the expanded use of irrigation, fertilizers, chemicals, and the introduction of Bt cotton. The leading cotton-producing countries that have introduced the Bt cotton technology include China, India, and Mexico in the Northern Hemisphere, and Argentina, Australia, and South Africa in the Southern. There has been a notable expansion in cotton production in India since the implementation of its Bt cotton program in 2002. Although, recently, there are improvements in world cotton prices, historically prices have fluctuated wildly around a generally declining trend. Various studies have indicated that declining world cotton prices are not favorable to poor cotton-exporting countries. Several factors affect world cotton prices, including the improvement in productivity, the increase in the use of synthetic fibers, and subsidies from governments of developed countries. The world market for textiles and clothing is enormous and has been growing strongly. Recently, China has dominated the market. The world market share of the European Union is also substantial. As part of world trade liberalization, the MFA was dismantled at the start of 2005, rendering the world market for textiles and clothing more price responsive and competitive and presenting opportunities for supplies with high productivity. Suppliers that lose competitiveness can expect to suffer losses in market shares. 21 REFERENCES Baffes, J. 2004. Cotton: Market Setting, Policies, and Issues. (World Bank Policy Research Working Paper 3218) Washington, D.C.: The World Bank. Bedi. J., and C.B. Cororaton. 2008. Cotton-Textile-Apparel sectors of India: situation and challenges faced. (Draft Discussion Paper) Washington, D.C.: International Food Policy Research Institute. Centre for International Economics. 2002. Trade distortions and cotton markets: implications for Australia cotton producers. Canberra, Australia: Cotton Research and Development Corporation. Food and Agriculture Organization (FAO). 2004. Cotton: impact of support policies on developing countries, a guide to contemporary analysis. (Trade Policy Brief No. 1) Rome: FAO. Food and Agriculture Organization (FAO). 2006. Cotton Commodity notes. Available at: http://www.fao.org/es/esc/en/15/304/highlight_307.html. Gillson, I., C. Poulton, K. Balcombe, and S. Page. 2004. Understanding the Impact of Cotton Subsidies on Developing Countries. London: Overseas Development Institute. Goreaux, L. 2003. Prejudice caused by industrialized countries subsidies to cotton sector in western and central Africa. (Background document to the submission made by Benin, Burkina Faso, Chad, and Mali) Geneva, Switzerland: World Trade Organization. International Cotton Advisory Committee (ICAC). 2002. Production and Trade Policies Affecting the Cotton Industry. Washington, D.C.: ICAC. International Trade Statistics. 2006. Available at: http://www.intracen.org/tradstat/welcome.htm Martin, W. 2004. Textile and clothing policy note: implications for Pakistan of abolishing textile and clothing export quotas. (Manuscript) Washington, D.C.: The World Bank. Nordias, H. 2004. The global textile and clothing industry post the agreements of textiles and clothing. (Discussion Paper No. 5) Geneva, Switzerland: World Trade Organization. Orden, D., A. Salam, R. Dewina, H. Nazli, and N. Minot. 2006. The impact of global cotton and wheat markets on rural poverty in Pakistan. (Final project report prepared for the Pakistan Poverty Assessment Update) Manila, Philippines: Asian Development Bank Islamabad Resident Mission. Poonyth, D., A. Sarris, R. Sharma, and S. Shui. 2004. The impact of domestic and trade policies on the world cotton market. (Food and Agriculture Organization [FAO] Commodity and Trade Policy Research Working Paper No. 8) Rome: FAO. Reeves, G., D. Vincent, and D. Quirke. 