Report No. 18857-IN India Cotton and Textile Industries Reforming to Compete (in Two Volumes) Volume II: Annexes January 14, 1999 Rural Development Sector Unit South Asia Region Document of the World Bank Abbreviations andAcronyms AICCIP All India Cotton Coordinated improvement Project AMPA Agricultural Product Markets Act ATC Agreement on Textiles and Clothing BIFR Board for Industrial and Financial Reconstruction CACP Commission for Agricultural Costs and Prices CCI Cotton Corporation of India CIB Central insecticides Board CICR Central Institute for Cotton Research CIRCOT Central Research on Cotton Technology CV Coefficient of Variation EC Act Essential Commodities Act ELS Extra-Long Staple EOUs Export Oriented units EPCG Export Promotion Capital Good Scheme FAO Fair Average Quality FC(R)A Forward Contracts (Regulation) Act GATT General Agreement on Tariffs and Trade GOI Government of India HYO Hank yarn Obligation ICDP Intensive Cotton Development Program ICMF Indian Cotton Mills Federation IPM Integrated pest Management ITMF International Textile Manufacturers Federation IPR Intellectual Property Rights MSCCGMF Maharashtra State Cotton Cooperative Growers' Marketing Federation MCPS Maharashtra Cotton Procurement Scheme MEP Minimum Export Price MFA Multi-Fiber Agreement MMF Man Made Fibers MOA Ministry of Agriculture MODVAT Modified Value Added Tax MSP Minimum Support Price NPC Nominal Protection Coefficient NTC National Textile Corporation NTSD Non-transferable Specific Delivery Cotton Contracts OGL Open General License OTC Office of the Textile Commissioner QRs Quantitative Restrictions RBI Reserve Bank of India SITRA South Indian Textile Research Association VABAL Value Added Based Advanced License Weights and Measures qtl - quintal (100 kg) kg - kilogram bale - 170kgs ha - hectare mt - metric ton Vice President: Mieko Nishimizu Director: Edwin Lim Sector Unit Leaders: Ridwan Ali/Michael Baxter Staff Members: Benoit BlarellDina Umali-Deininger INDIA COTTON AND TEXTILE INDUSTRIES: REFORMING TO COMPETE Volume II Table of Contents Annex 1: Textile Industry Policy A. Evolution of India's Cotton Textile Policy .................................................................. 1 B. Textile Industry: Historical Perspective 1950s to 1985 ................................................. 1 C. Towards Liberalization: 1985-97 ..................................................................5 D. External Trade Reforms ..................................................................9 B. Appendix Tables ................................................................. 14 Annex 2: Cotton Sector Policy A. Cotton Production Policies ..................................................................1 B. Input Policies ................................................................. 3 C. Cotton Trade and Processing Regulations ..................................................................5 D. Risk Management ..................................................................8 E. External Trade Policies ..................................................................8 F. Appendix Tables .............. 10 Annex 3: Structure and Performance of the Textile Industry A. Introduction ..................................................................1 B. Structure of the Textile Industry ..................................................................1 C. Export Performance ..................................................................2 D. Domestic Demand for Textile Products ..................................................................4 E. Performance of the Spinning Industry ..................................................................6 F. Man-Made Fiber Products ................................................................. 14 G. Performance of the Weaving Sector .................................... ............................. 15 H. Handloom Sector ................................................................. 19 I. Structure and Performance of the Apparel Industry .......................................... ............. 19 J. Appendix Tables & Figures ................................................................. 25 Annex 4: Cotton Production Performance and Prospects A. Introduction ..................................................................1 B. Cotton Production Performance ..................................................................1 C. Technical Constraints to Future Productivity Growth .................................................... 6 D. Changing Composition of Cotton Output ................................................................. 12 E. Appendix Tables .15 Annex 5: Structure and Performance of Cotton Markets A. Introduction ..................................................................1 B. Marketing of Seed Cotton and Cotton Lint ................................................................. 1I C. Cotton Ginning and Pressing Industry ..................................................................7 D. Cotton Grading and Classification ................................................................. 10 E. International Marketing of Cotton ................................................................. 13 F. Cotton Market Performance ................................................................. 19 G. Risk Management in the Cotton Sector ................................................................. 22 H. Appendix Tables & Figures ................................................................. 28 Annex 6: MFA Phaseout and the Cotton and Textile Policies A. Introduction ..................................................................1 B. Overview of the Indian Cotton Textile Industry ............................................................ I C. MFA and Its Implications for India ..................................................................2 D. Methodology ..................................................................4 E. Policy Reform Simulations ..................................................................6 F. Economic Impact of Policy Reforms ..................................................................9 G. Summary and Conclusions ................................................................. 15 H. Appendix: Structural Specifications of the GTAP Model ............................................. 17 Annex 7 Policy Options for Liberalizing the Cotton Textile Industry A. Introduction .................................................................1I B. Methodology .................................................................1I C. Income Distribution Impacts of Cotton Textile Liberalization ....................................... 5 D. Summary and Conclusions .7 E. Appendix: Simulation Model for the Cotton Sector Liberalization in India ................ 10 List of Tables Annex 1 Table A 1.1 Policy Environment in the Textile Industry, 1997 .6 Table Al.2 Excise Duties for Cotton Textile Products, 1997 .8 Table Al.3 Government Programs to Assist Handlooms .9 Table A 1.4 Import Tariffs on Cotton and Man-Made Products .13 Appendix Table Al.1 Textile Exports, Rs million ...................................................................... 14 Appendix Table A1.2 Inia's Commitments on Textile Imports under the Agreement with the USA &EU .................................. . ................ 15 Appendix Table A 1.3 Effective Rates of Excise Duties on Man-Made Staple Fibres Rs per Kg .................................................... 16 Annex 2 Table A2. 1 Average Annual Rate of Yield Growth of ICDP and Non-ICDP Districts in Haryana and Maharashtra......................................................... 3 Table A2.2 Annual Growth Rates in Ginning and Pressing Charges in Real Terms in Selected States .......................................................................7 Table A2.3 Cotton Production and Export Quotas ........................................................ 9 Appendix Table A2.1 Minimum Support Price of Seed Cotton of Fair Average Quality, 1990-91 to 1994-95, Rs per quintal ............................................................. 10 Appendix Table A2.2 GOI Expenditures under the Intensive Cotton Development Program, Rs million ...................................................................... 10 Appendix Table A2.3 Cotton Yields in ICDP and Other Districts in Haryana and Maharashtra ...................................................................... 11 Appendix Table A2.4 Ginning Fees in Selected States, 1990 rupees ............................................. 11 Appendix Table A2.5 Pressing Fees in Selected States, 1990 rupees ................. ........................... 12 Appendix Table A2.6 Selective Credit Control against Cotton Lint and Seed Cotton, 1975 to 1994 ...................................................................... 13 Annex 3 Table A3.1 Exports of Cotton Textile Products, $ million (1994/95 dollars) .3 Table A3.2 Direction of Export Trade in Cotton Textiles: 1987-94. 4 Table A3.3 Per Capita Household Consumption of Textiles at Different Income Levels, 1992 .4 Table A3.4 Spinning Capacity and Production Share by Ownership. 6 Table A3.5 Cotton Yarn Exports by Count .....................................................7 Table A3.6 Evolution of the Index of Modernization, 1983-95. 7 Table A3.7 Output to Installed Spindleage .8 Table A3.8 Output per Spindle by Category of Spinning Mills, 1994. 8 Table A3.9 Delay Analysis of Spinning Mills .9 Table A3.10 Productivity Range by Class of Spinning Mills Adjusted to 40s Count, 1994 .9 Table A3.11 Mill Age and Productivity .10 Table A3.12 Cotton Yarn Production and Hank Yarn Deliveries .10 Table A3.13 Hank Yarn Obligation Transfer Prices .10 Table A3.14 Index of Manufacturing Cost Component of Spinning Mills Using Thailand as Reference .12 Table A3.15 Capacity and Production by Firm Type and Yarn Count .13 Table A3.16 Effective Protection of Cotton Spinning Mills by Yarn Count and Type of Firms, 1995 .13 Table A3.17 Competitiveness of Different Categories of Mills by Yarn Counts at Domestic Cotton and Yarn Prices, Rs/kg .13 Table A3.18 Competitiveness of Different Categories of Mills by Yarn Counts at International Cotton Prices, Rs./kg .13 Table A3.19 Impact of Excise Duties on Market Prices for Different Yarn Types .15 Table A3.20 Comparison of Cost of Production of Gray Cloth, $US/Yard of 63 Inch-Wide Cloth, 1993 .17 Table A3.21 Public Textile Enterprises: Net Profit and Losses, Million Rupees, Net Worth and Number of Employees .18 Table A3.22 Apparel Manufactures by Type and Location .19 Table A3.23 Index of Apparel Productivity Using Hong Kong as Reference ................... 20 Table A3.24 Cost per Standard Minute Produced of a Standard Clothing Item, Selected Countries, 1995 .20 Table A3.25 Machinery and Investment Levels by Apparel Export Firms in Selected Countries (No. of Units) .21 Table A3.26 Number of Machinery Installed by Type in Apparel Export Firms, Number of Units .................................................... 21 Appendix Table A3.1 Per Capita Consumption and Percentage Share in Private Final Consumption, Expenditure of Textiles in Selected Countries And Regions, 1993 ...................................................................... 25 Appendix Table A3.2 Per Capita Cloth Consumption and Real GDP And Mid Year population in India .............................................................. 25 Appendix Table A3.3 Percentage Share of Purchases of Selected Man-Made Fiber Products by Income Groups .............................................................. 26 Appendix Table A3.4 Cotton Textile Exports, $Million, Constant 1994/95 Dollars ..................... 26 Appendix Table A3.5 Percentage Share of Apparel Exports by Fiber Type .................................. 26 Appendix Table A3.6 Yarn Production by Count, Million Kgs ..................................................... 27 Appendix Table A3.7 ITMF Computation of manufacturing Costs for 30s Combed Ring Spun Cotton Yarn at March 1995 Prices, US Cents/Kg ............................. 27 Appendix Table A3.8 Production of Man-Made Fibers in India, OOOmt .27 Appendix Table A3.9 Installed Capacity for Man-Made Staple Fiber Production, OOOmt 28 Appendix Table A3.10 Production of Man-Made Filament Yarn, OOOmt .28 Appendix Table A3.11 Installed Capacity for Man-Made Filament yarn Production, OOOmt .29 Appendix Table A3.12 Public Textile Enterprises: Net Profit and Losses, Constant 1993/94 Million Dollars and Number of Employees .29 Appendix Table A3.13 Effective Protection of Yarn Exporting Firms, 1994/95 .30 Appendix Table A3.14a Production Cost of 20s Count Yarn, Rs/Kg . ............................................... 30 Appendix Table A3.14bProduction Cost of 30s Count Yarn, Rs/Kg . ............................................... 30 Appendix Table A3.14c Production Cost of 40s Count yarn, Rs/Kg . ................................................ 31 Appendix Table A3.15 Yarn Exports by Count, OOOmt . .................................................................. 31 Appendix Table A3.16 Total Apparel Export Firms and their Percentage Distribution by Region ...................................................................... 31 Appendix Table A3.17 Prices of Yarn and Filaments, Rs/Kg .......................................................... 32 Appendix Table A3.18 Productivity Levels of Apparel Firms ......................................................... 32 Annex 4 Table A4.1 Source of Cotton Output Growth by Region and All India. 3 Table A4.2 Outbreaks of Major Pests by Year and State. 6 Table A4.3 Crop-wise Consumption of Pesticides in India .6 Table A4.4 Total Requirement and Supply of Quality Seeds, 1994/95, 100 Percent Replacement Rate .11 Table A4.5 Cotton Production, Area and Yield Instability .12 Table A4.6 Production of Cotton by Staple, Region and State, 1991-92 .13 Appendix Table A4.1 Trend Growth Rates of Production of Major Crop Groups, Pre- and Post-Green Revolution, percent per year .15 Appendix Table A4.2 Cotton Area, Production and Yield by Region and All India ..................... 16 Appendix Table A4.3 Selected Cotton Variety Characteristics .17 Appendix Table A4.4 Irrigated Area under Cotton Cultivation by State, 000 ha and Percentage Share of Total Cultivated Area (values in parenthesis) ............ 19 Appendix Table A4.5 Integrated Post Management Approaches for Cotton in Selected Countries ..................................................... 20 Annex 5 Table A5.1 Markets Cess on Kapas in Selected States .................................................. 3 Table A5.2 Percent Distribution By Grade Cotton Procured by GSCCGMF ................ 6 Table A5.3 Cotton Ginning and Pressing Factories in India, 1995 ................................ 8 Table A5.4 Degree of Cotton Contamination in Selected Countries ............................. 8 Table A5.5 Typical Set of Specifications of Different Grades of a Variety .................. 11 Table A5.6 Agency-wise Pattern of Cotton Exports, 000 bales, 1988/89 to 1993/94 .15 Table A5.7 Farmer's Share of the Final Price of Cotton Lint and Cotton Seed under Different Marketing Systems .20 Table A5.8 Percentage Seasonal Price Increase of Selected Cotton Varieties ................ 21 Table A5.9 Cotton Associations Permitted to Trade Cotton NTSD Contracts, 1995 .23 Table A5.10 The Correlation of Monthly Average Spot Prices for Main Cotton Varieties, 1984-1995 .25 Appendix Table A5.1 Cotton Procurement by Different Agencies, 10,000 bale of 100kg .28 Appendix Table A5.2 Minimum Support Price of Seed Cotton of Fair Average Quality, 1990- 91 to 1994-95, Rs per Quintal .28 Appendix Table A5.3 Profits and Losses of CCI, 1978-79 to 1993-94 (Rs. Million) .................... 30 Appendix Table A5.4 CCI Statewise Purchases and Buyer-wise Sales (Bales of 170kg) ............. 31 Appendix Table A5.Sa Ginning Fees in Selected States, Rs/bale of 170kg ..................................... 31 Appendix Table A5.5b Pressing Fees om Selected States. Rs/bale of 170kb .................................. 31 Appendix Table A5.6 Farmers' Share in Gross Income Earned by Cotton Corporation of India and Maharashtra Monopoly Procurement Scheme, Rs million .................. 32 Annex 6 Table A6.1 Commodity Aggregation in the GTAP Model .5 Table A6.2 Experimental Design and Specification of the Shocks. 6 Table A6.3 Assumptions Used in the Projections: Cumulative Percentage Growth Rates over the Period 1992-2005 .8 Table A6.4 Change in Composition of GDP, Exports, Imports and the Trade Balance for India: 1992-2005 .9 Table A6.5 Impact of Domestic Reforms on Indian Output, Exports, and Imports in the Year 2005 .11 Table A6.6 Impact of Removing Cotton Export Tax on Indian Output, Exports, and Imports in the Year 2005 .12 Table A6.7 Impact of Removing Cotton Textile Export Tax on Indian Output, Exports, and Imports in the Year 2005 .13 Table A6.8 Impact of Removing the Hank Yarn Obligation on Indian Output, Exports, and Import in the Year 2005 .13 Table A6.9 Impact of Increased Productivity of Apparel on Indian Output, Exports, and Imports in the Year 2005 .14 Table A6. 10 Impact of Domestic Reforms on India's Welfare in Year 2005 (1992 $U S m ill) .................................................... 15 Table A6. 11 Impact of Indo-US and-EU Agreement on Indian Output Exports, and Imports in the Year 2005 ...................................................................... 16 Appendix Table A6.1 Shares Used in Splitting the Cotton/Textile/Wearing Apparel Factors ...... 19 Annex 7 Table A7. 1 Initial Values for Simulation .2 Table A7.2 Price Elasticities .3 Table A7.3 Summary of Policy Reform Scenarios. 5 Table A7.4 Changes in Net Income & Export Values. 6 Table A7.5 Changes in Net Income & Export Values. 7 Appendix Table A7.1 Initial Value Table for Simulations .13 Appendix Table A7.2 Elasticity Table for Base Simulation Case .................................................. 13 Appendix Table A7.3 Different Sets Elasticities Used for Sensitivity Analysis ............................ 14 List of Figures Annex 1 Figure A1.1 Expanding Share of Powerlooms in Cloth Production in India. 2 Figure Al.2 Increasing Share in Spinning Capacity of the Independent Mill Sector .4 Figure A1.3 Cotton Yarn Exports Exceed Export Ceilings .10 Annex 2 Figure A2.1 GOI Expenditures on ICDP .2 Annex 3 Figure A3.1 Structure of the Cotton Textile Industry. 2 Figure A3.2 Indian Cotton Apparel, Fabric and Yarn Exports by Volume. 3 Figure A3.3 Per Capita Availability of Cloth, Sq.M per Capita. 4 Figure A3.4 Consumption Shares of Different Textile Fibers in Selected Countries, 1992 .5 Figure A3.5 Spindle Capacity in India .6 Figure A3.6 Production of Yarn by Count in India. 7 Figure A3.7 Nominal Protection Coefficients of Yarn for Selected Counts .12 Figure A3.8 Expanding Share of Powerlooms in Cotton Fabric Production in India .15 Figure A3 .9 Fabric Exports by Subsector .16 Figure A3.10 Rapid Growth in India's Apparel Exports .22 Appendix Figure A3. 1 Wholesale Price indexes for Textiles, Manufactured Products and Primary Articles, 1981/82=1 00.33 Appendix Figure A3.2 Value of Consumption of Cloth per Person for a Period of 30 Days for Each Fractile Group .33 Appendix Figure A3.3.. Average Price Paid for Cloth by Different Income Groups, Rs per Meter .................................................... 34 Appendix Figure A3.4.. Production of Blended yams and 100% non Cotton and Man-Made Filament Yams .................................................... 34 Annex 4 Figure A4.1 Cotton Lint Production in India, 1974-75 to 1996-97, million mt .............. 1 Figure A4.2 Cotton Lint Production by Region .....................................................2 Figure A4.3 Cotton Lint Yields by Region and All India, kg/ha ..................................... 3 Figure A4.4 Cultivated Area under Hybrid Cotton in Selected States, 1979-93 and Percent Share of Hybrid Area in Total Area in 1992-93. 4 Figure A4.5 Relative Share of Irrigated and Rainfed Cotton Areas in the Different Regions in India, 1981/82 to 1991/92 .4 Figure A4.6 Average Cotton Lint Yields in Selected Countries and Indian Regions, 1992-94 .5 Figure A4.7 Cotton Lint Production and Area of Selected Major Cotton Producers in the World, TE94 ..................................................... 5 Figure A4.8 Staple Length and Final Output .................................................... 12 Figure A4.9 Cotton Output by Staple Length .................................................... 13 Figure A4. 10 How does Indian Staple Standards Compare to International Norms ......... 14 Annex 5 Figure A5.1 Private Traders Dominate Cotton Seed Procurement, TE 1993-94 ............ 1 Figure A5.2 Quality Premium between J-34 Superior Grade Purchased by an EOU and the Average Market Price .................................................... 12 Figure A5.3 Indian Cotton Exports, Volume and Value, 1984/85 to 1993/94 ................ 15 Figure A5.4 Cotton Exports are Generally Below Export Quotas ................... ............... 15 Figure A5.5 Nominal Protection Coefficients for MCU-5, S-4 and J-34 Cotton Lint (Exportable Hypothesis) ...................... .............................. 16 Figure A5.6 Percentage Share of Cotton Seed Oil in Edible Oil Market, 1990-94 ........ 16 Figure A5.7 Nominal Protection Coefficients for Cotton Seed (Importable Hypothesis) .17 Figure A5.8 Nominal Protection Coefficient for Seed Cotton (Exportable Hypothesis) .................................................... 17 Figure A5.9 Cotton Import, Volume and Value .................................................... 18 Figure A5.10 Gross Processing Margins for J-34 and F-414 ............................................ 19 Figure A5.11 Seasonal Price Indexes of Selected Cotton Varieties .................................. 21 Figure A5.12 Inter-annual Price Volatility for Selected Cotton Varieties, 1983-94 ......... 22 Figure AS.13 Seasonal Variation for Selected Cotton Varieties (% of the average crop- year price) .22 Appendix Figure A5.1 Ration of the Wholesale Price Index of Seed Cotton to the Wholesale Price Index of Oilseeds, Cotton, Pulses, and Sugar, Khandasari and Gur ........... 33 Appendix Figure A5.2 All India Oil cake and Cotton Oil Cake Exports and Domestic Cotton Oil Cake Production .34 Annex 7 Figure A7. 1 Simplified Commodity Flows in the Cotton-Textiles Industry 2 List of Boxes Annex 1 Box A1.1 Main Features of Bilateral US and EU Agreements ................................... 11 Box A1.2 The Agreement on Textiles and Clothing and the Phase-out of the MFA .................................................... 12 Annex 4 Box A4. 1 Cotton Production Data in India: Which Estimate is Best? ....................... 2 Box A4.2 Cotton Research and Extension in India ..................................................... 8 Box A4.3 Cotton: Marketing and Distribution of Improved Seeds ............................ 10 Annex 5 Box A5. 1 Agricultural Product Markets Act and Mandi Operations .......................... 2 Box A5.2 Expenditures Financed by Market Cess Levied on Market Yards in the State of Andhra Pradesh .....................................................3 Box A5.3 Cotton Grading in Other Cotton Producing Countries ................................ 14 Box A5.4 Mexico's Cotton Price Support Scheme .................................................... 