51710 Quality and Marketing of Cotton Lint in Africa Africa Region Working Paper Series No. 121 October 2008 Abstract T his paper presents an analysis of the objective outcome of cotton sector liberalization in Sub-Saharan Africa in terms of quality, marketing, and pricing. It gives an overview of the world cotton market with a particular focus on the increasingly stringent demand for quality, and provides a detailed comparative description and performance assessment of quality control (seed cotton grading and lint classification) methods and their impact on prices in nine Sub-Saharan African cotton-producing countries (Benin, Burkina Faso, Cameroon, Mali, Mozambique, Tanzania, Uganda, Zambia, and Zimbabwe), as reflected in the export price differentials. The paper examines the lint marketing strategies and how market signals are passed on to producers in the countries under review. It elaborates on some of the major opportunities and challenges faced by cotton sectors in Sub-Saharan Africa. The price of cotton lint is primarily linked to fiber characteristics but also to non-quality factors such as the way it is marketed internationally. African cotton has two comparative advantages in the world market: the intrinsic quality of its fiber and the fact that it is handpicked. However, the elimination of contamination of raw cotton by foreign matter stands out as the first priority for quality improvement. Sector structure has important repercussions for quality, and performance varies widely throughout SSA. Quality and marketing of cotton lint affect the producer price for seed cotton. Quality improvement could potentially increase lint prices by up to 15 percent, which could translate into a 30 to 40 percent increase in farmer prices. Thus, quality management should be considered one of the most important areas of improvement for SSA cotton exporting countries Author Affiliation and Sponsorship Gérald Estur, World Bank Consultant gmpestur@aol.com The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with potential for improving economic performance and social conditions in Sub-Saharan Africa. The Series publishes papers at preliminary stages to stimulate timely discussion within the Region and among client countries, donors, and the policy research community. The editorial board for the Series consists of representatives from professional families appointed by the Regions Sector Directors. For additional information, please contact Paula White, managing editor of the series, (81131), Email: pwhite2@worldbank.org or visit the Web site: http://www.worldbank.org/afr/wps/index.htm. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s), they do not necessarily represent the views of the World Bank Group, its Executive Directors, or the countries they represent and should not be attributed to them. COMPARATIVE ANALYSIS OF ORGANIZATION AND PERFORMANCE OF AFRICAN COTTON SECTORS QUALITY AND MARKETING OF COTTON LINT IN AFRICA By Gérald Estur October 2008 Foreword C otton, like all commodities, is differentiated by quality parameters for the purposes of trade. Cotton fiber is the raw material for the textile manufacturer who transforms cotton into yarn, and then into fabric for apparel, household goods, or industrial products. Cotton quality needs are quite varied depending on the final product, and the quality differences affect the price as well as the value that manufacturers will get from the cotton. Increasingly, quality and performance demands are being placed on the entire textile pipeline, from raw material to the end product. Cotton fiber represents about 50 percent of the cost of yarn. The price of cotton lint is primarily linked to fiber characteristics but also to non-quality factors such as the way it is marketed internationally. In the abundant literature that has been dedicated to cotton in recent years, there is no synthetic or operational document that can show what has been the objective outcome of cotton sector liberalization in Sub-Saharan Africa in terms of quality, marketing, and pricing. One of the rationales for cotton reform in Africa is to increase the linkage between world prices for lint and producer prices for seed cotton. Quality and marketing of cotton lint affect the producer price, which is by far the largest component to the cost of lint production in Africa. Therefore, quality management is one of the most important performance criteria for African cotton exporting countries. This study examines the quality control and marketing methods of lint in nine Sub- Saharan African cotton-producing countries (Benin, Burkina Faso, Cameroon, Mali, Mozambique, Tanzania, Uganda, Zambia, and Zimbabwe), their impact on export prices, and how market signals are passed on to producers. The objective of this study is to provide a comparative basis to assess the relevance of institutional structures that have emerged across the continent to deal with lint quality and marketing. The study attempts to analyze the link between structural reforms and sector performance by describing changes in institutional structures for quality control (seed cotton grading and lint classification), as well as quantifying changes in the quality of seed cotton and cotton lint, as reflected in the export price differentials. The report is organized in four sections. Section 1 presents key elements of the current world market context, including an overview of the world cotton market and a particular focus on the issue of cotton quality. Section 2 provides a detailed comparative description and performance assessment of quality control and its impact on prices for individual countries under review. Section 3 examines lint marketing strategies in Sub-Saharan African countries. Section 4 summarizes the studys main findings, draws some general lessons and presents some of the major opportunities and challenges faced by cotton sectors in Sub-Saharan Africa. i Acknowledgements T his study on lint quality and marketing performance in Sub-Saharan Africa was carried out by Gérald Estur, consultant, as part of the Comparative Analysis of Organization and Performance of Cotton Sectors in Sub-Saharan Africa: Lessons from Reform Experience, a study conducted by a World Bank team led by Patrick Labaste (Lead Agricultural Economist, SD Department, Africa Region, World Bank), and composed of David Tschirley (MSU), Colin Poulton (Imperial College London), Nicolas Gergely (consultant), John Baffes (DEC, World Bank) and Duncan Boughton (MSU). Interviews were conducted in September and October 2006 with representatives of major ginning companies and international merchants involved in the production and trade of cotton in the countries under review. Preliminary findings were discussed at a workshop held in Washington in November 2006. A second workshop was held in Washington in April 2007 to present the emerging findings from the comparative analysis. Additional information was collected between May and December 2007. The author wants to thank the various individuals in the cotton business, in SSA and elsewhere, who have provided advice and information for this study. The author would like to particularly acknowledge the inputs and contributions provided by Ms/Mssrs Rafiq Chaudhry and Armelle Gruère (ICAC), Axel Drieling (CSITC, Faserinstitut Bremen), Hervé Gruson (Sodecoton), Boubacar Diagne and Seïdina Oumar Diaby (CMDT), Augustin Zagre and Joël Rodolphe Ky (SOFITEX), Joe Kabissa (TCB), Pierre-Henry Deren (SOTRACOT), Damase Buchi (Dunavant), Hans Windsor Muzoora (CDO), Gabriel Paposseco and Licianía Raquel de Casatanheira e Cossa (IAM), Alidou Amadou Soulé (SONAPRA). This report has benefited from the findings of the country reports and the author is thankful to the peer reviewers - Patrick Labaste, Colin Poulton and David Tschirley (MSU) who have provided very useful guidance and feedback on the first draft of this report. The views expressed in this report are those of the author who is sole responsible for all errors and omissions. ii Contents ABBREVIATIONS ..................................................................................................................... V EXECUTIVE SUMMARY ..................................................................................................... VII 1. COTTON MARKET CONTEXT ................................................................................... 1 1.1 Overview of the World Cotton Market .................................................................. 1 1.1.1 World Cotton Supply and Demand .................................................................. 1 1.1.2 Cotton Trade .................................................................................................... 1 1.2 Increasingly Stringent Demand for Quality ........................................................... 5 1.2.1 Cotton Quality ................................................................................................. 5 1.2.2 Fiber ................................................................................................................. 6 1.2.3 Only Fiber ........................................................................................................ 8 1.2.4 More than Fiber ............................................................................................... 9 1.3 Impact of Quality on Cotton Prices ...................................................................... 10 2. QUALITY PERFORMANCE OF AFRICAN COTTON ........................................... 12 2.1 Quality of African Cottons..................................................................................... 12 2.1.1 Overview of the Cotton Sector in Sub-Saharan Africa .................................. 12 2.1.2 Characteristics of Cottons from SSA ............................................................. 14 2.2 Quality Control Systems and Performance by Country ..................................... 16 2.3 Impact of Quality on Export Prices ...................................................................... 19 3. MARKETING STRATEGIES OF AFRICAN COTTON .......................................... 23 3.1 Marketing Systems ................................................................................................. 23 3.2 Marketing Performance ......................................................................................... 25 4. OPPORTUNITIES AND CHALLENGES FOR AFRICAN COTTON ................... 27 4.1 Quality ..................................................................................................................... 27 4.2 Marketing ................................................................................................................ 30 ANNEX 1: PRIMER ON LINT CONTAMINATION .......................................................... 33 ANNEX 2: COTTON QUALITY BY COUNTRY ................................................................ 34 ANNEX 3: LINT MARKETING BY COUNTRY ................................................................. 50 ANNEX 4: REFERENCES ...................................................................................................... 59 Figures Figure 1: World Cotton Imports ................................................................................................. 2 Figure 2: Cotton Imports by Country - % of world imports (2006/07) ...................................... 3 Figure 3: World Cotton Exports ................................................................................................. 3 Figure 4: World Cotton Exports (2005/06 & 2006/07) .............................................................. 4 Figure 5: Cotton Exports by Country - % of world exports (2006/07) ....................................... 4 iii Figure 6: Cotton production in Sub-Saharan Africa ................................................................. 13 Figure 7: Average Fiber Length by Country in mm (2005/06) ................................................ 16 Figure 8: Estimated Premium for Top Type of Lint in 2006/2007 by Country ........................ 20 Figure 9: Estimated Premium for Top Type of Lint in 2006/2007 by Country ........................ 20 Figure 10: Estimated Premium for Top Type of Lint Mid-90s and 2006/2007 by Country....... 21 Figure 11: Estimated Premium Weighted Average Basis by Country ....................................... 23 List of Tables Table 1: Higher Grades Preferred Fiber Properties .................................................................. 11 Table 2: Comparison of Fiber Properties .................................................................................. 14 Table 3: Summary Indicators of Seed Cotton Quality Control Systems and Performance ...... 16 Table 4: Summary Indicators of Lint Quality Control Systems and Performance ................... 16 Table 5: Price Differentials for Grade (US cents per pound) ................................................... 22 Table 6: Price Differentials for Staple Length (US cents per pound) ....................................... 22 Table 7: Extraneous Contamination by Origin (2004-2005 averages) ..................................... 36 Table 8: Share of Lint based on Grade ­ Benin ........................................................................ 34 Table 9: Share of Lint based on Staple Length ­ Benin ........................................................... 34 Table 10: Indicators of Quality Performance - Benin ................................................................. 35 Table 11: Share of Lint based on Grade ­ Burkina Faso ............................................................ 36 Table 12: Share of Lint based on Staple Length ­ Burkina Faso ............................................... 36 Table 13: Indicators of Quality Performance ­ Burkina Faso .................................................... 37 Table 14: Share of Lint based on Grade ­ Cameroon ................................................................. 38 Table 15: Share of Lint based on Staple Length ­ Cameroon .................................................... 38 Table 16: Indicators of Quality Performance ­ Cameroon ......................................................... 39 Table 17: Share of Lint based on Grade ­ Mali .......................................................................... 40 Table 18: Share of Lint based on Staple Length ­ Mali ............................................................. 40 Table 19: Indicators of Quality Performance ­ Mali .................................................................. 41 Table 20: Share of Lint based on Grade ­ Mozambique ............................................................ 41 Table 21: Share of Lint based on Grade ­ Mozambique (2005/06)............................................ 42 Table 22: Share of Lint based on Staple Length ­ Mozambique (2005/06) ............................... 42 Table 23: Indicators of Quality Performance ­ Mozambique .................................................... 42 Table 24: Share of Lint based on Grade ­ Tanzania................................................................... 43 Table 25: Share of Lint based on Staple Length ­ Tanzania ...................................................... 43 Table 26: Indicators of Quality Performance ­ Tanzania ........................................................... 44 Table 27: Share of Lint based on Grade ­ Uganda (2006/07) .................................................... 45 Table 28: Share of Lint based on Staple Length ­ Uganda (2006/07) ........................................ 45 Table 29: Indicators of Quality Performance ­ Uganda ............................................................. 46 Table 30: Share of Lint based on Grade ­ Zambia ..................................................................... 47 Table 31: Indicators of Quality Performance ­ Zambia ............................................................. 47 Table 32 - Indicators of Quality Performance ­ Zimbabwe ....................................................... 49 Table 33: Indicators of Marketing Performance ­ Benin ........................................................... 50 Table 34: Indicators of Marketing Performance ­ Burkina Faso ............................................... 51 Table 35: Indicators of Marketing Performance ­ Cameroon .................................................... 52 Table 36: Indicators of Marketing Performance ­ Mali ............................................................. 53 Table 37: Indicators of Marketing Performance ­ Mozambique ................................................ 54 Table 38: Indicators of Marketing Performance ­ Tanzania ...................................................... 55 Table 39: Indicators of Marketing Performance ­ Uganda ........................................................ 56 Table 40: Indicators of Marketing Performance ­ Zambia ......................................................... 57 Table 41: Indicators of Marketing Performance ­ Zimbabwe .................................................... 58 iv ABBREVIATIONS AFIS Advanced Fiber Information System AFZ African Franc Zone BPA Bukalasa Pedigree Albar (Uganda) CDO Cotton Development Organization (Uganda) CDT Cotton Development Trust (Zambia) CFDT Compagnie Française pour le Développement des Fibres Textiles CFR Cost and Freight CIF Cost, Insurance, and Freight CMB Cotton Marketing Board CMDT Compagnie Malienne pour le Développement des Textiles (Mali) CSITC Commercial Standardization of Instrument Testing of Cotton cts/lb Cents per Pound DAGRIS Développement des Agro-Industries du Sud DPCQ Direction de la Promotion de la Qualité et du Conditionnement des produits agricoles (Benin) ECGA Eastern Cotton Growing Area (Tanzania) ELS Extra Long Staple ESA East and Southern Africa est. Estimated EU European Union Euro FOB Free on Board FOT Free on Truck GM Good Middling GO Good Ordinary GOT Ginning outturn ratio GPT Gram per Tex HVI High Volume Instrument IAM Instituto de Algodão de Moçambique (Mozambique) ICA International Cotton Association ICAC International Cotton Advisory Committee ICE InterncontinentalExchange (formerly IRMA Institut de Recherche de Maroua (Cameroon) ISO International Organization for Standardization ITMF International Textile Manufacturers Federation kg Kilogram LS Light Spotted M Middling mike Micronaire mm Millimeter mt Metric ton N/A Non available Ne English yarn count NTA NTarla (Mali) OE Open End v PE Polyethylene PP Polypropylene RCI République de Côte d'Ivoire RG Roller Gin SG Saw Gin SGS Société Générale de Suveillance SITC Standardized Instrument for Testing of Cotton SJV San Joachin Valley (California) SLM Strict Low Middling SM Strict Middling SODECOTON Société de Développement du Coton du Cameroun (Cameroon) SOFITEX Société Burkinabé des Fibres Textiles (Burkina Faso) SONAPRA Société Nationale pour la Promotion Agricole (SONAPRA) SSA Sub-Saharan Africa STAM Station Anie Mono (Togo) TCA Tanzanian Cotton Association TCB Tanzanian Cotton Board UK Ukiriguru (Tanzania) U.S. United States UEMOA Union Economique et Monétaire Ouest-Africaine UNCTAD United Nations Conference on Trade and Development US$ U.S. dollar VLS Very Light Spotted WCA West and Central Africa WCGA Western Cotton Growing Area (Tanzania) WIS Wakefield Inspection Services WP World Price vi EXECUTIVE SUMMARY C otton fiber represents about 50 percent of the cost of yarn and there is a direct correlation between specific quality characteristics of the fiber and those of the yarn. Traditionally, the price of cotton was largely determined by factors such as staple length, grade, color and micronaire. Those factors are still the major determinants of price but spinners today are also interested in other fiber properties that affect the quality of their yarns and the efficiency at which they produce those yarns. As the textile industry has been striving to improve quality and efficiency through automatic high-speed machinery, new technologies place increasingly severe technical demands on textile fibers, raising the importance of other properties of cotton: strength, uniformity, maturity, fineness, elongation, neps, short fiber content, spinning performance, dyeing ability and cleanliness. All else being equal, spinners pay a higher price for longer, finer, and stronger cotton lint that is white, bright, and fully mature. Fiber matching all the required properties would receive a very significant premium over the industry standard price (tied to the Cotlook A Index,) as high as 10 cents/lb, assuming low contamination and excellent marketing and shipping conditions. Cotton is produced in 34 of the 42 SSA countries, with 30 countries exporting their output. Production increased from 616,000 tons in 1980/81 to a record of 1.75 million tons in 2004/05, accounting for 6.5 percent of the world output and 16 percent of world exports. 