2001. Trade distortion and cotton market: implications for Australian cotton producers. (Report) Canberra, Australia: Cotton Research and Development Corporation. Sumner, D.A. 2003. The impact of U.S. cotton subsidies on cotton prices and quantities: simulation analysis for WTO disputes. (Background paper) Brazil: World Trade Organization. Tokarick, S. 2003. Measuring the impact of distortions in agricultural trade in partial and general equilibrium. (Working Paper WP/03/110) Washington, D.C.: International Monetary Fund. Townsend, T. and A. Gutichounts. 1994. “A Survey of Income and Price SupporPrograms.” Beltwide Cotton Conferences, Proceedings, Cotton Economics and Marketing Conference, National Cotton Council, Memphis, TN. Townsend, T. 2002. Government measures and the world cotton industry. Paper presented at the 11th Australian Cotton Conference, Brisbane, 13–15 August. United Nations Commodity Trade Statistics, various issues. 22 3. PRODUCTION, PRICES, AND EMERGING CHALLENGES IN THE PAKISTAN COTTON SECTOR Abdul Salam 3.1. Introduction Pakistan is the fourth largest producer of cotton. Table 2.2 in Section 2 indicates that Pakistan contributed 8 percent to the world production of cotton in 2004–2006. It produces about 2.3 million tons of cotton. Cotton is the principal cash crop of Pakistan. It is second to wheat, which is the country’s staple food, in terms of area. Area annually planted under cotton is around 3 million hectares and accounts for 15 percent of the total cropped area. In 2005–2006, it contributed 24 percent to the total value added of agricultural crops (Pakistan Economic Survey, Statistical Supplement 2005–2006 [Government of Pakistan, 2006.]). Textiles, Pakistan’s largest industry and a major source of employment in manufacturing, depends on cotton farming for its supply of raw material. Cotton and its made-ups contribute 65 percent of the foreign exchange earned from merchandise goods. Increased cotton production in the recent past has helped in curtailing imports of edible oils, as cottonseed yields valuable vegetable oil for the domestic industry. It also provides feed for livestock and dairy farming. Cotton picking, a highly labor-intensive activity, is an important source of employment for rural women, providing supplementary income to rural farm and nonfarm households. In view of its extensive forward and backward linkages, the cotton crop occupies a unique position in the rural economy of Pakistan. Its performance holds the key not only for the growth and development of agriculture sector but also for the robust growth in the overall economy. A good cotton crop is imperative for the sustainable development of agriculture, food security, and poverty alleviation efforts at the micro- and macrolevels. We review the important features, basic characteristics, and performance of cotton—its production, marketing, and prices from 1991 to 2005—in this section to learn from experience and draw lessons for the future course of action. We present the historical data on cotton area, production, yields, the cotton harvest season, and cotton staple length in Section 3.2. We describe the salient features of cotton-growing farms in Section 3.3. We discuss the changes in cotton trading policy and marketing overtime and their impact on prices in Section 3.4. We examine the balance sheet of cotton, its composition, and changes in Section 3.5. We discuss behavior of cotton prices in the domestic and world markets in Section 3.6. The wedge between domestic and international prices and implications thereof for intersectoral resource transfers is also highlighted. We explore the implications of varying cotton incomes on its cultivation in Section 3.7. Emerging challenges in cotton cultivation in Pakistan, which include reducing variations in annual cotton production, the role of plant protection, and the need for IPM, the cultivation of Bt cotton and its requirements, contamination in cotton, and rising prices of inputs, are highlighted in Section 3.8. The section concludes with observations and recommendations. 