27 Annex 1 Textile Industry Policy A. Evolution of India's Cotton Textile Policy AL.1 The textile industry is one of the largest and most important sectors in the economy in terms of output, foreign exchange earnings and employment in India. The textile industry includes all subsectors--spinning, weaving, knitting and garmenting and uses different raw materials--cotton, jute, wool, silk, man-made and synthetic fibers. In 1995/96, the textile industry accounted for about one-fifth of total industrial output, about 7 percent of GDP and more than a quarter (US$9 billion) of total commodity exports. Although the share of the man-made fiber subsector is growing, cotton still captures about 66 percent of the raw material mix in the industry--the target material mix of the Ministry of Textiles is a 55 percent cotton/45 percent man-made fibers. Cotton-based textile exports -- fabrics, yarn, garments and made-ups-- accounted for 20 percent of total export value ($6 billion). Made-ups consist of non-apparel textile based products such as carpets, bedspreads, blankets, towels, curtain materials, etc. In the last five years, cotton-based exports increased at a rate of 13 percent per year in real terms (Appendix Table A3.4). The textile industry is also one of the biggest employers in the economy, employing nearly 38 million people. A1.2 Past government emphasis on import substitution and the protection of labor interests primarily shaped the structure of textile industry. The general economic policy of self reliance dating to the 1950s extended to the textile industry. Exports were treated as a marginal outlet for surpluses. Government concern for adequate access by the poor to their "preferred" cotton clothing--synthetic clothing was the rich person's apparel --also shaped tariff and fiscal policies which discriminated against the increased domestic use of synthetic fibers. The large labor base in the textile industry and the GOI's strong concern for employment produced several pieces of legislation which imposed restrictions on the operations of the large composite mills to protect the operations of the labor-intensive handloom industry. In the face of changing technologies, markets and competitive advantages, these regulations are imposing serious employment and operational rigidities in the composite mills, while leaving the backdoor open for the growth of the decentralized small-scale enterprise sector (powerlooms and knitting, yarn and fabric processing units). Beginning in the mid- 1980s, expanding international market opportunities brought a relaxation of some of the domestic industrial regulations, fiscal policies, and trade policies. A1.3 This Annex briefly reviews the policy instruments governing the cotton textile industry in India and their evolution through time. The next section traces how key policies from the 1950s to 1985 shaped the present structure of the textile industry. The third section summarizes the recent policy reforms (1985-1997) affecting domestic textile enterprise operations and trade. The last section reviews the progress achieved in external trade reform. B. Textile Industry: Historical Perspective 1950s to 1985 Decline of the Composite Mills, Rise of Powerlooms Al.4 Policies designed to promote the unorganized sector and protect the welfare of poor consumers led to the decline of the composite mills. Unlike the trend in most textile producing countries, the output of vertically integrated composite mills in India steadily declined over time, Annex I Page 2 of 16 both in absolute terms and in shares of the market (Figure Al. 1.). This decline is due to a package of policies which affected all aspects of mill operations from pricing, employment to technology modernization. These are described below. A1.5 Restrictions on loom capacity Figure Al.1: Expanding Share of Powerlooms in Cloth expansion and automatic loom installation Production in India were imposed on composite mills in the 35 l950s. These were designed to protect 30 employment within the mill sector as well as a 25 promote greater employment in the 1520 handloom sector. Restrictions on capacity 10 - expansion were only relaxed under the 1985 5 new Textile Policy (see page 5). The 0 restriction on the installation of automatic looms was finally lifted in 1977. Its major 00 ? impact is that up to the present, only 30 n Mill a3 Pow erloom percent of the looms in India is automatic-- 0 Handloom * Knitting the lowest in the world, in contrast to the Source: Office of the Textile Commissioner, Compendium of Textile world average of 80 percent. Statistics, Various Issues; ICMF, Handbook of Statistics of the Cotton Textile lnd2ustiy, various issues. A1.6 A multitude of labor regulations under the Factories Act of 1948, which governed labor relations in the organized sector, discriminated against the mill sector. The Factories Act is applicable to all establishments employing 10 or more workers in processes which use power or 20 or more workers in processes which do not use power. Subsumed under this Act are several employment regulations, it includes the Payment of Wages Act, 1936; Minimum Wages Act, 1948; Workmen's Compensation Act, 1946; Employees' State Insurance Act, 1948; Employees' Provident Pension Fund and Miscellaneous Provisions Act, 1952; Employees' Pension Scheme, 1971; Maternity Benefit Act, 1961; Payment of Gratuity Act, 1972; Trade Union Act, 1926; Industrial Employment (Standing Orders) Act, 1946; Employment Exchanges (Compulsory Notification of Vacancies) Act, 1959; Apprentices Act, 1961; Contract Labor (Regulation and Abolition) Act, 1970; Equal Remuneration Act, 1976; and Inter-State Migrant Workmen (Regulation of Employment Conditions of Service) Act, 1979. Among others, these regulations vested strong powers on labor unions and fostered employment rigidities that discouraged investments in more efficient, but more capital intensive technologies. They increased labor costs significantly in the mill sector-- the difference in wages between the organized and unorganized sectors range from 50 to 60 percent.' This inability to modernize hampered the composite mill sector's ability to effectively compete with the fast growing, low cost powerlooms (see below). A1.7 The GOI introduced price controls and later dual pricing, as a means of ensuring low priced clothing to poor consumers. Composite mills were directed to produce coarse cloth at statutory prices. This policy, implemented in varying intensities between 1965 and 1978, imposed a serious financial burden on weaker mills and was a key factor in what is now known as the "sick mill problem."2 The increasing number of sick mills coupled with strong pressure from unionized labor were major factors contributing to the government take-over of many bankrupt I P. Anubhai and V.L. Mote, 1994,"India's Textile Industry: A Case Study of Subsectoral Restructuring," in S. Meyanathan (ed.), Managing Restructuring in the Textile and Garmnent Subsector, Examples form Asia; Washington, D.C.: EDI. 2 D. Mazumdar, 1981, "The Development of the Indian Textile Industry and Its Three Sectors," Studies in Employment and Rural Development Series, No. 76, World Bank, Washington, D.C. Annex I Page 3 of 16 mills, and used them to produce the bulk of the cheap mill cloth for sale. As a result, public sector units increased from 24 in 1967 to 176 in 1985. With the elimination of the dual pricing policy, the public mills took up the exclusive responsibility for production of cheap cloths. A1.8 Differential taxation across sectors increased costs for the mill sector. The mill sector was subject to excise duties, while the handloom and powerloom sectors (until 1974) were exempted from paying any duties. For example, for "Medium A" cloth produced by mills in 1974, the excise duty was Rs 0.179 per sqm. In contrast, powerlooms paid Rs 0.0452 per sqm for units with less than 5 looms and Rs 0.0565 per sqm for units employing 25-49 looms.3 In addition, the mill sector was subject to "progressive" taxation, whereby higher taxes were levied on finer varieties of mill cloth. A 1.9 The GOI imposed artificial compartmentalization of the textile industry. There were five categories of production, based on the raw material used: (i) made wholly or in part of cotton, including cotton yarn, hosiery or rope; (ii) made wholly or in part of jute, including jute, twine and rope; (iii) made wholly or in part of wool, including wool tops, woolen yarn, hosiery, carpets, and druggets; (iv) made wholly or in part of silk, including silk yarn and hosiery, and (v) made wholly or in part of synthetic, artificial (man-made) fibers. Although consumer preference began shifting towards synthetics or mixed blends, the compartmentalization of the industry prevented enterprises from adequately responding to the new demands in the market. A1.10 In contrast, unfettered by the policies and regulations burdening the composite mill sector, powerlooms posted exceptional growth. Powerlooms, by design or default, were exempted from many of the regulations and obligations which handicapped the mill sector. Being small and scattered, they were not subject to the Factories Act and thus were able to keep wages and overheads low. Until 1985, they benefited from lower excise taxes, lower power tariffs and, reportedly, lower power tariff collection rates. The net effect of these exemptions was to lower powerloom costs by about 8-10 percent over the mills. The textile industry strike in Bombay in 1982 also opened an important window of opportunity for the powerloom sector. Lasting about 15 months, it created a sudden vacuum in the supply of fabrics which the powerlooms quickly filled. The cost advantages afforded by the various exemptions and the greater flexibility of operations enabled powerlooms to out-compete the mills. As a result, by 1985, powerlooms had cornered nearly half of the fabric market (Figure Al.1) Sustained Protection of the Handloom Industry Al.11 The physical restriction of mill sector output and differential taxation were two instruments used to protect the handloom industry. Highly labor intensive by nature and characterized by strong traditional cultural underpinnings, the interests of the handloom sector are a major driving force shaping textile policies. There are various estimates of the number of people employed in the handloom sector. The most recent estimate from the Textile Commissioner's office (1997) is 12.4 million people employed in 3.8 million handloom units. Due to its labor intensity and the government's continued concern for both employment generation and the protection of labor, many of the restrictions imposed on the composite mill sector, and later expanded to the powerloom sector, are designed to protect the handloom industry. Restrictions on capacity expansion and the installation of automatic looms by composite mills were intended for the protection of the handloom sector. A1.12 The rationale for the Hank Yarn Obligation of 1974 was to ensure an adequate supply of reasonably priced yarn to the handloom industry. The Hank Yarn Obligation requires spinning 3 Ibid. Annex I Page 4 of 16 mills to process 50 percent of their deliveries in hank form, with at most 85 percent in counts of 40s and below. Although the hank yarn obligation is exclusively intended for the handloom industry, leakages of between 15 to 25 percent to powerlooms have been reported. On a cost basis, it is less costly for powerlooms to use dyed hank yarn than dyed cone yarn. In addition, cotton and yarn export restrictions were imposed to ensure adequate domestic supply and keep the domestic prices of cotton and yarn, including hank yarn, at reasonable levels. The handloom sector also benefited from excise duty concessions. Rapid Growth in the Less Constrained Spinning Industry A1.13 A less constrained economic environment promoted the growth and modernization of the spinning industry. Inherently capital intensive4 and producing an intermediate product, the spinning industry was less targeted and less affected by the strict labor laws, strong labor unions, and other operational restrictions (e.g. price controls). Unlike weaving, where the addition of loomage was restricted at some point, no such restriction was placed on the spinning industry. It was allowed to grow and modernize, generally tailoring to the needs of the growing powerloom market. Government priority to value addition, however, translated into restrictions on yarn exports. To ensure an adequate supply of yarn to the weaving sector, especially the powerlooms and handlooms, and to check price increases in the domestic market, cotton yarn exports were and remain controlled (see page 9). Cotton yarn is defined as yarn containing more than 90 percent cotton by weight. Blended yarns are exempted from export controls, since they contain less than 90 percent cotton. A1.14 Despite the hankyarn obligation and restrictions onyarn exports, rapid demand growth from the weaving sector attracted the entry of new firms and capacity into the spinning industry. Between 1951 and 1985/86, spindlage Figure Al.2: Increasing Share in Spinning Capacity of the more than doubled from 11 million Independent Mill Sector, 1951 to 1995/96 (spindles) spindles to 26 million spindles. This 25 growth was largely driven by 20 r - - - - independent mills, whose capacity , 15 expanded by nearly a magnitude of ten j10 10 . -- from 1.84 million spindles in 1951 to r 13.4 million in 1985/86. This dramatic change in the structure of the spinning E industry continued till the 1990s (Figure E 4 ' A 2 Al.2). co IR m Independent -0- Conrposite Discrimination Against Man-Made S,ource: ICMF, Handbook of Statistics on Cotton Textile Fibers Industy, various issues. A1.15 The policy bias against the man-made fibers industry arose from two major GOI concerns. On the demand side, the prevailing view in government was that man-made fibers were preferred by rich consumers and thus its taxation would not affect poorer consumers who preferred cotton products. On the supply side, there was the apprehension during the initial years that man-made products could disrupt the cotton economy in many states. For these reasons, man-made products were subject to higher excise duties relative the similar cotton-based products (fibers, yarns, fabrics). This raised their prices relative to cotton products, depressing demand. Another factor increasing the local costs of manufacturing synthetic fiber and filament 4 The ratio of capital to labor cost in the spinning industry is estimated at 10:1 (ICCI and R. Jaikumar, 1995, "Beyond the Multifiber Agreement, Strategies for the Indian Apparel Industry"). Annex I Page 5 of 16 yarn was the uneconomic plant sizes, in part due to the licensing regime which limited entry and compartmentalized the industry by type of fiber. C. Towards Liberalization: 1985-97 A1.16 The last decade is characterized by the progressive relaxation of policies imposed on the textile industry and a greater emphasis on improving efficiency and competition. An important turning point in the development of the textile industry is the Textile Policy of 1985 which began to relax some of the restrictive policies handicapping the textile industry. A second milestone is the far-reaching economic liberalization program of the GOI beginning in 1991, which placed major emphasis on export-led growth. In line with the general policy of liberalization, several measures were undertaken to eliminate/reduce controls and bring about greater transparency in the textile sector. The textile industry was de-licensed as per the Statement of Industrial Policy, 1991 and the Textile (Development and Regulation) Order, 1992. Further reforms were also pursued on the fiscal front. The tax differentials between and within the weaving sectors and between cotton and man-made fibers were reduced. Table Al.1 summarizes the policy framework governing the textile industry as of 1997. A1.17 India's signing of the General Agreement on Tariffs and Trade (GATT) in 1995 prompted greater liberalization on the external front. The Agreement on Textiles and Clothing (ATC) included as part of the final act of the GATT provides for a dismantling of the discriminatory quota regime practiced under the Multi-Fiber Agreement (MFA) within ten years, creating new windows of opportunities for increase textile exports. It also prompted the opening of Indian markets to textile product imports. (See Annex 6 for detailed analysis of the impact of MFA phase-out on the Indian textile industry). Trade policy continues to be revised to make it more consistent with GATT commitments. These reforms were also driven by changing consumer preferences in the domestic and world markets. Recently completed bilateral agreements between India and the US and India and the EU also spell out the program for increased entry of Indian textile products into these countries and of increased entry of imports from these countries into India. The following section describes the major policy developments during this period. Textile Policy of 1985 A1.18 The GOI 1985 Textile Policy reversed some of the major restrictions on the textile industry. Based on the recommendations of the Expert Committee on Textiles, the main elements of the Textile Policy included: (i) dismantling the sector approach to the industry, retaining a special role for non-power technology; (ii) adopting a multi-fiber orientation and fiber flexibility; (iii) providing adequate raw materials at reasonable and stable prices; (iv) reducing the level of duties on synthetic raw materials; (v) removing entry and exit barriers; (vi) emphasizing modernization and technology and machinery imports at international prices; and (vii) making Indian textiles more competitive in the world market.5 5 P. Anubhai and V.L. Mote, 1994, "India's Textile Industry: A Case Study of Subsectoral Restructuring," in S.D. Meyanathan (ed.), Managing Restructuring in the Textile and Garment Subsector, Examples From Asia, Washington, D.C.: EDI, World Bank. Annex I Page 6 of 16 TableA1.1: Poliy pnvironment in the Textile Industry, 1997 Domesbic Regulations S Textile (Development and Weaving Min. of Industry Removed licensing requirement. Regulation) Order, 1992 Spinning * The Factories Act, 1948/Labor Weaving Min. of bndustry, State No clear "exiutrpolicy Policies Spinning govt. * Small-scale Reservation, 1977 Garment Min. of Industry Ceiling recently raised from Rs 6 to 30 million KniAing son fPxed ivestment on garent m r * Tax Pol icy Spinning Min. of Finance Differential taxation between cone & d ank yat , Weaving handlooms and powerlooms/mills, coEC on and MMF MMF * Hank Yar Obligation, 1974 Spinning Min. of Textiles Except yae for expoi hosiery yan, >60 counts, blended & industrial yarm * Handlooms (Reservation of Handloom Min. of Textiles Covers 22 products Articles of Production) 1985 * Janata Cloth Subsidy Scheme Handloom Min. of Textiles Subsidy financed by addl. 15% duty on textile F products Trade PoTicies * Export Quotas Apparel Apparel Export Yarn counts > 40s, EOU and EPCG yarn, Fabrics Promotion Council, Min. processed yam, & bilateral ya i exports Yarn of Textiles exempted, Fabric and garment export quotas under ATC and bilateral agreements allocated by OTC based on pre-set criteria a Import restrictions Apparel Min. of Commerce EOUs and EPCGs exempted Fabric __ S Import Tariffs Apparel Min. of Commerce Reduction subject to US and EU bilateral Fabri agreements A1.19 Thea 985 Textile Policyr illustrated GOI's initial attempts to relaxtheregulatoryeburden on the composite mill sector. These included the elimination of the compartmentalization of the industry, the lifting of restrictions on composite mill loom capacity expansion, and the equalization of taxation among composite mills, powerlooms and independent processing units. At thssstnea time, the GOI supported, with limited success, the structural adjust1,ent of poor performing goveenment-owned mills and the exit of non-viable units. The Board for Industrial and Financial Reconstruction (BIFR) was constituted under the Sick Industrial Companies Special Provisions Act, 1985 to decide on the fate of sick enterprises--liquidation or rehabilitation. A sick enterprise is defined as a company that has been registered for 5 years and has negative net worth (accumulated losses exceeding equity plus reserves). A Textile Moderfization Fund Scheme was created by the Industrial Development Bank of India (IDBI) in 1986 to meet the modednization needs of ailing textile mills. This involved modenization assistance at concessional interest rates. Prior to its discontinuation in August 1991, Rs 8.8 billion had been disbursed to assist 307 mills. A Textile Worker's Rehabilitation Fund Scheme (TWVRFS) was established to provide a safety net for workers affected by the closure of mills. A1.20 ThzeTextile Policy of 1985 also marked th2ebeginning of the reduction of thehbias against man-made fibers. The 1985 Textile Policy included the flexibility to reduce fiscal levies on man- made fibers and yarns and intermediates used as inputs for their production. This was intended to facilitate the increased absorption of man-made and blended fabrics for which consumer's displayed increasing preference (See Annex 3). A1.21 In 1991, recognizing that the restructuring of public sector enterprises under structural adjustment would be accompanied by redundancy and unemployment, the GOI established a National Renewal Fund (NRF). The NRF consists of three funding instruments: (i) financing of voluntary retirement schemes (VRS), (ii) designing and implementing retraining projects to impart new skills to workers to reduce the period of involuntary unemployment, and (iii) for initiating unemployment insurance. According to VRS guidelines, money from the NRF is meant Annex I Page 7 of 16 to finance 45 days pay for every year of completed service evaluated at the latest salary (private sector schemes are often more generous). Under the NRF scheme for retraining and redeployment, a worker is provided a Rs 40 per day stipend for fifty days while attending vocational training programs (appliance repair, plumbing, auto repair, carpentry, welding, tailoring, driving, etc.). Easing Firm Entry A1.22 The Statement of Industrial Policy, 1991 and the Textile (Development and Regulation) Order, 1992 eased entry into the spinning and weaving industry. The Statement of Industrial Policy 1991 eliminated the need to obtain licenses for new capacity in the mill sector. It also removed the restrictions on large companies coming under the Monopolies and Restrictive Prevention Act to make new investments and provided for some automatic clearances for foreign investment proposals. Under the Textile (Development and Regulation) Order 1992, certification of powerlooms became automatic, except for units which did not come under the small scale industry category and which were proposed to be located within 25 km of a city with a population exceeding 1 million. Similarly, units putting up new spindlage required only a submission of an information memorandum with the Textile Commissioner, unless the unit was not small-scale and is located within 25 km from the periphery of standard urban limits. A1.23 The knitting and apparel industries remain reserved for small-scale enterprises. In the case of the apparel industry, GOI recently raised the ceiling on fixed assets in plant and machinery from Rs 6 million to Rs 30 million,6 subject to the acceptance of an export obligation of 50 percent of new or additional production annually to be achieved within a period of three years, and not less than 50 percent of this obligation is to be achieved by way of non-quota exports. Large scale enterprises (investments greater than Rs 30 million) are required to assume an export obligation of 50 percent of new or additional production to be achieved within three years.7 The stringent export obligations impose an additional burden on large enterprises and make entry more difficult. Discriminatory Tax Policy Against Man-Made Fibers Improving A1.24 The last 5 years was marked by increasing rationalization of the tax policy in the textile industry. The excise duty differentials between cotton fiber and man-made staple fibers and between cotton yarn and man-made yarn were reduced. In July 1996, tax reforms included: (i) the reduction of the excise burden on polyester fiber yarn by 10 percentage points to 40 percent, (ii) merging of excise rates of synthetic filament yarns at 20 percent, (iii) merging of excise duties on cotton, man-made staple, and man-made filament fabrics (with some exemptions) at 10 percent; (iv) the Textiles and Textile Articles Act which provided for an ad valorem 15 percent duty was repealed; and (v) merging of the existing additional excise duty in lieu of sales tax of fabrics ranging from 5 to 20 percent to 10 percent. The modified VAT (MODVAT) scheme, a rebate scheme for taxes paid on inputs introduced in 1986/87, has also been extended to cover the fabric stage. The credit of duty paid on yarn would be available on a notional basis and the credit of duty paid on other inputs along with capital goods would be available on the basis of duty paying documents. In February 1997, additional tax reductions were introduced. These included the further reduction of excise duties on polyester filament yam from 40 to 30 percent and the reduction of the mean excise duty on blended synthetic yarn to 18 percent. 6 Small scale investment ceiling excludes land and building costs. 7 The export requirement was reduced from 75% to 50% in 1997. ICICI and R. Jaikumar, 1995. Annex I Page 8 of 16 A1.25 Although these reforms will ease the price disparities between cotton and man-made products, existing duty levels continue to hinder the industry's flexibility to respond to-changing domestic and international trends in consumer preferences (Annex 3). Implementation of More Scale-Neutral Taxation A1.26 The elimination of lower tax privileges by small-scale enterprises using cotton yarn and texturized man-made yarn as of February 1997 works toward creating a more level playingfield between mills and powerlooms. In the past, this policy served to discriminate against the mill sector, making mill-made products less competitive. Table Al.2: Excise Duties for Cotton Textile Products, February 1997. Cotton lint - o | 2%¢l /| Man-Made staple fibers (polymides, 200/l polyester, acrylic, others)ll Yarrns Cotton Cone Yarn 5% 2% Cotton Hank Yarn 1. Handloom: Plain Reel 0% 0% 2. Handloom: Cross Reel d for: counts < 25s 0% 0% counts > 25s 3.45% 15% Cotton yarn with synthetic fiber 20% 2% If manufacture which does not 0% 15% use power Yarn of synthetic staple fibers 18% (polyester, acrylic, modacrylic) Synthetic Filament Yarn and sewing 1. Nylon filament yarn for 10% 15% threads, monofilament nylon-cotton sarees < 60 deniers 70% 2% 2. Polyester filament yarn for 10% >60 deniers 40% low priced fabricsf Artificial filament yarn and viscose 20% 2% filament yarn Polyester Filament Yarn 30%0/o 2% Nylon filament yarn >750 deniers 30% 2% <750 deniers Blended Yarn 20% ° 2% _ Fabric Cotton Fabrics 10% lOo%e 1. 