85 percent of SSAs production is exported. However, the expansion in African production came mostly from increases in area devoted to cotton, especially in the CFA Franc zone, rather than from rising yields like in the rest of the world. SSA is a marginal player in terms of processing of cotton At the national level, African cotton is relatively homogeneous in fiber characteristics, due to similar growing conditions and the low number of varieties planted in most countries, despite the fact that cotton is grown without irrigation by small farmers. However, variability within bales is greater than for developed countries because the production of several farmers can be mixed in a single bale. African cotton has two comparative advantages in the world market: the intrinsic quality of its fiber and the fact that it is handpicked and should consequently have lower contamination with vegetative matter. While most African cotton is suitable for the medium-high level of ring spinning, progress on quality during the last decade has been slower than in competing origins and not as fast as required by the spinning industry. Due to fears regarding the contamination of raw cotton by foreign matter African handpicked cotton now trades at a discount to its machine-picked cotton competitors. The elimination of contamination stands out as the first priority for quality improvement, as it can potentially increase lint prices by up to 15 percent. However, a reputation for high contamination is difficult to overcome. Technical solutions to eliminate contamination are well-known and should be relatively easy to implement by the producers themselves, provided that producers are motivated by appropriate financial incentives and properly trained to the recommended methods. However, getting ginning companies to take those steps in highly competitive systems has proven quite difficult. A rough cost/benefit analysis of a quality improvement program indicates that removing contamination through screening prior to ginning vii costs less than one cent per pound, whereas uncontaminated cotton would potentially fetch a premium of at 5 cents per pound. Fair trade and organic cottons can provide an opportunity for small ginners working in close connection with producers to meet the required specifications for those niche markets. Still, the market potential for traditional non-contaminated cotton lint is much larger. While most African cotton is saw ginned, roller ginning produces lint that is longer, has fewer short fiber, seed coat fragments and neps. As a result, roller ginned upland cottons fetch a premium of 1 to 2 cents per pound over the price of the same cottons ginned with saws. A detailed costs/benefits analysis should be undertaken to assess whether roller ginning is a valid alternative to saw ginning. African cotton sectors must focus on making dramatic improvements in the reliability of seed cotton grading. The critical control point in the marketing chain as far as quality is concerned is the point of primary marketing. Most African cotton is classed through visual and manual inspections and instrument- classing data are available in some countries on a sample basis. Instrument testing of cotton is gradually replacing hand classing the world over. Instrument testing will enable SSA countries to sell their cotton based on reliable and comparable test results, avoid price discounts and claims, and secure or even improve their market share. Sector structure varies widely throughout SSA and has important repercussions for quality. Repeated defaults or the shipment of inferior quality by some will eventually affect all the exporters and growers of an entire country. Licensing of ginners and exporters is therefore a useful disciplinary tool if the process is transparent and based on clear and equitable rules. Panterritorial, panseasonal price mechanisms pass extremely high levels of financial risk to cotton companies, who are often ill-equipped to monitor and manage those risks. The development of direct sales from ginners to spinners through commissioned agents should be encouraged as a way to save the cost of intermediation by merchants and acquire unbiased market knowledge. The existence of different export channels could be an advantage for farmers who will benefit from some degree of competition. However, consolidation of export volumes should be encouraged as efficient marketing requires the ability to supply several thousands tons of consistent quality throughout the seasons. Properly run auctions provide excellent price transparency but depend on competition. A potential tool for improving the marketing of African cotton is electronic trading. In sum quality improvement could translate into an increase in selling prices of up to 10 US cents per pound of lint over the Cotlook A Index. At typical producer prices of 25 to 32 US cents per kg of seed cotton, an increase of 10 cents per pound in the price of lint that is fully passed on to farmers would increase farmer prices by 30 to 40 percent. As a result, quality management should be considered one of the most important areas of improvement for SSA cotton exporting countries. Within SSA Zimbabwe and Zambia receive the highest premiums and Tanzania and Mozambique the lowest. Over the last decade, premiums increased in Zambia (+ 5 cents/lb), Cameroon (+ 4.5 cents/lb), Mozambique (+ 2 cents/lb), and Burkina Faso (+1 cent/lb). Progress made in Zambia and Cameroon is clearly due to the successful reduction in contamination and stickiness. In contrast, differentials vis-à-vis the viii Cotlook A Index declined by 1.5 cents/lb in Tanzania, 1 cent/lb in Uganda and Zimbabwe, and 0.5 cents/lb in Benin and Mali. The declines reflect increased competition between ginners in all three ESA countries, and lax seed cotton grading and contamination in the two WCA countries. Based on the actual quality produced, average export price differentials range from minus 2 to plus 4 cents per pound with respect to the Cotlook A Index in 2006/07. Zambias concentrated system delivers the best performance. Zimbabwes premium is still high, but falling as a result of increased competition. Cameroon, with a national monopoly largely free of political meddling, also performs well, though not up to the standards of Zambia. The other national and local monopolies show a highly variable performance, while Tanzanias competitive system is, along with Mozambique, the worst quality performer. ix 1. COTTON MARKET CONTEXT 1.1 Overview of the World Cotton Market 1.1.1 World Cotton Supply and Demand Cotton is an annual crop, grown in more than 100 countries on about 2.5 percent of the world's arable land. Cotton is also a heavily traded agricultural commodity, with over 150 countries involved in exports or imports of cotton. World cotton production and consumption are trending higher, and the industry is being transformed by new technologies, particularly biotechnology. The world cotton industry has experienced dramatic changes over the last five decades. Production increased fourfold, rising from 6.6 million tons in 1950/51 to surpass 27 million tons in 2004/05, and has remained nearly as high in the following two seasons. With the total area under cotton stable, all the growth in world cotton production since the 1940s came from improved yields. This can be seen particularly in the world production share of the four largest producing countries (i.e. China, the United States, India, and Pakistan) which rose from 48 percent in 1970/71 to 72 percent in 2006/07. World textile fiber consumption is driven by three major economic variables: income, population growth, and fiber prices. World final demand for textile fibers has increased from 7.6 million tons in 1950 to 56 million tons in 2004. The January 2005 elimination of quotas that limited trade in textile and apparel for more than 30 years is leading to a shift in textile and apparel production toward China and other developing countries. Cotton consumption is boosted by the resulting lower retail prices for textile and apparel products. Cotton consumption per capita has been almost constant since 1960, while total textile fiber consumption per capita more than doubled. As a result, cottons share of world textile fiber use fell from more than 70 percent in the 1950s to about 40 percent by 2005. World mill use of cotton rose from 6 million tons in the early 1950s to 26 million tons in the mid-2000s. Mill consumption grew at an average annual rate of 6 percent since 1998/99. The bulk of the increase occurred in China, but important expansions were also registered in India, Pakistan, and Turkey. As a result, processing of cotton continued to shift from developed countries to developing countries and their share of world mill consumption rose from 58 percent in 1980/81 to 91 percent in 2006/07. The share of the four largest consuming countries (i.e. China, India, Pakistan, and Turkey) in world consumption reached 71 percent in 2006/07, up from 53 percent in 1980/811. Chinas share of world mill use rose to 41 percent in 2006/07, sharply up from 23 percent in 1998/99. 1.1.2 Cotton Trade Among the top seven cotton producing countries, only Uzbekistan does not also rank among the top seven consuming countries. An average of 31 percent of world output 1 China, USSR, India and the USA were the top 4 consuming countries in 1980/81. 1 was traded from 1980/81 through 2004/05. The share rose to 39 percent in 2005/06 but fell back to 31 percent in 2006/07. Relative to other agricultural goods, the cotton export market is concentrated with an unequal distribution of market share among exporters. The cotton import market was previously less concentrated2 but a significant structural change occurred with the sharp increase in Chinese imports. The world cotton trade is not highly concentrated by the standards of industrial markets and the international cotton shipping industry is highly competitive. Some 500 firms are engaged, at least in part, in the cotton trade and the 20 largest cotton organizations handle about one-third of world production. World imports rose dramatically from 5.7 millions tons in 2000/01 to an estimated 9.7 millions tons in 2005/06 due to the widening gap between demand and production in China. However, imports declined to 8.2 million tons in 2006/07. Figure 1: World Cotton Imports Source: ICAC To feed the expansion of its textile industry, Asia has become the leading importing region. China was a net exporter of cotton from 1998/99 through 2000/01 but became a net importer in 2001/02. Chinese imports skyrocketed from 50,000 tons in 2000/01 to 4.2 million tons in 2005/06. China accounted for 43 percent of world imports in 2005/06. However, as Chinese imports dropped to 2.3 million tons in 2006/07, their share in world imports fell to 29 percent. Conversely, as world industrial consumption of cotton is shifting to developing countries, the contribution of developed countries to total cotton imports dropped from 71 percent in 1984/85 to 12 percent in 2005/063. 2 Meaning with a more equal distribution of market share between importers 3 14% in 2006/07. 2 Figure 2: Cotton Imports by Country - % of world imports (2006/07) Source: ICAC World cotton exports reached 3.6 million tons in 1926/27 and that level was not surpassed before the early 1950s. Exports climbed to a new record high of 9.7 million tons in 2005/06 but declined to 8.1 million tons in 2006/07. The United States plays the leading role in cotton exports. Raw cotton was the largest merchandise export for the U.S. from 1803 through 1937, and the U.S. has been the largest cotton exporter since 18344. Figure 3: World Cotton Exports Source: ICAC 4 Exception: 1985/86 3 Due to the rapid decline of the U.S. textile industry since 1997/98, the share of US. production destined for export increased from 40 percent to 75 percent in 2006/07. U.S. exports reached a record 3.82 million in 2005/06, with a market share of 39 percent, well ahead of Uzbekistan (1 million tons), India (750,000 tons), Australia (630,000 tons), Brazil (430,000 tons), and Greece (355,000 tons). Total exports from SSA countries5 amounted to 1.5 million tons, or 15 percent of world exports, including 1 million tons from the CFA zone. World exports fell by 16 percent in 2006/07. Exports declined in the U.S. (-26 percent), Uzbekistan (-4 percent), Australia (- 26 percent), Brazil (-34 percent), and SSA (-15 percent). In contrast, Indian exports increased by 28 percent. Figure 4: World Cotton Exports (2005/06 & 2006/07) Source: ICAC Figure 5: Cotton Exports by Country - % of world exports (2006/07) Source: ICAC 5 Including ~100,000 tons exported to other SSA countries 4 Yearly variations in supply can cause wide fluctuations in price. The estimated nominal value of world exports dropped from a high of US$13 billion in 1994/95 to US$6 billion in 2001/02 and rebounded to approximately US$12 billion during marketing season 2005/06 (August-July). Nevertheless, cotton represents a very small share of the world trade in terms of value. In UNCTAD export statistics (by product), cotton ranked 170th by average 2004-2005 values, accounting for 0.1 percent world product exports in 20056. 1.2 Increasingly Stringent Demand for Quality 1.2.1 Cotton Quality Like all commodities, cotton is differentiated by quality parameters for the purposes of trade. Cotton fiber is the raw material for the textile manufacturer who transforms cotton into yarn and then into fabric for apparel, household goods, or industrial products. ISO defines quality as "the totality of characteristics of an entity that bear on its ability to satisfy stated and implied needs." A quality cotton is one that meets the needs and expectations of its customers: the spinners. Buyers expect to receive the quality they bought, nothing less will do7. All cottons are not created equal. There are many different types of cottons, referred to as growths or varieties, and many other characteristics that determine the use for the cotton, and subsequently its value. The prices of cottons with similar fiber properties tend to be equivalent at a particular location and time. Cotton quality requirements can vary substantially depending on the final product, and the quality differences affect the price that manufacturers are prepared to pay and the value that they can get from the cotton lint. Price differentials are quite wide, with a ratio of about 1:4 between the lowest and the highest quality. Increasingly, quality demands are being placed on the entire textile supply chain, from the raw material to the end products. Thus, the importance attached to lint quality by international spinners has risen in recent years. Chemical fiber performance has become a benchmark by which spinner judge cotton. Consequently, the main challenge for cotton has become competing with artificial fibers, mainly polyester, on both price and quality. Making this an even greater challenge is the fact that chemical fibers are generally easier to process, more versatile, and stronger than cotton fiber. Additionally, modern textile industry machinery requires the same characteristics of cleanliness and homogeneity from cotton fiber as those offered by artificial fibers. As a result, certain quality attributes have assumed increasing importance in recent years, alongside conventional parameters. 6 US$11.4 billion. 7 In this context ,,low quality does not mean the same as ,,substandard, which denotes that the quality is below a given or generally accepted standard. If buyer and seller are agreed on a particular quality and the seller supplies exactly that, then the term substandard does not apply whether the quality is low or not. 5 In short, the increasingly stringent demand for quality cotton can be articulated in the following motto: "fiber, only fiber, but more than just fiber." 1.2.2 Fiber As cotton is a natural and seasonal product, characteristics such as its intrinsic quality (the fiber properties), cleanliness and contamination, as well as the cotton's homogeneity can vary greatly due to genetic, environmental, harvesting, and ginning factors. Such variability impacts processing performance, costs, quality, and utilization throughout the entire cotton textile chain, from the farm to the end-product. In contrast, all synthetic fibers within a given lot are identical and totally free of contamination. Fiber properties primarily depend on the varieties grown, agro-climatic conditions, and crop management practices. Variety is the most important factor. Variety determines nearly all of the lint quality parameters and most of the agronomic ones. The environment restricts the varietal potentiality in the expression of fiber properties. Climatic conditions and cultural practices are critical for length parameters, maturity, and strength. Cotton fiber quality is at its best on the day the cotton boll opens. Nothing can be done afterwards to improve the quality of cotton but all subsequent operations can alter it for the worse. Better fiber quality translates into better yarn quality and higher processing efficiency. Among the fiber properties, staple length has the greatest influence on spinning performance. Both fiber fineness and fiber strength are generally correlated with staple length. According to the International Cotton Advisory Committee (ICAC), the world cotton supply can be divided into six categories based on commonly perceived competitive relationships between cottons of differing quality, variety, and geographic origins: extra-fine, fine, high-medium, medium, coarse count, and waste/padding. The categories are roughly parallel to staple length categories8 but are designed to incorporate more than just staple length information because two cottons of equal length might actually have significantly different spinning characteristics. Extra-fine, fine, and high-medium cottons are typically used in ring spinning for the production of combed yarn9. Medium cotton is typically used in ring spinning for the production of carded yarn10. Coarse count cotton is typically used for producing open- end yarn11. 8 Short (under 13/16"); medium (13/16 to 1"); medium-long (1-1/32" to 1-3/32"); long (1-1/8" to 1- 11/32") and extra-long (1-3/8" and longer). 9 Combed yarns are stronger, more uniform, smoother, purer and have greater shine than carded yarn. Medium combed yarns are used by circular knitters and hosiery manufacturers. Fine combed yarns, produced with the worlds finest cottons, are used in higher-end sheeting and shirting fabrics. 10 Ring Spun Carded Yarn is typically used for knitting and weaving, with a complete range of coarse to fine counts. 11 "Open End" refers to the system of rotor spinning based on twisting the yarn without turning the bobbin. Open end used are typically used to produce denim jeans as well as socks, woven clothing items and blankets. 6 The most commonly produced and traded cotton lint variety in the world belongs to the species Gossypium hirsutum, which is also known as upland cotton. Extra Long Staple (ELS) cotton used for producing very fine yarns12 come from another species13 and account for less than 5 percent of world cotton trade. Following the global trend toward improving yarn quality, the market share of medium and higher grades is rising, while the share of shorter ("coarse count") upland cotton is declining. Medium and higher grades of upland cotton now account for an estimated 75 percent of world trade, or some 7 million tons, and are typically used in ring spinning. In ring spinning, staple length is the most important fiber property, followed by strength and fineness. In rotor spinning, strength is the most important fiber property, followed by fineness and length. Ginning can affect lint quality. The gin stand, whether saw or roller, pulls the fiber from the seed and is the heart of the ginning system. The capacity of the system and the quality and potential spinning performance of the lint depend on the operating condition and adjustment of the gin stand. If gin stands are operated above the design capacity, the quality of the cotton may be reduced. Short fiber content increases as saw speed increases, causing yarn imperfections. Roller ginning, when compared to saw ginning, produces a superior fiber with excellent spinning potential because it does the least amount of damage when separating the fiber from the seed. Roller ginning has better outturn14 and produces lint that is longer15, has fewer short fibers, seed coat fragments and neps. However, roller ginning is a slower and more expensive16 process that can only be used for ELS cottons and medium-long17 upland varieties. Roller ginned lint contains more trash as saw ginned lint is cleaner, but contains finer particles of trash that are more difficult to remove. Cotton classification is the application of official standards and standardized procedures developed for measuring the physical attributes of raw cotton that affect the quality of the finished product and the processing efficiency. Those attributes include color18 and leaf grade19, preparation20, fiber length21, length uniformity22, fiber strength23, 12 Ne 80 and above. 13 Gossypium barbadense. 14 1 to 2% higher ginning outturn ratio. 15 1-1/32 to up to 1-1/16 inch staple length longer. 16 Higher electricity and maintenance costs 17 1-1/8 inch and above. 18 Color is determined by the degree of reflectance (Rd) and yellowness (+b). Reflectance indicates how bright or dull a sample is and yellowness indicates the degree od color pigmentation. 19 A combination of color, cleanliness and preparation of lint. 20 Preparation refers to the degree of smoothness or roughness with which the cotton is ginned, and the relative neppiness of the ginned lint. 21 Staple length is the average length of the longer one-half of the fibers (Upper Half Mean Length ­ UHML). It is reported in both 100ths 32nds of an inch. 22 Length uniformity is the ratio between the mean length and the upper half mean length of the fibers. 23 The strength reported on terms of grams per tex is the force in grams required to break a bundle of fibers one tex unit in size. The tex unit is equal to the weight in grams of 1,000 meters of fiber. 7 elongation, micronaire24, trash content, and extraneous matter content. The grade of cotton25 is mainly associated with harvesting methods, storage and ginning practices. 1.2.3 Only Fiber Cotton prices are not solely determined by intrinsic fiber properties. The cleanliness of lint refers specifically to the presence of vegetal matter other than lint (trash), while contamination refers to the presence of non-plant matter (foreign matter). Both cleanliness and contamination depend on harvesting methods, handling, storage, transport, and ginning practices. In contrast, chemical fibers are totally free of contamination. Contamination of lint by non-vegetable foreign matter is the most serious problem confronting cotton spinners around the world. Foreign matter, short fiber content26and contamination are the outcome of growing, harvesting, and ginning practices. Fibrous neps27 and seed coat fragments28 result from ginning and are mostly variety dependent. In contrast, synthetic fibers are totally free of contamination. Contaminated cotton causes disruptions in the spinning process which increases the cost of spinning and reduces the quality of yarn and end products. With modern automated high speed textile machinery, foreign matter cannot be spotted until yarn or fabric is dyed and there are no cost-effective means of removing contamination. As a result, contamination leads to the downgrading of end products or even to the rejection of an entire lot. Cotton that is contaminated, or that is suspected of being contaminated because of the origin, can only be sold at substantial discount to compensate the user for inspecting and cleaning the cotton before spinning. Price differentials for cotton with the same fiber characteristics range from 5 percent to 30 percent, depending on their degree of perceived contamination by extraneous matter, stickiness, and seed coat fragments. These discounts are usually applied indiscriminately to all cotton originating from an area or a country considered to be affected by contamination. Contamination by foreign matter is more serious with handpicked cotton. Seed cotton picked by hand is cleaner, and the fiber obtained has fewer neps and a lower short fiber content than cotton picked by machine, which must be cleaned more vigorously because it has more plant residues. Handpicked cotton should therefore normally be purchased at a premium over machine picked cotton. However, handpicked seed cotton often gets contaminated during picking, storage, handling, or transport, and the presence of foreign matter in the fiber offsets the theoretical advantage conferred by manual picking. As contamination of raw cotton by foreign matter is the main concern for quality yarn and fabric producers, spinners tend to prefer machine picked cotton to handpicked cotton. As a result, handpicked cotton has lost its advantage over the past 24 Micronaire is a measure of fiber fineness and maturity. 