3.2. Cotton Production 3.2.1. Overall Pakistan There are two major cotton-producing provinces in Pakistan. Table 3.1 indicates that the province of Punjab accounts for about 80 percent of total cotton crop area and total cotton production in the country, whereas the province of Sindh accounts for about 20 percent. The provinces of the Northwest Frontier Province and Balochistan have a combined share of less than 1 percent. 23 Table 3.1. Cotton area and production (% distribution) Area Production Average Punjab Sindh Others* Pakistan Punjab Sindh Others* Pakistan 1985–1990 76.1 23.8 0.1 100.0 84.3 15.7 0.0 100.0 1991–1995 82.2 17.7 0.1 100.0 86.2 13.8 0.0 100.0 1996–1999 79.1 20.5 0.4 100.0 75.9 23.7 0.4 100.0 2000–2005 80.1 18.7 1.2 100.0 77.1 22.0 0.9 100.0 Source: Calculated from the data reported in Agricultural Statistics of Pakistan 2004–2005 (Government of Pakistan, 2006). Note: *Northwest Frontier Province and Balochistan. Cotton is not a year-round crop. Based on the arrival of seed cotton over the period 1993–2005, Figure 3.1 shows that within the crop year between July and June, about 20 percent of the cotton harvest arrives in October. The arrival of cotton reaches a peak of above 30 percent of production in November but starts to slow down in December to 20 percent and further down to 10 percent in January. Figure 3.1. Arrival of seed cotton (average 1993–2005) 0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 July August September October November December January February March April May June P er ce n t Source: Pakistan Cotton Ginner's Association One of the important qualities of cotton is the staple length. Until 1980, Table 3.2 shows that nearly 80 percent of cotton types produced in Pakistan were of medium staple length. This staple length ranges between 1 and 1–3/16 inches (20.64 to 25.40 mm). From the 1980s onward, the cotton staple length improved to medium long, which ranges between 1–1/32 and 1–3/32 inches (26.19 to 27.78 mm). During this period, more than 50 percent of cotton production was medium long. In the first half of the 1990s, the share of medium staple length dropped considerably, whereas that of long staple (between 1– 1/8 and 1–5/16 inches or 28.57–33.34 mm) increased. However, from 1994 onward, the share of long staple has subsided and that of medium staple recovered. In this period, medium long staple dominates production. 24 Table 3.2. Staplewise production of cotton (% distribution) Period Short (<1-3/16 inches /20.64 mm) Medium (1–1-3/16 inches /20.64– 25.40 mm) Medium Long (1-1/32–1-3/32 inches/26.19– 27.78 mm) Long (1-1/8–1-5/16 inches/28.57– 33.34 mm) Total 1947–1970 13.6 82.5 7.4 0.4 100.0 1970–1980 6.2 77.5 15.3 1.0 100.0 1980–1990 2.7 26.4 55.4 15.5 100.0 1990–1991 1.0 4.1 59.5 35.4 100.0 1991–1992 1.0 9.0 61.1 29.0 100.0 1992–1993 1.0 3.1 73.8 22.2 100.0 1993–1994 1.1 23.4 71.4 4.1 100.0 1994–1995 1.3 20.1 73.0 5.6 100.0 1995–1996 0.9 53.5 45.4 0.2 100.0 1996–1997 1.1 37.1 58.9 2.9 100.0 1997–1998 1.2 14.0 72.0 12.8 100.0 1998–1999 1.2 37.2 59.0 2.7 100.0 1999–1900 0.0 33.0 63.0 4.0 100.0 2000–2001 0.0 12.4 83.4 4.2 100.0 2001–2002 0.5 28.5 67.9 3.2 100.0 Source: Pakistan Central Cotton Committee, 2006. Table 3.3 shows the performance of cotton in terms of crop area, production, and yield in Pakistan and in the two major cotton-producing provinces. For all of Pakistan, the total area under cotton was 2.66 million hectares in 1990–1991 and increased to 3.23 million hectares in 2004–2005. However, yearly fluctuations in response to various natural, economic, and other relevant factors explain the expansion in area. Before scaling new heights in 2004–2005, the maximum area of 3.149 million hectares was recorded in 1996–1997, whereas the minimum of 2.65 million hectares was reported in 1994–1995. The coefficient of variation for cotton area during the period 1991–2005 is estimated at 5.7 percent. The average annual growth rate in cotton area during this period is 0.8 percent and is statistically significant. 