100% cotton fabric of value 5% 5% Man-made Fabrics up to Rs 25/sqm Knitted Fabrics 2. Narrow fabrics with 15% elastometric yam or rubber thread 3. Unprocessed cotton or man- 0% made fabric not pile, terry towelling, chenille or with elastomertric yarn or rubber thread 4. fabrics produced with out 0% power or steam 5. knitted cotton fabrics Blended Fabrics 10% 10%e Processed man-made fabrics 10% 10%' Note: a - additional duty is computed on the applicable basic duty; b - the effective rate of duty after concessions; c - on cotton lint, there is no sales tax, but there is a purchase tax; d - if purchased by apex handloom cooperatives, government handloom corporations, or other specified bodies; e - additional excise duty is taken in lieu of sales tax; f - yarn to be used for program duly approved by the Development Commissioner of Handlooms; Source: Office of the Textile Commissioner, "Compendium of Textile Statistics, 1994," Bombay; D.N. Kohli, P. Avasthi, and M.R. Sharma, "1994-95 Central Excise Tariff," Delhi: A Cen-cus Publication; GOI, Memorandum Explaining the Provisions in the Finance Bill, 1996. Annex I Page 9 of 16 Table A 1.3: Government Programs to Assist Handlooms o~~~~~~~~~~~~~~~~~~~~~~R-g, MOm__ 1I.ntegrated handloom Janatacloth scheme Weavers service Thrift fund' HandloomAct1985 village development scheme centers scheme 2. Loom to loomless weavers Special rebates IIHT Group insurance .Hank yarn obligation scheme 3. Handloom development MDA Modemization of Health Package Excise exemption of centers looms schemes hank yam 4. Workshed-cum housing Silk yam bank scheme Margin money for scheme destitute weavers 5. Project package scheme Yarn at mill prices 6. Loan for coop members Managerial subsidy 7. Credit support Source: K. Srinivasiulu, 1996, "i985 Textiie Policy and Handloom Industiy, Policy, Promises and Performance",Economic and Political Weekly, pp. 3198-3206 Protection of Handlooms Remains Top Priority A1.27 In addition to legal and fiscal protection, the GOI has implemented a variety of other measures to support the handloom industry. These included capital, credit and operational subsidies, training programs, and social security schemes. (Table A1.3). Some measures came about in response to the short term crisis. For example, to shield the handloom industry from rising prices resulting from cotton production shortfalls in 1994, the GOI introduced the Hank Yarn Scheme, which consisted of a per unit subsidy of Rs 15/kg on 20 million kg of hank yarn. A budget of Rs 300 million was set aside for the purpose and is financed by a 15 percent tax on all textile items. This was channeled through handloom cooperatives and corporations. The Janata Cloth Scheme, financed through revenues generated by the additional tax, was also set up to assist handloom weavers. A1.28 The production of certain textile items is reservedfor the handloom industry under the Articles of Production for Handlooms Act (1985). It covers 22 items, including certain types of sarees, dhotis, gamcha, lungis, and table mats/napkins. The Act was judicially challenged and remained inoperative until it was upheld by the Supreme Court in 1994. By January 1995, 72,553 powerlooms had been inspected by enforcement officers and 656 criminal cases had been filed. D. External Trade Reforms Value Addition Drives Yarn Export Policies A 1.29 GOI priority for value-addition shaped cotton and yarn export policies in India. To ensure an adequate supply of raw materials and keep raw material prices low, cotton lint and yarn exports were subject to export quotas (see Annex 2 for discussion of cotton lint export quotas). In the case of yarn, the Ministry of Textiles sets the export ceilings yearly. The Cotton Textiles Export Promotion Council (Texprocil), under the Ministry of Textiles, is responsible for the allocation of export entitlements among exporters. Yarn exports until 1994 were also subject to a minimum export price. As a result of these restrictions, cotton yarn exports remain a very small percentage of total production. In 1994/95, they accounted for only 11 percent of total output. A1.30 Arbitrary revisions of the export ceilings to check changes in domestic prices hurt exporters. Until 1993, the ceiling on cotton yarn exports was 100 million kg. In 1994, the ceiling was raised to 130 million kg. In 1995, the ceiling was revise three times. Initially set at 100 million kg, it was suddenly reduced to 75 million kg in the middle of the year, exempting yarn in counts above forty. The 75 million was further divided between 52 million kg for Multi-fiber Annex I Page 10 of 16 Agreement quota countries and 23 million kg for non-quota countries. The sudden sharp reduction in quotas set off a panic as exporters were prevented from meeting their contractual deliveries. Later in the year (November) the ceiling was raised to 80 million kg. A1.31 Exemptions to the export quotas are driving the spectacular yarn export growth. Although the export quotas have been enforced for a long time, total exports have exceeded the ceiling due to numerous exemptions. As of August 1996, the following categories are exempted from the ceiling: * exports by 100 percent Export Oriented Units (EOUs); * exports against imports of machinery and capital goods under the Export Promotion Capital Good Scheme (EPCG); * exports against cotton imports under the Advance Licensing Scheme; * exports against cotton imports under Open General License (OGL); * exports of processed yarn, viz. mercerized, bleached, dyed and melange yarn; * exports to quota countries; and * exports of yarn above 40s counts. A 1.32 India now ranks as the third largest exporter ofda cown yarnk i the wirld, laftt Figure A1.3: Cotton Yarn Exports Exceed Export Ceilings exporter of cotton yamrnm the world, after Pakistan and China. Exports from EOUs are the main driving force for export growth. By 1995, . 100 percent EOUs accounted for nearly half of 2 l yarn exports, totaling 109.14 million kg (Figure 20< 't A1.3). 150 A1.33 GOI incentives promoted the expansion Eii of EQ Us. Principal benefits to EOUs include: a 5 tax holiday for a block of five years in a period of ° eight and the free import of raw materials and Co Co capital goods. Currently, they are permitted to sell up to 25 percent of their output domestically Source: ICMF, Handbook of Statistics on Cotton Textile *ndustry, various issues. provided that at least 30 percent of the raw material is of Indian origin. The 1995 estimates from SITRA show that EOUs account for about 22 percent of spindlage in the spinning sector. Multi-Fiber Agreement and Textile and Apparel Exports A1.34 The quotas under the Multi-fiber Agreement (MFA) are more binding for textile and apparel exports. The MFA provided a framework under which developed countries impose a quota on the export of yarn, textiles and apparel from developing countries. Quotas are applied typically on a bilateral basis, under the threat of unilateral restraints to be imposed by the importing country. The quotas emerge from bilateral negotiations and are specific to particular product categories, as defined by fiber and function. Text Box Al.1 describes some of the key features of the US and EU treaties. The MFA allowed for discrimination not only against specific fibers and products, but also among exporting countries. Al.35 The quotafor apparel exports are announcedfor three-year periods and are subject to specific criteria. The current quota allocation policy for textiles follows several systems: past performance entitlements (55 percent), manufacturer exporter entitlements (15 percent), ready goods export entitlements (15 percent), non-quota exporter's entitlements (10 percent), and powerloom exporters entitlements (5 percent). The garment quota allocation system follows knolex I ii ft 16 Tuxt Box A1.1 Main Features of Bilateral US and EU Agreements Indo-USA Agreement. The concessions offered by the USA included the abolition of the quota level equivalent to 100 million Swm and a five percnt inmease in the base levels of some categories of fabrics and made-ups. A clause in the agreemeotwas also inserted allowing India to levy advaloran rate or specific duty, whichever was higher, on imports of textiles or clothing. In exchange, the GO! will providC market access by remnoving restrictions and phasing out tariffs on textiles/clothing items. As a result of the agreement, India opened its market beginning January 1995 to almost all fibers, yarns, fabrics, made- ups and garments. Imports of all fibers, yams and specified industrial fabrics are allowed under OGE. All other items arc allowed under special import lieense. Indo-EU Agreement. In return for market access and tariff bindings offered by India to the EU, the community offered several concmssions, Exports of handloom garnents will be outside the quota regime. Flexibility of 7.000knt for the period 1995.97 and 8,000mt for 1995-2000 has been provided, over and above the existing flexibility allowed under the bilateral arrangement For two new categories, namely artificial fiber yarn (cat. 23) awd nightwear (cat. 24), the base level provided for the 1995 quota is 50 percn higher than the prevaUiing level of exports to the EU. The EU made commitments to provide for a spectial outward processing taiff quota amounting to 10 prcent of overal quantitative limits under category 7 (ladies blouses). S (men's shirts), 27 (adies' skirts) and 29 (ladies' suits and ensembles). India catgorically indicated its intention to discourage entry of low-priced and poor quality textile products from the EEC into India, This is to be achieved through a mechanism of fixing a specific duty rate as an altermative to ad valorem rate and providing for charging the duty, which ever is higher. Source- 'Textile Industry, A Review 1994-95," 7he lndas&n Textile Afonftotr Vol. 12, March 1996 pp. 42-5 1. three systems: past performance entitlement (70 percent), manufacturer exporter entitlements ( 15 percent), and non-quota exporters entitlements (15 percent).8 Al.36 Quota rents hinder entry of new firms. As quotas are largely controlled by established exporters, new entrants must pay quota transfer prices in the open market (which is legitimate) to enter the business. It is reported that quota premiums range from Rs 30 to Rs 60 per piece of apparel.9 The quota rents also reduce the profitability of new enterprises, because of the high quota transfer prices. New Export Opportunities under the ATC A 1.37 MFA phase-out under ATC opens new nriarket opportunities for India. In thc U1runguay Round of the GATT Agreement, the Agreement on Textiles and Clothing (ATC) worked out the phase-out of the MFA over a ten-year period beginning in 1995. Text Box A1.2 lists the main features of the ATC. Under the ATC agreement, quota growth rates will increase in three stages: 16 percent during 1995-1997, 25 percent during 1998-2002, and 27 percent for 2003-2005. At each of the three phases, the quotas for portions of product categories (16 percent, 17 percent and 19 percent of the 1990 export volumes respectively) will be removed from the MNFA. The ATC, however, gives considerable leeway to importing countries in their choice of products and subsectors (yarn, fabrics, clothing and textile products) for liberalization. Annex 6 examines in greater detail the impact of the MFA phase-out on the Indian textile industry. Import Policies A1.38 The econorr7ic liberalizationi program71 initiated in 1991 andcf the signing ot the (fiATT' brought a dramatic reduction of cottoni textile import restraints. The import of raw cotton. polyester and viscose staple fiber, man-made filament varns and textiles has been placed under Open General License (OGL). In the case of cotton, this was a major break in past policies x\here 'Textile Industr' in India. Present Status and Future Prospects'' O(uuteriv [to(omni, ReplI,rt te ltlic, of Puiblic Opinion. Vol. 38. No. 2. pp. 27-40. 9 Industrial Credit and In\estment Corporation ot India. Ltd. I and R. Jaikumar. I195. . and the \lulti-Fiber Agreement. Strategies for the Indian Apparel Industr%. 1(C11. Annex I Page 12 ol 16 until t 991. COttO imports were canalized through the Cotton Corporation of India. For textile products, market access into India was expanded in 1994, in separate treaties with the USA and the Ft-U. In exchange for increases in its MFA quotas, India agreed to a comprehensive liberalization of its textile import policies. Prior to 1994, textile and garment imports were in practice banned. except for some intermediate textiles and garments under special arrangements. In 1995, this translated to the freeing of imports of textile yams, synthetic fibers and some selected industrial fabrics (See Appendix Table Al.2). Selected textile fabrics (synthetic and cotton) and made-ups and a long list of apparel items were also allowed to be imported under a new special license issued to exporters (Special Import License, SIL). Special import licenses are granted to trading houses and certain other exporters as a share of their export earnings. SILs are tradable and can be used to import items on a special list of products whose import would otherwise be banned or restricted. It was also agreed that these products will be completely free from import licensing at specified future dates. Most cotton fabrics, however, which account for the bulk of Indian fabric production, and about half of apparel tariff lines were omitted from the treat, and are subject to continuing quantitative restrictions (QRs). Although these reforms were negotiated with the USA and EU. they are multi-lateral, because they apply to all countries exportilg to India also. Text Box A1.2. The Agreement on Textiles and Clothing and the Phs-out of the MFA A Trhree Stage Integration Proeess. The GAT provides the legal framework for the phasing out of the MA3nd its integration into the GATT/WT0 firwework by the year 2005. The 10 year transition l id has been divid phases. On January 1, 1995, importing countries were rquired to ingr a ce po f le and c h categories by removing them frm the MFA. Any xisting quotas restricting imports s are to bea pcnnanently. and these producs had to represent at least 16% of 1990 import volunx. Morover, products sdd to include at least one category from each of the followingsubsectors: yans, fabics, clothing, d othertexte . rates of remaining quotas were to increase by 16 pecnt diring the first phase, i.e. a quota subetto a 10 percent growth ratf in 1994, will grow by 11.6 percent during 1995-97, The scoond stago begins on January 1998-2001, another 17 percent of 1990 import volume will be chosen from a product list for ingtimon, export igrowt rates will be increaied by 25 percent. From January 2002-2005, another 18Me will be integrated and the growtl rates will be incrased to27 perent. On January 005, all remaining products, representing at least 49 percent of 1990 import volmne, will be integrated. For products which remain on the list, the MFA bilateral framework contius. This permits differenial growth rates to continue to apply to different exporting countries and different products until the end of the integration process, because initial growth rates differd at the start of the intcgration process. Safeguards. Once a product category has been inegated, importing countries are not ailowed to reitoduce quotas restricting imports. They are allowed, however, to use the safeguard measures under the GAIT Article XIX during th phase-out period, This allows countries to take acion in the event of sudden impot surges that threaten their domestic industries, In addition, importing countries can in^itite antdWumpg action agitu y t Tariffs remain high post-WR. Tarif for textiles and clothing will remain high despite the Uruguay Round (UR) agreement. The average tariff for textiles and clothing declined'from 16 percent to 12 percent post UR, but is tee times the average post-JR tariff for all goods of 4 percent. Source: M. Majmudar, 1996, "Me MFA Phase-out and EU ClothingSourcing: Forecasts to 2005,' TextIlk Outlook International, Mlarch 1996. 13 39 In Inie with the liberalization program, import tariffs on textile raw materials. Intel-nlediates and J1finished procucts have been reduced significanitly. As of shown in Table Al .4, b\ June 1996 . tariffs on cotton-based products were close to or below the 1998 EU/USA Textile I reaty targets and the GATT 2001 bildings. For man-made textile raw materials and products, an evcn more dramatic reduction in import tariffs has been pursued. For man-made fibers and \ irlns. tarifts were cut from a range of 50 to 150 percent to 30 percent in 1996. Annex I Page 13 of 16 Table A1.4: Import Tariffs on Cotton and Man-Made Products nbsed Pout Cotton Apparel 52% 40% 50%* 50% 25% Cotton fabric 52% 40% 30%* 40% 25% Cotton yai n 27% % 25% 35% 20% Cotton lint 0% 100% 35% 20% Raw cotton 0% 150% Man-Made Fiber Products Apparel 52% NB 50% 50% 40% Fabrics 52% NB 30% 40% 25% Nylon, Polyester, and Viscose Filarment yarn 32% 40% 30% 35% 200% Other man-made filament yarn 30% 400/o 30% 35% 1 20% Artificia and synthetic fibers 27-32% 40o%O 25No 355 20 DMT, PTA, & MEG 25% 2540% 20% I/ includes 2% special duty. Source: GOI and G. Pursell, 1996, "Indian Trade Policies Since the 1991/92 Reforms," forthcoming; GOI, Memorandum Explaining the Provisions in the Finance Bill, 1996. Annex I page 14 of 16 E. Appendix Tables Appendix Table Al.1: Textile Exports, Rs million. Item 90/91 91/92 92/93 93/94 94/95 95/96 Fabrics 11,965 17,067 20,492 23,863 32,093 36,729 Millmade 6,167 7,149 8,033 9,510 12,009 14,492 Powerloom 3,233 6,474 7,710 9,310 14,046 16,610 Handloom 1,227 1,872 2,730 2,817 3,180 2,553 Knitted 1,338 1,572 2,019 2,227 2,858 3,075 Made-ups 7,455 11,362 16,463 22,933 29,141 33,539 Millmade 2,554 3,130 3,249 3,306 3,654 4,891 Powerloom 2,055 3,212 5,542 9,403 13,633 16,175 Handloom 2,846 5,020 7,606 10,161 11,794 12,360 Knitted . 66 63 60 113 Yarn 5,114 9,824 11,619 15,999 26,085 33,583 Sewing Thread 49 67 47 54 31 49 Garments 30,175 43,815 64,756 85,820 97,782 103,543 Woven 20,892 29,958 41,299 55,507 67,549 66,968 Knitted 9,283 13,858 23,457 30,314 30,233 36,575 Total Cotton Textiles 54,758 82,134 113,377 148,670 185,132 207,444 Rayon/Synthetics excl garments 6,361 11,061 14,363 18,468 24,847 31,227 Fabrics 3,464 6,155 8,081 11,775 14,992 17,265 Spun Yarn 649 1,772 2,320 2,557 4,441 7,289 Filament Yarn 117 599 900 450 1,202 1,888 Made-ups 1,253 1,727 2,023 2,648 2,900 3,503 Fibers 540 357 554 395 638 534 Other Items 339 451 485 644 675 748 Woolen exci garments 923 1,667 3,125 4,484 3,600 4,389 Non-cotton garments 16,431 19,009 23,652 28,427 41,591 44,521 Silk (fabrics, made-ups, garments) 4,359 6,756 7,200 7,891 9,375 8,461 Total Textiles 82,833 120,626 161,716 207,940 264,543 296,040 Total Merchandise Exports 331,662 447,882 498,330 711,339 843,310 1,085,208 Source: Textile exports from ICMF, Handbook of Statistics on Cotton Textile Industry, New Delhi, various issues. Total merchandise exports from World Bank, India Sustaining Rapid Economic Growth (1997 CEM). Annex I page 15 of 16 Appendix Table Al.2: India's Commitments on Textile Imports under the Agreements with the USA & EU. I ~~~ ~~QRs I Tariffs Date effective Jan 1, 95 May 95 Jan 1, 98 Jan 1, 2000 Jan 1, 95 Jan 1, 98 Jan 1, 99 Jan 1, 2000 Jan 1, 2002TTan 1, 2005 COTrON (Ch.52) . . .. _ Yams* Free 65 40 20 A few fabrics (denim & SIL Free 65 40 30 prints) A few more of the above SIL Free 65 45 40 35 Most cotton fabrics No commitment No commitment WOOL (Ch.51) _ Wool tops, etc. Free _ 65 35 20 Wool yarns* Free 65 40 20 Wool fabrics SIL Free 65 40 30 25 MAN-MADE FILAMENT TEXTILES (ch.54) Fibers | Free | | 165 4 135 0 _ 20 l__ YaFrs Free 65 40io _ 20 _ 20 Most fabrics SIL Free _ j 65 f 40 j 30 Specified other fabrics SIL Free 65 f 40 | J 20 Source: G. Pursell, "Indian Trade Policies Since the 1991/92 Reforms," forthcoming. Annex I page 16 of 16 Appendix Table 1.3:Effective Rates of Excise Duties on Man-Made Staple Fibres, Rs. per Kg Synthetic Fibres Effective Dates Cellulosic Fibre Polyester ._______ . Fibre Acrylic Polypropylene Others From To of less of 1.5 Not more Fibre Fibre than 1.5 D & than 2 D D above 01-03-84 27-08-85 5.00 45.00 + 17.50@ - 37.50 28-08-85 05-09-85 5.00 25.00 + 17.50 - 37.50 06-09-85 28-02-86 5.00 25.00 + 10.00 - 37.50 01-03-86 21-05-86 5.00 25.00 ** 10.00 - 37.50 22-05-86 31-12-86 5.00 25.00 *** 10.00 5.68 37.50 01-01-87 28-02-87 5.00 25.00 *** 10.00 5.68 37.50 01-03-87 30-04-87 7.00 25.00 10.00 5.68 37.50 Polypropylene Others Nylon Spandex Fibre Staple Fibre 01-05-87 29-02-88 7.00 25.00 10.00 7.95 37.50 - 01-03-88 28-02-89 8.35 15.66 8.35 8.30 39.15 - a) 5.22 b) 8.35 01-03-89 19-03-90 8.35 16.44 ++ 8.77 5.00 41.11 - 20-03-90 14-12-90 10.20 10.20 ## 9.00 5.40 41.58 - 15-12-90 24-07-91 12.60 12.60## 11.09 5.40 41.58 - 25-07-91 28-02-92 13.13 13.13## 11.55 17.19 43.32 56.88 01-03-92 28-02-93 15.60 13.66 15.60 17.87 45.05 59.16 01-03-93 28-02-94 14.95 12.65 14.95 17.25 46.00 14.95 0 1-03-94 & Onwards Ad-Velorem 23% 23% 23% 23% 23% 23% 11.50% Note: @ If produced from indigenous acrylonitrile. For other acrylic fibre duty rates is Rs. 30/-per Kg .w.e.f. 24-80 to 27-8-85. + If produced out of polyester waste the aggregate excise duty is Rs. 33.75 per Kg. w.e.f. 4-2-82 to 5-9-85. ** If produced out of recycled polyester waste the aggregate excise duty is Rs. 14.00 per Kg. w.e.f. 6-9-85 to 28-2-86. @* * If produced out of polyester waste the aggregate excise duty is Rs. 20.00 per Kg. w.e.f. 1-3-86 to 31-12-86. ++ If produced out of recycled polyester waste the aggregate excise duty is Rs. 10.00 per Kg. w.e.f. 17-7-89 to 19-3-90. ## If produced out of recycled polyester waste the aggregate excise duty is Rs. 9.60 per Kg. w.e.f. 15-2-90 to 24.7.91. Rs. 10.00 per Kg. w.e.f. 25-7-91 to 29-2-92. * For Nylon fibre at Rs. 6.00 from 1-1 -76 to 24-7-91 Rs. 56.88 from 25-7-91 to 28-2-92. Rs. 59.15 from 1-3-92 to 27-2-93 Rs. 14.95 from 28-2-93 to 28-2-94. $ Regular Viscose Fibre # Polynosic & HWM Fibre $$ Acetate Staple Fibre attracts excise duty at the rate of Rs. 5/-per Kg. a Concession duty @ Rs. 5.22 per Kg. is payable if used for blending with cotton in the manufacture of Cotton Yam. b concessional duty @ Rs. 8.35 per Kg. is payable if used for blending otherwise. From 28-8-85 onwards Polyester Fibre used for manufacturing of Handloom Fabrics duty is nil if the content of the Fibre should be more than 40% but less than 70% by weight. Spandex yam effective from 27-6-94. Source: Association of Synthetic Fiber Industry, 1995, 1994-1995 Handbook of Statistics on Man-Made Synthetic Fiber/Yarn Industry, Bombay: ASPC Annex 2 Cotton Sector Policy A. Cotton Production Policies Cotton Price Support A2.1 There is a GOI cotton minimum price, but due to high market prices, procurement to support cotton prices has been undertaken only intermittently. The GOI minimum support price (MSP) dates back to the Second World War when the British government prescribed statutory floor and ceiling prices for cotton lint. Price ceilings were removed in 1967, but the statutory minimum price for each cotton lint variety was retained to protect farmers. In 1972-73, the MSP for cotton lint was converted to a MSP for seed cotton, since farmers marketed the latter. The Commission for Agricultural Costs and Prices (CACP), formerly the Agricultural Prices Commission, sets the minimum support price for each cotton variety. Appendix Table A2.1 lists the MSP for various cotton varieties from 1990/91 to 1994/95. The MSP is fixed by the Textile Commissioner for the Fair Average Quality (FAQ) grade of each variety of seed cotton on the basis of recommendations from the CACP. The Cotton Corporation of India (CCI) is responsible for procuring cotton from the free market to support these prices (see Annex 5 for more detailed description of CCI operations). Although there exists a MSP, actual seed cotton prices in the markets are generally higher by as much as 30 to 50 percent, even during the peak harvesting season. Over the last decade, CCI procured cotton only during the years 1985/86 (1.3 million bales), 1986/87 (0.5 million bales); 1989/90 (1,600 bales) and 1996/97 (11,000 bales). In most instances, the minimum support price serves more as a "reference price." A2.2 Maharashtra has a unique price support program enforced through the Maharashtra Monopoly Procurement Scheme. The Government of Maharashtra launched its Cotton Monopoly Procurement Scheme in 1972/73 under the Maharashtra Cotton (Procurement, Processing and Marketing) Act, 1971. The objectives of the Scheme are: (i) to ensure fair and remunerative prices to the cotton growers in the state by eliminating middlemen, and (ii) to bring about stability and growth in the overall production of cotton in the state. This scheme assures cotton growers a guaranteed variety-specific price for seed cotton fixed for the season. Prior to 1986/87, the state govermment guaranteed prices for each variety at about 10-15 percent above the support prices announced by the Govermment of India. This led to large financial losses and prompted a change in procurement policy. Purchases are now made at support levels only, with the state government advancing bonuses to restore price parity with neighboring states and avoid smuggling. At the end of the season, profits are distributed in a 75:25 ratio to growers and the Price Fluctuation Fund. Losses are always debited to the Price Fluctuation Fund. Under the provisions of Maharashtra Raw Cotton (Procurement, Processing, and Marketing) Act, 1971 - Section 33-3, 25 percent of the difference between the final price and the guaranteed price is set aside in a Price Fluctuation Fund (PFF). A2.3 MSCCGMF implements the Maharashtra Monopoly Procurement Scheme. When the MCPS was first introduced, the responsibility for cotton procurement was assigned to the Maharashtra State Cooperative Marketing Federation (MSCMF). Cotton cultivators were prohibited from selling kapas to any other buyer except the Federation and its affiliated Annex 2 Page 2 of 13 cooperative societies. In the 1980s, the responsibility was transferred to the Maharashtra State Cooperative Cotton Growers' Marketing Federation (MSCCGMF). Intensive Cotton Development Program A2.4 The Intensive Cotton Development Program (ICDP) is a Centrally Sponsored FigureA2.1:GOIExpendituresonICDP. Scheme initiated in the early 1970s to improve g 120 cotton productivity through the adoption of new @ 100 technology. The instruments used to enhance 80K. farmer adoption of improved technologies o 601 included field demonstrations, subsidized g 40 distribution of key inputs, and extension support. 2!0 The scheme currently covers 11 states. In o 0 1995/96, actual government expenditures in the 9 o c . program reached Rs 123.1 million (Appendix 4511° SOP se \ ' o. (1A6 Table A2.2). ICDP is operated on a cost sharing basis with state govermments. Prior to 1991/92, |_ Budgeted + Actual| the funding pattern of ICDP was 50:50 between Source: GOI, Budget Documents, various issues. Center and State. This was revised to 75:25 in 1992/93. As a result of the change in cost sharing proportions, GOI expenditures in the program more than doubled in real terms (Figure A2.1). An analysis of the ICDP expenditures in two states, Haryana and Maharashtra, show that together they account for about 35 percent of total expenditures in 1993/94 (the latest year for which state level data were available). A2.5 Periodic technical changes in the scheme reflect changing priorities in the sector. The program till the mid-1980s focused primarily on promoting the adoption of improved seed varieties. Funds were mainly spent on certified seeds, production of breeder and foundation seeds, together with some support for plant protection and demonstrations. In 1986/87, to meet the excess demand for long and medium staple cotton, the program was adjusted to give a thrust to the distribution of certified seeds of medium and long staple cotton. A2.6 ICDP was modified in 1989/90 to give special attention to plant protection measures and increased production of exportable types, viz., superior, long and desi cotton. Efforts were also made to (a) increase production of irrigated cotton at the tail-end of the canals, and (b) encourage use of delinted seeds of improved varieties/hybrids in larger areas. Government assistance was channeled to the production and distribution of foundation and certified seeds and use' of acid- delinted seed, pheromone traps, distribution of integrated pest management (IPM) kits, plant protection, tractor mounted sprayers, sprinklers, etc. In 1994/95, the emphasis on more environmentally friendly approaches to pest management increased. This is reflected in the increased expenditure share that goes to IPM demonstrations. In Maharashtra and Haryana, for instance, till 1993/94, more than 90 percent of ICDP funding was spent on input subsidies, but since then a significant reallocation has taken place and now over a third of expenditures is used to finance demonstrations. A2.7 ICDP's impact on yield growth shows mixed results. District level yields of the ICDP program beneficiaries were compared to non-participating districts in two states, Haryana and Maharashtra. Both ICDP and non-ICDP districts posted modest yield growths during the period 1980/81 to 1992/93. However, the ICDP districts-- selected for their lower yields than the state average -- in Haryana performed only slightly better than non-ICDP districts. In Maharashtra, Annex 2 Page 3 of 13 ICDP districts performed slightly worse than non- Table A2. 