25 Grade is not an intrinsic value; it is a commercial value based on a visual assessment of lint color, trash content and preparation. 26 SFC is defined as percentage of fibers shorter than ½- inch in length. SFC has a very high correlation with the Lower Half Mean Length. 27 Neps are cotton fibers tangled into a knot. Neps in the lint are undesirable because they appear as defects on the yarn and fabrics. 28 Also known as seed coat neps. 8 25 years and now trades at a discount to machine picked cotton. Some quality- conscious mills even refuse on principle to buy any manually harvested cotton to avoid the risk of quality claims downstream. 1.2.4 More than Fiber Along with the fiber characteristics, some other criteria, like reputation and other marketing characteristics generally not included in the contracts, can have a lasting influence on cotton prices. However, non-quality premiums and discounts are difficult to quantify because each shipper and spinner may have a different opinion on a specific growth or origin. Pricing of lint is significantly influenced by the way cotton is marketed and shipped. The spinning industry today is especially concerned about consistency in shipments. Customers require homogeneous and reliable year-round shipments, with consistent cotton characteristics, standardized bales (same size and density) wrapped in cotton cloth. The homogeneity of deliveries depends on seed cotton grading, lint classification, and bale allotments. The traditional method of cotton classification is through manual grading. Based on appearance and feel, manual grading includes determinations of color grade, leaf grade, staple length, preparation, and the identification of foreign or extraneous matter. These determinations are made by trained cotton classers based upon visual comparisons with physical and descriptive standards. As manual classing depends on human perceptions of sight and touch, it is somewhat subjective in nature. Several important quality attributes cannot be measured manually or visually. Manual classing methods are not harmonized and results are commonly rated not to be sufficiently reliable and precise. Despite those limitations, manual classing is still the actual base for trading cotton. However, with the increasingly stringent demand for quality, it is becoming more important to obtain reliable and complete information about the actual quality of cotton lint. Cotton growers use such data to obtain a fair price for their cotton and to improve cotton quality through breeding and research. Ginners use them to optimize the ginning process. Traders need them to price the cotton and to satisfy their customers. Finally, they benefit spinners by helping to ensure smooth running processes, to produce the required quality of yarn, and to minimize the raw material costs. Cotton testing instruments have been used on a small-scale basis by researchers and spinners since the 1960s. However, it was not until the early 1990s that the cotton industry began to accept instrument classification on a wide-scale basis. Current instrument classification includes measurements for upper-half-mean length, length uniformity index, fiber strength, micronaire, color Rd/+b, and percent area trash. An estimated 30 percent of world produced bales are instrument classed today, and the percentage is expected to increase to more than 60 percent in the next five years according to ICAC. The global cotton trade is moving towards standardization and harmonization through the use of high-volume instrument testing to satisfy customer demands. The task force for commercial standardization of instrument testing of cotton (CSTIC) was formed in December 2003 to establish a worldwide acceptable, adoptable, and reliable 9 instrument-based cotton quality assessment system for commercial purposes. The CSTIC task force is encouraging trading of cotton based on instrument data and supporting the introduction of instrument test results in trading and arbitration rules. Australia, Brazil, China, Uzbekistan, and the U.S. have fully implemented or are very close to fully implementing instrument classification on 100 percent of their cotton production and can offer bale per bale Standardized Instrument for Testing of Cotton (SITC) data. Thus, the lack of reliable cotton quality data for each bale negatively impacts the price of cottons that is classified manually. The spinning industry today is especially concerned with the consistency of lay downs29. Availability of bale per bale instrument testing30 data prior to shipment helps merchants customize their deliveries to their buyers needs, and allows spinners to plan lay downs without having to wait for the arrival of the cotton. Other positive non-quality criteria include same gin shipments with bale selection and container loading at the gin, short transit time, and good condition of bales upon arrival. In marketing, perception is often more important than fact. Trust and reputation matter in the cotton business and the market rewards origins and shippers that have a strong record of delivering certain quality characteristics with consistency, while respecting contract terms. Premiums and discounts attached to internationally traded cotton derive partly from the reputation of national origins. 1.3 Impact of Quality on Cotton Prices Cotton fiber represents about 50 percent of the cost of yarn and there is a direct correlation between specific quality characteristics of the fiber and those of the yarn. Traditionally, the price of cotton was largely determined by factors such as staple length, grade, color, and micronaire. Those factors are still the major determinants of price but spinners today are also interested in other fiber properties that affect the quality of their yarns and the efficiency at which they produce those yarns. As the textile industry has been striving to improve quality and efficiency through automatic high-speed machinery, new technologies place increasingly severe technical demands on textile fibers, raising the importance of other properties of cotton: strength, uniformity, maturity, fineness, elongation, neps, short fiber content, spinning performance, dyeing ability, and cleanliness. All else equal, spinners pay a higher price for longer, finer, and stronger (more resistant) cotton lint that is white, bright, and fully mature. As spinners run their mills throughout the year, they are usually willing to pay a premium to guarantee the supply of cotton they prefer to use as a regular component in their blends. Customers want cotton shipments to be homogeneous from the first bale of a sale to the last, with 29 Set of bales of cotton laid down together in the opening room of a spinning mill. 30 HVI is a registered trademark, USTER. The ICAC Task Force on Commercial Standardization of Instrument Testing of Cotton has adopted the acronym SITC (Standardized Instrument for Testing of Cotton). 10 consistent characteristics, free of contamination, and wrapped in cotton cloth with bale per bale instrument classification data. The recognized benchmark for international cotton prices, the Cotlook A Index, is based on the representative offering price for a "basket" of the medium grade cotton most commonly traded internationally. Those quotations refer to a common quality, contractual, and geographical31 basis: Middling 1-3/32" with standard micronaire (3.5 - 4.9) and strength (25-30 gpt). Terms: Cost and freight Far East, payment by Letter of Credit at sight, including 1 percent agents commission. Lint of this quality is typically used in ring spinning for the production of ring spun carded yarns32. The daily Index on a given day is the average of the cheapest five quotations from a selection of 18 medium grades, including four from WCA (Benin, Burkina Faso, Côte dIvoire, Mali) and one from ESA (Tanzania SG1). The spread between the cheapest and the highest quotations is usually from 5 to 8 cents per pound. The fastest growing and most profitable market for upland cottons is for higher grades and finer cottons used in the production of ring spun combed yarns33 which are subsequently utilized in the woven and knitted apparel sector. In that segment, the modern high-speed machinery requires better fiber characteristics to operate at maximum efficiency and spin high quality yarns. In addition to requiring longer, cleaner, whiter, brighter, stronger, and finer fiber, this higher segment of the market is more demanding of other fiber properties, such as elongation and neps. It also demands a lower variance in fiber properties, notably greater uniformity of length and lower short fiber content. Table 1: Higher Grades Preferred Fiber Properties Classing grade Strict Middling White (21-2) Staple length 1-1/8 inches (1.13 inches / 28.6 mm) Micronaire 3.8 - 4.2 Strength 30 grams per tex Color White Reflectance Rd 75 Yellowness +b < 9 Length uniformity ratio 83% Short fiber content 5% Elongation 6% Maturity 88% Fineness 180 millitex (µg/m) Neps < 200 / gram 31 Cost and Freight Far East. 32 Ring spun carded yarn is typically used for knitting and weaving, in a large range of coarse to fine counts. 33 Ne above 40. 11 Cotton Outlook is also publishing daily quotes for 17 "higher grades"34, including seven from WCA35 and four from ESA36. The spread between the cheapest and the highest quotations for higher grades is about 10 cents. Quotations for medium and higher grades overlap, and the highest quotation among medium grades is generally 3 to 5 cents above the cheapest higher grade. Typical premiums and discounts for grade, staple length, micronaire, and strength length uniformity are derived from the official schedules published by USDA for US upland cottons. Buyers may value fiber properties differently, depending on the final product and the production technology being used. Non quality premiums and discounts are difficult to quantify because they are somewhat subjective as shippers and spinners may have different opinions on each growth. Fiber matching all the above-listed properties would receive a very significant premium over the Cotlook A Index, possibly 10 cents, assuming low contamination and excellent marketing and shipping conditions. Conversely, cottons that do not meet those benchmarks would be discounted. 2. QUALITY PERFORMANCE OF AFRICAN COTTON 2.1 Quality of African Cottons 2.1.1 Overview of the Cotton Sector in Sub-Saharan Africa Cotton plays a major role in the economic development in Sub-Saharan Africa. Cotton is grown in 34 of the 48 SSA countries on an estimated 10 million farms. Total cotton area under cultivation is approximately 5 million hectares (15 percent of world cotton area in 2004/05) but the average yield stagnates below 360 kg of lint per hectare, less than half of the world average. Cotton production in SSA increased from 616,000 tons in 1980/81 to a record of 1.75 million tons in 2004/05, accounting for 6.5 percent of the world output. However, the expansion in African production came mostly from increases in area devoted to cotton, rather than from rising yields which are seen in the rest of the world. Due to lower international prices, production fell by 27 percent to 1.275 million tons in 2006/07. The African Franc Zone was the main driver behind this increase in production. 34 Strict Middling 1-1/8" or 1-5/32" 35 Benin, Burkina Faso, Cameroon (2), Chad 1-5/32, Côte dIvoire, Mali. 36 Tanzania RG1, Uganda RG1, Zambia SM 1-1/8, Zimbabwe SM 1-1/8. 12 Figure 6: Cotton production in Sub-Saharan Africa Source: ICAC SSA is a marginal player in terms of processing of cotton. Domestic consumption declined from 350,000 tons in 1997/98 to 265,000 tons in 2006/07, the lowest in 25 years, and only one percent of world mill use. 30 SSA countries are exporters of cotton. Based on average export values in 2004- 2005, cotton was among the top 3 export products in seven SSA countries and the share of cotton exports in total product export earnings exceeded 5 percent in ten SSA countries37. 85 percent of SSA production is exported. Exports from SSA countries rose from 400,000 tons in 1980/81 to 1.5 million tons in 2005/06 (August-July) and declined to 1.35 million tons in 2006/0738, accounting for about 16 percent of world exports. In the segment of medium and high-medium cotton, SSAs market share exceeds 30 percent. Nevertheless, SSA is the price-taker most exposed to price volatility. In export markets, African upland cotton is facing competition at three levels. First, African cotton, like any cotton, is competing with manmade fibers. Second, African cotton is competing against non-African growths. Machine-picked cottons from the U.S., Australia, and Brazil, as well as handpicked cottons from India and Uzbekistan are the main competitors of African cottons. Finally, African cotton is competing against other African growths and against offers for the same growth and quality from different shippers. 37 Burkina Faso, Benin, Mali, Zimbabwe, Togo, Chad, Zambia, Tanzania, Uganda, Cameroon 38 Total of exports by country including about 100,000 tons exported to other SSA countries. 13 2.1.2 Characteristics of Cottons from SSA Africa has a natural advantage to grow cotton, which is a tropical plant. Cotton production has similar characteristics across Sub-Saharan African countries that impact quality. All production belongs to the species Gossypium hirsutum, also known as upland cotton39. Upland cotton is grown on small-scale farms, and is almost entirely rain fed. Production is labor intensive, using manual or ox-drawn implements and relatively few purchased inputs, whereas it is capital intensive in a developed country. All seed cotton is harvested by hand, and mostly saw-ginned. At the national level, despite the fact that cotton is grown without irrigation by small farmers, African cotton is relatively homogeneous in fiber characteristics. This is due to similar growing conditions and the low number of varieties planted in most countries. However, variability within bales is greater for developed countries because the production of several farmers can be mixed in a single bale. Some parameters, micronaire and length uniformity in particular, vary according to varieties, weather conditions, growing areas, and production practices. Based on U.S. universal standards, classification of upland African cottons ranges from Strict Low Middling to Good Middling, and most production is classed as Middling or Strict Middling. Staple length ranges from medium to medium long (1-1/16 to 1-3/16 inches) and most production is classed 1-1/8 inches. Micronaire is within a rather narrow range (3.5 to 4.5) and fiber strength varies from resistant to very resistant (27 to 32 grams per tex). As shown in Table 2, the fiber properties of most African upland cotton are above those of the medium grades used in the calculation of the Cotlook A Index, the benchmark for international cotton prices, although they do not always match the specifications required for the production of finer ring spun combed yarns40 used in the woven and knitted apparel sector. Table 2: Comparison of Fiber Properties Fiber property Cotlook A Index Typical African Lint for finer yarns upland cotton Grade Middling - white Strict low middling to Strict middling - white Good middling Staple length 1-3/32 inches 1-1/6 to 1-3/16 inches 1-1/8 inches (27.8 mm) (27­30.2 mm) (28.6 mm) Micronaire 3.5­4.9 3.5­4.5 3.8­4.2 Fiber strength 25­30 grams per tex 27­32 grams per tex 30 grams per tex Progress has been slower than that of competing origins (Australia, Brazil, and, more recently, India) and has not matched the demands of the spinning industry. 39 Except some 40,000 tons of irrigated ELS cotton produced in Sudan. 40 Ne 40 and above. 14 This is because, as mentioned previously, the main competitive advantage of African cotton comes from manual harvesting. Seed cotton picked by hand is cleaner, and the fiber obtained has fewer neps and a lower short fiber content than machine-picked cotton, which must be cleaned more vigorously at the ginnery, since it has a higher content of vegetable matter. Therefore, handpicked African cotton should normally be purchased at a premium price over machine-picked cottons. Most African cottons were sold at a premium over Australian cottons until the mid 1980s when spinning mills were not as sensitive to contamination by foreign matters as they are today. Over time, the theoretical advantage conferred by manual picking has been lost because of the presence of foreign matter in the fiber. As contamination of raw cotton by foreign matters is the main concern for quality yarn and fabric producers, spinners tend to give preference for machine-picked cottons over handpicked cottons. The share of mechanically harvested cotton41 exceeds 55 percent of the export market. As a result, African handpicked cotton now trades at a discount to its machine-picked cotton competitors. Therefore, the elimination of contamination thus stands out as the first priority for quality improvement in SSA. Measures to combat contamination at the producer level have been taken only recently. Quality improvement strategies have produced positive results in Zambia, Cameroon, and Burkina Faso. Roller ginning produces lint that is longer, has fewer short fiber, seed coat fragments, and neps. As a result, roller ginned upland cottons fetch a premium of 1 to 2 cents per pound over the price of the same cottons ginned with saws. Prices of African cottons on world markets are also penalized by the way they are marketed and shipped. Most African cotton is classed through visual and manual inspections and instrument- classing data are available in some countries on a sample basis. Instrument testing of cotton is gradually replacing hand classing, and the world is in the process of adopting standard systems and procedures for the operation of cotton testing centers. As several countries can offer bale per bale SITC data for all their production, the lack of reliable cotton quality data on each bale negatively impacts African cotton prices. Cotton is sold on types42 (generally national types), not on description43. Types are defined by each country based on its own criteria. Shipments of African upland cottons are considered less reliable and have longer transit times than those of their major competitors. They lack homogeneity and consistency in terms of quality44 and packaging45. Buyers frequently complain about the poor condition of bales on arrival46. 41 U.S., Brazil, Australia, Greece, etc... 42 Cotton is sold on basis of exporter's private type or sample for grade and color. On Description/Type sales, the staple, micronaire and strength (if applicable) are separately guaranteed. 43 Described in terms of Universal Standards such as SM Middling, light spotted, etc. 44 Variations in quality within the same shipment and from one shipment to the other. 45 Variation in bale sizes, densities and covering materials. 46 Country damage. 15 2.2 Quality Control Systems and Performance by Country Annex II reviews the quality control systems and performance in each study country. Major indicators are summarized in Tables 3 and 4 below. Table 3: Summary Indicators of Seed Cotton Quality Control Systems and Performance Country No. of varieties No. of Grades Grading rating Contamination Benin 1 2 Lax Moderate Burkina Faso 3 3 Average Moderate Cameroon 3 2 Strict Moderate Mali 5 3 Very lax Serious Mozambique 8 2 Lax Moderate Tanzania 3 2 Very lax Very serious Uganda 2 2 Very lax Serious Zambia 3 3 Strict Low Zimbabwe 2 4 Very lax Moderate Sources: SONAPRA, SOFITEX, SODECOTON, CMDT, IAM, TCB, CDO, DUNAVANT, C. Poulton, International Traders. Table 4: Summary Indicators of Lint Quality Control Systems and Performance Country Classification % Instrument % STD 0/SM % 1-1/8" Average rating testing or better or longer length (mm) Benin Average 5% 55% 75% 28.4 Burkina Faso Good Sample basis 25% 80% 28.4 Cameroon Good 0 55% 75% 28.5 Mali Average 6% 10% 98% 28.5 Mozambique Poor 0 1% 69% 28.3 Tanzania Average Sample basis 1% 19% 27.8 Uganda Average Sample basis 30% 97% 28.6 Zambia Good 80% 85% 75% (est.) 28.5 (est.) Zimbabwe Mixed Sample basis 35% (est.) 80% (est.) 28.6 (est.) Sources: SONAPRA, SOFITEX, SODECOTON, CMDT. IAM, TCA, CDO, DUNAVANT, C. Poulton, International Traders. Figure 7: Average Fiber Length by Country in mm (2005/06) Source: ICAC 16 In Benin, seed cotton grading and lint classing remained regulated and carried out by the government following the liberalization of ginning. However, lint quality has remained irregular since liberalization, with the proportion of production classed in Standard 047 (the highest quality) dropping from 82 percent in 1996/97 to less than 40 percent during the following three seasons, and fluctuating between 43 percent and 62 percent over the past five years. The percentage of production classed as Standards 2 to 4 (very low qualities) declined from 17 percent to 2 percent. Staple length improved significantly over the past five years. Thus, the overall quality of Benin cotton is considered good but not reliably so. Burkina Faso has been able to significantly improve cotton quality despite the enormous growth in production over the past decade. Progress has also been made in reducing contamination. Staple length has improved substantially, with the proportion of cotton classed 1-1/8" rising from 20 percent in 1995/96 to 80 percent in 2005/06. As in other African countries, the proportion of seed cotton classed as 1st grade is higher than it should be as inspectors prefer to avoid conflict or personally take advantage from upgrading the seed cotton. All national lint production is classed manually and visually by the major ginning company SOFITEX in one location according to staple length and national types. Instrument testing is done on a sample basis, and classification reliability is considered good. Two main varieties are grown in Cameroon, one medium staple and one longer staple. The expansion of a longer staple variety has enabled Cameroon to enter the segment of fine cottons. With a staple length of 1-5/32", BLT variety is sold at a premium of about two cents/lb over the traditional medium staple variety. Grading in Cameroon is also stricter than in most other African countries and is consistent with lint grading. Although Cameroon cannot offer SITC, it does conduct micronaire tests for each bale, unlike most other African countries. In the early 1990s, Cameroon cotton was very affected by stickiness. The reputation of cotton suffered and prices were sharply discounted. In response to this, SODECOTON developed a quality improvement strategy which has been paying off. As stickiness mostly comes from late harvest, incentives for early picking were offered to farmers which have led to stickiness virtually disappearing. Yet, the reputation of stickiness is persisting, which confirms that it is difficult to regain a good reputation once it has been lost. Seed cotton grading in Mali is done by producer associations and is very lax. Large price differentials between the grades result in 99 percent of the crop being purchased as 1st grade, regardless of the lint classing results, whereas less than 90 percent actually deserve that rating. Grading was stricter when it was the responsibility of CMDT and based on lint classification results. All national lint production is classified manually and visually by CMDT in three different locations, according to staple length and the same national types. This leads to occasional problems of consistency in classing results. In addition, poor maintenance of gins, related to uneven cash flow and profitability in recent seasons, has also affected lint quality. These substantial issues leave Mali with the reputation of being among the most contaminated origins. Despite this trend, Mali, like much of WCA, has increased its average staple length over the 47 African Standard 0 is close to Universal Standard Good Middling, Standard 1 to Strict Middling, and Standard 2 to Middling. 