25 Table 3.3. Area, production, and yield of cotton in Pakistan All Pakistan Punjab Sindh Area Production Yield Area Production Yield Area Production Yield 000 hectares % change 000 bales % change Kilogram/ hectare % change 000 hectares % change 000 bales % change Kilogram/ hectare % change 000 hectares % change 000 bales % change Kilogram/ hectare % change 1990–1991 2,662 9,628 615 2,125 8,501 680 537 1,125 356 1991–1992 2,836 6.5 12,822 33.2 769 25.0 2,287 7.6 11,417 34.3 849 24.8 548 2.0 1,403 24.7 435 22.2 1992–1993 2,836 0.0 9,054 −29.4 543 −29.4 2,438 6.6 8,237 −27.9 574 −32.3 397 −27.6 816 −41.8 349 −19.7 1993–1994 2,805 −1.1 8,041 −11.2 487 −10.2 2,249 −7.8 6,523 −20.8 493 −14.2 555 39.8 1,517 85.9 465 33.0 1994–1995 2,653 −5.4 8,697 8.2 557 14.4 2,244 −0.2 7,410 13.6 561 13.9 406 −26.8 1,282 −15.5 537 15.5 1995–1996 2,997 13.0 10,595 21.8 601 7.8 2,463 9.8 8,720 17.7 602 7.2 529 30.3 1,862 45.2 598 11.5 1996–1997 3,149 5.1 9,374 −11.5 506 −15.8 2,540 3.1 7,103 −18.5 475 −21.0 601 13.6 2,250 20.8 636 6.4 1997–1998 2,960 −6.0 9,184 −2.0 527 4.2 2,348 −7.6 6,817 −4.0 494 3.8 600 −0.2 2,336 3.8 662 4.0 1998–1999 2,923 −1.3 9,790 6.6 569 7.9 2,283 −2.8 6,628 −2.8 494 0.0 630 5.0 2,134 −8.6 576 −13.0 1999–2000 2,983 2.1 11,240 14.8 641 12.5 2,329 2.0 8,804 32.8 643 30.2 634 0.6 2,377 11.4 637 10.7 2000–2001 2,928 −1.8 10,732 −4.5 623 −2.7 2,386 2.4 8,540 −3.0 608 −5.3 524 −17.4 2,141 −9.9 695 9.0 2001–2002 3,116 6.4 10,613 −1.1 579 −7.1 2,526 5.9 8,046 −5.8 541 −11.0 547 4.4 2,443 14.1 759 9.3 2002–2003 2,794 −10.3 10,211 −3.8 621 7.3 2,208 −12.6 7,664 −4.7 590 9.0 543 −0.7 2,412 −1.3 755 −0.5 2003–2004 2,991 7.1 10,048 −1.6 571 −8.1 2,387 8.1 7,702 0.5 549 −7.0 561 3.3 2,243 −7.0 680 −10.0 2004–2005 3,229 8.0 14,600 45.3 769 34.6 2,518 5.5 11,149 44.8 753 37.2 635 13.2 3,017 34.5 808 18.8 Mean 2,924 10,309 599 2,355 8,217 594 550 1,957 597 SD 166 1,652 82 125 1,451 104 71 602 143 C.V. % 5.7 16.0 13.7 5.3 17.7 17.5 13.0 30.8 24.0 Growth rate % 0.8 1.4 0.6 0.5 0.1 −0.4 1.4 6.8 5.4 t-statistics 2.9 1.7 0.8 1.6 0.1 −0.3 1.8 5.8 7.8 Minimum 2,653 8,041 487 2,125 6,523 475 397 816 349 Maximum 3,229 14,600 769 2,540 11,417 849 635 3,017 808 Source: Agricultural Statistics of Pakistan 2003–2004 (Government of Pakistan, 2003, 2004, 2005 and 2006); Long Term Trends in Area and Production of Crops, APCOM Series No. 204 (Agricultural Prices Commission of Pakistan, 2003); Pakistan Journal of Agricultural Economics, Statistical Appendix (2001); and Federal Bureau of Statistics, various issues. Note: Growth rates calculated from the coefficient of a time variable in the semilog equation ln Y = a + bt, where Y is the dependent variable; C.V. = coefficient of variation; SD = standard deviation. 26 The average cotton yield over the period was 599 kilogram/hectare with a coefficient of variation of 13.7 percent. The trend growth rate of cotton yield is 0.6 percent per year but is not statistically significant. The small coefficient of time trend in cotton yield also reflects the large variation in yearly cotton yield. The inadequacy of plant protection measures may partly explain this. In view of its susceptibility to a number of insect pests resulting in wide fluctuations in yield and potential output loss, cotton farming in Pakistan has become a high-risk proposition, involving heavy expenditures for pest control. During 1991–2005, cotton yield varied from 487 kilogram/hectare to 769 kilogram/hectare. The highest cotton yield of 769 kilogram/hectare was in the 1991–1992 crop year. It led to a record production of 2.18 million tons (12.82 million bales 5). This bumper harvest generated much optimism about the future prospects of cotton and related sectors in the country. However, the widespread breakout of CLCV in the subsequent years dampened this euphoria. The cotton yield touched its lowest level at 487 kilogram/hectare in 1993–1994. The production during this year was at its lowest