1: Average Annual Rate Of Yield ICDP districts (Table A2.1).1 A number of factors Growth (80/81-92/93) Of ICDP And Non-ICDP Districts In Haryana And Maharashtra may contribute to ICDP's weak performance, including poor design, spreading resources too thinly, and limited flexibility and implementation capacity at Haryana 2.65% 1.19% the local level. Further research may throw more light Maharashtra 1.68% 2.35% on specific shortcomings, but based on the limited Source: computed. evidence available from Maharashtra and Haryana, ICDP does not appear to have made any appreciable difference to productivity growth at the district level. A2.8 Proposed Cotton Technology Mission. To ensure the sustained growth of cotton production, the GOI is proposing to establish a Cotton Technology Mission. The mission will focus on several areas, including: strengthening cotton research to increase productivity (i.e. development of improved seeds and improved nutrient and pest management) and promoting the increased modernization of cotton processing (i.e. ginning and pressing technologies) B. Input Policies Seed Policy A2.9 The production, multiplication, and distribution of cotton seeds in India are governed by several policies and regulations. These include: Seed Act (1966), Seed Control Order (1983), Breeder Seed Supply 1986, New Seed Policy of 1988, Plants, Fruits, and Seeds Order 1989, and the Cotton Control Act A2.10 Seed Act of 1966. The Seed Act of 1966 provides the basis for regulating seed sales. Under the Act, which became effective in 1968, labeling of seeds for sale became compulsory, but certification remained voluntary. For seed certification, prior notification of the variety/hybrid is an essential requirement. The Central Sub-Committee on crop standards approves the release and notification of a multi-state variety or hybrid, while the concerned state is charged with the authority of the release of a single state variety or hybrid. A2.11 Many private firms prefer to release seeds without certification to speed up dissemination. The seed certification process is cumbersome and time-consuming. Thus, most private companies enter their hybrids and varieties in official trials, but at the same time start marketing their seeds as truthfully labeled seeds. Companies have their own network for testing in different agro-climnatic zones. Thus planting material reaches farmers sooner than would be the case if they had to wait for completion of the notification and release procedures. A2.12 Manpower in the seed certification agencies is weak There are 19 state seed certification agencies. Most of the personnel of these agencies are drawn from the respective Departments of Agriculture, but do not necessarily have adequate knowledge and training on seed certification. Frequent transfers further hurt the quality of work. IMinistry of Agriculture notes that the poor performance of ICDP districts in Maharashtra could also be attributed to the socio-economic conditions of farmers and prevailing agro-climatic conditions. 2 P. K. Agrawal, 1995, "Seed Industry Regulations in Relation to Seed Industry Development in India," in D. Gisselquist (ed.) "Easing Barriers to Movement of Plant Varieties for Agricultural Development. Annex 2 Page 4 of 13 A2.13 Seed Control Order, 1983. This Order brought seed sales within the purview of the Essential Commodities (EC) Act. As per the EC Act, all seed dealers and storage points require a license to operate. The Order also requires dealers and storage points to display stocks and prices daily and gives authority to inspectors freely enter and inspect the premises, draw samples, and stop the sale of seeds if they do not conform to norms and standards. Due to a legal challenge, the Order was not enforced for many years. In 1994, however, the Supreme Court upheld the legality of Order, and it became operative in July 1994. A2.14 Breeder's Seed Supply, 1986. This legislation liberalized the multiplication of publicly developed seeds. Prior to 1986, public sector institutions supplied breeder seeds to the National Seed Corporation (NSC), a parastatal, for multiplication. The resulting foundation seeds were then sold to the private sector for the production of certified seeds. This arrangement slowed the diffusion of improved seeds considerably. Recognizing the problem, the GOI in this landmark decision, opened the multiplication of publicly-bred self pollinated varieties to private seed companies. This policy change alleviated the shortage of certified seeds and has been a major factor stimulating the growth of the private seed industry in India. A2.15 New Seed Policy of 1988. This policy liberalized the import of improved seed& The GOI restricted the import and export of seeds until 1988. The objective of the New Seed Policy, 1988 was to allow farmer access to the best seed available regardless of origin. Consequently, import procedures were liberalized and many seeds and planting materials became freely importable. The new policy also encouraged foreign companies to invest in the seed industry. Cotton seeds and planting material, however, were not included. Cotton seed imports are allowed on a "case by case" basis for multi-location testing under the All India Coordinated Cotton Improvement Project of the Indian Council for Agricultural Research to evaluate their suitability to the country. A2.16 Plants, Fruits, and Seeds Order 1989. Exports and imports of seeds are regulated by this Order. Within the Ministry of Agriculture, the Directorate of Plant Protection, Quarantine and Storage regulates the import and export of seeds. At present, there is no seed testing laboratory in India that is accredited by the International Seed Testing Association. A2.17 Cotton Control Act. The Cotton Control Act was established to maintain varietal purity and to regulate diseases. The implementing agencies are both the Central and State Departments of Agriculture. Under the act, these agencies can direct production of selected varieties in selected areas, and also prohibit mixing of varieties. Because implementation of such physical controls is virtually impossible, the Act is has not been enforced. It is estimated that there are about 100 varieties in India, of which 20 to 25 account for 90 percent of the crop. In contrast, China cultivates 7 varieties, USA has 9 and Egypt has 8.3 Moreover, mixing of seeds of different varieties is ubiquitous in India. A2.18 Proposed Plant Variety Protection Act, 1993. In accordance with the Trade Related Intellectual Property (TRIPs) Agreement under the GATT, India is required to enact a Plant Breeder's Rights legislation by the year 2005 At present, there is no legislation which provides for plant variety protection or ownership of varieties. According to the existing Indian Patent Act of 1970, no lifeform is patentable. This is being rectified by a proposed Plant Varieties Act (PVA) formulated in 1993 and currently under consideration by Parliament. The key features of the Act include: * Protection of new varieties for a duration 15 years; 3 Indian Raw Cotton Export," Business Standard, August 26, 1996. Annex 2 Page 5 of 13 * Establishment of a variety evaluation agency/system for varieties applying for patent protection; * Seeds of patented varieties to be deposited with the National Gene Bank to be publicly available for development of new varieties; * Compulsory licensing of seed producers, with government empowered to issue additional licenses to other producers if current license holder fails to produce seeds required by farmers; * Compulsory certification of patented varieties; * Farrners will be allowed to re-use and sell seeds grown from seeds of protected varieties; and, * Variety owners or seed dealers are to provide compensation to growers if the protected variety does not perform as advertised. A2.19 Although the proposed PVA is designed to protect the plant breeder's rights and thus encourage private sector participation, two provisions are being criticized by the private sector. First, the authority given to government to issue additional licenses if the current license holder fails to produce an adequate supply of seeds contradicts the principle of plant breeder's proprietary rights. Second, forcing variety owners or seed dealers to compensate growers if a protected variety does not perform as advertised is a highly inefficient mechanism to discourage fraud. It would be more efficient to address this under existing legal avenues for dealing with and penalizing fraudulent behavior. Pesticide Policy A2.20 Pesticide Regulations. Two legislations provide the basis for regulating pesticide use in India: the Insecticide Act, 1968 and the Insecticide Rules, 1971. These legislations were enacted to regulate the import, manufacture, sale, transport, distribution and use of pesticides to prevent risks to human beings or animals, and govern all matters connected therewith. The central agencies were established under the Act: the Central Insecticides Board (CIB) and the Registration Committee (RC). The CIB headed by the Director General, Health Services, GOI consists of 29 members who advise the Central Government and State Governments on matters relating to: (i) the risks to human beings or animals involved in the use of insecticides and the safety measures necessary to prevent such risks; and (ii) the manufacture, sale, storage, transport and distribution of the insecticides with a view to ensure safety to human beings or animals. The RC consisting of a Chairman, and five members (including the Drugs Controller of India and Plant Protection Adviser to Government of India) is responsible for registering pesticides after verifying their efficacy and safety to human beings and animals. While the registration of pesticides and formulations and matters connected therewith is the Center's responsibility, the burden of field enforcement of the provisions of the Act falls on the shoulders of the States and the Union Territories. C. Cotton Trade and Processing Regulations A2.21 A number of restrictive policies and regulations historically governed cotton marketing. In recent years, one has been officially repealed (Cotton Transport Act). Most of the other regulations officially remain, but their enforcement has been lifted for the time being (Cotton Control Order 1955, 1986, 1995; Selective Credit Controls, Cotton Ginning and Pressing Act, Cotton futures trading under the Forward Contracts (Regulation) Act, 1952). Most of these regulations flow from the Essential Commodities Act, 1955 (EC Act). Cotton, whether ginned or un-ginned, is classified Annex 2 Page 6 of 13 as an essential commodity under the EC Act. The Act empowers the Textile Commissioner of the Ministry of Textiles to control the product mix, a large array of prices, stock levels and marketing and movement of cotton in the country. Cotton (Control) Order, 1955,1986 and 1995 A2.22 The Cotton (Control) Order, introduced in 1955, conveys vast powers to the GOI to regulate cotton marketing in the country. These powers are vested with the Textile Commissioner and are enforced only during emergency situations. These include the right to: * confiscate cotton; * fix storage limits for traders, ginners and mills, * restrict movement of seed cotton and cotton lint, * specify maximum/minimum quantities that spinners can buy, * issue licenses to traders and ginners to conduct any business in cotton, and * fix requirements of various licensees to maintain records of stocks, receipts, and sales, and file periodic returns. A2.23 Some of these provisions, particularly those related to storage and movement controls, licensing and filing of returns, are frequently revised, lifted and reintroduced. Others are less often used, but their mere presence deter private investment in the sector. Prior to 1994, the Textile Commissioner was empowered to put restrictions on cotton stocks held only by yarn (mill) manufacturers. But due to a sharp production shortfall, the order was expanded to cover all cotton buyers (ginning and pressing factories, private traders, etc) except cotton farmers in 1994. OnApril 1994, the Textile Commissioner prohibited mills from holding stocks of Indian cotton in excess of their 3 month consumption during the previous year. Private trade and ginning and pressing factories were barred from holding cotton, pressed or unpressed, in excess of 110 percent of the quantity held on the last day of the corresponding month in the previous year or 10 mt, whichever was greater. These storage limits, however, were rescinded in October, but reintroduced for private traders and ginning and pressing factories in December 1994. In February 1995, the limit was further relaxed when ginning and pressing units were included in the storage limit exemptions. Finally in June 1995, the storage limit was completely eliminated. CCI and cooperative federations were exempted from these restrictions throughout. A2.24 Continually changing provisions increase trading uncertainties and raise costs. Instances such as this, where there is frequent tinkering with the private stocking function in response to short term production variations, raise uncertainties which inflate transaction costs (partly because those who can stock at least cost are not allowed to do so) and discourage investments. The overall costs can be quite large since private trade handles almost 85 percent of the cotton marketed in the country. These controls also limit competition since the ceiling set for new entrants is a meager 10 mt, which is an uneconomic size for entering and investing in the business. It is also not clear just how much these types of restrictions contribute to check hoarding, which is the main GOI justification for imposing these controls. Field visits found some wholesale traders obtaining multiple licenses under different names to circumvent the storage limitation. This increases costs further, since the trader has to maintain and report separate records for each license. In practice, therefore, there is little to curb excessive stock holding, if a trader so desires. Cotton Transport Act A2.25 This Cotton Transport Act of the GOI requires a license for the transportation of cotton. It imposes restrictions on the free transport of both seed and lint cotton in order to avoid mixing of varieties. The Act was rescinded in May 1995. Annex 2 Page 7 of 13 Cotton Ginning and Pressing Factories Act, 1925 A2.26 The Cotton Ginning and Pressing Act, 1925, until its lifting in 1997, and various rules framed by state governments reduce incentives to modernize and improve quality. The Act covers the allotment of press marks to ginneries, filing of returns on bales pressed, and perhaps most important, maximum rates that can be charged by ginning and pressing factories. Under the Ginning Act(s), various state governments announce conversion charges specific to each variety and regions. The fixed charges which is based on weight, however, create an incentive for ginners to retain trash in the lint, rather than exclude it. Although a premium for gins equipped with pre-and post-cleaning equipment is embedded in the notified charges, it is reportedly inadequate to encourage gins to improve ginning processes or modernize. Trend analysis for a sample of three states shows that in some cases, notably Punjab, these rates decreased in real terms between 1985/86-1993/94. In others there Table A2.2:Annual Growth Rates in Ginning were only marginal real increases (Table A2.2). and Pressing Charges in Real Terns in Appendix Table A2. 1 shows the ginning andSectdtas(18/-934 pressing charges in real 1990 rupees. ginning_and Punjab -2.4% -0.7% A2.27 The repeal of the Ginning and Pressing Maharashtra 0.7% 1.0% Factories Act is under consideration by Parliament. IAndhra Pradesh 0.3% -0.3% By 1996, however, all states have rescinded related p legislation. Punjab, in 1995, was the first state to officially lift controls on ginning and pressing fees. In 1996, the other states followed. As an incentive for modernization, CCI provided separate cotton export quotas for firms using modern ginning and pressing technologies. Selective Credit Controls A2.28 Selective credit policies restricted private storage of lint and yarn. TheReserve Bank of India (RBI) through selective credit controls under Section 21 of the Banking Regulation Act regulates the cost of credit, maintenance margins, volume of credit, and loan purposes against selected commodities. These controls, adjusted frequently in light of emerging commodity balances and prices, influence the stock holding behavior of private traders. A2.29 In October 1996, the RBI lifted the selective credit controls on cotton and kapas. Selective credit controls have been applied to seed and lint cotton on a fairly regular basis since 1965 (Annex Table A2.2). Until October 1996, a minimum margin of 30 percent on stocks and warehouse receipts, and a credit ceiling of 110 percent of the peak level of credit maintained by the client in any of the three years ending October 1994/95, was in force. These percentages are, however, changed frequently, as much as 4-5 times a year. The public sector and cooperative units are exempt from these regulations. In 1985, private cotton mills (including spinning units) were exempted, and since then these controls have applied only to cotton traders and ginning and pressing units. Since private trade handles most of the cotton marketed in India (85 percent excluding Maharashtra), these controls, together with physical stock limits under the ECA Act, provide GOI with powerful tools to regulate India's cotton trade. On October 21, 1996, "in the context of the favorable supply situation," the RBI lifted the credit controls on private traders and ginning and pressing units, which completely liberalized access to credit for all market participants.4 Banks are free to fix prudential margins on advances against sensitive commodities. 4RBI, 'RBI Monetary Policy for the Second Half of 1996-97," October 19, 1996. Annex 2 Page 8 of 13 D. Risk Management The Forward Contracts (Regulation) Act, 1952 A2.30 The Forward Contracts (Regulation) Act, 1952 (FC(R)) Act governs the trading of commodity futures contracts in India. After an almost 31 year hiatus, the GOI lifted the ban on the trading of cotton futures contracts in February 1997. Commodity futures contracts are forward contracts that are tradable and standardized in their obligation to make or take delivery of a fixed quantity and quality of a commodity at a specific location, on a specific future date or time. Futures contracts are backed by margin money which insures against default. Cotton was the first commodity for which futures contracts were introduced in India, with the first organized market formed in 1921. In the 1920s and 1930s, futures and option contracts on several types of cotton were actively traded in two different associations, and futures trade in the largest of these, the East India Cotton Association (which organized export-oriented trade) was of international importance. Futures trade in cotton again became important after a short interval in the 1940s during which all futures trade was banned. From 1952 to 1966, several exchanges in the country showed an active trade in futures in raw cotton and also, between 1964 and 1970, in kapas. A2.31 Prior to 1997, only NTSD contracts in raw cotton were allowed The NTSD cotton contract is a type of forward contract between two parties in which a commodity of specific grade, has to be delivered to a specified location during a pre-determined time frame at a predetermined price. Neither buyer or seller can transfer the contract to another party and financial settlement is not allowed. Grade, location and delivery dates cannot be renegotiated after the contract has been signed. At present, eight associations5 are recognized and allowed to trade NTSD contracts by the Forward Markets Commission, which implements the FC(R) Act.6 A2.32 Restrictions on cotton fiutures trading fostered active illegalfutures trading. In response to the legitimate hedging needs of cotton gins, a significant illegal trade in seed cotton and cotton lint futures is reported to have developed with its main center in Surendranagarin, where V-797 cotton futures are traded primarily around harvest time, and in other centers. This illegal trade appears to be relatively well-organized: participants pay margins in case their positions move against them, and default rates are reportedly quite low. E. External Trade Policies Cotton Lint Export Quotas A2.33 Priority given to value addition by the GOI drives export restrictions on cotton lint. As for most agricultural commodities, India's extemal trade policies for cotton have aimed at domestic self-sufficiency with external trade largely being a "residual" to balance domestic demand and supply. The implicit underlying principles governing cotton export policies in India may be summarized as: * aimed at export value-added export rather than raw cotton, * to achieve this, keep cotton prices low for domestic industry, and * any exports of cotton be "surplus" to domestic requirements. 5 An association has to apply for recognition from the central government. Taking into account public and the industry's interests, the government may or may not approve the application. In general, recognition is granted for only a short period at a time --six months to three years. The central government also retains the right to withdraw its recognition at any time. For more details about commodity exchanges and commodity forward and futures trading, see World Bank 1996, "Managing Price Risks in India's Liberalized Agriculture: Can Futures Markets Help?," Report No. 15453-IN. 6 See Annex 5 for more detailed discussion on the operations of the cotton associations. Annex 2 Page 9 of 13 A2.34 Cotton exports, till recently, were dominated by government and cooperative agencies. The export quotas are announced yearly, and has ranged from 8,000 mt to 303,600 mt. The quota volume fluctuates significantly from year to year, depending on the local supply and demand situation (Table A2.3). Its share of domestic Table A2.3: Cotton Production And Export Quotas production has been as low as 1 percent, rising to 15 Yew 'a percent in some years. In May 1995, the CCI was Om0 m,flL -- appointed the sole canalizing agency for the export of 84-85 1446.2 50.2 3% staple cotton. The policy was subsequently modified 85/86 1483.6 235.1 16% to permit the State government agencies and State 86/87 1173.9 90.4 8% cooperative marketing federations to export staple 87/88 1084.9 7.5 1% cotton. It is only in rare instances, when CCI and the 88/89 1486.5 36.7 2% state agencies and cooperatives fail to fulfill their 89/90 1940.4 251.8 13% quotas fully that the government allows private traders 90/91 1673.2 228.7 14% to meet the deficit. Only small quotas for non- 91/92 1651.4 23.0 1% spinnable short and non-staple varieties are allowed 92/93 1969.1 303.6 15% for all exporters, including the private sector. It was 93/94 1820.7 96.9 5% only during the 1995-96 season that cotton exports 94/95 2354.5 31.6 1% were opened to private trade. Export allocations are 95/96* 270.3 now awarded through auction. * September to August Source: Office of the Textile Commissioner. A2.35 The GOI-imposed minimum export price (MEP) was lifted in 1994-95. Prior to 1994-95 the GOI imposed an MEP, which was fixed by the Textile Commissioner for each cotton variety, taking into consideration cotton prices in international markets. All export contracts need to be registered. The export certificates covering the quantities and qualities to be exported by each agency are allocated by the Textile Commissioner. In 1995-96 season, the GOI finally eliminated the MEP. Cotton Import Policy A2.36 Prior to 1991, cotton lint imports were canalized through the CCI. In 1970, CCI was established and became the sole canalizing agency for importing cotton. Imports were still subject to quota restrictions and tariff barriers as existed earlier. Domestic allocations of imported cotton were assigned by the Textile Commissioner to mills and trading of allocations was banned. A2.37 The Advance Licensing Scheme 1986-87 permitted textile mills to import directly. In order to ensure exports of yarn and fabrics, the GOI modified its policy of canalizing all cotton imports through CCI and permitted textile mills to import cotton directly against their export commitments of yarn, cloth and made-ups under Advance Licensing. Input-output norms were fixed for different cotton products. A2.3 8 Cotton imports were finally freed under zero tariffs in 1994. In 1994, in the face of production shortfalls and rising domestic prices, cotton lint imports were placed on OGL, that is freely importable, with a zero import duty.7 G. Pursell and A. Sharna, 1995, "Indian Trade Policies Since the 1991/92 Reforms," mimeo, World bank. Anmex 2 Page 1 0of 13 F. Appendix Tables Appendix Table A2.1: Minimum Support Price of Seed Cotton of Fair Average Quality, 1990-91 to 1994-95, Rs per quintal Basic Staple Micronaire length in Value millimeters 1990-91 1991-92 1992-93 1993-94 1994-95 (2.5% Span Variety length) Assam Comilla 7.0-8.0 - 600 690 780 870 Bengal Deshi 6.8-7.2 525 600 690 780 870 Wagad 20 5.5-6.0 550 625 720 810 900 V-797 22 4.2-4.8 585 660 760 855 950 Jayadhar 22-23 5.0-5.6 585 660 760 855 950 J-34/Bikaneri 23 3.64.4 605 