17 past decade, with the proportion of cotton classed 1-1/8" rising from 13 percent in 1996/97 to 96 percent in 2005/06. One final issue in Mali is the fact that the widespread use of plastic materials in the picking, storing, transporting, and baling of cotton adds to its contamination. The expanded use of colored plastic materials, which are easier to detect visually than white plastic materials, has been an improvement. However, the efforts to educate farmers have had limited impact so far because they have no financial incentive to deliver uncontaminated cotton. Mozambique currently grows eight varieties of cotton. However, the traditional varieties are now outdated and being progressively replaced in the new concession areas by imported varieties with longer staple and much higher ginning outturn. Established concessionaire companies buy seed cotton on the basis of two grades. Pirate buyers do not grade seed cotton. Lint classing is based on visual and manual inspection, and is centralized, with exporting companies sending samples to the Instituto de Algodão de Moçambique for testing. Current procedures are outdated and inconsistent. The overall reputation of Mozambique cotton remains poor and the efforts of the new players to improve the quality may be penalized by the persisting low quality in the traditional cotton growing areas. Some buyers, however, do distinguish between old and new suppliers. Yet, anecdotal evidence suggests that Mozambique cotton remains moderately contaminated. In Tanzania, about half of the cotton production is saw ginned and the other half roller ginned. Roller ginned cotton fetches a premium of one cent per pound over saw ginned cotton with the same grade and type. Yet, other aspects of Tanzanias system have prevented it from building on this structural quality advantage. Theoretically, buyers and ginners are obliged to purchase seed cotton on the basis of two grades. In practice, grading has disappeared at the first point of purchase and different grades are purchased and ginned together. Due to ginning overcapacity, buyers and ginners give priority to volume and purchase seed cotton regardless of quality. The proportion of lower grade cotton has almost certainly increased since liberalization due to the initial decline in input use and mixing of seed cotton varieties. In recent years, steps have been taken to rectify both of these failings, with uncertain results. In this competitive system, lint classification is centralized at the Tanzania Cotton Board. The proportion of upper grades fell from 45 percent in the early 90s to a low of 17 percent in 1994/95 after liberalization, and rebounded to about 80 percent in 2005/06. Tanzanian cotton has the reputation of being among the most seriously contaminated origins in the world. As a result, the reputation of Tanzania cotton has suffered and exports have shifted to a lower segment of the market. In the 1970s, Uganda cotton was considered the best upland cotton in Africa. The country still occupies a special niche-market position by supplying roller ginned lint, and fiber quality is still considered relatively high. Yet, the quality control system at the primary collection level has largely broken down. Ginning overcapacity has led to a scramble for cotton and a large number of buyers purchase seed cotton regardless of quality, forcing others to do the same. Indiscrimate buying is correlated with high levels of contamination and because Ugandan cotton is reputed to be very contaminated, the premium its lint attracts has shrunk during the last decade. 18 Zambia has delivered the best quality performance in Africa over the past five years. In this concentrated system, the two major ginners have given incentives to producers to deliver higher quality seed cotton, providing them with good quality planting seeds and crop protection chemicals, as well as extension services and technical advice. Seed cotton grading in three grades is strict and consistent. The dominant ginning company provides farmers with cotton bags for picking cotton and has established manual cleaning stations at the entrance of the gins to manually remove contaminants. Its main competitor has put more emphasis on changing farmer behavior, and claims to achieve similar low levels of contamination without the cleaning stations. As a result of these efforts, Zambian cotton over the past five years has gained the reputation of being among the least contaminated in Africa. Smaller ginners, however, do not perform as well in terms of contamination as the two industry leaders. Quality was a high priority in Zimbabwe when the pre-liberalization Cotton Marketing Board had a monopoly on cotton marketing. Seed cotton was divided into four grades with very strict grading standard and pricing closely related to lint classing results. Contamination was a major criterion during the buying process. A first visual inspection at the buying station was followed by another at the gin before processing. The use of non-cotton picking bags was prohibited and bales were wrapped in cotton fabric. The high quality was maintained from 1995 through 2001 but this trend has reversed since 2002. Though the established companies have attempted to maintain quality, they have been heavily affected by the collapse in seed cotton grading at collection points, with much lower proportions of lint classed in the top grades. Meanwhile, while Zimbabwe cotton had the reputation of being among the least contaminated origins, some mills now complain that it is seriously contaminated. Such cotton is likely supplied by newly established ginners. 2.3 Impact of Quality on Export Prices Quality performance of a cotton sector is best measured by the estimated average premium (or discount) over (or below) the Cotlook A Index (in US cents per pound of lint). The benchmark for upland cottons is currently California San Joaquin Valley Acala48, followed by Australian SM 1-1/8". These machine-picked cottons fetch premiums of about 10 and 9 cents, respectively, over the Cotlook A Index. As shown in Table 2, the fiber characteristics of most African upland cottons are superior to those of the Cotlook A Index. As mentioned prior, this is again due to the fact that African upland cotton is all handpicked and subsequently cleaner, with fewer short fiber and neps than cottons from most other origins. For these reasons, African cotton could fetch a premium of up to ten cents per pound on international markets if it could develop a reliable reputation for lint uncontaminated with foreign matter. 48 SJV Acala is long (1.15 inch 1-1/8" full, very strong (32-34 gpt), micronaire 4.2. This cotton is shipped year-round, very consistent in deliveries, wrapped in cotton cloth, with bale per bale HVI classification, short transit time and very reliable shipments. However, SJV cotton is neppy, with a rather high short fiber content, micronaire is sometimes on the high side, and stickiness is frequent. 19 Figure 8 is based on quotations published by Cotton Outlook, corroborated in discussions with international traders, and reflects the prices at which top types of lint from each country were offered for sale (CFR main Asian ports) in 2006/07. Figure 8: Estimated Premium for Top Type of Lint in 2006/2007 by Country Source: ICAC Figure 9 compares the premiums for the top type of the countries under review with those of other origins quoted in Cotton Outlook. Figure 9: Estimated Premium for Top Type of Lint in 2006/2007 by Country Source: ICAC 20 As there are no major differences in basic fiber parameters between African cottons, price differentials between different origins primarily reflect their level of contamination (real or perceived). In 2006/07, the average premium of the quotation for the top type of each country in Cotton Outlook over the Cotlook A Index ranged from 0.5 to 7 cents per pound, with Zimbabwe and Zambia receiving the highest premium and Tanzania and Mozambique the lowest. Figure 10 shows trends in lint prices for top types of African cottons from the mid- 1990s to 2006/07. The estimates of past differentials relative to the Cotlook A Index derive from discussions with major international merchants and from the personal experience of the author who was involved in trade at that time. Over the last decade, premiums increased in Zambia (+ 5 cents/lb), Cameroon (+ 4.5 cents/lb), Mozambique (+ 2 cents/lb), and Burkina Faso (+1 cent/lb). Although none of those cotton sectors belong to competitive systems. Progress made in Zambia and Cameroon is clearly due to the successful reduction in contamination and stickiness, which would have been difficult or impossible to achieve in a competitive system. In contrast, differentials vis-à-vis the Cotlook A Index declined by 1.5 cents/lb in Tanzania, 1 cent/lb in Uganda and Zimbabwe, and 0.5 cents/lb in Benin and Mali. The declines reflect increased competition between ginners in all three ESA countries, as well as lax seed cotton grading and contamination due to poor management in the two WCA countries. Figure 10: Estimated Premium for Top Type of Lint Mid-90s and 2006/2007 by Country Source: ICAC Among the liberalized cotton sectors of ESA, the quality premium enjoyed by Zambian lint has risen in recent years, but the premiums attached to Tanzanian and Ugandan lint and, more recently, Zimbabwean lint have all fallen. In Mozambique two opposing trends have been seen. In the traditional cotton heartlands, quality is low and either static or declining. By contrast, in the newer concession areas, newer entrants to the sector have achieved slightly higher and rising quality. 21 There is no clear relationship between the premium/discount for the top type of each country and the actual realized export prices received. For countries where the shares of cotton sold under the various types in each country and the premiums/discounts received for each of these types are known, it is possible to calculate a theoretical weighted average contract price. The average export contract price is known for some countries where there is only one ginning company. Yet, the actual realized price is not equal to the average contract price because the actual quality and quantity shipped may differ from the quality and quantity sold, which translates into price adjustments. The theoretical average quotation by country49 is calculated based on the following data: Average premium for the quotation of the top type, as shown in Figure 8, Usual world market price differentials for grade compared to middling (depending on cotton origin), as shown in Table 5: Table 5: Price Differentials for Grade (US cents per pound) Good Middling Strict Middling Middling Strict Low Middling + 1.5 to + 2.5 ct/lb + 0.75 to + 1.0 ct/lb BASE - 0.5 to ­ 2.0 cts/lb Usual world market price differentials for staple length, relative to 1-3/32" (depending on cotton origin), as shown in Table 6: Table 6: Price Differentials for Staple Length (US cents per pound) 1-5/32" 1-1/8" 1-3/32" 1-1/16" + 1.5 to + 2.0 cts/lb + 0.5 to + 1.0 ct/lb BASE - 1.75 to ­ 4.0 cts/lb Actual 2005/06 classing data for WCA countries and most recent available data or estimates for ESA countries. To calculate the theoretical average export price of the crop, a deduction of one cent per pound is applied to the weighted average quotation to reflect the usual difference between the sellers offering price and the actual negotiated contract price. Based on these calculations, theoretical weighted average export price differentials compared to the A Index range from minus 2 to plus 4 cents per pound50. 49 Assuming that the whole crop was sold on a given day. 50 Actual season-average contract prices can be different from the above theoretical averages depending on the timing of sales. In addition, final average realized prices can be different from the contract prices depending on the actual quality and weight shipped and on eventual claims. Nevertheless, these calculations represent the current best estimates of average market quality premiums earned by each sector. 22 Figure 11: Estimated Premium Weighted Average Basis by Country In conclusion, Zambias concentrated system delivers the best performance, comparable to what Zimbabwe delivered prior to the entry of new competitors. Zimbabwes premium is still high, but falling as a result of increased competition. Cameroon, with a national monopoly largely free of political meddling, also performs well, though not up to the standards of Zambia. The other national and local monopolies show highly variable performance, while Tanzanias competitive system is, along with Mozambique, the worst quality performer. Mozambiques poor performance reflects the legacy of a nearly unregulated local monopoly system with original concession companies uncommitted to productivity and quality; while quality is likely better among the newer affiliated ginners, it will take time for them to overcome the countrys poor reputation. Uganda, with a competitive structure that is heavily regulated to eliminate direct competition for cotton, does better, though some of this difference is due to the larger share of roller ginned cotton in that country. 3. MARKETING STRATEGIES OF AFRICAN COTTON 3.1 Marketing Systems Marketing systems in cotton producing countries reflect different historical, social, and geographical influences. In a number of Sub-Saharan African countries former colonial powers introduced cotton cultivation to create a sustainable tax base. As cotton became an important source of foreign exchange, some newly independent governments took control of marketing and exports. The wave of economic liberalization and restructuring programs in the 1990s resulted in the transformation of most of the former single-channel marketing systems into totally or mostly free marketing systems. However, government or semi-monopolistic structures continue to play an important control role in several African countries, especially in WCA. Although fixed producer prices and price controls at the farm gate still prevail in some countries, no country continues to control export by limiting the freedom of exporters to set sales price themselves. 23 In entirely open systems, producer prices, collection marketing, and export marketing are entirely free. Controls are managed by private industry structures or by joint government and industry structures such as cotton boards, authorities or institutes in which the private sector has a meaningful say. In countries with mixed systems, governments influence over the marketing system remains strong, with internal controls on the marketing and pricing of cotton, although exports are entirely run by the private sector and face an entirely free world market. Until the mid-1980s most lint produced in Africa was sold by national cotton companies and marketing boards to international merchants or to spinners through commissioned agents. Vertical integration between traders, ginners, and exporters has increased in order to supply spinners with the large volumes of lint they require on a regular basis. This has led to major international cotton trading companies developing relationships with African exporters and investing directly in ginning operations. Today, two types of companies supply lint to the world market out of SSA: independent ginners sell lint to international cotton merchants, while ginning companies affiliated with such merchants ("affiliated ginners") sell lint to or through their parent companies51. International cotton merchants thus play a leading role in the marketing of African lint. They buy cotton from independent ginners or receive it from their affiliated ginners, sell it to textile mills or other merchants, hedge price risks and arrange shipments. Most sales contracts between ginners and international merchants are private contracts concluded on a one-to-one basis52, business is generally concluded either via a phone call or face-to-face between the buyer and the seller. The verbal commitment is reconfirmed in writing by telex, fax, or email. A formal signed written contract spells out all important provisions of the contract agreement. Some ginning companies invite selected authorized buyers to tender by fax or e-mail. Written contracts incorporate all the major terms of sales. Quantity is specified in metric tons. Cotton is sold either "certified shipping weights final" or "net landed weights final". African cotton quality is generally sold "on types" on the basis of the national or exporters types or samples for grade and color. Some sales are done "on description"53. Staple length, micronaire, and sometimes strength, are separately guaranteed. The sales contract price is usually in U.S. cents per pound in ESA countries and in euros per kilogram in the Franc Zone countries. The most common delivery terms are FOT54 ex-gin, FOB55, and CFR56. Shipment terms can be for one specified month or several months. Typically, payments by letter of credit are required. The rules of arbitration applicable to the contracts are usually those of ICA (International Cotton 51 In Africa, these parent companies include Dunavant, Cargill, Plexus, Dagris, Reinhart, and others. 52 Cottco in Zimbabwe and cotton Tchad in Chad sell directly to International Spinners through Commissioned Agents in importing countries. 53 Described in terms of Universal Standards such as Middling, Strict Middling, etc. 54 Free on Truck. 55 Free on Board. 56 Cost and Freight. 24 Association) for ESA countries and of AFCOT (Association Française Cotonnière) for Franc Zone countries. African cotton is mostly offered at fixed price. The sales price of a fixed price contract is final at the conclusion of a sale and does not change, regardless of fluctuations in the market. When cotton is offered on call, the price is based on premiums or discounts relative to the Cotlook A Index or to a certain month of the ICE cotton futures. On call pricing necessitates that the buyer and seller agree to a volume and delivery date, and agree that the final price will remain unfixed until it can be judged according to a pre- determined schedule. Some merchants offer guaranteed minimum price contracts, allowing the ginner to participate in market increases over a stipulated base price, should that occur. The guaranteed minimum price is lower than the market price on the same day. In WCA, forward sales contracts at fixed price have been used extensively for decades, primarily as a way to secure input and crop financing. Sales are contracted in euros per kilogram FOB, offsetting the exchange risk. Ginners usually base their prices on the Cotlook A Index57, valued at the forward exchange rate for the shipping period considered. About 50 percent of anticipated crop is generally sold before ginning starts, which can occur up to 18 months prior to shipment. Small ginners in ESA generally do not have the opportunity to sell their production forward and are only active on the spot market. Sales in the export markets are usually concluded through local commission agents or merchants/importers. Direct sales between spinners and ginners are exceptional as most importers prefer to deal with reputable international cotton merchants. In contrast, African ginners usually sell lint directly to domestic spinning mills. Merchant-to- merchant sales of African cotton are frequent. 3.2 Marketing Performance Annex 3 reviews the indicators of the marketing systems in each study country. Affiliated ginners are present in all countries in this review except Cameroon and Mali, which continue to operate national monopoly sectors. Most lint from SSA is handled by independent ginners. These include very large companies ­ the national monopolies of Mali and Cameroon (CMDT and SODECOTON, respectively) and the former national monopolies in Burkina and Benin (SOFITEX and SONAPRA, respectively) ­ as well as smaller private companies, most of them in ESA. The efficiency of marketing should be assessed in comparing the actual average contract price with the marketing-season average world price adjusted for quality and location. Unfortunately, the data necessary to quantify the respective performances of the marketing systems in Sub-Saharan Africa is scarce. Marketing audits can be done for a few parastatal companies in WCA58. However, it is not possible to assess the 57 Or the quotation for the African Franc Zone in Cotton Outlook. 