level of 1.37 million tons (8.04 million bales). This sharp decline of 37 percent in production resulted entirely from the fall in yield, as the crop area, 2.81 million hectares, was only marginally less than that of 1991–1992. The low cotton crop prompted serious effort among policymakers, planners, and researchers to screen and develop varieties with resistance or tolerance to leaf curl virus. The authorities also launched a massive educational campaign to apprise growers of various (preventive and curative) measures to protect the crop and to obtain good production. Because of the interventions launched in the wake of severe CLCV infestation and poor yield in the 1990s and the hard work of farmers, cotton yield recovered. Nevertheless, it remains vulnerable to a host of insect pests. Sustained pest control intervention is, therefore, necessary; otherwise pest infestation will inflict heavy crop losses, and cotton yield will be subjected to high annual fluctuations. Cotton production subsequently recovered to 1.80 million tons (10.6 million bales) in 1995–1996 but fell again during the next two years and reached 1.561 million tons (9.2 million bales) in 1997–1998. However, during the next couple of years its production staged a recovery. Production reached 1.9 million tons (11.2 million bales) in 1999–2000. Falling production marked the next four years, owing to either contraction in area, fall in yield, or a combination of both. Then, an all-time production record of 2.48 million tons (14.6 million bales) was reached in 2004–2005. This represented a jump of 45.3 percent in production over the previous year. The improvement in production was due to a 34.6 percent increase in yield, from 571 kilogram/hectare in 2003–2004 to 769 kilogram/hectare in 2004–2005. The gains in yield were largely due to favorable weather and low pest infestation. Cotton area, which expanded by 8 percent, also contributed to the record production. Additional production arising from higher cotton yield will have a significant effect on farmers’ income even if prices are low. For example, notwithstanding a fall in the average domestic price of 27 percent over the level in 2003–2004, the additional production of seed cotton in 2004–2005 was worth rupees (Rs) 47.6 billion (US$0.79 billion). Out of this amount, Rs 40 billion (US$0.66 billion) is due to higher cotton yield largely arising from favorable weather condition and less pest infestation. This contributes to the income and well-being of farmers and others who depend on cotton production, especially the workforce engaged in cotton picking. Moreover, if the additional produce of cotton lint is valued at the average world price of Index B cottons of 51.20 US cents/pound in 2004–2005, it will translate to US$809 million. If the value of the additional cottonseed of $224 million (calculated at the domestic market price of Rs 375/40 kilogram) is included, the gross value of additional production will be US$1,033 billion. The contribution of a higher cotton yield is US$854 million. These figures show the first round and direct effects of higher cotton output in 2004–2005 to the economy, whereas their indirect and multiplier effects through the forward linkages may have additional effects on the overall national economy. Furthermore, there may be favorable effects on the balance of trade through both import substitution and higher exports. 5 In Pakistan, 1 bale is 170 kilograms. 