680 785 885 985 Narrna G.Cot 12 23.5 4.2-5.0 605 680 785 885 985 AK/Y-I 24 4.8-5.2 615 690 795 895 995 F-414/M- 24 -25 3.84.2 620 695 800 900 1000 777/Agatti Digvijay 'A' (Guj) 25 4.04.5 630 700 790 885 975 AHH468 25 -26 3.84.2 675 760 875 985 1090 1007 27 3.7-4.0 695 780 900 1000 1110 LRA-5166 27 4.04.5 695 780 900 1010 1120 JHKY-I/MECH 30 3.84.2 750 840 950 1050 1200 H4 30 3.6-4.2 750 840 950 1050 1200 Shankar-6(Sau.) 28 -29 3.7-4.2 750 840 955 1055 1205 Shankar-6(Guj.) 30 3.74.2 765 860 975 1080 1230 MCU-5(South) 33 3.0-3.5 770 865 980 1085 1235 DCH-32(M.P.) 33 -34.5 3.0-3.3 775 870 985 1090 1245 DCH-32(Mah.) 34.5 3.0-3.2 775 870 985 1090 1245 DCH(South) 39 3.0-3.5 820 920 1040 1150 1300 Suvin 40 3.2-3.6 1140 1280 1450 1600 1830 Source Office of the Textile Commissioner, Govemment of India. Appendix Table A2.2: GOI Expenditures under the Intensive Cotton Development Program, Rs million Year Expenditures Budgeted Actual 1988/89 13.6 13.6 1989/90 37 20.5 1990/91 40.9 40.8 1991/92 41.5 39.6 1992/93 107.2 107.3 1993/94 152 93 1994i95 152.5 118.1 1995/96 152.6 123.1 1996/97 153.4 na Source: GOI, Budget Documents, various years Annex 2 Page 11 of 13 Appendix Table A2.3: Cotton Yields in ICDP and Other Districts in Haryana and Maharashtra Haryana Maharashtra Year ICDP Other ICDP Other districts districts districts districts kg/ha kg/ha kg/ha kg/ha 80-81 364 270 73 88 81-82 372 284 78 104 82-83 379 290 101 107 83-84 251 193 38 63 84-85 379 255 94 92 85-86 389 275 122 124 86-87 426 298 51 61 87-88 305 175 104 96 88-89 353 250 66 110 89-90 466 285 130 154 90-91 430 280 95 136 91-92 474 344 59 84 92-93 473 349 na na Source: Haryana and Maharashtra agricultural departments. Appendix Table A2.4: Ginning Fees in Selected States, 1990 rupees State 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 Punjab: All centres 95.59 90.77 88.03 81.98 79.39 87.00 95.83 91.87 86.26 Madhya Pradesh: Khandwa 136.59 134.18 124.71 116.14 119.96 121.55 127.05 116.79 114.06 Khargone 138.95 136.41 126.79 118.08 116.21 121.55 127.05 117.47 110.30 Gujarat: Mehsana 108.33 110.24| 113.07 111.89 121.09 121.38 119.47 115.44 123.58 Surendranagar 100.09 106.45 98.94 na na 118.58 114.30 120.05 125.71 Maharashtra: Selected centres 140.12 139.87 134.03 140.73 142.23 146.45 146.48 156.18 155.42 Andhra Pradesh: Adilabad 129.53 123.00 114.32 106.47 131.21 130.90 134.52 114.08 144.14 Note: Charges were deflated using the WPI. Source: East India Cotton Association. Annex 2 Page 12 of 13 Appendix Table A2.5: Pressing Fees in Selected States, 1990 rupees State 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93, 93-94 Punjab: All centres 60.95 57.88 56.24 52.38 55.13 60.00 65.94 59.12 55.30 Madhya Pradesh: Khandwa 84.50 82.87 78.25 75.15 81.59 90.00 87.92 81.48 81.10 Khargone 85.89 84.19 79.47 75.15 77.18 89.00 74.73 82.28 80.36 Gujarat: Mehsana 69.27 68.40 66.02 63.77 69.46 70.00 70.34 83.88 79.63 Surendranagar 67.88 66.43 61.74 na na 68.00 68.58 81.48 81.10 Maharashtra Selected centres 80.69 70.77 70.46 69.57 79.39 83.05 85.28 83.88 82.58 Andhra Pradesh: I Adilabad 80.35 76.30 73.36 68.32 88.21 82.00 83.53 72.70 81.84 Note: Charges were deflated using the WPI. Source: East India Cotton Association. Annex 2 Page 13 of 13 Appendix Table A2.6: Selective Credit Controls against Cotton Lint and Seed Cotton, 1975 to 1994 Date of RBI Parties to whom Minimum Interest Rate Directive Applicable Ceiling Level of Credit Minimum Margin per annum April 6, 1985 Other than Cotton Upto the peak level in any of the a. 60% of stocks 17.5% Mills years, 1980-81, 1981-82 & 1982-83 b. 45% against warehouse receipts __ _ Mills Subject to margin only a. 25% of stocks 17.5% October 25, Other than Cotton Same as earlier a. 45% of stocks 1985 Mills b. 30% against 17.5% warehouse receipts Mills Exempted from selective credit controls April 6, 1986 All parties exempted from selective credit controls August 14, 1987 Other than Cotton Upto peak level in any of the years, a. 45% of stocks Mills 1982-83, 1983-84 & 1984-85 b. 30% against 16.5% - reduced to 16% _______ _____ ____________ _ I . warehouse receipts on October 8, 1988. April 16, 1990 All parties exempted from selective credit controls 17% raised to 18.5% on July 3, 1991 October 8, 1991 Other than Cotton Upto 85% of peak level in any of the a. 45% of stocks Mills years, 1987-88, 1988-89 & 1989-90 b. 30% against 20% warehouse receipts February 10, Other than Cotton Upto 85% of peak level in any of the a. 60% of stocks 1992 Mills years, 1988-89, 1989-90 & 1990-91 b. 45% against 19% warehouse receipts April 22, 1992 Other than Cotton Same as earlier a. 75% of stocks Mills b. 60% against 19% - reduced to 18% on warehouse receipts October 9, 1992 December 10, Other than Cotton Same as earlier a. 60% of stocks 1992 Mills b. 45% against 18% warehouse receipts January 19, Other than Cotton Same as earlier a. 45% of stocks 1993 Mills b. 30% against 18% warehouse receipts October 11, All parties exempted from selective credit controls 1993 February 18, Other than Cotton upto of peak level in any of the year, a. 45% of stocks 1994 Mills 1989-90, 1990-91 & 1991-92 b. 30% against 17% warehouse receipts May 16, 1994 Other than Cotton 85% of peak level in any of the years, a. 60% of stocks Mills 1990-91, 1991-92 & 1992-93 b. 45% against 15% warehouse receipts October 18, Other than Cotton Upto of peak level in any of the years a. 45% of stocks 1994 Mills 1990-91 & 1991-92 & 1992-93 b. 30% against Deregulated warehouse receipts December 26, Other than Cotton Same as earlier a. 60% of stocks 1994 Mills b. 45% against Deregulated warehouse receipts Source: Reserve Bank of India, Bombay. Structure and Performance ,A,nnex tof the Textile Industry A. Introduction A3.1 The structure of the textile industry is both complex and uniquely Indian. Unlike the textile industries in other countries, the Indian textile industry is characterized by: (i) the co- existence of a broad spectrum of production techniques--from hand-operated to sophisticated automated technology; (ii) a dualistic manufacturing structure dominated by a fast expanding decentralized or "unorganized" small-scale manufacturing segment and a declining vertically integrated, large-scale "composite" mill segment; (iii) a predominance of cotton as a primary raw material; (iv) the existence of a large public sector (20 percent of domestic fabric production), composed mainly of nationalized and sick mills taken over by the government; and (v) a predominantly small-scale apparel sector. During the last decade, the industry displayed rapid growth in output and exports. But existing structural weaknesses and the current regulatory environment will increasingly hamper the industry's ability to sustain this performance. A3.2 This Annex reviews the status and performance of the spinning, weaving and apparel industries and identifies the structural and policy related constraints that could hinder the capacity of these industries to take advantage of emerging market opportunities. This Annex begins with a description of the structure of the Indian textile industry. A discussion of domestic and export demand trends follows. The remaining sections review the performance of the cotton spinning, weaving and apparel subsectors and the man-made fiber industry. B. Structure of the Textile Industry A3.3 The textile industry is composed of spinning, weaving, fabric processing and garment- making units (Figure A3. 1). Enterprises in the textile industry are also classified as belonging to the "organized" and "unorganized" sectors. The organized sector consists of composite mills and spinning mills. The composite mills are vertically integrated mills covering the full array of textile processing operations--spinning, weaving, dyeing, printing and finishing. They account for about 7 percent of total cloth output in the sector. In 1994, there were 266 composite mills, of which 149 were in the private sector, 35 in the public sector and 2 in the cooperative sector. A3.4 The spinning sector consists of two major segments: the composite mills and independent mills. These independent spinning mills, mostly in the private sector, were the major driving force in the growth of the spinning industry. By 1994/95, production capacity in the spinning industry reached 30 million spindles. While the production capacity of composite mills steadily declined beginning in the early 1980s, independent mill capacity more than doubled from 8.9 million spindles in 1980/81 to 19.4 million in 1994/95. A3.5 The "unorganized sector" consists of powerloom, handloom, knitting and yarn and fabric processing units. They are called such because of the highly decentralized nature of their operations. Production is generally coordinated following two systems: the master-weaver and the loom-owner systems. In the master-weaver system, the fabric supplier, upon accepting an order from a garment manufacturer, purchases yarn and arranges for their dyeing and weaving. The resulting "gray cloth" is then sent to a processing house for dyeing and printing. All these Annex 3 Page 2 of 34 "subcontractors" are paid on a fixed per meter conversion basis. In the loom-owner system, the loom owner coordinates all the processing activities. The unorganized sector currently-produces the bulk of fabrics (90 percent) in India. Within the unorganized sector, powerlooms account for about 60 percent of total output. A3.6 The apparel industry consists mainly of small scale units. The apparel industry is composed of about 77,000 enterprises. It is dominated (80 percent) by cottage and small industry operations with 11-20 sewing machines. Medium-sized (21-49 machines) and large firms (greater than 50 machines) occupy only about 14 percent and 6 percent of the market respectively. The apparel industry employs about 1.8 million people, mostly paid on a piece rate basis. Figure A3.1: Structure of the Cotton Textile Industry. |Export Domestic Market i M-anufacturers: Garment & Other TexCie Products Processing Hue Processing House (Yarn dyed ypwrdie!deing, (manual, dyeing, cloth) printingfinf shing) printing,finishing) ( gray cloh (gray cloth)| Compcosite NMis . A (Spinning, weaving, Powerloom Knitting H dloom printing, finishing) z l ~~~~~~Machiine Dycing|HadDyig.. . ' Ji _ . _ ll~~~nits p (cone yarn) (hank yarn) |Export | |SinngMil | (cotton ntE - cotnseJ |GnigMills| "Unorganized Sector" C. Export Performance Spectacular Growth in Textile Exports A3.7 Textile exports boomed during the last decade. Textile export revenues during the last five years in real terms grew (12 percent) about 25 percent faster than total merchandise exports (Table A3.1). By 1995-96, it amounted to $8.8 billion (1994-95 dollars) or close to one-third of total merchandise exports. Fabric and garment exports make up the major share (about 70 percent) of these revenues. Cotton textile exports alone (fabrics, yarn, garments and made-ups) totaled $6 billion dollars. Moreover, cotton export revenues in real terms grew at a slightly faster rate (13 percent) than aggregate textile exports. By volume, textile exports displayed dramatic growth. Cotton apparel exports (pcs) increased annually by 16 percent between 1986 and 1993, while cotton fabric exports (sqm) grew by 11 percent between 1986-95. Cotton yam exports (mt) expanded at an even higher rate of 27 percent per year between 1986-95 (Figure A3.2). Annex 3 Page 3 of 34 Approximately 70 percent of cotton yarn is Table A3.1. Exports of Textile Products, exported by spinning mills, the rest is exported $ million (199.4/95dollars) by yarn traders, who receive a 1 to 2 percent commission. A3.8 The rapid growth in exports of cotton moton P roducts _ __45.0_ _17 _ yarn, garments and made-ups explain the Millmade 384.0 413 3 spectacular growth in exports. Yarn export Powerloom 201.3 474 17 value in real termns increased by 24 percent Handloom 76.4 73 1 annually in real terms, garment exports by 11 Knitted 83.3 88 3 Made-ups 464.2 953 17 percent and made-ups by 17 percent. Made- Millmade 159.0 139 -4 ups include 24 identifiable trade descriptions, Powerloom 128.0 461 33 such as bed sheets, blankets, carpets, curtains, Handloom 177.2 352 16 Yarn 318.4 958 24 dhurries, bags, table cloths, and towels. Fabric Thr a 3.1 1.0 -19 exports by powerlooms compensated to a Garments 1878.9 2952 11 large extent the decline in mill export values. Woven 1300.8 1909 10 Knitted 578.0 1043 13 A3.9 Increasing cotton textile product Subtotal Cotton 3409.6 5915 13 consumption in developed countries, the main Rayon/Synthetics 396.1 890 17 market for India's exports, contributed to excl. garments 57.5 125 Woolen excl. 57.5 125 17 rapid export growth. Demand expansion in garments developed countries for cotton textile products Non-cotton garments 1023.1 1269 7 has been fueled by the shift in consumer All Merchandise 5157.6 3441 12 preferences for cotton textiles and the greater Source: Textile data are from ICMF, 1996, Handbook of Statistics on availability of superior quality raw cotton at Cotton Textile Industby, New Delhi, Index of Unit Value of relatively lower prices. Manufactured Exports (MUV) deflators from World Bank. A3.10 ButIndia remains a minor player in the world textile market. India's share of exports in the global cotton textile trade is only 10 percent. It is lower if blended and non-cotton fabrics are Figure A3.2: Indian cotton apparel, fabric and yam exports included. For exports of textiles of all types, India by volume. accounts for 3.2 percent of exports of developing 1800 countries. 1600 - 1400 r Export Patterns Shaped by MFA 1200 A3.11 The Multi-fiber Agreement (MFA) E 1000o X - primarily shaped India's textile and garment c 800 I export patterns. (See Annex 7 for more detailed o 600 discussion of MFA). Prior to the 1991, the bulk of @ ------- ----------- Indian cotton textile exports (yarn, fabric, and 200 -- ------------ -- made-ups) went to the EC (41.5 percent), USA 200 (16.5 percent), and the Eastern Bloc countries led 0 by the former USSR (12.5 percent). But by 1993, 86 87 88 89 90 91 92 93 94 95 India fulfilled 100 percent of its quota allocation in ..*.Apparel,pcs {}Yarn, mt the EU for gray and bleached textiles and cotton - Fabrics, sqm yarn. In the US market, the quota utilization was 93 Source: ICCI and R. Jaikumar, 'Beyond the MFA Agreement: Strategies for Indian Apparel Industry," and percent in 1993. Binding export ceilings in ICMF, 1996, Handbook of Statistics on Cotton Textile traditional quota markets (EU and USA) led to Industy. increased regional diversification of Indian textile exports. Consequently, by 1994, the share in Indian exports to Western Europe declined to 34 percent, that to the USA to 10 percent. Annex 3 Page 4 of 34 A3.12 Recent export growth came Table A3.2: Direction of Export Trade in Cotton Textiles: 1987-94, mainly from non-quota countries. Rs million. New markets included: (i) non-quota _ countries (e.g. Australia, New Australia,NewZealandand 171 325 2,157 countris (e.g.Australa, New Canada Zealand, and Canada) where market Japan, South Korea and Taiwan 494 731 5,547 opportunities were seized within the Entrepot centers 1,579 4,156 18,716 MFA structure; (ii) Asian economies Sub-total [i] 2,244 5,212 26,420 (e.g. Japan, South Korea, and West Europe and USA [ii] 5,587 11,034 28,873 (e.g. Japan, South Korea, and ~~~Sub-total of [ii + [ii] 7,831 16,246 55,293 Malaysia) which until recently were Grand total of all destinations 9,447 19,000 65,407 untapped potential markets; and (iii) Source: Indian Cotton Mills Federation, Handbook of Statistics on Cotton entrepot centers where intermediate Textile Industry, 1995. textile products were processed for re-export (e.g. Dubai, Singapore, Hong Kong, Bangladesh, Mauritius, Sri Lanka, Taiwan, and the Canary Islands). Exports to entrepot centers accounted for about a third of the increase in the value of exports between 1990 and 1994 (Table A3.2). D. Domestic Demand for Textile Products Large Growth Potential A3.13 Domestic demand growth potential is large. Despite the export boom, the Indian textile industry is primarily domestic oriented. Domestic consumption of yarn and fabrics still accounts for about 89 and 90 percent of total domestic output, respectively. The tTable A3.3:Per Capita Household Consumption of Textiles at Different Income Levels, 1992. large domestic market and rapidly . ;J, rising domestic incomes therefore open important market opportunities <6,000 6.7 2.1 10.0 for the textile industry. The high 6,000 to 19,999 7.5 1.7 3.9 13.1 income elasticity of demand for textile 20,000 to 39,999 7.6 2.6 5.2 15.5 products in India is confirmed by the 40,000 to 59,999 9.7 3.5 7.2 20.4 Market Research Wing of the Textile 1> 60,000 110.3 4.3 8.1 22.7 Source: ICMF, Handbook of Statistics of Cotton Textile Industry 1995 from Committee which report that per Market Research Wing, Textile Committee capita consumption of textiles increased significantly across income groups Figure A3.3: Per capita availability of cloth, sq.m per capita. (Table A3.3). Households earning Rs 40,000 or 30 more in 1992 consumed more than double (20.4 25 in/household) those of households earning less 20 than Rs 6,000 (10 m/household). A shift E 15 towards higher value textiles and most likely ' 10 better quality textiles is also indicated by the rising price per unit paid as incomes increase (Appendix Figure A3.2). Sustained GDP growth and the associated rising incomes will . . . be thus key factors which will boost textile demand in the domestic market. IN Cotton D Non-Cotton 1 Blended & lWxed A3.14 The potential expansion of domestic gource: Office of Textile Commissioner, 1994, Compendium o] textile consumption is large as current Textile Statistics. consumption levels are very low. In 1993, per capita consumption was estimated at 2.8 kgs per year. This is lower than the developing country average of 3.8 kgs and about one-sixth of average developed country per capita consumption levels (Appendix Table A3.1). This is partly attributable to the fact that per capita cloth availability (historical per capita consumption levels Amnex 3 Page 5 of 34 are not available) in India stagnated in the late 80s (Figure A3.3). The average annual growth rate was 1.5 percent. Consumption stagnated despite the fact that (i) population was increasing at an annual rate of 2 percent (Appendix Table A3.2), (ii) real per capita incomes (as measured by per capita GDP) increased by 4.4 percent per year, and (iii) prices of textiles relative to other commodities were declining (Appendix Figure A3.1). A3.15 Consumption growth in the 1990s improved considerably. Per capita cloth availability expanded twice as fast (3 percent per year), reaching 26.2 sq. m per capita in 1993/94. A3.16 Several factors may explain the apparent stagnation of per capita textile consumption despite rising incomes and declining relative prices of textiles with respect to other commodities. First, it is partly due to the increased substitution of cotton textile products with man-made and blended products in textile consumption. Although cotton cloth dominates consumption, the share of man-made products is increasing steadily. Per capita non-cotton cloth consumption nearly tripled in the last 15 years to 6.7 sq.m. Its share of total cloth consumption doubled from 13 percent in 1980/81 to 26 percent in 1993/94 (Figure A3.2). This occurred despite the fact that cotton textiles are generally cheaper than their non-cotton and blended counterparts (Appendix Figure A3.2). A3.17 The higher durability associated with man-made fibers, the increased "comfort factor" associated with mixed blends, and improvements in relative costs due to reductions in domestic duties on man-made fibers facilitated these substitutions. The complementary roles between cotton and synthetic fibers arise from the different physical characteristics of each fiber. For example, a cotton-polyester blended fabric combines the greater strength and superior resistance of a pure cotton fabric, yet exhibits greater moisture absorption capacity than an all polyester product. These complementary characteristics Figure A3.4: Consumption shares of different textile fibers in are generally highly desired in "comfort selected countries, 1992. clothing." 100%. 90% A3 .18 Second, low priced cotton textiles may 80% I behave as "inferior goods" with consumers 70%.I switching to higher priced man-made and 60%l blended textiles at a disproportionate rate as 40% 1 income rises. These factors clearly indicate a 30%a growing consumer preference for man-made 20%. and blended products. This trend towards 10%- increased man-made fiber consumption 0% parallels the trend in other countries (Figure x = ' A3.4). India's share of cotton consumption is about double those of Indonesia, Malaysia and Japan and about triple those of Korea and g Cotton u Wool M Cellulosics o Synthetics Germany. It slightly exceeds China's, but it still larger than its neighbor, Pakistan. Source: ICAC, Washington, in ICMF, 1995, Handbook of Statistics in Cotton Textile Industry. Man-Made Fiber Products: Not the Rich Man's Product 3.19 One of the major factors shaping government man-made fiber policy has been the belief that man-made fiber products are the preferred choice of rich consumers. This translated into discriminatory tax policies against man-made fibers and the institution of other barriers to entry to protect the cotton sector (e.g. licensing). A study by the Market Research Wing of the Textile Committee, however, found the contrary. Of the households purchasing synthetic textile products, 70 to 80 percent earned less than Rs 40,000 year (Appendix Table A3.3). Over 50 Annex 3 Page 6 of 34 percent of households purchasing polyester, nylon and viscose products earn less than Rs 20,000 per year. E. Performance of the Spinning Industry A3.20 This section reviews the production and export performance of the Indian spinning industry. It examines the manufacturing efficiency of the industry and its implications on the competitiveness of the spinning sector in a more liberalized cotton trade environment. Production Trends A3.21 Subject to less government Figure A3.5: Spindle capacity in India. interventions, large investments both 35-; in terms of new capacity and better 30- technology flows into the spinning 25- industry, reflecting its considerable perceived economic potential (Figure A3.5). Almost all capacity additions 15i and modernization in the spinning 0 sector since 1985 occurred in the | private sector. This was facilitated by a soft loan scheme in 1976 and the - e Lo U 1010CDCD1-I' C C 0Q0C0 1 Textile Modernization Fund Scheme 0 0 0 ('4 o 1 introduced in 1986. Although the fund was intended for the whole Source: ICMF, Handbook of Statistics in the Cotton Textile Industry, various issues. industry, the spinning mills were the major beneficiary of the fund, in particular the more efficiently managed plants. In addition, about 50 export-oriented spinning mills were set up in response to various GOI export incentives (viz. tax free import of raw materials and low import duties on foreign machinery). As of early December 1995, there were 65 projects including new spinning mills and expansion of existing mills in the cotton textile sector. About 48 mills with an aggregate spinning capacity of 1.1 million spindles were Table A3.4: Spinning Capacity and Production Share by Ownership, 1995 scheduled for completion ~~j during the second half of % l 1995/96. These Public 5.92 50 11.0 0.5 valued at ~~~~(19.9%) investments are valuedat Cooperative 3.25 55 7.4 0.1 about Rs. 22.6 billion (10.7%) ($674 million). Private 21.24 93 82.6 99.4 r________ (70.3%) Seventeen percent of Source: ICMF, 1995. these mills are export- oriented units. Another 12 mills with an aggregate capacity of 24.4 million spindles are scheduled for completion in 1996/97. Current capacity stands at close to 32 million spindles, 70 percent of which is in the private sector. In 1995, the private sector accounted for 83 percent of production and almost all yarn exports (Table A3.4). A3.22 About one-third of total spinning capacity is used purely for the production of cotton yarn. The remainder is used for the production of blended yarns (polyester:viscose, polyester: cotton, polyester:wool; viscose:cotton) and small quantities of 100 percent viscose spun yarns. Out of the nearly 30 million spindles in the country, over 70 percent (21 million spindles) are suitable for the production of 100 percent cotton and cotton:viscose yarns only. The balance can use man-made fibers like polyester staple fiber, acrylic stable fiber, and viscose staple fiber in Annex 3 Page 7 of 34 combination with other fibers or exclusively. The current Figure A3.6:Production of cotton yarn by count stock of technologies therefore impose some (short-term) in India, rmllion kg. bounds on cross-fiber flexibility. 1600 A3.23 Low and medium count cotton yarns accountedfor 1400 _ most of output growth Lower count cotton yarns (<30s 1200 count) production, which make up over 50 percent of total .71000 cotton yarn output, grew by 5.4 percent per year between ' 800 - 1985/86 and 1996/97. Production growth was boosted by * 600 the rapidly increasing demand from the expanding 400 powerloom sector. Medium count cotton yarn output (3 Is 200 to 40s) grew rapidly, at 4.2 percent per year (Figure A3.6). 200 Production of higher count cotton yarns (>40s) experienced 0 . _ more modest growth during this period (1.5%); their share 0 (0 s (0 0 a of total production remains small (11 percent in 1996/97) Ois- 30s . 31s - 40s and is declining. 