58 SODECOTON in Cameroon and CMDT in Mali. 25 marketing performance of private ginning companies as they are unable to divulge information on their actual sales. Nonetheless, ginning companies affiliated to large international merchants in ESA countries are certainly more efficient than pure ginners. Merchants carefully select sellers to guarantee contract performance with their buyers. Large parastatal companies in WCA are considered reliable by merchants, defaulting only in good faith, while private independent ginners have mixed reputations. Merchants generally consider it much easier to purchase cotton in WCA countries than from independent ginners in ESA countries because the offer is not atomized between numerous trading companies59, and the volume is sufficient to ensure year-round shipments, while quality standards are relatively consistent. In contrast, small independent ginners in ESA are generally not in a position to guarantee consistent or year-round shipments, and big volumes. Most independent African ginners have little knowledge of the world cotton market. They lack access to market information and to futures markets and usually receive most of the market information from their clients, the merchants. Their sales strategies are not formalized and tend to be passive or non-existent. Additionally, their know-how of price risk management tools is minimal. As lint is sold FOT or FOB to international merchants, ginners ignore the actual selling price of their lint as well as intermediary costs, and direct feedback is not received from the end-users on the quality of their lint. Therefore, the cotton marketing systems may send biased signals to ginners and growers. Thus, international cotton merchants are in a much stronger negotiating position when dealing with independent ginners. Ginning companies in WCA countries are very exposed to intra-seasonal price and exchange rate volatility. The system of guaranteeing a pan-seasonal producer price prior to planting transfers the price risk from producers to ginners. Exchange rate fluctuations create an additional risk that needs to be managed by ginners. International cotton prices are expressed in U.S. dollars but African cotton ginners and exporters pay most of their production expenses, mainly the producer price, in domestic currencies. Fluctuations in the ,,basis60 between African cotton spot prices and New York futures cannot be hedged, and are a major inhibitor to hedging of African cottons. Subsidiaries of international merchants involved in ginning are generally more aware of the market demand and likely to pass on market signals to producers. Although, they do not have the same exposure to international price fluctuations as part of the market risk is taken by the mother company who has the ability to hedge. Forward sales at fixed price in Euro per kilogram are an effective way of mitigating risks, although high percentages of forward sales increase the risk of not being able to deliver the contracted quality and may lead to overselling. Smaller independent ginners in ESA are generally not in a position to guarantee the volume and the quality of their production before it is ginned, are not able to store lint for an extended period of time, and therefore seldom engage in forward sales. They primarily deal with price and 59 Though exceptions may exist: for example, supply is more atomized in Benin than in Zambia. 60 Arithmetic difference. 26 exchange rate risk by adjusting their buying price over the course of the season and by selling the lint as it is ginned. 4. OPPORTUNITIES AND CHALLENGES FOR AFRICAN COTTON 4.1 Quality African cotton has two comparative advantages in the world market: the intrinsic quality of its fiber and the fact that it is handpicked. Staple length is the most important cotton fiber property and SSA has been able to follow the trend of the market toward longer fiber as most African production now reaches the typical benchmark of 1-1/8". However, though most African cotton is suitable for the medium-high level of ring spinning, progress on quality during the last decade has been slower than in competing origins and not as fast as required by the spinning industry. The trend in spinning technology toward more automation and higher speeds makes improvements in quality and consistency a vital issue for the future of African cotton sectors. Management of lint quality has become more important as spinners have imposed more stringent demands for quality and for greater accuracy of measurement of fiber properties. Yet, the impact of quality on the pricing of cotton is generally not well-known by ginners and even less by producers. Quality improvement could translate into an increase in selling prices of up to 10 US cents per pound of lint over the Cotlook A Index. At typical producer prices of 25 to 32 US cents per kg of seed cotton, an increase of 10 cents per pound in the price of lint that is fully passed on to farmers would increase farmer prices by 30 to 40 percent. As a result, quality management should be considered one of the most important areas of improvement for SSA cotton exporting countries. Contamination with foreign matter is the most serious problem facing African cotton, and is crucial in pricing. The elimination of contamination stands out as the first priority for quality improvement, as it can potentially increase lint prices by up to 15 percent. However, a reputation for high contamination is difficult to overcome. Technical solutions to eliminate contamination are well-known61 and should be relatively easy to implement by the producers themselves without requiring heavy investments and a lot of technical assistance, provided that producers are motivated by appropriate financial incentives and properly trained to the recommended methods. Eliminating contamination is possible provided that the right signals are sent to farmers through a more rigorous pricing system, discounting contaminated seed cotton, or 61 Solutions to eliminate contamination are well-known have proven successful in Zambia, resulting in an increase in the premium for the top type of Zambian lint relative to the Cotlook A Index by 5 cents per pound over a period of five years. There, the two dominant ginning companies have been equally successful in largely eliminating contamination, using very different approaches: Dunavant relies primarily on the manual removal of contamination at the gin, while Cargill has succeeded in changing farmer behavior to avoid contamination. 27 giving a premium to uncontaminated cotton. However, getting ginning companies to take those steps in highly competitive systems has proven quite difficult. A rough cost/benefit analysis of a quality improvement program indicates that removing contamination through screening prior to ginning costs less than one cent per pound, whereas uncontaminated cotton would potentially fetch a premium of 5 cents per pound. African cotton sectors must focus on making dramatic improvements in the reliability of seed cotton grading. Eventually seed cotton should be paid on the basis of the price of lint obtained after ginning. The critical control point in the marketing chain as far as quality is concerned is the point of primary marketing. Most contamination of lint by foreign matter originates prior to primary marketing, typically from the use of polypropylene bags for transporting seed cotton from the field to the collection point. Improving quality requires giving incentives to farmers to undertake good quality practices and further sorting prior to ginning to achieve better uniformity. Growing fewer varieties in a country makes it easier to maintain homogeneity of quality, though proper controls and good classification can ensure good performance even when several varieties are grown. More seed cotton grades is generally good, but only if these are strictly linked to lint classification outcomes. Compared to saw ginning, roller ginning has a higher ginning outturn, and produces fiber that is longer and of a better quality, and commands a premium of up to 2 cents per pound, but the process is slower and only suitable for ELS and medium-long upland varieties. So far, only saw gins are used in WCA, although the staple length of most cotton is now long enough for roller ginning. A detailed costs/benefits analysis should be undertaken to assess whether roller ginning is a valid alternative to saw ginning. International cotton merchants put great importance on the reliability of lint classification in a country, independent of the actual quality of the lint. High quality cotton that is not reliably classed will not fetch the premium that it should deserve. Instrument classing of lint is still rarely used in Africa, and would certainly improve the competitiveness of African cottons because spinners require it and major competitors use it. It is also a way to improve the consistency of the fiber properties delivered over a sales contract. Instrument testing will enable SSA countries to sell their cotton based on reliable and comparable test results, avoid price discounts and claims, and secure or even improve their market share. When using cotton testing instruments for classification, several critical elements necessary to maintain precise and accurate results include calibration standards, laboratory conditions, sample conditioning, and instrument verification procedures. Nevertheless, sellers cannot expect a significant increase in price from instrument testing because many spinners are equipped with their own SITCs. Its more likely to be a way to avoid their cotton to be discounted for not being able to offer bale per bale instrument testing data like many of their competitors do. Spinners are not willing to pay a premium for SITC classed African cotton, but they would definitely pay a premium for cotton free of contamination. Without any doubt, cotton classed manually with zero contamination would sell at a much higher price than cotton classed by instrument but without such a guarantee. 28 Monopoly systems that have an obligation to purchase all seed cotton may find it difficult to refuse poor quality deliveries. It was expected that the move to free-market systems would encourage quality because buyers would impose their own price differentials, better linked to the final market requirements. In fact, what actually occurred following liberalization was that cotton quality, and therefore values, deteriorated in several countries due to the decline in support and quality control services. In some countries liberalization at first translated into uncontrolled change. To a ginner, quantity is more important than quality, especially when there is ginning overcapacity. Therefore, highly competitive markets for seed cotton, where numerous buyers scramble to secure the purchase of a sufficient volume of seed cotton to cover their fixed costs are bad for quality control. Many small ginners are simply not aware of the impact of quality on prices. Competition between buyers undermines the incentives for producers to do the extra work required for delivering high quality seed cotton. As a result, many small growers argue that quality does not pay because the price for good or bad cotton is about the same. The major quality control problem in the highly competitive sectors is the inability to enforce grading at primary marketing. If one buyer tries to do this, farmers can choose to take their seed cotton to an alternative buyer, who will purchase it anyway, irrespective of quality. Quality can pay provided that ginners can achieve higher quality at a reasonable cost. High quality requires that ginners can control their supply chain, which implies some form of contract farming. Harvest and immediate post-harvest practices are the most important determinants of the quality of seed cotton that the farmers deliver. "Active quality control", in which better than average quality (uncontaminated cotton) at the farm gate receives a premium, is preferable to "passive quality control" in which people are penalized62 for doing something wrong. The reputation of a given origin and the efforts to improve quality has an important effect on economic outcomes, particularly on producer prices. Premiums and discounts attached to internationally traded cottons derive partly from the reputation of national origins. It is always much easier to destroy a good reputation than to establish it, let alone re-gain it. Some form of collective coordination regulation, albeit public or private, is required to protect the quality reputation of one origin. Quality improvement requires a concerted effort. A comprehensive strategy to improve quality in SSA must involve researchers, producers, ginners, and transporters alike to improve fiber properties through research and better production practices. This will in turn reduce the variability of lint quality through more rigorous seed cotton grading and lint classification, control of contamination through capacity building and price incentives, as well as through the optimization of quality management in ginning. Researchers and breeders have an important role to play in quality improvement. Producers and ginners need education on quality requirements, emphasizing the relation of quality to prices received. Efforts must also be made to generalize the use of cotton cloth wrappers for bales, develop container loading at the gins, and optimize export logistics. 62 Through price discounts. 29 Even totally free systems require a certain amount of control. Competition reinforces the need for independent quality control and certification before shipment. Official controls can help to maintain or improve export quality, provided the buyers are convinced of their accuracy and reliability, and the process can be handed over to qualified private sector quality controllers63. Official licensing of ginners and exporters can also be a useful disciplinary tool, and professional associations can impose a certain degree of self-discipline on their membership. Capturing the potentially significant price premiums available to African cotton while simultaneously sharing some of that benefit with farmers requires much stronger farmer organizations than are currently found in SSA. In concentrated systems, these organizations would have to bargain with the large ginners for remunerative prices on the basis of solid knowledge of world prices, realized export prices quality premiums obtained, and cost structure borne by ginners. In competitive systems, these organizations would need to focus on training farmers regarding the benefits of increased quality, and monitoring prices paid by companies to ensure the transmission of quality premiums. 4.2 Marketing Many options exist for independent ginners in SSA to improve marketing performance. Forward sales, the most common marketing method in the cotton business, are the easiest and most effective marketing strategy to cover risks. The flexibility and effectiveness of such sales can be enhanced if they are supplemented by the use of market instruments such as futures and options; the application of these instruments can be specified in the physical contract arrangements with merchants. Ginners can also spread their risks by committing proportions of the total production to different marketing options: cash sales after ginning and other options requiring commitment prior to harvest: forward sales at fixed price, ,,on call at price-to-be-fixed contracts, and minimum guaranteed price contracts. Direct sales from ginners to spinners through commissioned agents can save the cost of intermediation by merchants while improving the quantity and quality of market knowledge available to the ginners. Stronger farmer organizations can also work to the advantage of the sector if they allow more systematic contractual trade relationships between the two parties. Examples could include more formalized contract farming relationships which specify pre- determined volumes of seed cotton to buy/sell and with precise quality specifications (such as zero contamination). Stronger farmer organizations might also be better informed regarding world markets and able to negotiate a pricing approach that is tied to world prices and which recognizes world market premiums for quality. Considerable institutional strengthening and training will be required first to reinforce producer organizations. The existence of different export channels could be an advantage for farmers who will benefit from some degree of competition. However, labeling and consolidation of 63 Such as Cargo Control, SGS and WIS. 30 export volumes should be encouraged as efficient marketing requires the ability to supply several thousands tons of consistent quality throughout the seasons64. There has been substantial growth in the use of cooperative marketing pools during the last 20 years. Marketing pools are usually operated by farmer-owned cooperatives, but they can also be offered as a marketing option by ginners and merchants. Individual farmers place some or all of their cotton lint under the control of the managers of the marketing pool. The managers sell cotton from the pool, and each farmer receives an average price for the season adjusted for the particular quality delivered 65. Marketing pools are popular because they offer substantial economies of scale in marketing. A marketing pool representing several hundred thousand bales of cotton can afford to employ professional managers whose full-time jobs are to market cotton. They can also tailor cotton sales and deliveries to the quality and shipping preferences of individual textile mills. In contrast, a single producer with only a small quantity of cotton to sell cannot afford to hire a professional marketing manager and cannot easily meet the quality demands and timely delivery requirements of textile mills. Marketing pools also offer the advantage of a form of price risk management, since each producer will receive the average price achieved by the entire pool and does not have to worry that his or her particular cotton was sold when the market price was lowest in the season. Importers and spinners need to be reasonably certain that contracts will be correctly executed whether the market rises or falls: repeated defaults or the shipment of inferior quality by some will eventually affect all the exporters and growers of an entire country. Licensing of ginners and exporters is therefore a useful disciplinary tool if the process is transparent and based on clear and equitable rules. Properly run auctions provide excellent price transparency but depend on competition. A first prerequisite for attracting and concentrating competition is that the entire crop is sold in this way66. A potential tool for improving the marketing of African cotton is electronic trading. E- trade platforms can be a very effective means of transparent price discovery brought about by real time multilateral bids and offers and online contracting can reduce transaction costs. To realize improvements such as the ones recommended above, ginners in SSA need to be informed and trained to better understand the world cotton market and prices, master cotton trade rules and regulations, and understand how to use risk management techniques based on futures and options contracts. 64 A typical spinning mill in Asia consumes about 1,000 tons per month; to be included in its blend would require the ability to supply 300 tons monthly. 65 Marketing pools are offered in the United States, Brazil, Argentina, Colombia, Greece, Israel, Australia and other countries where farmers sell lint instead of seed cotton. In 2006, an estimated 40% of U.S. cotton was sold through some form of cooperative marketing pool, including those operated by farmer-owned marketing cooperatives, as well as by the pools operated by private merchants under contract to a group of farmers in a particular region. 66 Sudan relies on auctions to market its entire cotton crop. 31 Since the cotton market is a highly volatile one, and exchange rate movements also have an impact on the sustainability of the sector, it is important to improve risk management mechanisms in SSA. Pan territorial, pan seasonal price mechanisms pass extremely high levels of financial risk to cotton companies, who are often ill-equipped to monitor and manage those risks. Farmers, cotton companies, and banks can benefit from capacity building on this issue, particularly since it includes information about the global market, relationships between local and international prices, and quality premiums. Competition between buyers/ginners of seed cotton may increase the producer price, especially when there is overcapacity in ginning. However, the immediate impact of increased competition between ginners/exporters is to put pressure on the selling price of lint, as buyers (spinners) always take advantage of competing offers from several sellers to obtain lower prices. A monopoly (state-owned or private) is generally in a better position to protect its selling price67. However, in a competitive market, ginners can benefit from the more attractive prices that can be offered by merchants having to cover their "short"68 positions. So far, the cotton marketing system is not sending the right signals to growers and ginners while African cotton producers do not receive the incentives to deliver the fiber quality and characteristics to the end user's, (i.e. the spinner), wants and needs. Additionally, most African ginners lack access to market information and futures markets and as lint is sold FOT or FOB to international merchants, ginners are deficient in direct feedback from the end users on the quality of their lint. In contrast, a single channel agent can bypass intermediaries and avoid the negative impact of competing offers on the selling price. Cotton sectors should benefit from the market linkages brought by international merchants through better reaction to market signals, quality improvement, and risk hedging. Subsidiaries of international merchants involved in ginning are generally more aware of the market demand and likely to better pass on market signals to producers. They do not have the same exposure to international price fluctuations as a part of the market risk is taken by the mother company which has the ability to hedge cotton price and exchange rate risks. Yet, more efficient marketing systems do not necessarily translate into higher producer prices because international merchants are likely to transfer profits from their local ginning operations to their European headquarters. As long as the negotiating power of producers remains weak, private ginners will only pay the lowest producer price needed to secure the volume they need to cover their fixed costs. Fair trade and organic cottons can provide an opportunity for small ginners working in close connection with producers to meet the required specifications for those niche markets. Notwithstanding the market potential for traditional non-contaminated cotton lint is much larger. 67 Until the mid-1980s, most of the African Franc Zone cotton production was sold directly to end-users through CFDT/Compagnie Cotonnière, and paid on a commission basis. Liberalization of lint marketing had a negative short-term impact on the daily selling prices of WCA cottons in the international market. 68 Forward sales to spinners not backed by physical purchase. 