27 To summarize, the average cotton harvest in Pakistan during the period of 1991–2005 is 1.75 million tons (10.3 million bales) with a coefficient of variation of 16 percent. The average annual growth rate in production for the past 15 cotton crop years is 1.4 percent. Given the wide fluctuations in cotton yield leading to high variability in production, the major challenge in cotton farming is how to stabilize the crop yield at a reasonable level (e.g., 650 kilogram/hectare). Fluctuations in cotton yield are due to changes in weather conditions and pest infestation. Whereas the former is beyond the control of farmers, intervention through pest control is imperative. A science/knowledge-based approach for adequate control of insect pests is therefore critical. Furthermore, improved pest scouting and dissemination of information among the farmers, along with judicious use of pest control methods and technology, can help save tremendous cotton yield reductions and prevent income losses. Pest control measures will also improve the quality of the product and bring higher prices. Thus, adequate pest control in cotton farming can make a valuable contribution in strengthening the economy and in alleviating poverty. Accordingly, it must be accorded high priority in the strategy of agricultural development in general and cotton production in particular. In addition, as will be discussed later in the section, producer incentives in the form of insurance/guarantees in producer prices at the harvesttime may be helpful, although the government cannot intervene much with world prices. 3.2.2. Punjab There are several cotton-producing districts within the Punjab province. Table 3.4 gives a list of major cotton-producing districts in Punjab with their average percentage contribution to the overall cotton production in the province. The top five cotton-producing districts are Rahim Yar Khan (13.9 percent of cotton production), Bhawalpur (13.0 percent), Vehari (10.4 percent), Bhawal Nagar (8.7 percent), and Lodhran (8.6 percent). Table 3.4. Major cotton-producing districts in Punjab (1996–2005 average % distribution of area and production) District Area Production Rahim Yar Khan 12.9 13.9 Bahawalpur 11.0 13.0 Vehari 10.0 10.4 Bahawalnagar 8.2 8.7 Lodhran 8.2 8.6 Khanewal 7.8 8.0 Muzaffargarh 7.6 6.8 Multan 6.9 6.4 Rajanpur 5.6 7.2 D.G. Khan 4.0 4.8 Pakpattan 3.1 2.2 Sahiwal 3.6 2.7 Okara 1.6 1.2 Jhang 2.6 1.7 T.T. Singh 1.9 1.3 Faisalabad 1.8 1.1 Layyah 1.7 1.1 Kasur 0.6 0.3 Others 1.0 0.6 Total 100.0 100.0 Sources: Averages calculated from the data were originally supplied by the Provincial Departments of Agriculture and reported in its price policy reports on seed cotton by the Agricultural Prices Commission of Pakistan (2004). 28 In view of the dominant shares of Punjab in area and production of cotton, its production performance has a major bearing on the overall situation in the country. The average area under cotton in the province over the period is 2.36 million hectares, with a coefficient of variation of 5.3 percent. The low value of the coefficient of variation is indicative of the small annual variation in area devoted to cotton production in the province. It also reflects the concentration of cotton cultivation in certain areas/regions, primarily the southern region of the provinces where cotton is the dominant crop. The trend growth rate in cotton area in Punjab is estimated at 0.5 percent per year. The cotton yield in Punjab during the period ranged from a minimum of 475 kilogram/hectare in 1996–1997 to a maximum of 849 kilogram/hectare in 1991–1992. The record cotton yield in the province in 1991–1992 resulted in the highest level of cotton production: 1.94 million tons (11.42 million bales) in the province during the 1990s. However, this could not be sustained in the subsequent years, as cotton farming in the province witnessed the emergence of leaf curl virus and other pests with devastating effects on production. As a result, cotton yield experienced wide swings over the period. The average yield for the period was 594 kilogram/hectare, with a coefficient of variation of 17.5 percent, indicating wide yield variations. The trend growth rate of cotton yield over the period was −0.3 percent. However, because of wide yearly fluctuations, this is not statistically significant. The production of cotton in the Punjab province has ranged from a minimum of 1.11 million tons (6.52 million bales) in 1993–1994 t