41s-60s >60s Export Trends Source: ICMF, Handbook of Statistics of the A3.24 Cotton Yarn exports boomed in the 1990s. Export Cotton Textile Indust,y, various issues; OTC volumes of ls to 40s count cotton yarns increased by an average of 24 percent per year between 1990 and 1995 (Table A3.5). Exports of over 60s counts similarly grew by 22 percent per year, while 41s-60s grew by 19 percent per year. Cotton yarn exports remain dominated by lower counts. In 1995/96, about 65 percent of yarn exports were of below 30s counts. Table A3.5: Cotton Yarn exports by count, A3.25 Although yarn export quotas remains in now place, the increasing number of export exemptions 1990 67.9 7.3 7.9 81.2 makes the quotas less binding. During 1997, of the 19921 105.5 7.6 9.0 122.1 515 thousand mt of yarn exports, only 38 percent 1993 145.1 8.8 12.0 165.9 were made under the quota; exports by EOUs and 1994 188.4 11.4 16.2 216.0 EPCGs accounted for 50 percent, exports against 1995 203.4 20.9 19.6 243.9 advance license 2 percent, exports against cotton Ave. AnnualGrowth Rate imports under OGL 1 percent, and other exports (to 90-95 1 24% 1 19% 22% 24% non-quota countries, processed yarn) 11 percent. Source: ICMF, 1996, Handbook of Statistics of the Cotton Textile Industiy; East India Cotton Association, 1997/98, Manufacturing Efficiency Cotton news & Statistics, No. 48 A3.26 The productivity of the spinning sector can be measured in various ways. In this section, the index of modernization and the output per spindle are used to measure machine efficiency, while the operative hours per output of yarn and the number of operatives per spindle are used to measure labor productivity. A downtime Table A3.6: Evolution of the Index of Modernization, analysis, which partly reflects managerial or 1983-95. X-efficiency, is also performed. Finally, the ~Jtld impact of the hank yarn obligation on the productivity of the spinning mills is evaluated. 31.3.86-31.3.91 13% 2.6%p.a. 31.3.91-31.3.95 28% 7.1%p.a. A3.27 Modernization investments in the Note: In 1981-82 some composite mills were re-classified as spinning industry largely occurred in the spinning mills. Hence, to avoid misleading numbers, the independent sector. The installation of new starting point for the early 1980s has been taken to be 31 March 1983. equipment between time periods as a share of Source: S. Chaudury, "Cotton Sector Study," Background the existing stock of equipment is one working paper, World Bank. Annex 3 Page 8 of 34 indicator of the rate of industrial modernization. This index of modernization is measured as the net increase in installed capacity (spindleage) to starting year capacity (Table A3.6). In India, the evolution of the index over time displayed cyclical Table A3.7: Output to Installed Spindleage pattern. Modernization accelerated in the early 1980s, (Kgs yrn/s pindle/year) decelerated in the second half of the decade, and again cpu. y ; ;W7 99t accelerated in the 1990s. The investments in new Italy 52 52 61 96 equipment, however, occurred mainly in independent Germany 59 56 63 96 USA 83 78 62 120 spinning mills, rather than composite mills. Spindleage Korea 103 101 119 133 actually declined in the composite mill sector (See Taiwan 101 71 70 91 Annex 1, Figure A1.2). China n.a. 104 186 115 Pakistan 96 87 105 196 A3.28 Aggregate Productivity. Productivity per Thailand n.a. 78 83 83 spindle in India remains low by international India 58 54 55 World 59 73 75 78 standards. One indicator of aggregate productivity is - Note: figures are for yam containing no less than the ratio of annual total output to the stock of installed 85% cotton. spindleage. Although it is an approximate measure, Source: S. Chaudury, "Cotton Sector Study," because it does not take into account time utilization Background working paper, World Bank. and spreads around the average, it illustrate trends across countries. This ratio for India stagnated at around 55 during the last 30 years (Table A3.7). In contrast, there were significant technological improvements in other countries, leading to higher productivity and better quality. One reason for the low productivity value in India is the large number of idle spindles in the country. A3.29 A South Indian Textile Research Association (SITRA) survey found that 60 percent of spinning mills operate below established productivity norms. In a survey of 269 (largely private) spinning mills in 1994, SITRA ranked mills with respect to their composite productivity index (CPI). The CPI is the ratio of the standard operative hours to produce an established weight of yarn (HOK) to the actual HOK. SITRA's standard HOK was set at 88.8 gm per spindle per 8 hours, adjusted to 40s counts to facilitate cross-firm comparison. It should be noted that SITRA norms are still below international norms. For example, staffing requirements for a well- managed mill in Western Europe or Table A3.8: Output per Spindle By Category of Spinning Mills, 1994 North America would be 30 percent (Grams per 8-hr adjusted to 40s) less than that of the SITRA standard for labor productivity. The mills are ?! "4 > ;;- . Top 17 7% 86.7 96.0 79 classified into 6 categories (Top, Very Very High 22 8% 86.1 1025 74.7 High, High, Average, Low, Very Low High 68 25% 81.2 97.7 64.5 productivity). Using these standards, Average 76 28% 76.2 89.9 49.8 60 percent of the firms surveyed had VeryLow 46 17% 69.2 83.6 60.3 average productivity levels at or below Note: P.S. denotes production per spindle SITRA norms (Table A3.8). Source: SITRA, 1994 A3.30 Productivity is low despite high capacity utilization levels. The majority of mills surveyed by SITRA showed high capacity utilization rates. Of the total mills surveyed, 95.5 percent of the mills worked 7 days a week and the remainder 6 days a week. The average spindle utilization was 91 percent of the calendar hours available, after making necessary allowances for routine maintenance and cleaning. These capacity utilization levels compare well to developed country norms (90 percent) and international norms (85 percent) for spinning. Of the 9 percent loss in machine utilization time - unscheduled down-time - the causes varied considerably across categories of mills (Table A3.9). The main reasons identified by SITRA for under-utilization include: (i) additional time (over SITRA norms) for cleaning, routine maintenance, repairs and Annex 3 Page 9 of 34 count change, (ii) Table A3.9: Del Analysis of Spinning Mills power shortages, . " ~ '~"~~;~~.~5- ~ J ~ . and (iii) poor . x management as Top 1.(3%)09(2% 1.5 (39%) 3.8 Very High 2.1 (37%) 1.1 (19%) 2.5 (44%/6) 5.7 reflected by High 2.2 (37%) 1.1 (19%) 2.6 (44%) 5.9 absenteeism, back- Average 2.2 (25%)0 j 1.4 (16°%) 5.2 (59%) 8.8 stuff shortage and Low 2.7 (25%) 1.4 (13%) 6.8 (62%) 10.9 other causes. As VeryLow 3.2 (18%) 1.7 (10%) 12.8 (72%) 17.7 Note: Total time loss is the percent downward deviation from what is considered by SITRA as the expected, external standard utilization level of available calendar hours. This time loss is distributed across the three factors were types of causes and the percentage is given in parentheses. proportionately Source: SITRA, 1994 more important for better mills than the others, while bad management was proportionately of lower importance. This reflects another aspect of productivity in the form of better X-efficiency. A3.31 Labor Productivity. The SITRA survey indicates low labor productivity in 60 percent of firms. Labor productivity can be measured in terms of HOK (i.e., number of operative hours required to manufacture 100 kgs of yarn) or OHSAM (i.e., number of operatives per 1,000 spindles). In the SITRA survey, both are expressed in terms of standard 40s count yarn. The SITRA standard is 24.0 for HOK and 2.7 for OHSAM. Of the mills surveyed, 85 percent of mills (High, Average, Low and Very Low groups) fell below this standard (Table A3.10). Mills surveyed, on average, required 33.2 operative hours to produce 100 kgs of yarn, and engage 3.45 operatives per 1,000 spindles adjusted to 40s. The range of variation across mills in the two labor productivity measures is extremely high - over 50 percent from the average in both directions. This is partly attributable to labor laws. However, there is a discernible improvement across all categories between 1992 and 1994, especially for the last 20 mills which have a much greater catching-up to undertake. Table A3. IO: Productivity Range by Class of Spinning Mills Adjusted to 40s cou 1994. T~-,. OHSAM 2.0 2.5 2.9 3.5 4.0 4.9 OKToplO 11 - 2 a Avg. A 1992 18-21 21-23 24 - 25 34 - 35 _36 S SS-87 1994 14-19 19 - 21 21 - 23 32-33 33.3 49 - 83 OHSAM 1992 1.9 -2.2 2.3 - 2.4 2.6 - 2.6 3.4 - 3.7 3.76 5.6-7.7 1994 1.6- 2.0 2.0 -2.3 2.4 -2.6 3.2 - 3.5 3.45 5.0- 7.4 Notes: (i) Classified in the categorization used by SITRA; (ii) Grouped from the detailed listing provided in the SITRA survey, the ranges of mills taken from the ranking used by SITRA, which is that of composite productivity index. Source: S. Chaudury. 1996, "Cotton Sector Study," Background paper, World Bank. A3.32 Factor Productivity and Mill Age. Factor productivity is higher in newer mills. A comparison of labor productivity and productivity per spindle by age of mill shows a negative relationship between productivity levels and mill age (Table A3.11). The improvements in productivity of new mills can be largely explained by the relaxation of restrictions on capital goods imports, which allowed new firms to access modern machinery more easily. For mills established after 1990, the disparity in production per spindle is smaller compared to older mills. At the same time it is also clear that appropriate modernization enabled some even very old mills to maintain high levels of productivity. Annex 3 Page 10 of 34 Impact of the Hank Yarn Obligation A3.33 The Hank Yarn Obligation (HYO) requires spinning mills to process 50 percent of their deliveries in hank form or Table A3. 11 Mill Age and Productivity transer h atfon o 0 :9 ;X; ts:4rk,.*Tr:,,, ~~~~~transfer the obligation to : -L - - ;cis-- - -;. o. other firms. Hank yarn is a Before 1940 44 57.3 34.1 19.9 57.5 76.3 94.5 type of yarn used by 1940 to 1949 14 46.8 32.0 19.9 66.4 76.8 86. handlooms. It is reeled 1950 to 1959 53 53.3 34.1 19.2 49.8 76.1 90.2 1960 to 1969 61 54.7 35.2 21.1 55.8 76.5 102.5 manually and therefore 1970 to 1979 25 70.3 34.8 16.2 57.9 79.7 91.9 requires a much larger 1980 to 1989 48 83.2 29.8 15.5 54.5 80.6 97.7 labor complement 1990 & after 12 40.2 24.5 14.4 73.7 85.0 96.0 Notes: (i) HOK is labor productivity - number of operative hours required to produce 1,000 compared to auto-widig kgs of standard 40s count. The higher the value, the lower the productivity. (ii) P.S. is the set-ups. The hank yarn production per spindle in gms per 8 hours working - or kgs for an 8,000 operating hour year. obligation is calculated on The higher the figure the better the utilization from a given set of spindles. Source: S. Chaudury, 1996, "Cotton Sector Study," Background paper, World Bank. the basis of a percentage of the net qualifying Table A3.12: Cotton Yarn Production and Hank Yarn amount after exemption of exports, own Deliveries, million kg consumption (for composite mills) and hosiery -.;.y* - Xg>-......... -.60,000 10,000 20,000 40,000 40,000 60,000 Polyester * Sarees 25 31 28 84 10 6 * Shirtings 29 36 24 89 7 4 * Dress material 29 32 23 84 10 6 Nylon sarees 33 33 23 90 6 4 Viscose Dress material 22 30 22 74 15 11 Source: C. S. Gokhale and V. Katti, 1995, 'Globalizing Indian Textiles, Threats and Opportunities," Bombay: Tecoya Disseminators, Appendix Table 18. Appendix Table A3.4: Cotton Textile Exports, $ million, contant 1994/95 dollars. Item 90/91 91/92 92/93 93/94 94/95 95/96 Ave. Annual Growth Rate Fabrics 745 758 819 797 1,022 1,047 8% Millmade 384 317 321 318 382 413 3% Powerloom 201 287 308 311 447 474 17% Handloom 76 83 109 94 101 73 1% Knitted 83 70 81 74 91 88 3% Made-ups 464 504 655 764 926 953 17% Millmade 159 139 130 110 116 139 -4% Powerloom 128 143 222 314 434 461 33% Handloom 177 223 304 340 376 352 16% Knitted 2 2 2 3 Yarn 318 436 464 535 831 958 24% Sewing Thread 3 3 2 2 1 1 -19% Garments 1,879 1,945 2,589 2,868 3,114 2,952 11% Woven 1,301 1,330 1,651 1,855 2,151 1,909 10% Knitted 578 615 938 1,013 963 1,043 13% Total Cotton 3,410 3,646 4,532 4,968 5,896 5,915 13% Rayon/Synthetics excl garments 396 491 574 617 791 890 17% Fabrics 216 273 323 393 477 492 19% Spun Yarn 40 79 93 85 141 208 33% Filament Yarn 7 27 36 15 38 54 34% Made-ups 78 77 81 88 92 100 6% Fibers 34 16 22 13 20 15 -10% OtherItems 21 20 19 22 22 21 1% Woolen excl garments 57 74 125 150 115 125 17% Non-cotton garments 1,023 844 945 950 1,325 1,269 7% Silk (fabrics,madeups, garments) 271 300 288 264 299 241 -2% Total Textiles 5,158 5,354 6,464 6,948 8,425 8,441 12% Total Merchandize Exports 20,651 19,880 19,920 23,768 26,857 30,942 9% Source: Textile data are from ICMF, 1995, Handbook of Statistics on Cotton Textile Industry, New Delhi, Indes of Unit Value of Manufactured Exports (MUV) deflators from World Bank. Appendix TableA3.5: Percentage share of apparel exports by fiber type. Cotton (%/6) Man-Made (%) Wool (%) Year Volume Value Volume Value Volume Value 86 82.2 75.4 15.1 19.4 2.8 5.2 87 85.8 81.1 12.1 15.3 2.0 3.6 88 84.6 78.6 13.4 17.9 2.0 3.6 89 78.7 68.3 19.2 28.1 2.0 3.6 90 76.5 64.0 22.4 34.3 1.1 1.7 91 78.1 68.4 21.1 31.1 0.3 0.5 92 80.5 70.9 19.0 27.7 0.5 1.2 93 83.4 . 74.8 14.9 21.8 1.7 3.4 Source: Apparel Promotion Council Annex 3 Page 27 of 34 Appendix Table A3.6: Yarn production by count, million kgs. Year Is - los lls.-20s 21s-30s 31s-40s 41s-60s 61s - 80s >80s Total 85/86 134 317 271 324 121 61 25 1253 86/87 139 329 281 337 126 64 26 1302 87/88 141 334 285 342 128 65 26 1321 88/89 144 277 248 419 133 55 34 1310 89/90 160 313 247 436 122 61 33 1372 90/91 197 354 277 456 129 64 33 1510 91/92 211 342 255 440 113 54 35 1450 92/93 234 336 276 471 132 43 31 1523 93/94 266 367 299 471 137 45 37 1622 94/95 263 363 300 438 134 45 43 1586 Growth _I_ 85/86 - 9.18 1.69 0.84 4.24 0.75 -4.31 5.17 2.93 194/95 __ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ Source: ICMF, Various Issues, Handbook of Statistics on Cotton Textile Industry. Appendix Table A3.7: ITMF Computation of Manufacturing Costs for 30s Combed Ring-Spun Cotton Yarn at March 1995 prices, US cents / Kg Item USA Japan Italy Korea Brazil Thailand India Waste 42 47 46 47 38 47 38 Labor 53 100 89 18 21 10 5 Power 18 59 23 20 21 23 33 Stores 14 17 13 14 13 17 12 Financial charges 154 117 129 126 182 123 137 Mfg. expense 281 340 300 225 275 220 225 Raw materials 215 232 227 230 201 231 204 Total cost 496 572 527 455 476 451 429 Total cost without raw 281 340 300 225 275 220 225 materials Notes:(i) Waste cost is the wastage at various stages net of recoveries on sale of salvaged saleable wastes. (ii) Fin Charges = Depreciation plus Interest on fixed and working capital. (iii) Mfg. Expenses denote manufacturing expenses and is the sum of waste, labor, power, stores and financial charges. (iv) The cost of raw cotton is based on March 1995Cotlook prices, except for Brazil and India where local prices are used as quoted by the domestic market. (v) Raw material cost is the cost of 1 kg of cotton. The cost of wastage is already taken separately as an expense, so here only I kg cost of cotton is represented against an output of I kg of yarn (vi) Costs are computed as if manufactured from new equipment. Source: ITMF Appendix Table A3.8: Production of man-made fibers in India, 000 mt. Year Viscose Acrylic Polyester Poly Propylene Total 80-81 82.66 10.06 22.43 115.15 81-82 84.2 13.61 29.71 127.52 82-83 49.29 16.16 25.77 91.22 83-84 82.78 16.59 26.87 126.24 84-85 102.02 20.88 38.98 161.88 85-86 89.98 21.82 42.83 154.63 86-87 96.3 23.06 65.63 184.99 87-88 119.45 22.09 79.43 220.97 88-89 125.7 26.34 111.58 263.62 89-90 147.65 30.48 127.12 305.25 90-91 160.17 42.48 134.94 9.93 337.59 91-92 158.08 46.99 135.94 1.05 341.01 92-93 162.45 55.03 161.82 1.07 379.3 93-94 183.34 68.35 200.09 1.56 451.78 Ave. annual growth rate 85/86- 93/94 l 9% 17% 18% 13% Source: OTC, Compendium of Textile Statistics 1994. Annex 3 Page 28 of 34 Appendix Table A3.9: Installed capacity for man-made staple fiber production, 000 mt Year Viscose Acetate Polyester Acrylic Poly Propylene Total 80-81 97.4 0.8 33.58 16 147.78 81-82 97.67 0.8 35.83 16 150.3 82-83 97.67 0.8 36.93 16 151.4 83-84 97.67 0.8 44.37 16 158.84 84-85 107.67 0.8 44.99 20 173.46 85-86 107.67 0.8 43.27 22.5 174.24 86-87 112.7 0.8 128.51 22.5 264.51 87-88 112.7 0.8 146.22 22.5 282.22 88-89 116.7 0.8 204.06 36 357.56 89-90 157.85 0.8 230.06 48 436.71 90-91 176.05 0.8 230.06 63 43 469.91 91-92 186.58 0.8 230.06 63 43 480.44 92-93 186.58 0.8 230.06 87 53 504.44 93-94 222.95 0.8 230.06 87 58 540.81 Ave annual growth rate 85/86-93/94 1 10% 0% 17% 1 22% 14% Source: OTC, Compendium of Textile Statistics 1994. Appendix Table A3.10.: Production of man-made filament yarn, 000 mt. Year Viscose Polyester Nylon Poly- Total Propylene 80-81 41.36 10.69 20.73 72.78 81-82 41.05 15.51 23.14 79.7 82-83 33.19 24.69 25.51 83.39 83-84 35.42 47.99 30.64 114.05 84-85 32.95 55.22 34.54 122.71 85-86 42.04 67.43 39.38 148.85 86-87 44.64 81.43 36.72 162.79 87-88 45.93 111.46 34.32 191.71 88-89 44.36 142.97 35.76 223.09 89-90 49.24 156.48 38.73 244.45 90-91 50.94 185.29 39.82 276.05 91-92 52.69 207.6 30.86 291.15 92-93 47.95 246.54 32.47 326.96 93-94 53.02 288.94 37.33 379.29 Ave. annual growth rate 85/86- 93/94 1 3% 19% -1% 12% Source: OTC, Compendium of Textile Statistics 1994. Annex 3 Page 29 of 34 Appendix Table A3.11: Installed capacity for man-made filament yarn production, 000 mt. Year Viscose Acetate Polyester Nylon Poly Propylene Total Mar-81 41.84 2.4 9.05 21.74 75.03 Mar-82 43.39 2.4 15.29 25.29 86.37 Mar-83 43.39 2.4 27.25 29.99 103.03 Mar-84 43.39 2.4 33.71 30.5 110 Mar-85 43.39 2.4 41.87 34.02 121.68 Mar-86 44.29 2.4 43.34 41.92 131.95 Mar-87 45.19 2.4 70.84 43.52 161.95 Mar-88 54.99 2.4 74.18 47.52 179.09 Mar-89 54.99 2.4 107.05 68.2 232.64 Mar-90 53.21 2.4 158.67 94 9.25 317.53 Mar-91 60.41 2.4 177.36 97.6 9.25 347.02 Mar-92 68.87 2.4 234.81 108.18 11.35 425.61 Mar-93 66.54 2.4 236.81 109.38 39.35 454.48 Mar-94 68.04 2.4 265.31 115.7 39.35 490.8 Ave annual growth rate (%) 85/86- l 93/94 6% l 0% 26% l 16% 19% Source: OTC, Compendium of Textile Statistics 1994. Appendix Table A3.12:Public Textile Enterprises: Net Profit and Losses, constant 1993/94 million dollars and number of employees. Enterprise 89/90 90/91 91/92 92/93 93/94 Total Networth Employeesb Cawnpore Textiles -2.43 -2.69 -2.61 -1.37 -1.89 -10.98 0.47 1446 Elgin Mills -20.54 -21.02 -21.53 -17.26 -11.55 -91.90 0.40 4838 NTC L _ _ L E_=_ NTC, Ltd 0.22 1.18 1.32 0.08 -0.04 2.76 146.98 334 AP, Karnataka & Kerala -5.14 -5.03 -7.18 -19.24 -13.21 -49.79 24.07 11437 Delhi, Punjab & Rajasthan -3.79 -0.90 -4.59 -12.09 -9.60 -30.98 11.23 7618 Gujarat -18.70 -13.04 -15.73 -31.99 -26.27 -105.73 11.38 7791 MP -12.33 -11.75 -10.95 -25.25 -21.51 -81.80 13.19 12206 Maharashtra-North -15.26 -8.53 -10.15 -25.87 -25.19 -85.01 24.10 10770 Maharashtra-South -22.12 -9.99 -14.51 -25.77 -20.90 -93.29 22.86 13339 TN and Pondicherry 9.86 9.13 0.14 -7.24 -1.72 10.16 25.42 11442 UP -19.77 -19.53 -17.78 -32.73 -28.26 -118.07 19.02 1861 WB, Assam, Bihar& -19.91 -22.16 -19.02 -39.09 -32.67 -132.84 17.48 12003 Orissa Total |-129.90 -104.33 -122.61 -237.82 -192.82 -787.46 95,085 Note: a - equity and reserves, b -non-casual employees. Figures deflated using Index of Unit Value of Manufactured Exports (MUV) from the G-5 industrial countries (World Bank). Source: GOL, Public Enterprise Survey, various issues Annex 3 Page 30 of 34 Table A3.13: Effective Protection of Yarn Exporting Firms, 1994/95. EOUs Domestic Price Export Price Yarn Cotton Y Cotton EPC 20s count 90.0 60.7 90 69.2 1.41 30s count 102 69.1 102 76.3 1.28 40s count 110 70.9 79.3 1.27 Modern Domestic Price Export Price Yarn Cotton Yam Cotton EPC 20s count 73.8 54.6 89.0 62.3 1.02 30s count 94.9 62.2 91.8 68.7 1.41 40s count 101.2 63.7 99.0 71.4 1.36 Dated Domestic Price Export Price Yarn Cotton Yam Cotton EPC 20s count 73.8 54.6 89.0 62.3 1.02 30s count 94.9 62.2 91.8 68.7 1.41 40s count 101.2 63.7 99.0 89.4 1.36 Note: EPC= (Domestic price of yam-value of cotton at domestic prices)/ (border price of yarn-value of cotton at border prices) Source: Price data from H. Bhattacharya, 1996,"India Cotton Sector Study," Background working paper. Appendix Table A3.14a: Production cost of 20s count yarn, Rs/kg Present Cotton Price At Intemational Cotton Price E.O.U. MODERN DATED E.O.U. MODERN DATED Production Cost/k Raw Materials 60.7 54.63 54.63 69.2 62.28 62.28 Salaries/Wages 6.1 7.6 9.78 6 7.5 9.6 Salaries/Fuel/Energy 2.84 3.22 3.56 2.84 3.22 3.56 Stores/Spares 0.38 0.42 0.59 0.38 0.42 0.59 Packing Materials 0.9 0.95 1 0.9 0.95 1 Admin. Cost 1.14 1.38 1.77 1.14 1.38 1.77 Financial Charges 1.4 1.78 2.06 1.4 1.78 2.06 Depreciation 1.8 1.78 1.62 1.8 1.78 1.62 Total Cost/kg 75.26 71.76 75.01 83.66 79.31 82.48 Yam Price/kg 90 73.8 73.8 90 81.0 81.0 Profit/Loss per kg 14.74 2.04 -1.21 6.34 1.69 -1.48 Source: H. Bhattacharya, 1996, "India Cotton Sector Study," Background working paper. Appendix Table A3.14b: Production cost of 30s count yam, Rs/kg Present Cotton Price At International Cotton Price E.O.U. MODERN DATED E.O.U. MODERN DATED Production Cost/kg Raw Materials 69.1 62.19 62.19 76.3 68.67 68.67 Salaries/Wages 7.62 8.4 12.15 7.23 7.78 11.54 Power/Fuel/Energy 4.62 4.88 5.85 4.55 4.81 5.7 Stores/Spares 0.76 0.7 0.97 0.76 0.7 0.97 Packing Materials 0.9 0.95 1 0.9 0.95 1 Admin. Cost 2.02 2.42 2.9 2.02 2.42 2.9 Financial Charges 3.05 3.25 3.39 3.05 3.25 3.39 Depreciation 2.8 2.74 2.66 2.8 2.74 2.66 Total Cost/kg 90.87 85.53 91.11 97.61 91.32 96.83 Yam Price/kg 102 94.9 94.9 102 91.8 91.8 Profit/Loss per kg 11.13 9.37 3.79 4.39 0.48 -5.03 Source: H. Bhattacharya, 1996, "India Cotton Sector Study," Background working paper. Annex 3 Page 31 of 34 Appendix Table A3.14c: Production cost of 40s count yarn, Rslkg Present Cotton Price At Intemational Cotton Price E.O.U. MODERN DATED E.O.U. MODERN DATED Production Cost/kg Raw Materials 70.86 63.68 63.68 79.3 71.37 71.37 Salaries/Wages 9.6 12.2 14.47 9.2 11.81 15.19 Power/Fuel/Energy 7.75 8.22 8.55 7.75 8.22 8.55 Stores/Spares 1.14 1.22 1.41 1.14 1.22 1.41 Packing Materials 0.9 0.95 1 0.9 0.95 1 Admin. Cost 2.82 3.4 4.24 2.67 3.23 4 Depreciation 4.15 4.04 3.89 4.15 4.04 3.89 Total Cost/kg 101.45 98.15 102.19 107.52 105.28 110.36 Yarn Price/kg 110 101.2 101,2 110 99.0 99.0 Profit/Loss per kg 8.55 3.05 0.99 0.66 -6.28 -11.36 Source: H. Bhattacharya, 1996, "India Cotton Sector Study," Background working paper. Appendix Table 3.15: Yam exports by count, 000 mt Year Yarn counts, 000 mt Is - 40s 41s - 60s > 60s Total 1990 67.9 7.3 7.9 83.2 1991 103.3 9.0 7.7 120.0 1992 105.5 7.6 9.0 122.1 1993 145.1 8.8 12.0 165.9 1994 188.4 11.4 16.2 216.0 1995 203.4 20.9 19.6 243.9 Growth Rate _ 90-95 24.28% 19.04% 20.58% 25.01% Source: ICMF, 1995, Handbook of Statistics on Cotton Textile Industry, New Delhi. Appendix Table A3.16: Total Apparel export firms and their percentage distribution by region. Region Manufacturers Merchants Total North 34% 49% 48% West 18% 26% 25% South 46% 23% 25% East 1% 2% 2% Total Firms All India 2019 23739 25758 Source: S.R. Khanna, 1994, Textile Outlook Intemational. Annex 3 Page 32 of 34 Appendis Table A3.17: Prices of Yarn and Filaments, Rs/kg. Cotton Yarn Blended Yarn Filament Yarn Period@ Hank Cone Hosiery Cone Poly/Viscose Poly/Cotton VFY NFY POY Texturised (Wt.Avg.) (Wt.Avg.) (Avg.) (Avg.) (Avg.) (Wt.Avg.) (Wt.Avg.) (Avg.) Yarn (Avg.) Mar-89 41.31 50.85 47.13 76.29 103.00 108.04 171.76 139.67 164.22 Jun-89 44.49 52.50 49.25 90.94 108.56 108.30 179.76 146.92 155.09 Sep-89 46.22 55.56 52.81 100.77 123.67 141.12 198.66 167.34 165.82 Dec-89 45.72 53.92 52.19 99.67 117.56 132.93 211.67 161.59 165.44 Mar-90 47.32 56.41 52.52 105.21 125.44 127.69 253.99 155.56 173.86 Jun-90 46.79 54.27 53.13 104.23 119.11 120.81 192.67 153.83 168.05 Sep-90 45.45 56.63 53.21 106.48 121.11 117.58 183.34 171.92 175.03 Dec-90 46.64 55.40 53.60 104.87 116.11 114.01 192.60 181.00 175.91 Mar-91 48.01 55.31 53.18 104.87 119.13 110.99 207.67 172.00 179.64 Jun-91 50.11 58.49 55.91 104.76 114.87 109.72 204.82 167.59 175.72 Sep-91 57.19 62.81 64.81 112.64 116.87 144.73 197.72 184.96 187.22 Dec-91 64.57 67.27 72.29 114.84 106.25 165.90 213.14 163.88 177.35 Mar-92 60.49 70.93 71.65 120.06 122.00 164.76 214.79 168.63 187.27 Jun-92 63.03 68.93 71.04 122.48 118.40 163.21 196.91 166.21 169.78 Sep-92 63.69 70.73 71.38 130.47 126.50 169.36 196.96 159.90 167.13 Dec-92 63.05 71.61 69.79 131.72 126.13 187.78 194.94 160.03 163.65 Mar-93 63.31 69.32 69.63 127.24 133.13 194.08 183.28 144.35 172.48 Jun-93 63.99 73.26 69.63 129.94 127.63 196.40 182.34 156.43 165.45 Sep-93 64.60 74.99 69.69 137.16 136.75 196.31 182.69 155.95 164.90 Dec-93 65.53 75.71 69.75 137.66 136.13 197.48 189.86 150.55 168.37 Mar-94 83.88 88.28 78.79 139.19 135.84 198.16 194.94 146.31 166.53 Jun-94 82.60 89.38 92.33 135.94 150.12 190.40 178.79 150.33 162.15 Sep 94 83.31 92.24 87.79 154.52 153.12 188.15 171.18 153.83 162.49 Dec 94 91.71 102.84 91.85 149.87 153.12 190.50 159.80 153.20 161.27 Mar 95 97.36 114.79 109.41 149.30 150.42 201.12 164.04 153.20 163.28 Sep95 96.11 111.14 102.54 152.22 156.58 197.41 180.61 158.20 164.76 Jun 95 92.92 104.76 99.55 151.97 147.71 202.92 191.87 156.24 160.54 Dec 95 95.59- 108.17 102.30 153.07 159.58 207.35 221.52 158.44 160.77 Mar 96 95.78 108.08 105.84 150.68 147.17 218.74 253.74 150.31 156.57 Note:@ i) Figures upto March 93 relate to month end prices. ii) Figures from June 93 onwards relate to average of weekly prices for respective month. Source: OTC 1994, Compendiuim of Textile Statistics. Table A3.18. Productivity Levels of Apparel Firms (number of pieces per day) Ladies blouses Gents shirts Ladies dresses Ladies Skirts Trousers S. Korea 14.59 17.39 8.77 17.54 15.55 Taiwan 18.89 18.18 12.44 16.63 16.12 China 10.93 13.96 7.83 13.00 6.71 Thailand 16.97 19.75 12.19 20.47 13.08 India 10.18 9.12 6.25 9.62 6.84 Hong Kong 20.56 20.87 20.17 19.25 19.25 Source: S.R. Khanna (1993) The Challenge of Global Competition in the 1990s: an Agenda for Enhancing the Competitive Position of the Indian Textiles and Clothing Industry. ICRIER Mimeo Annex 3 Page 33 of 34 Appendix Figure A3.1: Wholesale price indexes for textiles, manufactured products and primary articles, 1981/82=100. 300 250- 200- K 150 - 100 50 C'J 'T CD 00 0 ('I 'I 0 CD LO r- CD a) co XD CD co co a) O> + Textiles -a- Manuf. Prod - Primary Articles Source: Reserve Bank of India Bulletin, various issues. Appendix Figure A3.2: Value of consumption of cloth per person for a period of 30 days for each fractile group 95-100 1 90-95 80-90 70-80 60-70 50-60 40-50 30-40 20-30 10-20 5-10 0-5 0 25 50 75 100 125 15 RsImo/capita * Rural u Urban Source: NSSO, National Sample Survey, 50th round. Annex 3 Page 34 of 34 Appendix Figure A3 .3: Average price paid for cloth by different income groups, Rs per meter. 90 8o 70 60 60-0 50 E40 30 20 '10 0 < n6,000 6,000- 10,000- 20,000- 40,000- > 60,00 9,999 19,999 39,999 59,999 + Cotton n- Non-cotton A. Mixed Source: Market Research Wing, Textiles Committee, in ICMf, 1995, Handbook of Statistics on Cotton Textile Industry. Appendix Figure A3.4: Production of blended yarns and 100% non cotton and man-made filament yarns. 