32 ANNEX 1: PRIMER ON LINT CONTAMINATION Contamination by foreign matter in cotton is the most serious problem confronting spinners around the world. This contamination includes even the smallest piece of plastic or polypropylene, and other natural foreign matter, such as animal and human hair, all of which disrupt the spinning process, thereby increasing the cost of spinning and reducing the quality of yarn. Seed-coat fragments are also particularly unwelcome contaminants for spinners. Most end users demand zero contamination in yarn, fabric and garments. There are no cost effective means of removing contamination once it is present in yarn or fabric. As a result, contamination leads to the downgrading end - products or even to total rejection of an entire delivery. Cotton that is contaminated, or even might be contaminated because of the origin, can only be sold at substantial discounts to compensate the user for inspecting and cleaning the cotton before spinning. For decades, spinners have produced evidence of all kinds of foreign matter found inside cotton bales which could often be eliminated by hand as long as spinning was still largely a labor-intensive processing technology. Following the extensive modernization of production equipment in recent years, processing steps which previously required the intervention of manual labor are increasingly performed by machines. The trend towards automation in textile manufacturing is not limited to any particular geographical area and will gain further momentum in the years to come. The biggest problem which automation presents to cotton relates to cleanliness. Automated equipment is not in position to detect contamination, let alone eliminate it. In the case of plastic material ­ one of the most vicious forms of contamination ­ the damage is becoming visible only when the time the fabric leaves the final finishing process, at which stage it is tool late to apply any remedy. It is not only affecting the quality and appearance of the final textile product, but may actually damage the processing machinery itself. As a result, price differentials for cottons with the same fiber characteristics can be very significant depending on the degree of perceived contamination by extraneous matters, stickiness and seed coat fragments. Cotton spinners pay the full price for lint when they are confident that the cotton lint is clean and does not contain impurities which could affect the spinning process and discount their end-products. If they have any doubt, they will offer only discounted prices for the 'suspect' cotton. These discounts, ranging from 5-30 percent of the price, are usually applied indiscriminately to all cotton originating from an area considered to be affected by contamination. Exporters offering a guarantee that their cottons are not contaminated would recieve a significant price premium. However, no exporter today is in a position to offer such a guarantee as it would imply that they would be responsible for all claims for quality defects on yarns or fabrics due to contamination. 33 The Impact of Picking Methods on Contamination Contamination by foreign matters is more serious with handpicked cottons. Cotton gets contaminated during picking, storage, handling or transport of seed cotton. The plastic strings of woven fertilizer bags, often used for picking and handling of seed cotton, represent a major contaminate. As a result, handpicked cottons in general, and those from India and Africa in particular, are discounted in the international market despite the fact that cotton spinners prefer the characteristics of the handpicked cottons, which are cleaner69, have fewer neps and a lower short fiber content than mechanically harvested cottons. Seed cotton picked by hand is cleaner, and the fiber obtained contains less fine trash, short fiber, neps and seed-coat fragments than machine-picked cotton, which must be cleaned more vigorously, since it has a higher residue of vegetal matter. However, the advantage conferred by handpicking has been lost because of the presence of foreign matter in the fiber. In practice, therefore, handpicked cotton nowadays trades at a discount to its machine-picked competitors, whereas it should enjoy a premium. Certain spinners even refuse on principle to purchase handpicked cotton, since they judge that the risks associated with contamination are too great considering the qualitative demands of their downstream clients. More than half of cotton exported worldwide is machine-picked. Machine-picked cotton is more liable to deteriorate by ginning practices. Defoliation forces some bolls to open before they are ready, which has consequences on fiber quality. Additional cleaning to remove extra trash also affects fiber quality. If it is not eliminated, the cotton is unfit for spinning. Seed cotton drying and lint cleaning result in machine- picked cotton contains more fine trash (pin leaf), short fiber, neps and seed-coat fragments than handpicked cotton, lowering its spinnability and the quality of final products. In contrast, handpicking is best as it provides a soft action on the seed cotton to be removed from burrs. Cotton could be contaminated with plant material also called organic matter during picking. Dry leaves stick to open bolls and get picked along with cotton. Other plant material like small dry branches or green leaves could also be picked along with cotton. Most of this organic material, usually called trash, is eliminated by using pre and post ginning cleaners. Such cleaners are installed at all gins and it is a routine process to clean cotton during ginning. No doubt machine-picked cotton has higher trash and requires extensive cleaning. Indeed, in the field, the first contamination risk occurs when collection is delayed, for example, due to competing cotton and cereal harvests. The later the cotton harvest, the more likely it is that the cotton flower will contain leaf pieces and twigs, or suffer from aphids and other insects (such as the white fly). Collecting cotton in the plastic bags that fertilizers come in brings significant risk of contamination as plastic strings often slide into the cotton. 69 In terms of trash (plant matter) 34 It is difficult to detect contamination at a mill during bale opening and spinning. Therefore, contamination could and should be avoided during harvesting, and could be detected by observation of the cotton during handling, storing, loading/unloading trucks before the ginning process. The development of methods and appropriate incentives to produce uncontaminated cotton will have the dual benefit of protecting growers against price discounts, and of enabling spinners to optimize the value of their production. The importance of primary operations cannot be overemphasized. Cotton of excellent inherent quality can be damaged by poor handling and storage at primary marketing levels. Such damage cannot be corrected at later stages of processing, or at significantly higher costs. The most common quality problems at primary buying stations are the result of mixing different types and varieties of cotton and contamination in storage by the introduction of extraneous matters, either accidentally or deliberately, to gain weight. The Importance of Contamination In an effort to shed more light on spinners perception of the problem of cotton contamination and foreign matter, the International Textile Manufacturers Federation (ITMF) has conducted regular surveys every other year since 1982 with spinning mills affiliated to its world-wide membership. The questionnaire identifies 16 sources of contamination70 as well as measuring the degree of contamination associated with each (non-existent/insignificant ­ moderate ­ serious). The presence of stickiness and seed coat fragments is evaluated separately (without quantifying the degree of contamination). The methodology of these surveys is statistically flawed and quite subjective. Nevertheless, perception and reputation are often more important than facts in the cotton market and the ITMF surveys do have an impact on cotton prices. 152 spinning mills participated in the 2005 survey and evaluated a total of 716 samples from 68 origins/varieties, of which 41 were considered for the survey. The other 28 were not ranked because there were less than 5 evaluations for each of them. A large Indonesian spinning mill, PT Apac Inti Corpora, collected comprehensive data on the nature and the extent of contamination in different origins of cotton, based on the systematic contamination removal from over 200,000 tons of cottons in the last seven years. The results of the study were presented at the 2006 Bremen Cotton Conference. Manual cleaning adds about 2 cents per pound to the cost of raw cotton. Cottons from different countries differ in degree of contamination, but the nature of contamination is similar. Fibrous type contaminants are the most dreadful because they are difficult to remove in the spinning process. Extraneous contaminants were found in all origins of cottons without any exception. Not one single cotton shipment in the last 7 years was found to be totally free of contamination. A level of less than one gram of fibrous contaminants per ton of raw cotton would be acceptable by spinners to satisfy their customers and avoid complaints and claims. 70 Falling into 5 categories: fabrics; strings; organic matter; inorganic matter; oily substances/chemicals. 35 Cottons from 15 origins were classed in 3 groups: Table 7: Extraneous Contamination by Origin (2004-2005 averages) Origins Bales Fibrous Contaminants contaminated contaminants (g/ton) (%) (%) Group I Australia, Brazil, China, Mexico, 20-35 60-65 2-4 (low) USA Group II Mozambique, Paraguay, 60-85 75-80 5-15 (moderate) Uzbekistan, West Africa, Zambia Group III India, Syria, Pakistan, Uganda, 90-100 80-85 20-100 (high) Zimbabwe Source: Vijayshankar, PT Apac Inti Corpora, 2006 All machine-picked origins are classed in the group of the least contaminated origins and China is the only handpicked origin in that group. 36 ANNEX 2: COTTON QUALITY BY COUNTRY Benin Only one variety is planted in Benin (STAM 18A). Seed cotton grading and lint classing remained regulated and carried out by the government after the liberalization of ginning in Benin. Seed cotton is graded in two grades by DPCQ (Direction de la Promotion de la Qualité et du conditionnement des produits agricoles). As liberalization led to ginning overcapacity, seed cotton grading is lax. The proportion of seed cotton classed as second grade is minimal and much lower than it should be. All national lint production is classed by the parastatal ginning company SONAPRA according to national types. Manual and visual classing is supplemented by a testing laboratory equipped with two SITCs and one AFIS. About 5 percent of production is instrument tested. In addition, SGS and Dunavant have one SITC each in Cotonou for checking parameters prior to shipment for the account of merchants or buyers. Lint quality has been irregular over the last seasons due to delays in the distribution of inputs and in the start of the marketing season (due to prolonged discussions between ginners and producers about the producer price). Late payments of seed cotton have also been a negative factor for producer motivation. The proportion of production classed in Standard 071 dropped from 82% in 1996/97 to less than 40% during the following 3 seasons, as liberalization of ginning led to overcapacity. During the last 5 seasons, the percentage of production classed in Standard 0 fluctuated between 43 and 62, and the percentage of Standard 1 ranged from 35 to 46, while the percentage of production classed in Standards 2 to 4 declined from 17 to 2. Table 8: Share of Lint based on Grade ­ Benin STD 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 0 82.1% 31.5% 40.2% 38.1% 50.9% 43.3% 47.8% 61.9% 48.1% 56.0% 1 16.0% 57.7% 51.3% 56.5% 44.1% 40.0% 46.3% 34.9% 46.5% 42.1% 2-4 1.9% 10.8% 8.5% 5.4% 5.0% 16.7% 5.9% 3.2% 5.4% 1.9% Source: SONAPRA, 2006 However, the staple length improved significantly: the percentage of production classed as 1-1/8" rose from 19% in 2001/02 to 76 percent in 2005/06, and the average staple length rose from 27.9 to 28.4 mm. Table 9: Share of Lint based on Staple Length ­ Benin Length 2001/02 2002/03 2003/04 2004/05 2005/06 1-5/32" 0.0% 0.1% 0.0% 0.0% 0.2% 1-1/8" 19.0% 28.7% 16.8% 34.7% 76.1% 1-3/32" 79.9% 70.5% 83.1% 64.9% 23.7% 1-1/16" 1.1% 0.8% 0.1% 0.5% 0.0% Source: SONAPRA, 2006 71 African Standard 0 is close to Universal Standard Good Middling, Standard 1 to Strict Middling, and Standard 2 to Middling. 34 The average results of instrument tests over the last 5 seasons confirm the improvement in staple length (from 1.09 to 1.12 inches). The average uniformity is very stable close close to 82 percent. The short fiber index is on the high side (above 10 percent), while elongation is rather low (below 6 percent). Strength is good (29.5 to 30.5 gpt). Cotton is bright (Rd = 75) but not very white (+b 9.5 to 10). Although the average micronaire improved from 3.7 in 2001/02 to 4.0 during the last 3 seasons, this parameter remains a cause of concern. According to the 2005 ITMF contamination survey based on 8 mill evaluations, Benin is among the origins least contaminated (13th), least affected by stickiness (20th) and least affected by seed coat fragments (13th). Table 10: Indicators of Quality Performance - Benin Quality Indicators Benin Seed cotton Main varieties 100% STAM 18A Trend Ageing No. of producers 325,000 Overall quality of seed cotton Mixed Quality trend Down Contamination Moderate Stickiness Moderate Grading Lenient No. of seed cotton grades 2 % purchased as 1st grade n/a Quality control Lax Lint Type of ginning 100% SG No. of gins 18 Ginning outturn 42% Quality of ginning Mixed (declining) Overall quality of lint Higher grades Reputation Good Quality trend Down Strengths Length, strength Weaknesses Contamination, Micronaire Seed coat fragments Moderate Classification (no. of labs) Manual (1 lab Parakou) Types National SITC 2 SONAPRA + 2 private % SITC-classed 5% Classification rating Average % top types ­ SM/Standard 0 55% (05/06) Average staple length 28.4 mm (05/06) Burkina Faso Three varieties are grown in Burkina Faso (FK 290; STAM 59A; FK 37). Burkina Faso is in the process of testing and releasing biotech cotton. Seed cotton is graded in three grades by the cotton companies. As in other African countries the proportion of seed cotton classed as top grade is higher than it should be 35 as inspectors prefer to avoid conflict or personally take advantage from upgrading the seed cotton. All national lint production is classed by the major ginning company SOFITEX according to national types. Manual and visual classing is supplemented by a testing laboratory equipped with one SITC line. SOFITEX started a quality improvement strategy in 1992 that seems to be paying off. Quality of cotton improved significantly despite the sharp jump in production. Progress has also been in the reduction of contamination and stickiness. SOFITEX has begun the procedure for being be ISO certified. The proportion of production classed in Standard 0 reached a record of 26 percent in 2005/06, while the proportion classed below Standard 1 fell to a low of 25 percent. Table 11: Share of Lint based on Grade ­ Burkina Faso STD 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 0 17.5% 9.0% 11.0% 16.5% 12.0% 16.7% 15.0% 12.5% 24.3% 25.7% 1 44.8% 36.0% 32.2% 34.0% 45.0% 47.2% 41.5% 38.6% 39.5% 49.4% 2 28.4% 50.5% 51.5% 44.1% 41.8% 35.0% 39.8% 42.8% 29.8% 24.0% 3/4 9.3% 4.5% 5.4% 5.4% 1.2% 1.1% 3.8% 6.2% 6.5% 0.9% Source: SOFITEX 2006 The proportion of cotton classed 1-1/8" rose from 19.5 percent to 86 percent in 2001/02, and after relapsing to 69 percent rebounded to 80 percent in 2005/06. The average staple length rose from 27.9 mm in 1995/96 to 28.4 mm in 2005/06. Table 12: Share of Lint based on Staple Length ­ Burkina Faso L72 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 36 19.5% 71.5% 77.6% 71.0% 68.2% 66.0% 85.9% 84.3% 69.1% 69.2% 80.4% 35 79.4% 25.7% 16.9% 23.3% 26.1% 32.5% 12.4% 11.1% 24.2% 23.3% 17.8% 34 1.1.% 2.8% 5.5% 5.7% 5.7% 1.5% 1.7% 4.6% 6.7% 7.5% 1.8% Source: SOFITEX 2006 The average results of instrument tests conducted in 2005/06 are quite satisfactory: staple length of 29.4 mm; uniformity ratio 82.6 percent; strength 31.5 gpt; elongation 6.8 percent; mike 4.0. Thermodetector tests indicate that 70 percent of samples have no or low stickiness and 30 percent are moderately sticky. However, in the 2005 ITMF contamination survey, Burkina Faso is ranked among the most contaminated origins (19th), the origins most affected by stickiness (10th), and among the least affected by seed coat fragments (16th). These ratings are based on a survey of 15 mill evaluations. 72 US standards: 32 = 1"; 33 = 1-1/32"; 34 = 1-1/16"; 35 = 1-3/32"; 36 = 1-1/8" 36 Table 13: Indicators of Quality Performance ­ Burkina Faso Quality Indicators Burkina Faso Seed cotton Main varieties FK 290; STAM 59A; FK 37 Trend Improving (GM under test) No. of producers 250,000 Overall quality of seed cotton Good Quality trend Up Contamination Moderate (improving) Stickiness Low Grading Average No. of seed cotton grades 3 % purchased as 1st grade n/a Quality control Good Lint Type of ginning 100% SG No. of gins 15 Ginning outturn 42% Quality of ginning Good Overall quality of lint Higher grades Reputation Good Quality trend Up Strengths Length, strength, micronaire Weaknesses Contamination Seed coat fragments Moderate Classification (no. of labs) Manual (1 lab Bobo Dioulasso) Types National SITC 1 % SITC-classed Samples Classification rating Good % top types ­ SM/Standard 0 25% (05/06) Average staple length 28.4 mm (05/06) Cameroon Two main varieties are grown in Cameroon, one medium staple and one longer staple (50 percent IRMA 1239A; 45 percent IRMA BLT). Cotton research is funded by SODECOTON. Seed cotton is graded in two grades by the national cotton company, SODECOTON. Grading is stricter than in most other African countries. 73 percent of seed cotton was classed as first grade in 2005/06. All national lint production is classed manually and visually by SODECOTON in Garoua. There is no SITC but micronaire tests are made for each bale. IRMA cottons are white and bright but staple length (1-1/8" shy) as well as strength are on the low side. Micronaire problems are occasional. The expansion of BLT has enabled Cameroon to enter the segment of fine cottons. With a staple length of 1-5/32" to 1-3/16", BLT is sold at a premium of about 2 cents/lb over the traditional IRMA variety. The image of Cameroon cotton has benefited from this move to a premium market. On average 90 percent of the lint production was classed in Standards 0 and 1 since 1995/96 (between 79 percent and 97 percent depending on the weather). 37 Table 14: Share of Lint based on Grade ­ Cameroon STD 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 0 29.4% 39.8% 32.5% 41.6% 44.3% 52.8% 44.2% 41.0% 47.4% 34.5% 54.6% 1 57.4% 48.7% 46.6% 48.2% 46.6% 42.9% 49.0% 53.5% 50.0% 52.9% 34.1% 2 12.8% 9.9% 20.5% 8.9% 7.7% 4.2% 6.6% 4.2% 2.4% 11.7% 11.3% 3-4 1.1% 1.5% 0.4% 1.3% 1.4% 0.1% 0.3% 1.3% 0.2% 1.0% 0.0% Source: SODECOTON 2006 Over the last decade, the average length of cotton rose from 27.4 mm to 28.5 mm (+4%) Table 15: Share of Lint based on Staple Length ­ Cameroon 73 L 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 37 9.1% 9.1% 9.5% 18.8% 20.0% 27.2% 26.2% 30.6% 32.2% 22.2% 15.1% 36 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.4% 38.0% 24.1% 59.5% 35 27.4% 45.6% 31.7% 36.3% 39.1% 46.8% 31.8% 60.2% 29.4% 50.2% 25.4% 34 63.5% 45.2% 51.8% 41.9% 38.4% 25.7% 41.7% 8.7% 0.4% 3.5% 0.1% 33 0.0% 0.0% 7.0% 3.0% 2.5% 0.4% 0.2% 0.0% 0.0% 0.0% 0.0% Source: SODECOTON 2006 In the early 1990s, Cameroon cotton was very affected by stickiness. Its reputation of cotton suffered and prices were sharply discounted. SODECOTON developed a quality improvement strategy, which has been paying off. As stickiness mostly comes from late harvest, incentives for early picking were offered to farmers. Stickiness has virtually disappeared but the reputation still sticks. This confirms that it is difficult to regain a good reputation once it has been lost. Nevertheless, in the 2005 ITMF contamination survey, based on 14 mill evaluations, Cameroon is ranked as the origin most affected by stickiness, in contradiction wit the opinion of the merchants trading it. Cameroon is ranked 11th among the least contaminated origins and 19th among the origins least affected by seed coat fragments. 73 US standards: 32 = 1"; 33 = 1-1/32"; 34 = 1-1/16"; 35 = 1-3/32"; 36 = 1-1/8"; 37 = 1-5/32" 38 Table 16: Indicators of Quality Performance ­ Cameroon Quality Indicators Cameroon Seed cotton Main varieties 50% IRMA 1239A; 45% BLT; 5% D 742 Trend Improving (BLT) No. of producers 350,000 Overall quality of seed cotton Good Quality trend Up Contamination Moderate Stickiness Low (improved) Grading Average No. of seed cotton grades 2 % purchased as 1st grade 73% (05/06) Quality control Strict Lint Type of ginning 100% SG No. of gins 7 Ginning outturn 42% Quality of ginning Good Overall quality of lint Higher grades Reputation Good Quality trend Up Strengths Color (IRMA), length (BLT) Weaknesses Length/strength (IRMA); reputation (sticky) Seed coat fragments Moderate Classification (no. of labs) Manual (1 lab Garoua) Types National SITC 0 % SITC-classed 0% Classification rating Good % top types ­ SM/Standard 0 55% (05/06) Average staple length 28.5 mm (05/06) Mali Two main varieties are currently grown in Mali: STAM 59A (70 percent) and NTA 90- 5 (15 percent). Research is continuously breeding, testing and releasing new varieties. Seed cotton is graded in three grades by the producer associations. Grading is very lax. Large price differentials between the grades result in 99 percent of the crop being purchased as top grade, regardless of the lint classing results, whereas less than 90 percent actually deserves that rating. Grading used to be stricter when it was under the responsibility of CMDT and based on the lint classing results. Contamination is not taken into account during the buying process, and little care is given to the cleanliness of cotton before it reaches the gin. The use of plastic materials for picking, storing, transporting and baling is widespread. The expanded use of colored plastic materials, which are easier to detect visually than white plastic materials, has been an improvement. However, the efforts to educate farmers have had a limited impact so far because there is no financial incentive for farmers to deliver uncontaminated cotton and no discount is applied for contaminated cotton. 39 All national lint production is classed manually and visually by CMDT in three different locations, based on the same national types. This leads occasionally to problems of consistency in classing results. A SITC line has been installed in Koutiala in 2005 and 6 percent of production was instrument classed in 2005/06. An adverse feature of Malian cotton is its creamy color, which has a tendency to increase during storage under hot and humid conditions. The lack of maintenance of the gins for budgetary constraints has affected the quality of lint and increased seed coat fragments. During the last eleven seasons, the percentage of production classed in Standard 0 fluctuated between 6 percent and 24 percent, and the percentage of Standard 1 ranged from 38 percent to 56 percent. These fluctuations are largely weather-related. Table 17: Share of Lint based on Grade ­ Mali Std 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 0 9.3% 13.7% 11.4% 8.3% 6.0% 19.7% 20.5% 23.9% 6.5% 11.2% 8.8 % 1 55.8% 44.0% 53.2% 47.2% 52.4% 54.3% 46.3% 51.9% 37.5% 49.5% 54.5% 2 33.7% 36.9% 33.3% 41.3% 38.7% 24.9% 30.7% 21.5% 51.7% 37.0% 35.0% 3-4 1.3% 5.4% 2.2% 3.3% 2.9% 1.1% 2.6% 2.6% 4.2% 2.3% 1.6% Source: CMDT The proportion of cotton classed 1-1/8" (3rose from 13 percent in 1996/97 to 96 percent in 2005/06. Conversely, the proportion of cotton classed 1-3/32" (34) dropped from 86 percent to 2 percent. The average staple length rose from 27.9 mm in 1995/96 to 28.5 mm in 2005/06. Table 18: Share of Lint based on Staple Length ­ Mali 74 L 95/96 96/97 97/98 98/99 99/00 00/01 01/02 02/03 03/04 04/05 05/06 38 0.0% 0.0% 0.0% 0.1% 0.2% 3.1% 0.2% 0.3% 0.0% 0.0% 0.0% 37 0.0% 0.2% 0.3% 2.7% 7.7% 6.0% 6.4% 6.0% 2.2% 1.4% 2.0% 36 13.4% 13.3% 28.8% 53.6% 78.5% 81.8% 89.9% 61.2% 89.6% 91.4% 96.0% 35 83.1% 85.7% 70.8% 43.6% 13.6% 9.0% 3.5% 31.0% 8.1% 6.8% 1.8% 34 3.5% 0.8% 0.1% 0.0% 0.0% 0.0% 0.0% 1.4% 0.0% 0.4% 0.0% Source: CMDT In the 2005 ITMF contamination survey, based on 16 mill evaluations, Mali is ranked 16th among the most contaminated origins and fourth among the origins most affected by stickiness but 21st among the origins least affected by seed coat fragments. 74 US standards: 32 = 1"; 33 = 1-1/32"; 34 = 1-1/16"; 35 = 1-3/32"; 36 = 1-1/8"; 37 = 1-5/32"; 38 = 1- 3/16". 40 Table 19: Indicators of Quality Performance ­ Mali Quality Indicators Mali Seed cotton Main varieties 70% STAM 59A; 15% NTA 90-5 Trend Slowly improving No. of producers 200,000 Overall quality of seed cotton Mixed Quality trend Down Contamination Serious (improving) Stickiness Moderate Grading Very lenient No. of seed cotton grades 3 % purchased as 1st grade 99% Quality control Lax Lint Type of ginning 100% SG No. of gins 17 Ginning outturn 42% Quality of ginning Declining (lack of maintenance) Overall quality of lint Higher grades Reputation Poor Quality trend Down Strengths Length, strength, mic Weaknesses Color, contamination Seed coat fragments Moderate Classification (no. of labs) Manual (3 labs Fana-Koutiala-Sikasso) Types National HVI lines 1 % HVI-classed 6% Classification rating Average % top types ­ SM/Standard 0 10% (05/06) Average staple length 28.5 mm (05/06) Mozambique Mozambique grows eight varieties of cotton. The traditional varieties REMO-40 and CA-324 are outdated and are progressively replaced with new varieties from abroad in the new concession areas, with longer staple and much higher ginning outturn. Seed cotton is purchased on the basis of two grades. Pirate buyers do not grade seed cotton. Lint classing is based on visual and manual inspection, and is centralized, with exporting companies sending samples to IAM for testing. Current procedures are outdated and not consistent. Lint grades improved since 2003. Table 20: Share of Lint based on Grade ­ Mozambique Grade 2000 2001 2002 2003 SM 0% 0% 3% 0% M+ 9% 30% 23% 47% M VLS 49% 60% 63% 41% SM LS 30% 9% 9% 12% Lower types 12% 1% 2% 0% Source: IAM 41 Table 21: Share of Lint based on Grade ­ Mozambique (2005/06) Extra I II III IV V Inferior 0% 1% 17% 61% 17% 2% 1% Source: IAM Staple length used to be rather short and irregular but has increased in recent years. Table 22: Share of Lint based on Staple Length ­ Mozambique (2005/06) 1-5/32" 1-1/8" 1-1/16" 1-1/32" Average (mm) 1% 68% 27% 3% 28.3 Source: IAM Cotton quality is mixed in Mozambique. The overall reputation remains poor and the efforts of the new players may be penalized by the persisting low quality in the traditional cotton growing areas, although some buyers, at least, do distinguish between old and new suppliers. Mozambique is not ranked in the 2005 ITMF contamination survey because it has been evaluated by only four mill respondents. According to PT Apac Inti Corpora, Mozambique cotton is classed in the group II of moderately contaminated origins. Based on 500 tons inspected in 2004 and 2005, 82 percent of the bales were found to be contaminated, on average 13 grams per ton, including 75 percent of fibrous contaminants. Table 23: Indicators of Quality Performance ­ Mozambique Quality Indicators Mozambique Seed cotton Main varieties A637/REMO40/CA324/ISA205/ STAM42/IRMA1243/CHUREZA/ALBAR Trend Up No. of producers 200,000 Overall quality of seed cotton Poor Quality trend Up Contamination Moderate Stickiness Moderate Grading Mixed No. of seed cotton grades 2 % purchased as 1st grade n/a Quality control Mixed Lint Type of ginning 100% SG No. of gins 12 Ginning outturn 33%-41% Quality of ginning Poor Overall quality of lint Medium grades Reputation Poor Quality trend Flat Strengths Weaknesses Short staple, irregular Seed coat fragments n/a Classification (no. of labs) Centralized (IAM) Types Private SITC 0 % SITC-classed 0% Classification rating Poor % top types ­ SM/Standard 0 1% (05/06) Average staple length 28.3 mm (05/06) 42 Tanzania A new variety, UK 91 was introduced in 2005/06 growing season and has now replaced the two varieties that were previously cultivated in the Western cotton growing area (95 percent of national production): UK 77 and UK 82. Theoretically, buyers and ginners are obliged to purchase seed cotton on the basis of two grades (AR and BR). In practice, grading has disappeared at first point of purchase and different grades are purchased and ginned together. Due to ginning overcapacity, buyers and ginners give priority to volume and purchase seed cotton regardless of quality. The proportion of lower grade cotton has almost certainly increased since liberalization due to the initial decline in input use and mixing of seed cotton varieties. In recent years, steps have been taken to rectify both of these failings, with uncertain results. In this competitive system, lint classification is centralized at the Tanzania Cotton Board. Tanzania cotton is sold on the world market on the basis of types (representing staple length Type I = 1-1/8"; type II = 1-1/32"; type III = 1-1/16") and national grades. Standard grades for WCGA are TANG (GM), GANY (Middling) and YIKA (GO). The proportion of upper grades (Middling or better) fell from 45 percent in the early 90s to a low of 17 percent in 1994/95 after liberalization, and rebounded to about 80 percent in marketing seasons 2005/06 and 2006/07. Table 24: Share of Lint based on Grade ­ Tanzania Mktg Season Tang (SM) M+ GANY (M) Below GANY YIKA/below 1990/91 4.1% 13.8% 22.1% 59.0% 0.9% 1991/92 4.1% 9.0% 14.1% 70.6% 2.2% 1992/93 3.5% 14.3% 27.9% 53.8% 0.5% 1993/94 1.4% 6.5% 37.1% 54.6% 0.8% 1994/95 0.0% 6.7% 16.4% 73.6% 9.3% 1995/96 0.0% 12.7% 21.0% 60.2% 6.1% 1996/97 0.0% 16.8% 34.5% 45.4% 3.4% 1997/98 1.1% 15.1% 28.8% 52.1% 2.9% 1998/99 0.3% 7.2% 24.5% 50.8% 17.25 1999/00 11.4% 29.2% 36.3% 20.7% 2.4% 2000/01 4.9% 24.1% 44.8% 26.1% 0.0% 2001/02 0.0% 8.2% 50.0% 41.8% 0.0% 2002/03 0.0% 8.5% 52.1% 38.3% 1.2% 2003/04 0.3% 10.8% 59.4% 29.4% 0.1% 2004/05 1.4% 3.2% 51.8% 14.3% 0.0% 2005/06 1.9% 41.6% 39.8% 16.0% 1.2% 2006/07 0.4% 23.4% 55.1% 20.6% 0.5% Source: TCB The staple length of Tanzanian cotton is rather short and irregular, despite the fact that about half of cotton production saw ginned and the other half roller ginned. Table 25: Share of Lint based on Staple Length ­ Tanzania Marketing Season 1-1/8" 1-1/16" 1-1/32" Average (mm) 2005/06 17.7% 69.8% 12.5% 27.82 2006/07 19.2% 69.3% 11.5% 27.84 Source: TCB 43 Roller ginned cotton fetches a premium of one cent per pound over saw ginned cotton with the same grade and type. Yet other aspects of Tanzanias system have prevented it from building on this quality advantage. Tanzanian cotton has the reputation of being among the most seriously contaminated origins in the world. As a result, exports have shifted to a lower segment of the market and the quality premium has fallen since liberalization. Tanzania is not ranked in the 2005 ITMF contamination survey because it has been evaluated by only three mill respondents. PT Apac Inti Corpora classed Tanzania cotton in the group III of seriously contaminated origins. Based on 750 tons inspected in 2004 and 2005, 97 percent of the bales were found to be contaminated, on average 43 grams per ton, including 86 percent of fibrous contaminants. Table 26: Indicators of Quality Performance ­ Tanzania Quality Indicators Tanzania Seed cotton Main varieties 50% UK 82; 25% UK 77; 25%UK 91 Trend Up No. of producers 350,000 Overall quality of seed cotton Poor Quality trend Down Contamination Very serious Stickiness n/a Grading Very lax No .of seed cotton grades 2 st % purchased as 1 grade 99% (est.) Quality control Poor Lint Type of ginning 55% SG / 45% RG No. of gins 18 SG / 34 RG Ginning outturn 34-37% Quality of ginning Mixed Overall quality of lint Medium/higher grades Reputation Bad Quality trend Down Strengths RG Weaknesses Mixed, contamination Seed coat fragments Serious with old equipment Classification (no. of labs) Manual (1 lab) Types National SITC 1 % SITC-classed Sample basis Classification rating Average % top types ­ SM/Standard 0 0.4% (06/07) Average staple length 27.85 mm (06/07) Uganda Only one variety (BPA) is cultivated throughout the country. Cotton varieties are replaced frequently and the multiplication mechanism is effective. 44 The country still occupies a special niche-market position by supplying roller ginned lint, and fiber quality is still considered relatively high. Yet the quality control system at primary collection level has largely broken down. Ginning overcapacity has led to a scramble for cotton and a large number of buyers purchase seed cotton regardless of quality, forcing others to do the same. The primary manifestation of the breakdown is contamination: because Ugandan cotton is reputed to be very contaminated, the premium its lint attracts has shrunk during the last decade. The former system of quality control, where seed cotton is purchased on the basis of two grades (AR and BR) has been maintained on paper. In practice, quality control at the primary level has broken down. Ginning overcapacity has led to a scramble for cotton and a large number of buyers purchase seed cotton regardless of quality, forcing others to do the same. Lint classing is centralized. CDO is operating one SITC. 92 percent was classed 1-1/8" in 2006/07. The average staple length of 28.6 mm is an unimpressive performance considering the fact that almost all cotton is roller ginned (3 percent was saw ginned in 2006/07) and, as such, at least one staple longer than the same cotton if it were saw ginned. Table 27: Share of Lint based on Grade ­ Uganda (2006/07) RG - GM RG - SM RG - M RG -SLM RG - LM SG - SM SG - M OTHER 2.6% 15.7% 43.2% 26.1% 6.1% 0.6% 2.6% 3.2% Source: CDO Table 28: Share of Lint based on Staple Length ­ Uganda (2006/07) 1-5/32" 1-1/8" 1-1/16" Average (mm) 5% 92% 3% 28.6 Source: CDO In the 1970s Uganda cotton was considered the best upland cotton in Africa. Uganda still occupies a special niche-market position by supplying mostly roller ginned lint, and quality is considered relatively high. Yet, as cotton is reputed as being very contaminated, and the premium attached to Ugandan lint has shrunk during the last decade. Uganda is not ranked in the 2005 ITMF contamination survey because it has been evaluated by only one mill respondent. According to PT Apac Inti Corpora, based on 4,600 tons inspected in 2004 and 2005, Uganda cotton is classed in the group III of seriously contaminated origins (96% of the bales were found to be contaminated, on average 20 grams per ton, including 83% of fibrous contaminants). 45 Table 29: Indicators of Quality Performance ­ Uganda Quality Indicators Uganda Seed cotton Main varieties 60% BPA 2000; 40% BPA 99 Trend Improving No. of producers n/a Overall quality of seed cotton Good Quality trend Down (recently reversed) Contamination Serious Stickiness Low Grading Average No. of seed cotton grades 2 % purchased as 1st grade n/a Quality control Lax Lint Type of ginning 97% RG / 3% SG No. of gins 36 RG / 1 SG Ginning outturn 36% Quality of ginning Mixed Overall quality of lint Higher grades Reputation Good Quality trend Down (recently reversed) Strengths RG, length, color, brightness Weaknesses mixed, contamination Seed coat fragments Low Classification (no. of labs) Manual (1 lab) Types Private SITC 1 % SITC-classed Sample basis Classification rating Average % top types ­ SM/Standard 0 30% (06/07) Average staple length 28.6 mm (06/07) Zambia Zambia has delivered the best quality performance in Africa over the past five years. Two main varieties are grown in Zambia (F135 and Chureza). These varieties are ageing. A new variety (CDT2) is coming up. In this concentrated system, the two major ginners have given incentives to producers to deliver higher quality seed cotton, providing them with good quality planting seeds and crop protection chemicals, as well as extension services and technical advice. Seed cotton grading in three grades is strict and consistent. The dominant ginning company (Dunavant) provides farmers with cotton bags for picking cotton and has established manual cleaning stations at the entrance of the gins to manually remove contaminants (primarily polypropylene). Its main competitor (Cargill) has put more emphasis on changing farmer behavior, and claims to achieve similar low levels of contamination without the cleaning stations. There is no national classing system for lint. According to Dunavant, in 2005/06, 94 percent of their production (an estimated 66 percent of the national crop) was classed as Strict Middling or better. Fiber characteristics are satisfactory. The estimated average staple length is 28.5 mm. 46 Table 30: Share of Lint based on Grade ­ Zambia 2001/02 2002/03 2003/04 2004/05 2005/06 GM 0% 1% 12% 5% 7% SM+ 64% 18% 39% 62% 72% SM 20% 35% 23% 20% 15% M 13% 24% 15% 3% 3% M Sp 2% 15% 6% 8% 2% SLM LM 1% 7% 5% 2% 1% Source: Dunavant 2006 As a result of these efforts, Zambian cotton over the past five years has gained the reputation of being among the least contaminated in Africa. Smaller ginners, however, do not perform as well in terms of contamination. In the 2005 ITMF contamination survey, Zambia is ranked among the most contaminated (20th) origins, but among the origins least affected by stickiness (6th) and least affected by seed coat fragments (14th). PT Apac Inti Corpora classed Zambian cotton in the group II of moderately contaminated origins. Based on 300 tons inspected in 2004 and 2005, 75 percent of the bales were found to be contaminated, on average 6.5 grams per ton, including 79 percent of fibrous contaminants. Table 31: Indicators of Quality Performance ­ Zambia Quality Indicators Zambia Seed cotton Main varieties 60% F135; 40% Chureza Trend Ageing No. of producers 280,000 Overall quality of seed cotton Very good Quality trend Up Contamination Low (improving) Stickiness Low Grading Good No. of seed cotton grades 3 % purchased as 1st grade 80% Quality control Strict Lint Type of ginning 95% SG; 5% RG No. of gins 13/1 Ginning outturn 40% Quality of ginning Mixed Overall quality of lint Higher grades Reputation Very good Quality trend Up Strengths Low contamination Weaknesses Seed coat fragments Moderate Classification (no. of labs) Manual/HVI (n/a) Types Private SITC 1 % SITC-classed 70-80% Classification rating Good % top types - Standard 0 80% (05/06) Average staple length (mm) 28.5 mm (est.) 47 Zimbabwe Albar varieties cultivated in Zimbabwe have excellent fiber characteristics compatible with modern spinning technology. Quton Seed Company, a subsidiary of the Cotton Company group, has exclusive rights to commercialize cotton varieties developed by the Government Cotton Research Institute in Kadoma and is also developing its own varieties. For years, cotton produced in Zimbabwe has enjoyed the reputation of being among the least contaminated origins in the world. Quality was a high priority when the Zimbabwe Cotton Company had a monopoly for cotton marketing. Seed cotton grading was very strict and pricing was closely related to lint classing results. Contamination was a major criterion during the buying process. A first visual inspection was done at the buying station and another inspection was preformed at the gin before processing. The use of non-cotton picking bags was prohibited and bales were wrapped in cotton fabric. The high quality was maintained during the effective "duopoly" years of 1995 to 2001 but quality has deteriorated since 2002. "Zimbabwe since the inception of major commercial production has always had a grading system, whereby seed cotton at point of purchase would be graded into four basic seed cotton grades A, B, C and D. This system ensured that lint quality after ginning would be homogenous, resulting in premiums for the lint product from clients and consequently, better returns to the grower.(...) With the influx of new players into the industry the grading standards have been undermined to the point of being non existent. Grading has disappeared at first point of purchase and different grades are purchased and ginned together. This has resulted in a marked deterioration in lint quality from new entrants. Established players such as the Cotton Company and Cargill have maintained these standards." (Mapara, 2006). These dramatic changes in seed cotton buying practices have already affected the premiums for Zimbabwe cotton in the international market. The major deterioration is not in the premium achieved by the top grades, but in the proportion of lint classed in the top grades. The estimated average staple length is 28.6 mm. In the 2005 ITMF contamination survey of fifteen mill evaluations, Zimbabwe cotton ranked among the least contaminated origins (5th), the origins least affected by stickiness (2nd) and the least affected by seed coat fragments (2nd). However, PT Apac Inti Corpora classified Zimbabwe cotton as having seriously contaminated origins. Based on 2,900 tons inspected in 2004 and 2005, 93 percent of the bales were found to be contaminated, with an average of 28 grams per ton, including 82 percent of fibrous contaminants. This cotton was likely supplied by newly established ginners. 48 Table 32 - Indicators of Quality Performance ­ Zimbabwe Quality Indicators Zimbabwe Seed cotton Main varieties 75% Albar SZ 9314; 15% Albar FQ 902 Trend Improving (GM under test) No. of producers >250,000 Overall quality of seed cotton Very good Quality trend Down Contamination Low (worsening) Stickiness None Grading Lax No. of seed cotton grades 1 to 4 % purchased as 1st grade n/a Quality control Eroding Lint Type of ginning 95% SG; >5% RG No. of gins 15>1 Ginning outturn 41% Quality of ginning Mixed Overall quality of lint Higher grades Reputation Very good Quality trend Down Strengths Color, brightness, low contamination Weaknesses Contamination rising Seed coat fragments Low Classification (no. of labs) Manual (? Labs) Types Private SITC 1 % SITC-classed Sample basis Classification rating Good (declining) % top types ­ SM/Standard 0 35% (est.) Average staple length (mm) 28.6 mm (est.) 49 ANNEX 3: LINT MARKETING BY COUNTRY Benin Table 33: Indicators of Marketing Performance ­ Benin Marketing Indicators Benin Seed cotton No. of ginners 7 Middlemen No Ginning season Dec-May Payment to farmers Mixed Average km collection pt to gin 60 km Lint / Export sales Contracting One-to-one / tenders No. of lint buyers 20 Offering Year-round Contract types Fixed forward & spot () Position FOB Cotonou Top type KABA-S 1"1/8 Price differential/Cotlook A + 2.5 cts/lb Trend Down Bale covering PP/cotton Bale weight 225 kg (est.) Average selling price (05/06) N/A Transportation of lint Truck % loaded in container at gin 0% (est.) Average distance gin-port 435 km FOT gin to CFR (cts/lb) 7.5 cts/lb Reputation of shippers Mixed Domestic sales No. of spinning mills (operating) 1 Consumption (tons) 3,000 % of crop 2% Trend Shrinking Pricing WP less transport 50 Burkina Faso Table 34: Indicators of Marketing Performance ­ Burkina Faso Marketing Indicators Burkina Faso Seed cotton No. of ginners 3 Middlemen No Ginning season Nov-May Payment to farmers 10 days Average km collection pt to gin 95 km Lint / Export sales Contracting One-to-one No. of lint buyers 12 Offering Year-round Contract types Fixed forward & spot () Position FOB Abidjan/Lomé Top type BOLA/S - BOBY/S 1"1/8 Price differential/Cotlook A + 3 cts/lb Trend Up Bale covering PP Bale weight 225 kg Average selling price (05/06) n/a Transportation of lint Truck % loaded in container at gin 0% (est.) Average distance gin-port 1,000 km FOT gin to CFR (cts/lb) 10.5 cts/lb Reputation of shippers Reliable Domestic sales No. of spinning mills (operating) 1 Consumption (tons) 2,000 % of crop <1% Trend Shrinking Pricing WP less transport 51 Cameroon Table 35: Indicators of Marketing Performance ­ Cameroon Marketing Indicators Cameroon Seed cotton No. of ginners 1 Middlemen No Ginning season Dec-April Payment to farmers n/a Average km collection pt to gin 85 km Lint / Export sales Contracting Tenders No. of lint buyers 10 Offering Year-round Contract types Fixed forward & spot () Position FOB Douala Top type PLEBE 1"5/32 (IRMA-S 1"1/8) Price differential/Cotlook A + 4 cts/lb (IRMA-S + 2 cts/lb) Trend Up Bale covering PP Bale weight 215 kg Average selling price (05/06) Cotlook A Index + 6% Transportation of lint Truck/train % loaded in container at gin 0 Average distance gin-port 1,300 km FOT gin to CFR (cts/lb) 11.5 cts/lb Reputation of shippers Reliable Domestic sales No. of spinning mills (operating) 1 Consumption (tons) 3,000 % of crop 3% Trend Shrinking Pricing WP less transport 52 Mali Table 36: Indicators of Marketing Performance ­ Mali Marketing Indicators Mali Seed cotton No. of ginners 1 Middlemen No Ginning season Nov-May Payment to farmers 10/15 days Average km collection pt to gin 95 km Lint / Export sales Contracting Tenders / one-to-one No. of lint buyers 15 Offering Year-round Contract types Fixed forward & spot (), Minimum guaranteed Position FOB Abidjan/Dakar/Lomé/Tema Top type SARAMA 1"1/8 (JULI-S 1"1/8) Price differential/Cotlook A + 2.5 cts/lb (+ 2) Trend Down Bale covering PP/PE/cotton Bale weight 227 kg Average selling price (05/06) Cotlook A Index Transportation of lint Truck/train % loaded in container at gin 5% (est.) Average distance gin-port 1,000 km FOT gin to CFR (cts/lb) 11 cts/lb Reputation of shippers Improving Domestic sales No. of spinning mills (operating) 2 Consumption (tons) 2,000 % of crop <1% Trend Stagnant Pricing WP less transport 53 Mozambique Table 37: Indicators of Marketing Performance ­ Mozambique Marketing Indicators Mozambique Seed cotton No. of ginners 9 Middlemen No Ginning season May-Sep Payment to farmers n/a Average km collection pt to gin n/a Lint / Export sales Contracting One-to-one No. of lint buyers n/a Offering Seasonal Contract types Spot Position FOT gin/FOB Beira Top type SM 1"1/8 Price differential/Cotlook A + 1 ct/lb Trend Up Bale covering Mixed Bale weight n/a Average selling price n/a Transportation of lint Truck/train % loaded in container at gin 0 (est.) Average distance gin-port n/a FOT gin to CFR (cts/lb) 8.5 cts/lb (est.) Reputation of shippers Mixed Domestic sales No. of spinning mills (operating) 0 Consumption (tons) 0 % of crop 0% Trend Stagnant Pricing 54 Tanzania Table 38: Indicators of Marketing Performance ­ Tanzania Marketing Indicators Tanzania Seed cotton No. of ginners 33 (54 buyers) Middlemen Yes Ginning season June - October (Western area) Payment to farmers Cash Average km collection pt to gin 100 km Lint / Export sales Contracting One-to-one No. of lint buyers 14 Offering Seasonal Contract types Spot Position FOB Dar es Salam / FOT gin Top type RG1 1"1/8 Price differential/Cotlook A +0.5 ct/lb (SG 1: -0.5 ct/lb) Trend Down Bale covering Mixed Bale weight SG 200-270 kg / RG 181 kg Average selling price n/a Transportation of lint Road / train % loaded in container at gin n/a Average distance gin-port 1,100 km FOT gin to CFR (cts/lb) 8 cts/lb Reputation of shippers Mixed Domestic sales No. of spinning mills (operating) 12 Consumption (tons) 15,000 % of crop 15% Trend Stagnant Pricing n/a 55 Uganda Table 39: Indicators of Marketing Performance ­ Uganda Marketing Indicators Uganda Seed cotton No. of ginners 20 Middlemen Yes Ginning season Dec-Apr Payment to farmers Cash Average km collection pt to gin n/a Lint / Export sales Contracting One-to-one No. of lint buyers 25 Offering seasonal Contract types Spot Position FOT gin/FOB Mombasa Top type RG1 1"1/8 Price differential/Cotlook A + 3 cts/lb Trend Down Bale covering Hessian cloth Bale weight 181-185 kg Average selling price n/a Transportation of lint Truck % loaded in container at gin 0 (est.) Average distance gin-port n/a FOT gin to CFR (cts/lb) 10.5 cts/lb Reputation of shippers Mixed Domestic sales No. of spinning mills (operating) 3 Consumption (tons) 2,000 % of crop 7% Trend Stagnant Pricing n/a 56 Zambia Table 40: Indicators of Marketing Performance ­ Zambia Marketing Indicators Zambia Seed cotton No. of ginners 7 Middlemen Few Ginning season May-Sep Payment to farmers Cash/max 10 days Average km collection pt to gin 120 km Lint / Export sales Contracting One-to-one No. of lint buyers 10 (est.) Offering Year-round Contract types Fixed forward & spot Position FOT gin/FOB Durban Top type SM 1"1/8 (GM) Price differential/Cotlook A + 6 cts/lb Trend Up Bale covering PP/cotton Bale weight 180/220 kg Average selling price n/a Transportation of lint Truck % loaded in container at gin n/a Average distance gin-port n/a FOT gin to CFR (cts/lb) 10 cts/lb Reputation of shippers Mixed Domestic sales No. of spinning mills (operating) 4 Consumption (tons) 12,000 % of crop 15% Trend Shrinking Pricing Export parity price (Sales tax 35%) 57 Zimbabwe Table 41: Indicators of Marketing Performance ­ Zimbabwe Marketing Indicators Zimbabwe Seed cotton No. of ginners 5 Middlemen Few Ginning season May-Sep Payment to farmers Cash Average km collection pt to gin 110 km Lint / Export sales Contracting One-to-one No. of lint buyers 10 (est.) Offering Year-round Contract types Fixed forward & spot Position CFR - FOB Durban - FOT gin Top type SM 1"1/8 Price differential/Cotlook A + 7 cts/lb Trend Down Bale covering Cotton Bale weight 210 kg Transportation of lint Truck / train % loaded in container at gin n/a Average distance gin-port 1,300 km FOT gin to CFR (cts/lb) 9.5 cts/lb Reputation of shippers Mixed Domestic sales No. of spinning mills (operating) 5 Consumption (tons) 24,000 % of crop 25% Trend Stagnant Pricing Subsidized 58 ANNEX 4: REFERENCES Association Cotonnière Africaine Journées qualité de lA.C.A. Presentations and reports. Cotonou, Bénin. November 8th ­ 10th, 2006 Association Française Cotonnière (AFCOT) Séminaire sur la Qualité. Abidjan, Cote dIvoire. June, 2002 Baffes, John. "Uganda Country Study", Draft Report, Comparative Analysis of Structure and Performance of African Cotton Sectors, World Bank, October, 2006 Bremen Cotton Exchange "Cotton Varieties by Origins" 2006 Boughton, Ducan. Tschirley, David. Pitoro, Raul. "Mozambique Country Study", Draft Report, Comparative Analysis of Structure and Performance of African Cotton Sectors, World Bank, October, 2006 Conrad P. Lyford, Sangyeol Yung and Don E. Ethridge. "The Mill-Level Price of Quality Cotton in the U.S." Texas Tech University Lubbock, Texas. 2004 Beltwide Cotton Conferences ­ San Antonio, Texas, January 5th ­ 9th Cotton Incorporated. "The Classification of Cotton" 2007 Faser Institut Bremen & Bremen Cotton Exchange. 28th International Cotton Conference Bremen Proceedings. March 22nd ­ 25th, 2006 Fok, Michel Bachelier, Bruno. "Identification dun plan daction damélioration de la qualité et de la valorisation de la qualité du coton dans les pays de lUMEOA"ONUDI. December, 2004 Gergely, Nicolas. "Benin Country Study" "Burkina Faso Country Study", "Cameroon Country Study" "Mali Country Study". Draft Reports, Comparative Analysis of Structure and Performance of African Cotton Sectors, World Bank, October, 2006 Poulton , Colin. "Tanzania Country Study", "Zimbabwe Country Study", Draft Reports, Comparative Analysis of Structure and Performance of African Cotton Sectors, World Bank, October, 2006 Hunter, Lawrence. "Cotton Fibre Properties: Their Impact on Textile Processing Performance and Costs" "CSIR Materials Science and Manufacture and Nelson Mandela Metropolitan University" Port Elizabeth, South Africa 59 International Cotton Advisory Committee (ICAC). "World Cotton Statistics" September, 2006 International Cotton Advisory Committee (ICAC) Fiber Quality Needs of the Modern Spinning Industry and Advances in Ginning Research. Report of an expert panel on ginning methods. September, 2001 International Cotton Advisory Committee (ICAC). Technical Seminar, 58th Plenary Meeting, Charleston, USA, October, 1999 International Textile Manufacturers Federation (ITMF). "Cotton Contamination Surveys" 1999, 2001, 2003, 2005 Larsen, Marianne Nylandsted. "Quality standard-setting in the global cotton chain and cotton sector reforms in sub-Saharan Africa" Institute for International Studies. Copenhagen, Denmark Mapara, H. Cottco. "Cotton in Zimbabwe" Harare, Zimbabwe. Marinus H.J. van der Sluijs, Stuart Gordon and Martin W. Prins."Australian cotton: How good is it really? - A report on the quality needs and perceptions of Australian cotton by domestic and overseas spinning mills" The Australian Cotton CRC and the Cotton Textile Research Unit, CSIRO Textile and Fibre Technology, Geelong. The Australian Cotton Grower April ­ May, 2004 Schindler, C.P. "The ITMF Contamination Survey" International Textile Manufacturers Federation, Zurich, Switzerland. 2005 Tschirley, David. "Zambia Country Study", Draft Report, Comparative Analysis of Structure and Performance of African Cotton Sectors, World Bank, October, 2006 "Extraneous Contamination in Raw Cotton Bales ­ A Nightmare to Spinners" Vijayshankar PT. Apac Inti Corpora, lndonesia "2006 Buyers Guide" Cotton Council International. "Cotton Trading Manual" Wood head publishing, 2005 "Manuel qualité pour les filières cotonnières" UEMOA. July, 2006 "Cotton Outlook", Various issues 60 Africa Region Working Paper Series Series # Title Date Author ARWPS 1 Progress in Public Expenditure Management in Africa: January 1999 C. Kostopoulos Evidence from World Bank Surveys ARWPS 2 Toward Inclusive and Sustainable Development in the March 1999 Markus Kostner Democratic Republic of the Congo ARWPS 3 Business Taxation in a Low-Revenue Economy: A June 1999 Ritva Reinikka Study on Uganda in Comparison with Neighboring Duanjie Chen Countries ARWPS 4 Pensions and Social Security in Sub-Saharan Africa: October 1999 Luca Barbone Issues and Options Luis-A. Sanchez B. ARWPS 5 Forest Taxes, Government Revenues and the January 2000 Luca Barbone Sustainable Exploitation of Tropical Forests Juan Zalduendo ARWPS 6 The Cost of Doing Business: Firms' Experience with June 2000 Jacob Svensson Corruption in Uganda ARWPS 7 On the Recent Trade Performance of Sub-Saharan August 2000 Francis Ng and African Countries: Cause for Hope or More of the Alexander J. Yeats Same ARWPS 8 Foreign Direct Investment in Africa: Old Tales and November 2000 Miria Pigato New Evidence ARWPS 9 The Macro Implications of HIV/AIDS in South Africa: A November 2000 Channing Arndt Preliminary Assessment Jeffrey D. Lewis ARWPS 10 Revisiting Growth and Convergence: Is Africa December 2000 C. G. Tsangarides Catching Up? ARWPS 11 Spending on Safety Nets for the Poor: How Much, for January 2001 William J. Smith How Many? The Case of Malawi ARWPS 12 Tourism in Africa February 2001 Iain T. Christie D. E. Crompton ARWPS 13 Conflict Diamonds February 2001 Louis Goreux ARWPS 14 Reform and Opportunity: The Changing Role and March 2001 Jeffrey D. Lewis Patterns of Trade in South Africa and SADC ARWPS 15 The Foreign Direct Investment Environment in Africa March 2001 Miria Pigato ARWPS 16 Choice of Exchange Rate Regimes for Developing April 2001 Fahrettin Yagci Countries ARWPS 18 Rural Infrastructure in Africa: Policy Directions June 2001 Robert Fishbein ARWPS 19 Changes in Poverty in Madagascar: 1993-1999 July 2001 S. Paternostro J. Razafindravonona David Stifel ARWPS 20 Information and Communication Technology, Poverty, August 2001 Miria Pigato and Development in sub-Sahara Africa and South Asia ARWPS 21 Handling Hierarchy in Decentralized Settings: September Navin Girishankar A. Governance Underpinnings of School Performance in 2001 Alemayehu Tikur Inchini, West Shewa Zone, Oromia Region Yusuf Ahmad 61 Africa Region Working Paper Series Series # Title Date Author ARWPS 22 Child Malnutrition in Ethiopia: Can Maternal October 2001 Luc Christiaensen Knowledge Augment The Role of Income? Harold Alderman ARWPS 23 Child Soldiers: Preventing, Demobilizing and November 2001 Beth Verhey Reintegrating ARWPS 24 The Budget and Medium-Term Expenditure December 2001 David L. Bevan Framework in Uganda ARWPS 25 Design and Implementation of Financial Management January 2002 Guenter Heidenhof H. Systems: An African Perspective Grandvoinnet Daryoush Kianpour B. Rezaian ARWPS 26 What Can Africa Expect From Its Traditional Exports? February 2002 Francis Ng Alexander Yeats ARWPS 27 Free Trade Agreements and the SADC Economies February 2002 Jeffrey D. Lewis Sherman Robinson Karen Thierfelder ARWPS 28 Medium Term Expenditure Frameworks: From February 2002 P. Le Houerou Concept to Practice. Preliminary Lessons from Africa Robert Taliercio ARWPS 29 The Changing Distribution of Public Education February 2002 Samer Al-Samarrai Expenditure in Malawi Hassan Zaman ARWPS 30 Post-Conflict Recovery in Africa: An Agenda for the April 2002 Serge Michailof Africa Region Markus Kostner Xavier Devictor ARWPS 31 Efficiency of Public Expenditure Distribution and May 2002 Xiao Ye Beyond: A report on Ghana's 2000 Public Expenditure S. Canagaraja Tracking Survey in the Sectors of Primary Health and Education ARWPS 34 Putting Welfare on the Map in Madagascar August 2002 Johan A. Mistiaen Berk Soler T. Razafimanantena J. Razafindravonona ARWPS 35 A Review of the Rural Firewood Market Strategy in August 2002 Gerald Foley West Africa P. Kerkhof, D. Madougou ARWPS 36 Patterns of Governance in Africa September Brian D. Levy 2002 ARWPS 37 Obstacles and Opportunities for Senegal's September Stephen Golub International Competitiveness: Case Studies of the 2002 Ahmadou Aly Mbaye Peanut Oil, Fishing and Textile Industries ARWPS 38 A Macroeconomic Framework for Poverty Reduction October 2002 S. Devarajan Strategy Papers : With an Application to Zambia Delfin S. Go ARWPS 39 The Impact of Cash Budgets on Poverty Reduction in November 2002 Hinh T. Dinh Zambia: A Case Study of the Conflict between Well Abebe Adugna Intentioned Macroeconomic Policy and Service Bernard Myers Delivery to the Poor ARWPS 40 Decentralization in Africa: A Stocktaking Survey November 2002 Stephen N. Ndegwa 62 Africa Region Working Paper Series Series # Title Date Author ARWPS 41 An Industry Level Analysis of Manufacturing December 2002 Professor A. Mbaye Productivity in Senegal ARWPS 42 Tanzania's Cotton Sector: Constraints and December 2002 John Baffes Challenges in a Global Environment ARWPS 43 Analyzing Financial and Private Sector Linkages in January 2003 Abayomi Alawode Africa ARWPS 44 Modernizing Africa's Agro-Food System: Analytical February 2003 Steven Jaffee Framework and Implications for Operations Ron Kopicki Patrick Labaste Iain Christie ARWPS 45 Public Expenditure Performance in Rwanda March 2003 Hippolyte Fofack C. Obidegwu Robert Ngong ARWPS 46 Senegal Tourism Sector Study March 2003 Elizabeth Crompton Iain T. Christie ARWPS 47 Reforming the Cotton Sector in SSA March 2003 Louis Goreux John Macrae ARWPS 48 HIV/AIDS, Human Capital, and Economic Growth April 2003 Channing Arndt Prospects for Mozambique ARWPS 49 Rural and Micro Finance Regulation in Ghana: June 2003 William F. Steel Implications for Development and Performance of the David O. Andah Industry ARWPS 50 Microfinance Regulation in Benin: Implications of the June 2003 K. Ouattara PARMEC LAW for Development and Performance of the Industry ARWPS 51 Microfinance Regulation in Tanzania: Implications for June 2003 Bikki Randhawa Development and Performance of the Industry Joselito Gallardo ARWPS 52 Regional Integration in Central Africa: Key Issues June 2003 Ali Zafar Keiko Kubota ARWPS 53 Evaluating Banking Supervision in Africa June 2003 Abayomi Alawode ARWPS 54 Microfinance Institutions' Response in Conflict June 2003 Marilyn S. Manalo Environments: Eritrea- Savings and Micro Credit Program; West Bank and Gaza ­ Palestine for Credit and Development; Haiti ­ Micro Credit National, S.A. AWPS 55 Malawi's Tobacco Sector: Standing on One Strong leg June 2003 Steven Jaffee is Better than on None AWPS 56 Tanzania's Coffee Sector: Constraints and Challenges June 2003 John Baffes in a Global Environment AWPS 57 The New Southern AfricanCustoms Union Agreement June 2003 Robert Kirk Matthew Stern AWPS 58a How Far Did Africa's First Generation Trade Reforms June 2003 Lawrence Hinkle Go? An Intermediate Methodology for Comparative A. Herrou-Aragon Analysis of Trade Policies Keiko Kubota 63 Africa Region Working Paper Series Series # Title Date Author AWPS 58b How Far Did Africa's First Generation Trade Reforms June 2003 Lawrence Hinkle Go? An Intermediate Methodology for Comparative A. Herrou-Aragon Analysis of Trade Policies Keiko Kubota AWPS 59 Rwanda: The Search for Post-Conflict Socio-Economic October 2003 C. Obidegwu Change, 1995-2001 AWPS 60 Linking Farmers to Markets: Exporting Malian October 2003 Morgane Danielou Mangoes to Europe Patrick Labaste J-M. Voisard AWPS 61 Evolution of Poverty and Welfare in Ghana in the October 2003 S. Canagarajah 1990s: Achievements and Challenges Claus C. Pörtner AWPS 62 Reforming The Cotton Sector in Sub-Saharan Africa: November 2003 Louis Goreux SECOND EDITION AWPS 63 (E) Republic of Madagascar: Tourism Sector Study November 2003 Iain T. Christie D. E. Crompton AWPS 63 (F) République de Madagascar: Etude du Secteur November 2003 Iain T. Christie Tourisme D. E. Crompton AWPS 64 Migrant Labor Remittances in Africa: Reducing Novembre 2003 Cerstin Sander Obstacles to Development Contributions Samuel M. Maimbo AWPS 65 Government Revenues and Expenditures in Guinea- January 2004 Francisco G. Carneiro Bissau: Casualty and Cointegration Joao R. Faria Boubacar S. Barry AWPS 66 How will we know Development Results when we see June 2004 Jody Zall Kusek them? Building a Results-Based Monitoring and Ray C. Rist Evaluation System to Give us the Answer Elizabeth M. White AWPS 67 An Analysis of the Trade Regime in Senegal (2001) June 2004 Alberto Herrou-Arago and UEMOA's Common External Trade Policies Keiko Kubota AWPS 68 Bottom-Up Administrative Reform: Designing June 2004 Talib Esmail Indicators for a Local Governance Scorecard in Nick Manning Nigeria Jana Orac Galia Schechter AWPS 69 Tanzania's Tea Sector: Constraints and Challenges June 2004 John Baffes AWPS 70 Tanzania's Cashew Sector: Constraints and June 2004 Donald Mitchell Challenges in a Global Environment AWPS 71 An Analysis of Chile's Trade Regime in 1998 and July 2004 Francesca Castellani 2001: A Good Practice Trade Policy Benchmark A. Herrou-Arago Lawrence E. Hinkle AWPS 72 Regional Trade Integration inEast Africa: Trade and August 2004 Lucio Castro Revenue Impacts of the Planned East African Christiane Kraus Community Customs Union Manuel de la Rocha AWPS 73 Post-Conflict Peace Building in Africa: The Challenges August 2004 Chukwuma Obidegwu of Socio-Economic Recovery and Development 64 Africa Region Working Paper Series Series # Title Date Author AWPS 74 An Analysis of the Trade Regime in Bolivia in2001: A August 2004 Francesca Castellani Trade Policy Benchmark for low Income Countries Alberto Herrou-Aragon Lawrence E. Hinkle AWPS 75 Remittances to Comoros- Volumes, Trends, Impact October 2004 Vincent da Cruz and Implications Wolfgang Fendler Adam Schwartzman AWPS 76 Salient Features of Trade Performance in Eastern and October 2004 Fahrettin Yagci Southern Africa Enrique Aldaz-Carroll AWPS 77 Implementing Performance-Based Aid in Africa November 2004 Alan Gelb Brian Ngo Xiao Ye AWPS 78 Poverty Reduction Strategy Papers: Do they matter for December 2004 Rene Bonnel children and Young people made vulnerable by Miriam Temin HIV/AIDS? Faith Tempest AWPS 79 Experience in Scaling up Support to Local Response December 2004 Jean Delion in Multi-Country Aids Programs (map) in Africa Pia Peeters Ann Klofkorn Bloome AWPS 80 What makes FDI work? A Panel Analysis of the February 2005 Kevin N. Lumbila Growth Effect of FDI in Africa AWPS 81 Earnings Differences between Men and Women in February 2005 Kene Ezemenari Rwanda Rui Wu AWPS 82 The Medium-Term Expenditure Framework: The April 2005 Chukwuma Obidegwu Challenge of Budget Integration in SSA countries AWPS 83 Rules of Origin and SADC: The Case for change in the June 2005 Paul Brenton Mid Term Review of the Trade Protocol Frank Flatters Paul Kalenga AWPS 84 Sexual Minorities, Violence and AIDS in Africa July 2005 Chukwuemeka Anyamele Ronald Lwabaayi Tuu-Van Nguyen, and Hans Binswanger AWPS 85 Poverty Reducing Potential of Smallholder Agriculture July 2005 Paul B. Siegel in Zambia: Opportunities and Constraints Jeffrey Alwang AWPS 86 Infrastructure, Productivity and Urban Dynamics July 2005 Zeljko Bogetic in Côte d'Ivoire An empirical analysis and policy Issa Sanogo implications AWPS 87 Poverty in Mozambique: Unraveling Changes and August 2005 Louise Fox Determinants Elena Bardasi, Katleen V. Broeck AWPS 88 Operational Challenges: Community Home Based August 2005 N. Mohammad Care (CHBC) forPLWHA in Multi-Country HIV/AIDS Juliet Gikonyo Programs (MAP) forSub-Saharan Africa AWPS 90 Kenya: Exports Prospects and Problems September Francis Ng 2005 Alexander Yeats AWPS 91 Uganda: How Good a Trade Policy Benchmark for September Lawrence E. Hinkle Sub-Saharan-Africa 2005 Albero H. Aragon Ranga Krishnamani Elke Kreuzwieser 65 Africa Region Working Paper Series Series # Title Date Author AWPS 92 Community Driven Development in South Africa, 1990- October 2005 David Everatt Lulu 2004 Gwagwa AWPS 93 The Rise of Ghana''s Pineapple Industry from November 2005 Morgane Danielou Successful take off to Sustainable Expansion Christophe Ravry AWPS 94 South Africa: Sources and Constraints of Long-Term December 2005 Johannes Fedderke Growth, 1970-2000 AWPS 95 South Africa''s Export Performance: Determinants of December 2005 Lawrence Edwards Export supply Phil Alves AWPS 96 Industry Concentration in South African December 2005 Gábor Szalontai Manufacturing: Trends and Consequences, 1972-96 Johannes Fedderke AWPS 97 The Urban Transition in Sub-Saharan Africa: December 2005 Christine Kessides Implications for Economic Growth and Poverty Reduction AWPS 98 Measuring Intergovernmental Fiscal Performance in May 2006 Navin Girishankar South Africa David DeGroot Issues in Municipal Grant Monitoring T.V. Pillay AWPS 99 Nutrition and Its determinants in Southern Ethiopia - July 2006 Jesper Kuhl Findings from the Child Growth Luc Christiaensen Promotion Baseline Survey AWPS 100 The Impact of Morbidity and Mortality on Municipal September Zara Sarzin Human Resources and Service Delivery 2006 AWPS 101 Rice Markets in Madagascar in Disarray: September Bart Minten Policy Options for Increased Efficiency and Price 2006 Paul Dorosh Stabilization Marie-Hélène Dabat, Olivier Jenn-Treyer, John Magnay and Ziva Razafintsalama AWPS 102 Riz et Pauvrete a Madagascar Septembre Bart Minten 2006 AWPS 103 ECOWAS- Fiscal Revenue Implications of the April 2007 Simplice G. Zouhon-Bi Prospective Economic Partnership Agreement with the Lynge Nielsen EU AWPS 104(a) Development of the Cities of Mali June 2007 Catherine Farvacque-V. Challenges and Priorities Alicia Casalis Mahine Diop Christian Eghoff AWPS 104(b) Developpement des villes Maliennes June 2007 Catherine Farvacque-V. Enjeux et Priorites Alicia Casalis Mahine Diop Christian Eghoff AWPS 105 Assessing Labor Market Conditions In Madagascar, June 2007 David Stifel 2001-2005 Faly H. Rakotomanana Elena Celada AWPS 106 An Evaluation of the Welfare Impact of Higher Energy June 2007 Noro Andriamihaja Prices in Madagascar Giovanni Vecchi AWPS 107 The Impact of The Real Exchange Rate on November 2007 Mireille Linjouom Manufacturing Exports in Benin 66 Africa Region Working Paper Series Series # Title Date Author AWPS 108 Building Sector concerns into Macroeconomic December 2007 Antonio Estache Financial Programming: Lessons from Senegal and Rafael Munoz Uganda AWPS 109 An Accelerating Sustainable, Efficient and Equitable December 2007 Hans P. Binswanger Land Reform: Case Study of the Qedusizi/Besters Roland Henderson Cluster Project Zweli Mbhele Kay Muir-Leresche AWPS 110 Development of the Cites of Ghana January 2008 Catherine Farvacque- ­ Challenges, Priorities and Tools Vitkovic Madhu Raghunath Christian Eghoff Charles Boakye AWPS 111 Growth, Inequality and Poverty in Madagascar, 2001- April 2008 Nicolas Amendola 2005 Giovanni Vecchi AWPS 112 Labor Markets, the Non-Farm Economy and April 2008 David Stifel Household Livelihood Strategies in Rural Madagascar AWPS 113 Profile of Zambia's Smallholders: Where and Who are June 2008 Paul B. Siegel the Potential Beneficiaries of Agricultural Commercialization? AWPS 114 Promoting Sustainable Pro-Poor Growth in Rwandan June 2008 Michael Morris Agriculture: What are the Policy Options? Liz Drake Kene Ezemenary Xinshen Diao AWPS 115 The Rwanda Industrial and Mining Survey (RIMS), June 2008 Tilahun Temesgen 2005 Survey Report and Major Findings Kene Ezemenari Louis Munyakazi Emmanuel Gatera AWPS 116 Taking Stock of Community Initiatives in the Fight June 2008 Jean Delion against HIV/AIDS in Africa: Experience, Issues, and Elizabeth Ninan Challenges AWPS 117 Travaux publics à Haute Intensité de Main d' Oeuvre August 2008 Nirina H. Andrianjaka (HIMO) pour la Protection Sociale à Madagascar : Annamaria Milazzo Problèmes et Options de Politique AWPS 118 Madagascar : De Jure labor Regulations and Actual August 2008 Gaelle Pierre Investment Climate Constraints AWPS 119 Tax Compliance Costs for Businesses in South Africa, August 2008 Jacqueline Coolidge Provincial Analysis Domagoj Ilic Gregory Kisunko AWPS 120 Umbrella Restructuring of a Multicountry Program October 2008 Nadeem Mohammad (Horizontal APL) Restructuring the Multicountry Norbert Mugwagwa HIV>AIDS Program (MAP) in Africa AWPS 121 Comaparative Analysis of Organization and October 2008 Gérald Estur Performance of African Cotton Sectors 67 WB21847 C:\Documents and Settings\WB21847\My Documents\Working Paper Series\WPS FY2009\Labaste\Cotton.quality and marketing in Africa.DEC 9.doc 12/09/2008 6:08:00 PM 68