800 600 0 CO40 K II0 ot f X C~ CO t CO j X Cb m C Blended yarn o 100% non-cotton spun yarn a Filanent Yarn Source: OTC, Compendium of Textile Statistics 1994, ICMF; ICMF, Handbook of Statistics on the Cotton Textile Industry 1995. Cotton Production A.nnex sPerformance and Prospects A. Introduction A4.1 Cotton is one of the most important cash crops in India. It is grown predominantly in rainfed areas (70 percent) in small farms--two-thirds of cotton farmers cultivate less than two hectares and only one percent cultivate more than 20 hectares. Cotton output posted one of the highest growth rates in the last 15 years (3.9 percent), supported by the near doubling of yields in the context of declining cultivated area. Despite the considerable yield growth that has already been achieved, Indian yield levels (246kg/ha in 1995-96) are extremely low by international standards. Domestically available varieties have potential yields of up to 5 times existing levels, but are unattained in field conditions due to economic, technical and institutional constraints. This chapter reviews the recent performance of the cotton production sector and examines the factors that could hinder the sustained productivity growth of the sector. B. Cotton Production Performance Production Trends A4.2 India is one the major cotton producers in the world. With 25 percent of total world cultivated area -9.1 million ha in 1996-97, it accounts for the largest area in the world. In contrast, cotton lint production in 1996-97, amounted to only 2.4 million mt or about 15.5 percent of world output (Text Box 4.1). As will be discussed below, India's weak production performance primarily results from very low average Figure A4. 1: Cotton Lint Production in India, yields, 246 kg/ha in 1996-97. Cotton production in India 1974-75 to 1996-97, million mt is concentrated in three major regions. They are: (i) the northern zone (Haryana, Punjab and Rajasthan); (ii) the 300 8 central zone (Maharashtra, Gujarat and Madhya 250 7 Pradesh); and (iii) the southern zone (Karnataka, Tamil 200 6 Nadu and Andhra Pradesh). For the purpose of this s 15OU 4E study, Maharashtra will be examined separately from the 4 rest of the states in the central zone because of its large 1oo share of production and unique marketing system (See so12 E Annex 5). 0 __ 0o ' 00t- < c A4.3 Among major crops, cotton output had the a o w , >, 1 second highest production growth rate during the last > > x X X X o 0 15 years. At the All-India level, cotton production A Yield + Prodn W Area expanded by about 500,000 mt from 1.27 million mt in Source: Agricultural Statistics at a Glance, MOA Triennium Ending (TE) 1981-82 to 2.02 million mt in TE 1995-96 (Figure A4.1) or at an annual rate of about 3.9 percent per year, second only to oilseeds (5.5 percent) (Appendix Table A4.1). Cotton grew faster than all cereals (2.6 percent per year) and sugarcane (2.8 percent per year) and about the same rate as plantation crops (3.8 percent) and fruits and vegetables. (3.9 percent). Annex 4 Page 2 of 21 A4.4 The North and Southern Figure A4.2: Cotton Lint Production bN Region. 1974-75 to 1995-96 regions displayed the fastest growth.-__ - Over the last two decades (TE 74-75 to 900 - - TE 95-96), output in the Northeni 800 1 Region (Haryana, Punjab and 600 Rajasthan) grew at an average rate of a 500 4.3 percent per year. more than 400 3 300 doubling from 311,000 to 777,000 mt 200 (Figure A4.2). Output in the Southern 100 Region (Andhra Pradesh, Kamataka, 0 X and Tamil Nadu) also grew by 4 <' ¢ 9 P %"' ' p percent per year, increasing from t e \e '\°r v %: e %Oe 9 154,000 to 514,000 mt during the same n North n&u.. Central . Maharashtra South period. Maharashtra posted an annual Note: North: Punjab, Haryana, Rajasthan; Central: Gujarat, Madhya Pradesh: growth rate of 3.4 percent. In contrast, South: Kamataka, Andhra Pradesh, Tamil Nadu cotton production in the Central Source: Agricultural Statistics at a Glance, various issues. Region (Gujarat and Madhya Pradesh) declined by 1.4 percent per annum during the same period. The Northern and Southern Regions account for 39 and 27 percent respectively of the total All-India output growth. Text Box A4. 1: Cotton Production Data in India: Which Estimate is Best? Estimating the al-India coton production level is a complex exercise, especially becas it deermines the ultimate export quota lcveis. Ih Cotton Advisory Board (CAB) is charged with estimang the supply, demand and opening stock of cotton oach year based on estimats provided by its membes. Its members include Textile Commissioner, joint secretaries ofthe Ministy of Texies and of the Ministry of Agriture, stat govem¢en officials in charge of agiculture, cotton growas, In"dia Cotton Milds Federation ICMF, N4ional TetiIl Copratin (NTC), Cotton Corporation of India (CCI).Mahasshm State Cotton Cowes Co-operaive Maketing Federaton (MSC F, All India Co-operve Cotton Gmwers Fedeation (AZOON), East India Cotton Asswciatxon (EICA), All India Powedoom Fedaios a otder associons. GOI sets the xport quota for cottn on the basis of producton estimates by the CAB. Production estimates submitted by each of the CAB members seem to be influenced by their respective vested intests. For example, repreentaie ftom the industry eport lower levels of domestic cotton rqiements in order to limit the size of the expon quota; cotton trader rpesentaives (eg., EICA) report much higher produdion levels in an eff to obtain larger cotton export quotas (See table below). Consequtly, th fal CAB estimates end to differ from the estimates made by the Dietrat of Cotton Development of the Minisy ofAgricultur In some case the difference has amounted to as much as 20 percent The same complication aiss in estimating the opeing stock of cotton. Offidal and Trade Ests of Cotton Production (thousend of bales of 170 kg each) Year DCD (A) CA (1B) EtCA (C) (B) - (A) (C) - (A) 198546 87.27 107.00 1 15.50 18.4% 24% 1988-9 87.44 106.00 106.00 17.5% 18% 1989-90 114.22 135.75 135.75 15.9% 16% 1990-91 97.59 117.00 117.00 16.6% 170 1991-92 97.14 119.00 120.78 18.4% 20% 1992-93 115.83 135.00 140.00 14.2% 17% Noste: DCD- Ditetore of Cotton Developmentb CAB, Cotton AdvisoLry Board Source: Office of the Textile Commissioner, 01; MOA. EICA. Improving Yield Performance A4.5 Yield improvements drove output increases. At the all India level, yields improved by about 3 percent per annum on average between 1974/75 and 1995/96, contributing for all of the additional cotton lint production. These national results however mask significant regional differences. In the North, where yields are highest, yields rose steadily (2 percent per annum between 1974/75 and 1995/96), with some evidence of yield stagnation in the most recent years (Figure A4.3). Despite this achievement, higher yields contributed relatively little to output Annex 4 Page 3 of 21 growth in the North, while area Figure A4.3: Cotton Lint Yields by Region and All-India, kg/ha expansion still played the major 500 role (Table A4.1). By contrast, yields in the southern region and 400 Maharashtra improved the fastest s 300 over the same period (yearly 200 average of 3.96 and 2.65 percent, respectively), explaining most of 100 the observed production 0 increases. The central region North Central Maharashtra South India recorded the most disappointing *TE74-75 MTE77-78 E]TE80-81 MTE82-83 *TE84-85 performance. The modest ETE87-88 gTE90-91 TE93-94 *TE95-96 improvement in yields achieved between 1974/75 and 1995/96 Note: Note: North: Punjab, Haryana, Rajasthan; Central: Gujarat, Madhya Pradesh; South: Karnataka, Andbra Pradesh, Tamil Nadu. (0.8 per cent per annum) was Source: Agricultural Statistics at a Glance, MOA, various issues; CMIE, 1997, insufficient to stem the rapid Agriculture. decline in area planted to cotton (1.4 Table A4. 1: Source of Cotton Output Growth by Region and All per cent per annum). India A4.6 The adoption of improved ~~¶ varieties and increased irrigated North 5 7 ill production contributed to yield and jCentral -228 171 -46 -12 area increases. Overall, 108 new Maharashtra 13 79 11 106 improved varieties were released in [I0ndia 7 101 6 86 India between 1967 and 1992. Over Source:Computed. 26 new improved varieties were introduced in the North,l 38 in the Central Region and Maharashtra, and 44 varieties the Southern Region. In Maharashtra and the South and Central Regions, the new varieties consisted mostly of higher yielding hybrids of longer staple cottons. In the Northern Region, the new varieties consisted mainly of higher yielding and shorter duration varieties which facilitated the introduction of cotton in the cropping pattern. These improved varieties exhibit potential seed cotton yields of 600 to 4,500 kg/ha. A4.7 The increased adoption of these new varieties contributed to rising yields. For example, by 1992-93, hybrids covered 36 percent (2.7 million ha) of the total cotton cultivated area in India, compared to 10 percent in 1980-81.2 In the Central Region (including Maharashtra), 36 percent of the 4.7 million ha of cotton area is planted to hybrids. As of 1996-97, Maharashtra had the largest area planted to hybrid cotton, covering 140,000 ha or 40 percent of cotton cultivated area in the state (Figure A4.4). The increased cultivation of improved varieties in Maharashtra is actively being promoted by the state government (See Annex 2). In the Southern Region, an even larger share, 61 percent, of the total 1.6 million ha of cotton area is planted with hybrids. Karnataka and Madhya Pradesh exhibited rapid hybrid adoption. As a percentage of l India is the only country in the world that grows all four cultivated cotton species: Gossypium barbadense, G. hirsutum, G. arboreum and G. herbaceum. The G. barbadense specie has long and fine fibers with staple lengths in excess of 32 mm and a Micronaire value below 4.0. The G. hirsutum specie is a "medium staple" with staple length ranging from 25 to 30 mm and Micronaire values from 3.8 to 5.0. The G. arboreum and G. herbaceum species are short and course with staple lengths of less than 25 mm and a Micronaire greater than 6.0. All four species are grown in the Southem region. The G. hirsutum, G. arboreum and G. herbaceum are grown in the Central Region (including Maharashtra). Only the G. hirsutum and G. arboreum species are grown in the Northem Region. 2 A.K. Basu and R.S. Paroda, 1995, Hybrid Cotton in India--A Success Story, Bangkok, Thailand: Asia Pacific Association of Agricultural Research Institutions. Annex 4 Page 4 of 21 total cotton area, hybrids cover, 69 Figure A4.4: Cultivated Area under Hybrid Cotton in Selected States, 1979-93 percent of cotton area in Karnataka, and Percent Share of Hybrid Area in Total Areain 1992-93 48 percent in Gujarat, 42 percent in Madhya Pradesh, 63 percent in Andhra 1200 - Pradesh and 11 percent in Tamil Nadu ooo- in 1992-93. 000 800 A4.8 Irrigated production, ha 600 48% 63% 69% especially in the north, facilitates the 400 am achievement of higher yields. In the 200 - 11% northern region, virtually all cotton is 0- (au ' E AP grown under irrigated conditions arat har nata (Figure A4.5). Irrigation explains the as much higher yields achieved, since improved seed varieties respond more U 1979-80 0 1984-85 C 1990-91 M 1992-93 favorably to irrigated conditions. In the central region, although total cultivated Source: East India Cotton Association, Indian Cotton Annual 1992-93 and A.K. area decreased by about 470,000 ha Basu and R S. Paroda, 1995, Hybrid Cotton in India--A Success Story. between 1980/81 and 1991/92, the share of irrigated over total cotton cultivated area increased from 23 to 34 percent. In the southern region, irrigated cotton only occupies only about 30 percent of cultivated area. Cotton production in Maharashtra remains a rainfed crop, contributing to the low observed yield levels. At the all-India level, only about 30 percent of total cultivated cotton area in India is irrigated. Figure A4.5: Relative Share of Irrigated and Rainfed Cotton Areas in the Different Regions in India, 1981/82 to 1991/92. 2.0 2.5 1.51 2.0) 1.5~~~~~~~~~~. ~1.0 oO' iMm' g X it g I 1. 20.5 . 3.0 2.0 21.0 0t[ t E oI o* itt 00 00 I0' ,% 00 00 .1 L ~ ~ 000 O O a3 M-IrrigX M-Rainfed 0 C-Irrig ff S-Rainfed Note: N - Northern region; C - Central region, M- Maharashtra, S - Southern region Source: Directorate of Cotton Development, MOA, GOI, India Annex 4 Page 5 of 21 A4.9 Despite major strides, Indian yield Figure A4M6: Average Cotton Lint Yields in Selected Countries and levels are very low by international standards. Indian Regions, 1992-94 Of the major cotton producers in the world, 1.2 India had one of the highest rate of yield 1 growth during the period 1981 to 1994, posting 0.8 the third highest growth rate (4.6 percent per year) after Brazil (7.7 percent per year) and - 0.6 Pakistan (4.7 percent per year).3 Despite the 04 ___________ considerable yield increases achieved so far, domestic yield levels remain extremely low 0.2 relative to the rest of the world. The average 0 yield of irrigated cotton in the Northern Region zz_ (India-IN) during 1991-92 to 1993-94 is less L a -g . X D a than or close to half of the other major irrigated m c cotton producers in the world, including China, Note: N -North, C-Center, M-Maharashtra, S-South Egypt, the United States, and Uzbekistan Source: FAO Production Yearbook, various issues (Figures A4.6). Rainfed cotton yields in Figure A4.7: Cotton Lint Production and Area of Selected Maharashtra (India-RM) and the Central (India- Major Cotton Producers in the World, TE 94 RC) and Southern (India-RS) Regions are 8.00 likewise considerably lower than rainfed cotton 7.00 yields in Brazil. As a result, although India has 6.00 the largest area planted to cotton in the world, its E 5.00 output level remains low compared to other . 4.00 major producing countries (Figure A4.7). 5r 3.00 A4.10 The gap between potential yields of u 2.00 improved varieties and observed field-level 1.00 yields is large. The benefits from increased o.oo * cultivation of new higher yielding varieties in Area Prodn India have not been maximized. For example, a * B * China 3 India * Pakistan popular "superior long" hybrid (H-4) has a E USA a Uzbekistan 1 Egypt potential seed cotton yield of 3,500 kg/ha under ideal research station conditions, which Source: FAO Production Yearbook, various issues translates to about 1,700 kg/ha under field conditions. In 1991-92, H-4 yielded 103 kg/ha in Andhra Pradesh, 335 kg/ha in Gujarat, and 113 kg/ha in Maharashtra. Another "superior long" hybrid (DCH 32) has a potential yield of 1700 kg/ha under field conditions, but is yielding 329 kg/ha in Karnataka and 460 kg/ha in Tamil Nadu. In the North, improved American varieties have potential yields of 450-3000 kg/ha, but average yield obtained is about 630 kg/ha in Punjab, 464 kg/ha in Haryana and 320 kg/ha in Rajasthan.4 The major technical factors contributing to the low yields in India are identified below. 3Yield growth rates per year in other major cotton producing countries over the 1981-1994 period were 1.1 percent in China, 1.8 percent in the United States, 2.1 percent in Turkey, -0.25 percent in Uzbekistan, and -0.46 percent in Egypt. 4 East India Cotton Association, Indian Cotton Annual 1992-93. Annex 4 Page 6 of 21 C. Technical Constraints to Future Productivity Growth A4.11 Several major factors constrain cotton yield levels and further productivity improvements. The most important of these are: * poor pest and disease management leading to resistance of major pests to available pesticides; * limited availability of improved seeds especially adapted to rainfed cultivation; * genetic drift due to poor on-farm seed management; and * access to reliable supply of water. Escalating Pest and Disease Problems A4.12 Insect pests are a major limiting factor in cotton production in India. Of the 162 insect species associated with cotton in India, about Table A4.2: Outbreaks of Major Pestsby Year and State 12 are of major importance and responsible for i - >7 l 50 to 60 percent of yield losses. In the Andhra Pradesh 87/88 84/85, 85/86 89/90 northern and central zones, bollworns, aphids Gujarat 86/87 and jassids are of major importance, while in Haryana 78,80, 83 the southern zone, the bollworm and whitefly Maharashtra 84/85, 85/86 pose serious problems. The bollworm causes Karnataka 84/85,85/86 damage through larval feeding on the fruiting Tamil Nadu late 60s, 70s 84/85, 85/86 structures. The whitefly causes damage by Punjab 93/94, 94/95 94 feeding of te cotton plnt reducingSource: S. Jayaraj, 1996, "Pesticides and Pest Management in feeding on the sap ofthe cotton plant reducing Cotton Production in India: Some Policy Issues," Working Paper. plant vigor and through spoilage of the cotton bolls. It is also the principal transmitter of the cotton leaf curl Table 4.3: Crop-wise Consumption of Pesticides in India virus. Major infestations of the cotton bollworm, cotton whitefly ; and the cotton leaf curl virus (of which the whitefly is the primary Cotton 44.5 vector), have resulted in considerable production losses.5 In 1986- Paddy 22.8 87, it was estimated that crop losses due to bollworms in the Sorghum 8.9 Fruits & Vegetables 7.0 Guntur district in Andhra Pradesh amounted to 59 percent of Wheat 6.4 potential output, valued at about $160 million ($126 million for Pigeonpea 2.8 cotton lint and $34 million of the cotton seed oil).6 The most Groundnut 25 Pearl millet1. recent bollworm outbreak occurred in Punjab in 1993/94 and Maize 1994/95, and contributed to the observed leveling off of yields in Sugarcane 0.7 the North during this period. Heavy late season rains in 1993 and Ragi 0.4 milder late season rains in 1994 created favorable conditions for Jute 0. Chickpea 0.2 bollworm infestation. In 1994, cotton producers and traders in Onion 0.2 Punjab were projecting losses of about 7,000 mt of cotton due to Rape & Mustard 0.2 the leaf curl virus. The increasing frequency of pest outbreaks in Others 0.3 cotton producing areas is illustrated by Table A4.2. Source: S. Jayaraj, 199 , "Pesticides A4. 13 Indiscriminate use of pesticides and crop management and Pest Management in Cotton Production in India: Some Policy practices favoring pest build-up are major factors contributing to Issues." 5 Personal communication, Dr. Nigel Armes, ICRISAT. 6 Nalin M. Kishore, 1992, "Pesticide Externalities, Comparative Advantage, and Commodity Trade, Cotton in Andhra Pradesh," World Bank Policy Research Working Paper 928, Washington, D.C.: World Bank. Annex 4 Page 7 of 21 the recent pest outbreaks. The cotton sector is the largest consumer (45 percent) of pesticides of all agricultural crops in India (Table A4.3).7 The GOI encouraged the use of pesticides for agriculture and public health since the 1940s. With the view that pesticides are an important input to agricultural production, GOI and state governments subsidized their distribution and sale. Recently, GOI tried to address the negative externalities associated with pesticide use--pesticide resistance, pest resurgence (when the destruction of its natural enemies results in the pest population increasing to levels greater than before treatment), public health hazards, environmental and food contamination by withdrawing certain direct subsidies to farmers. Central subsidies now apply only to special categories like Scheduled Castes and Scheduled Tribes for insecticides and fungicides. But a 25 to 50 percent subsidy on herbicides continue to be provided to farmers. Some state governments continue to subsidize pesticides. Moreover, poor quality control of pesticides being sold in the market and weak support services to farmers further aggravate pest problems. The UP Pesticide Testing Laboratory at Kanpur tested about 19,090 pesticide samples during 1969-95 and found that 9.4 percent of the samples were substandard. A4. 14 India primarily consumes domestically produced pesticides. The industry presently has installed capacity of 116, 000 mt per annum, of which 60 percent is produced by the organized sector and the remainder by 500-odd units belonging to the small-scale sector. Imports are currently only about 2,000 mt. In some states, the State Department of Agriculture, the Agro- Industries Corporation and Farmer's Cooperative Federation are manufacturing and /or selling pesticides. In UP, about 85 percent of pesticides are distributed and sold through the three outlets above. A4.15 Pesticide subsidies, coupled with improper pesticide application and use led to the development of secondary pests, pesticide (broad spectrum pyrethroids) resistance and pest resurgence problems. The bollworm outbreak in Andhra Pradesh in 1987/88, for example, was traced to synthetic pyrethroid pest resistance. In contrast to other cotton producing countries, Mali for example, where pesticides are applied about 4 to 6 times per season or in Egypt where the recommended application is 3-4 per season, farmers in India spray an average of 10 times per season, and in some parts of Punjab and Haryana as much as 25-35 times per season.8 Consequently, insecticides alone account for nearly 40 percent of the total cost of production. In some areas, the introduction of jassid resistant cotton and the excessive use of pyrethroids, which can destroy indigenous biological control agents, resulted in the outbreak of the secondary pests, the white fly which was more attracted to the new jassid resistant cotton. A4. 16 India is not the first country to encounter pest resurgence and resistance problems due to excessive use of pesticides. International experience attest to the need for urgent action. Excessive use of pesticides have resulted in pesticide resurgence and resistance in several major cotton producing countries, resulting in a drastic drop in production. Over-use of pesticides in Thailand, Australia, Mexico, and more recently Pakistan and China, has resulted in bollworm pesticide resistance. Problems of pest resurgence and the development of secondary pests have also plagued cotton production in Zimbabwe and Sudan. Due to a lack of appropriate advice largely resulting from weak extension systems, grower's initial reaction to poor control due to resistance is to increase dosage, frequency of application, or use of an ad-hoc mixture of two or more insecticides, all of which exacerbate the situation. As cotton crop losses escalate due to 7 The cotton sector accounts for 60 percent of insecticide and 5 percent of fungicide purchases. 8 Fred E.M. Gillham et al., 1995, "Cotton Production Prospects for the Next Decade," World Bank Technical Paper No. 287, Washington, D.C.: World Bank. Annex 4 Page 8 of 21 pest damage, farmers shift to alternative crops. In China, for example, bollworn resistance to pyrethroids resulted in substantial yield declines, from 860 to 660 kg/ha, and a production decline from 5.7 million to 4.5 million mt in the Shandong, Hebei and Henan provinces in 1992-93, even though there was an increase in area from 6.5 million to 6.8 million ha. A4.17 Poor crop husbandry practices aggravate pest problems. Continuous cotton cultivation, high density planting, excessive irrigation and fertilizer application, and the cultivation of alternative host plants contribute to the pest build-up. The whitefly, which was considered a rare pest prior to 1984, became a major pest in 1984/85 due to the excessive use of synthetic pyrethroids, high doses of nitrogenous fertilizer, close plant spacing, indiscriminate use of pesticides on preferred alternate hosts and favorable climatic conditions (high temperatures with scanty rainfall). The resulting loss in Andhra Pradesh ranged from 10-25 percent in 1984/85 and 20 to 45 percent in 1985/86. The bollworm outbreak in the state during the succeeding season was traced to excessive and indiscriminate use of broad spectrum pyrethroids, high doses of fertilizers, staggered sowing, high density planting, cultivation of varieties and hybrids which were highly susceptible, availability of alternate cultivated/wild host plants, extensive mono- cropping, ratooning, and inefficient spraying equipment.9 Integrated Pest Management A4.18 Integrated pest management is a management approach that encourages natural conltrol of pest populations by anticipating pest problems and preventing pests from reaching econonmically damaging levels. All appropriate techniques are used such as enhancing natural enemies, planting pest resistant crops, adapting cultural management, and using pesticides judiciously. 10 In recognition of the escalating pest problems in India, the GOI in the Eighth Five Year Plan (1992-97) launched the Integrated Pest Management Program (IPM) "based on the use of biotic agents for the control of pests and taking up treatment based on need rather than as prophylactic action." (Vol. II, p.4). Text Box 4.2: Cotton Research and Extension in India Cotton Research. The Indian Council of Agricultural Research (ICAR) coordinates research on al I crops in India. Cotton rcsearch is primarily conducted by State Agricultural Universities and [CAR institutes. The Central lnstitute for Cotton Rcsearch (CICR. Nagpur) mainly investigates production issues while the Central Research on Cotton Technology (CIRCOT, Bombay) focuses on quality issues. A few private research organizations also conduct cotton research, the most prominent being ? (SIRCOT). Research at all the centers is coordinated under the All India Cotton Coordinated Improvement Project (AICCIP). Cotton Extension The State Departnent's of Agriculturc is primarily responsible for extension, in gencral. Divisional Ot'ficers (DOs)are supervised by an associate director for extension. District Agricultural Ofticers operate under theDOs and supervise the Subject Matter Specialists $MSs) and Agricultural Extension Officers(AEOs). B3elow the AEO is the Village Level Extension Worker (VLEW), with one VLEW for every 5.000 farmers in irrigated areas and every 800 farmners inrainfed areas. VLEWs are the contact point for farmers and arc trained by-weekly bySMSs, Source: T'. Bell and E.Thigpen, 1996. "The Indian Cotton/rextile Industry, Policy Implications," Background Working Paper, World Bank. A4.19 The GOI's IPMprogram consists of three major components: research, technology and knowledge dissemination, and pest monitoring. Research pertaining to IPM in cotton is being conducted by the Indian Council of Agricultural Research Institutes, State Agricultural A.K. Basu and S.S. Naravanan. 1992, '-All India Coordinated Cotton Improvement Project: Achievements 1967-92. Central Institute for Cotton Research, Nagpur. pp. 176. 10 UJSDA, Agricultural Research Service. 1993. "USDA Programs Related to Integrated Pest Management." I'SDA Program Aid 1506. Beltsville MD. Annex 4 Page 9 of 21 Universities (SAUs) and a few private research organizations (See Text Box 4.2). Public sector research on cotton IPM in particular and cotton in general is coordinated under the All India Cotton Coordinated Improvement Project Research (AICCIP). Research areas include the development of resistant varieties, biological control agents, bio-pesticides, pest monitoring tools, improved crop husbandry and pesticide management practices (economic threshold levels of pesticide applications, fertilizer and water management). A4.20 Technology and knowledge transfer is being implemented using three mechanisms: demonstrations, training, and dissemination of user friendly publications. As part of the program, 19 Central Pest Surveillance Stations, 13 Central Plant Protection Stations, and 11 Biological Stations located in various parts of the country are engaged in demonstrating IPM technology to farmers, educating extension workers, and other state functionaries. The Central Plant Protection Training Institute under the Central Directorate of Plant Protection also provides training on plant protection consistent with the IPM approach. The Food and Agriculture Organization and Asian Development Bank are currently jointly funding IPM training schools for farmers in several states. SAUs in cotton growing areas also produce user friendly publications covering improved agronomic practices, which agricultural extension workers could discuss with farmers. Pest monitoring and forecasting are conducted by the Central Surveillance Stations and Central Plant Protection Stations located in various states in collaboration with the State Departments of Agriculture. A4.21 To complement the government IPM program several cotton IPM initiatives are being implemented by the private sector and international organizations. The Food and Agriculture Organization (FAO) and United Nations Development Program (UNDP) are supporting an Inter- Country IPM Program in India, which covers cotton and rice in the states of Gujarat and Tamil Nadu. Key elements of the FAO-UNDP Farmer Centered Agricultural Resource Management Projects include technology assessment, technology upgrading, demonstration and farmer training. The International Center for Research in the Semi-Arid Tropics (ICRISAT) and the Natural Resources Institute, UK, are also doing IPM research on major cotton pests in India. A4.22 Several factors contribute to the slow rate of adoption by farmers of improved pest management practices, hampering pest control efforts. Although considerable research output has been generated under various research programs under the AICCIP, the continuing incidence of pest outbreaks indicates that adoption of improved technologies and crop husbandry practices remains low. Several factors contribute to this low adoption rate. First, low cotton prices discourage cotton growers from adopting improved but likely more costly pest management techniques. Second, weak extension systems constrain the dissemination of improved pest management technologies, practices and knowledge among farmers. Third, pesticide subsidies increase the incentives for pesticide use, contributing to pesticide resistance and pest resurgence problems. Fourth, poor regulation and enforcement of quality standards in the pesticide industry result in the sale of substandard pesticides in the market. A4.23 Addressing the escalating pest and disease problems in cotton in India will require a multi-pronged effort of GOI, States, NGOs and the private sector to effectively implement IPM. Greater efforts need to be directed at: (i) improving financial incentives to cotton growers for IPM; (ii) improving technology dissemination systems and farmner education programs; (ii) creating a level playing field in pest management (i.e. removal of pesticide subsidies), (iii) reviewing and where necessary, strengthening pesticide legislation and enforcement of pesticide regulations. Annex 4 IPage l(of21 A4.24 Extensive experience in integrated pest management has been accumulated worldwide from which India could draw from. Appendix Table A4.5 lists some of the more environmentally friendly technologies and pest management practices that have been developed for cotton in Asia. The local adaptation of many of these technologies and pest management approaches would be extremely valuable in enabling the cotton sector to effectively deal with the escalating pest problem in India Text Box A4.3 Cotton: Marketing and Distribution of Improved Seeds India has a flourishing seed industry with public and private sector participation. Public sector agcncies generally produce and market certified publicly developed improved varieties and hybrids. They account for about 60 to 70 percent of certified seeds sold in the market. Several public agencies are involved in seed multiplication and distribution, including ICAR research institutes, National and State Seed Corporations, Central State Farms, CCI, and state governments. Several cotton grower's cooperative societies also produce certified seeds of cotton. The private sector is highly concentrated in the production and marketing of hybrid cotton developed in house or by the public sector, In 1993194. the private sector accounted for almost 80 percent of cotton hybrid seed sales. They include private seed companies and trade associations (e.g. South India Mills Association Cotton Development Research Association). Seed production and distribution in the country are governed by the Seeds Act 1966 and the Seeds (Control) Order 1984. There are four categories of seeds are sold in India: certified seeds, truthfully labeled seeds, market seeds and farrner seeds. Certified seeds are seeds which passed the standard certification standards for quality (purity, seed viability, moisture content, foreign matter, etc.) established by the GOI Seed Certification Agency. Truthfully labeled seeds are seeds which comply to above standards but have not been officially certified. Because of the length of tinme involved in obtaining seed certification, private sector firms prefer to market their seeds directly. Market seeds are seeds which do not meet the standards of quality established the Seed Certification Agency. They are seeds aside by gainers and sold through dealers, which sell at half the cost of certified seeds. The use of market seeds is concentrated in the Northern States of Punjab, Haryana, and Rajastha. A small proportion of Ibrmers set aside their own seeds after having their cotton custom ginned. Source: A.K. Basu, 1996, "Current Status of Cotton Seed in India," Background working paper. World Bank, FEGilIham, et al., 1995, "Cotton Production Prospects for the l'ext Decade," World Bank Technical Paper No. 287'A,F. Ferguson & Co., "t)raft Report on Cotton Prospects Study." 1994. New Delhi._ Cotton Seed Supply and Quality A4.25 Improved seeds are poorly adapted to India 's predominantlv rain7ed conditions. contributing to loit' yields. About 55 (19 percent improved varieties and 36 percent hybrids) percent of cotton cultivated area is already planted to high yielding varieties (Text Box 4.3). However, most new varieties developed are more ideally suited to irrigated than rainfed production systems. Consequently, these varieties do not perform to their maximum potential in the more resource constrained rainfed environment where most cotton is growni. This is highlighted by the much larger yield gap observed in the mostly rainfed Central and Southern Regions than in the irrigated Northern Region. A4.26 Poor farmer seed management contributes to lows yield perfbrmance. As mtich as 40 percentl I of Indian cotton growers use market seeds or own-farmer seed retention. These seeds are generally of mixed genetic purity and quality, which in turn reduce the quality of seed cotton output. Market seeds are cotton seeds set aside by ginners and sold through dealers. A4.27 Poor seed management is more a demand than a supply problemii. Seed availability is often cited as a major constraint, but recent official estimates indicate that overall seed availability is adequate. Conservative supply and demand estimates for seeds have been made by About 64 percent of cultivated area is planted * ith non-hybrid seeds that can be replanted \\ithout much loss of genetic purity. Only about 1/3 of farmers planting non-hybrid seeds use pure (certified or truthful>, labeled seeds) high qualit) seeds: the other 2/3 (about 40 percent) use market seeds or ovrn farmer seed retention Annex 4 Page II of 21 the task force on cotton Table A4.4: Total Requirement and Supply of Quality Seeds, 1994/95, 100 Percent production of the Department of Replacement Rate Agriculture and Cooperation of Ministry of Agriculture (Table lt A4.4). Estimates indicate that Improved Varieties the supply of hybrid seeds Northern zone 1.7 20 3400 exceeds demand by a Central zone 3.6 10 3600 Subtotal 5.3 7000 2128 30% comfortable margin (17 Hybrids percent). This is a positive Central & South 2 2.2 1 2.5 550 642 117% development for hybrids given Source: Ministry of Agriculture, GOI. that 100 percent replacement is required every year. Seed supply and demand of improved varieties, on the other hand, indicate that only 30 percent is replaced each year with high quality seeds (assuming 100 percent replacement rate). However, this is almost equal to the replacement rate needed to retain genetic characteristics for (non-hybrid) improved varieties since improved varieties generally do so at least over three generations. In the case of improved varieties, therefore, the central problem is not inadequate supply but with maintaining purity of the seed over succeeding generations. The task force noted that location specific seed shortages occur in some areas in the short term, but that such shortages could be attributed to infrastructure (weak transport, lack of road networks) and economic constraints (weak demand for seeds discourage dealers to carry sufficient stock). In the long-term, the strong and well-established Indian seed industry would have the capacity to respond to increases in seed demand. At the all-India level, the supply and demand imbalances are no longer as critical as they may have been in the past. A4.28 Poor market incentives in cotton marketing contribute to weak seed management. The incentives for maintaining the purity of seeds over succeeding generations are extremely limited in India and compound the constraints imposed by the structural features of cotton production and marketing in India. Good seed management in India is made inherently difficult by the small-scale structure of cotton production, and the fact that farners mostly trade raw cotton. The structure of the cotton production sector in India--mostly small farms--requires the bulking up of produce from several farmers prior to ginning. The mixing of produce, which may also consist of different varieties, in turn results in the mixing of cotton seeds, which are then sold back to farmers as market seeds. The smallholder production systems in India also imply that cotton is traded in kapas or raw cotton form, rather than lint as in other countries (e.g. USA, Australia). Even if the farmer had the kapas custom-ginned, the difficulty arises in trying to- sell small volumes of lint. A4.29 However, farmers and ginners are not financially encouraged to take the necessary actions to surmount these structural constraints to quality management. Low cotton prices combined with minimal use of grading standards and quality premiums limit farmers' incentives to produce better quality output (and not mix seeds) which pure quality seeds would provide.12 Poor financial incentives, in addition to administratively set ginning fees, also discourage cotton ginners to sort raw cotton and cotton seed in an effort to produce better quality output and improve the purity of market seeds they sell. 12 See Annex 5 for a detailed exposition of quality management in cotton marketing, and policy causes for low cotton prices. Annex 4 Page 12 of 21 Cotton Production Instability Table A4.5: Cotton Production, Area and Yield Instability A4.30 The Central Region and Maharashtra have the highest and increasing degree of production instability. The All-India cotton production remained between 1 and 2 1975-85 7.81 1478 14.99 million mt during the last two decades. Within these 1985-94 3.49 | 14.78 13.80 bounds, production instability, as measured by the Center coefficient of variation, increased only slightly from 11 to 1975-85 4.28 1 l5.83 12.71 13 percent (Table A4.5). However, this masks large regional 1985-94 17.49 33.17 28.42 differences in production instability. Production instability is 1975-85 5.13 20.09 19.64 much higher and increased significantly in the Central 1985-94 4.68 l 27.39 l 61.86 Region and Maharashtra, driven by increased crop yield South instability in those two regions and variability in cultivated 1978594 171.55 9.60 7.18 area in the Central region. In the south, production India instability declined due to greater yield stability which more 1975-85 5.13 1 10.73 1 9.59 than compensated for the drastic increase in area instability. 1985-94 5.36 l 13.35 l 9.99 Source: Computed. A4.3 1 Dependence on rainfed production, poor adaptability of available genetic material to rainfed conditions, and pest problems contribute to yield instability. In contrast to the central regions, the northern region exhibits more stable production, which could be attributed to the predominance of irrigated cultivation that allows greater stability of output for both traditional and higher yielding varieties. However, increasing pest problems in the Northern regions are contributing to increasing production instability. D. Changing Composition of Cotton Output A4.32 India is one of the few countries in the world that can produce a wide variety of cotton, with different staple lengths. Staple length, one basis for classifying raw cotton, is one of the most important determinants of quality. The longer Figure A4.8: Staple Length and Final Output the staple length, the higher the yarn count which Staple Final Product can be produced and therefore the finer the resultant fabric quality. The varieties of cotton produced in Extra long staple Sewing Thread Fine Shirting India are classified by the private sector into 5 Long Staple Fine Knits staple categories: Shirting Find Sheeting (Percale) * Superior long (27mm above, 1.1/6" & Medium Staple Broadcloth Print Cloth above); Sheeting (combed) * Long (24.5 to 26 mm, 31/32" to 30/32"); Corduroy * Superior Medium (22 to 24 mm, 28/32" to I Knitwear Short Staple Coarse Print Cloth 30/32"); Coarse Knitwear * Medium (20 to 21.5 mm, 25/32" to Socks 27/32"); and Sheeting (Muslin) Light Duck * Short (19 mm & below, 24/32" & below). Light Canvas Drill, Denim A4.33 The range of locally produced cotton Heavy Canvas staples enables India to produce a wide variety of Source: FE. Giliham et at., 1995. "Cotton Production cotton products. Staple length is a major Prospects for the Next Decade," World Bank Technical determinant of the type and quality of final product. Paper 287 As shown in Figure A4.8, short staple cottons for example, are generally used for knitwear, denim and heavy canvas, while longer staples are used Annex 4 Page 13 of 21 for producing higher quality products. India's comparative advantage in producing a broad range of cotton staples therefore provides the country great flexibility in expanding into a diverse range of textile product markets. Figure A4.9: Cotton Output by Staple Length A4.34 Regions tend to specialize in specific staples. The Northern region 2000 _ is the primary producer of shorter 1600 staples. In 1991-92, the North 1400 accounted for 73 percent of superior 1200 medium produced in the country, 45 E 1000 percent of the medium staples and 51 ° 800 percent of the short staples (Table 600 400 A4.6). The southern region 200 dominates the production of longer 0 staples (mostly DCH 32 and Suvin) 82-83/85-86 86-87/88-89 89-90/91-92 with 54 percent of the superior long output. The central region produces * Superior LongE Long D Superior Medium X Medium D Short a wider range of staples (mostly American and Desi varieties). The Source: East India Cotton Association, Indian Cotton Annual 1992-93. regional patterns of growth in cotton production (Figure A4.9) therefore also closely mirror the changing composition of cotton staple output. Table A4.6: Production of Cotton by Staple, gonad St e, 1991-92 A4.35 The staple composition of ~i cotton production in India has changed significantly. As the All-India r 73 4 51 output expanded, the relative shares of Haryana 21 __ different staples also changed (Figure - Rajasthan 11 45 28 A4.9). During the last 15 years, output Central 29 9 13 38 6 of superior long staples grew rapidly, Gujarati 27 7 9 33 by about percent ~~~Madhya Pradesh 2 24 6 increasing by about 8 percent annually. Maharashtra 17 89 6 16 That of superior medium increased by South 54 2 8 25 7 percent per year. In contrast, the Andhra Pradesh 23 2 25 production share of medium and long Karnataka 21 2 3 production ~~~~~~~~~~~Tarn!] Nadu, staples dropped drastically at a rate of All India 100 100 100 100 100 23 percent and 9 percent per year Source: East India Cotton Association, Indian Cotton Annual 1992-93. respectively. As of the TE 1991-92, superior medium staple accounted for the largest share of domestic output (50 percent), followed by superior long staple (30 percent). The long, medium and short staples accounted for 8, 1 and 11 percent respectively. A4.36 Traditionally, India was a major producer of short staple cotton. In the early post- Independence period, the bulk of Indian cotton production was of very short staple length. Till the end of the 1 960s, longer staple cotton was imported, while the surplus short staple cotton was exported. A strong research program focused on the development of local long-staple hybrids fostered increased domestic production, which eased cotton imports during the seventies. Imports in later years were mostly due to sharp domestic production shortfalls in some years. One weakness of the cotton research program which developed the long staple varieties was the lack of priority given to developing strength in the fibers. Fiber strength is also important in reducing breakage of the long fibers and therefore maintaining quality during processing. Annex 4 Page 14 of 21 A4.37 Indian cotton grading standards differ from international standards, and tend to be "shorter ". According to international standards, staple lengths up to 25.4 mm are termed "short," greater than 25.4 mm to 27 mm are termed "strict low middling," 28 mm to less than 32 mm are medium, greater than 32 to 35 mm are termed "long," and over 35 mm are "extra-long/fine." As shown in Figure A4. 10, the Indian staple length classifications (medium, superior medium, long, and superior long), indicated by the suffix "I", are shorter than the international standards. Thus by international standards, Indian cotton is more highly dominated by medium staples. As India becomes a more active exporter, the local adoption of international consistent stands will become more critical. It will also greatly facilitate the transmission of price signals through the marketing channel. Figure A4. 10: How does Indian Staple Standards Compare to Intemational Norms? 45 DCH 32, Suvin Pima, Egyptian 40 Et________________________________ MCU5 a 35 Gia75l 6 30 MCU7 1H4/6, S4/6 ! 25 J34,GI _ _ _ _ _ _ _ _ ~~20 Laxmi 10 I Lote:i-iiuiian ~ C 00 d. Note: I - Indian classification; W - intemational classification. Source: EICA and F.E.M. Gillhamn, et. al. 1995, "Cotton Production Prospects in the Next Decade," World Bank Technical Paper No. 287. Annex 4 Page 15of 21 E. Appendix Tables Appendix Table A4. 1: Trend Growth Rates of Production of Major Crop Groups, Pre- and Post-Green Revolution, percent per year Weight in Pre-Green Post-Green Post Green Significant Percent Index of Revolution Revolution Revolution Change between Contribution to Item Agricultural Period I Period II Period I & II All Crop Growth Production 1951-67 1968-81 1982-94 at 5% Level in Period II All Crops Index 100.00 2.40 2.20 3.30 Yes* 100.00 Food grains 62.29 2.00 2.10 2.80 No 53.40 Total Cereals 54.98 2.50 2.60 3.00 No 50.00 Coarse Cereals 10.79 1.70 0.70 0.70 No 2.30 Pulses 7.94 0.20 -0.40 0.90 No 2.20 Rice 29.74 Prod 2.8 2.2 3.6 No Area 0.6 0.8 0.5 No Yield 2.3 1.4 3.1 Yes Wheat 14.45 Prod 4.7 5.3 3.3 Yes Area 1.6 2.9 0.5 Yes Yield 3.0 2.4 2.8 No Non-Foodgrains 37.08 3.20 22.00 4.20 Yes 47.20 Oilseeds 12,64 2.50 1.00 5.80 Yes 22.20 Cotton 5.09 3.30 2.50 3.40 No 5.20 Sugarcane 8.11 4.00 2.60 2.90 No 7.10 Plantation Crops 2.29 2.90 4.10 3.80 No 2.60 Condiments & 2.59 0.80 2.00 3.40 No 2.70 Spices Fruits & Veg. 4.90 7.60 4.10 4.00 No 5.90 GDP Agriculture 2.00 2.00 3.00 Yes* (including forestry and fishing) *At 10% level of significance Note: Pre-green revolution growth rates for all the above crop groups are shown in Appendix Table 1. Source: D. Ahluwalia, 1996, "Agricultural Growth in India: Trends, Sources and Challenges," Draft Mimeo, World Bank New Delhi. Annex 4 Page 16of 21 Appendix Table A4.2: Cotton Area, Production, and Yield by Region and All India Category 79-80 80-81 81-82 82-83 83-84 84-85 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 94-95 95-96 AREA,000 ha North 1335 1322 1390 1518 1476 1202 1237 1311 1382 1490 1634 1646 1644 1697 1660 1649 2002 Central 2331 2167 2145 2076 1943 1908 1938 1889 1224 1653 1411 1529 1695 1619 1615 1682 1947 Maharashtra 2588 2667 2710 2648 2685 2685 2753 2693 2518 2628 2636 2730 2724 2480 2479 2760 3065 South 1803 1597 1753 1566 1560 1641 1548 1063 1294 1527 1606 1491 1557 1686 1530 1736 1995 Others 70 70 59 63 57 -54 57 -8 41 45 44 44 41 61 37 44 54 India 8127 7823 8057 7871 7721 7382 7533 6948 6459 7343 7331 7440 7661 7543 7321 7871 9063 PRODN, 000 mt North 387 377 404 444 316 390 446 559 470 606 786 677 774 805 591 685 777 Central 345 337 413 317 281 397 386 227 99 310 367 292 243 256 263 264 312 Maharashtra 288 216 249 274 138 249 338 151 250 233 376 320 196 307 446 425 476 South 268 252 266 237 341 401 302 278 275 330 402 375 430 419 433 457 514 Others 12 10 9 10 9 9 11 -41 -9 8 9 9 8 151 93 190 147 India 1300 1192 1340 1281 1086 1446 1484 1174 1085 1486 1940 1673 1651 1938 1826 2021 2226 YIELD, kg/ha North 290 285 291 293 214 324 360 427 340 407 481 411 471 474 356 415 388 Central 148 155 193 153 145 208 199 120 81 188 260 191 143 158 163 157 160 Maharashtra III 81 92 103 52 93 123 56 99 89 143 117 72 124 180 154 155 South 148 158 151 151 219 244 195 262 212 216 251 252 276 249 283 263 258 India 1601 152 166 163 141 19~6 197 1691 168 2072F 2521 2251 2161 2491 2831 263i 258 Notes: North: Punjab, Haryana, Rajasthan; Central: Gujarat and Madhya Pradesh; South: Kamataka, Andhra Pradesh, Tamil Nadu. Source: Ministry of Agriculture. Annex 4 Page 17of 21 Appendix Table A4.3: Selected Cotton Variety Characteristics Principal Planting Pickg. Modal Range in Fiber Spinng. Varieties Grown Dates Dates Staple Length Staples Fineness Counts Class mm's micronaire NORTHERN Punjab Desi May-Jun Sep-Dec 5 13.0-16.0 6.5-7.3 20s J-34 May-Jun Sep-Dec M 23.0-24.5 3.54.5 24s-28s F-414/H-777/Agatti May-Jun Sep-Dec SM 24.5-25.5 3.6-4.0 24s-28s Haryana Desi May-Jun Sep-Dec 5 13.0-16.0 6.5-7.3 20s J-34 May-Jun Sep-Dec M 23.0-24.5 3.54.5 24s-28s F-414/H-777/Agatti May-Jun Sep-Dec SM 24.5-25.5 3.6-4.0 24s-28s Rajasthan Desi May-Jul Oct-Nov 5 13.0-16.0 6.5-7.3 20s J-34 May-Jul Oct-Nov M 23.0-24.5 3.54.5 24s-28s F-414/H-777/Agatti May-Jul Oct-Nov SM 24.5-25.5 3.6-4.0 24s-28s Digvilay/G-1 1 May-Jul Oct-Nov SM 23.0-25.0 3.84.8 24s NORTHWEST COAST Gujarat Jun-Sept Dec-Mar 5 19.0-22.0 5.5-6.0 Wagad Jun-Sept Dec-Mar M 20.0-22.0 3.7-4.4 30s V-797 Jun-Sept Dec-Mar SM 21.0-23.0 4.2-5.0 CJ-73 Jun-Sept Dec-Mar SM 22.0-24.0 4.5-5.2 Gujarat-12 Jun-Sept Dec-Mar SL 23.0-25.0 3.84.8 24s Digvilay/G-l 1 Jun-Sept Dec-Mar SL 28.0-30.0 3.74.1 40s-50s Shankar-6 (Superior) Jun-Sept Dec-Mar SL 26.0-28.0 3.4-3.8 30s-40s Shankar-6 (Average) Jun-Sept Dec-Mar SL 29.0-31.0 3.84.2 44s-60s Shankar4 (Superior) Jun-Sept Dec-Mar SL 28.0-30.0 3.54.9 40s-50ss Shankar4 (Average) Jun-Sept Dec-Mar SL 34.0-37.0 2.9-3.3 60s-80s Varataxmi (Superior) Jun-Sept Dec-Mar SL 32.0-35.0 2.8-3.3 44s-60s Varalaxmi (Average) Jun-Sept Dec-Mar SL CENTRAL Madhya Pradesh 197/3 & Y-1 & Vimar May 20-Jul Oct 15-Apr M 22.0-24.0 4.0-5.3 26s L-147/A-51-9 May 20-Jul Oct 15-Apr SM 23.0-25.0 3.54.5 24s-28s H-4 May 20-Jul Oct 15-Apr SL 27.0-29.0 5.0-5.7 40s-50s Varalaxmi (Superior) May 20-Jul Oct 15-Apr SL 34.0-37.0 2.9-3.3 60s-80s Varalaxmi (Average) SL 32.0-35.0 2.8-3.3 44s-60s DCH-32 (Superior) May 20-Jul Oct 15-Apr SL 36.0-40.0 2.9-3.3 80s-lOOs DCH-32 (Average) SL 33 .0-36.0 2.7-3.2 60s-80s Maharashtra Gaorani-6 Jun-Sep Nov-Apr SM 21.0-23.0 4.5-5.2 197/3 & Y-1 & Virnar Jun-Sep Nov-Apr M 22.0-24.0 4.0-5.3 26s AK-235/AK-277 Jun-Sep Nov-Apr SM 22.0-24.0 4.6-5.2 AKH-4/AKH-5 Jun-Sep Nov-Apr SM 22.0-24.0 4.2-4.8 20s L-147/A-51-9 Jun-Sep Nov-Apr SM 23.0-25.0 3.5-4.5 24s-28s DHY-286/AKH-468 Jun-Sep Nov-Apr SM 23.0-25.0 3.6-4.2 24s-36s LRA-5166 Jun-Sep Nov-Apr L 25.0-26.0 3.5-4.0 40s 1007 Jun-Sep Nov-Apr L 25.0-27.0 3.5-4.5 24s-36s MECH-1 Jun-Sep Nov-Apr SL 27.0-29.0 3.6-4.3 H4 Jun-Sep Nov-Apr SL 27.0-29.0 5.0-5.7 40s-50s JKHY-1 Jun-Sep Nov-Apr SL 27.0-29.0 2.8-3.8 40s MCU-5 (Superior) Jun-Sep Nov-Apr SL 32.0-34.0 3.0-3.5 50s-60s Annex 4 Page 18of 21 Appendix Table A4.3 cont'd: Selected Cotton Variety Characteristics Principal Planting Pickg. Modal Range in Fiber Spinng. Varieties Grown Dates Dates Staple Length Staples Fineness Counts Class mm's micronaire Maharashtra cont'd MCU-5 (Average) SL 30.0-33.0 2.7-3.2 34s-50s Varalaxmi (Superior) Jun-Sep Nov-Apr SL 34.0-37.0 2.9-3.3 60s-80s Varalaxmi (Average) SL 32.0-35.0 2.8-3.3 44s-60s DCH-32 (Superior) Jun-Sep Nov-Apr SL 36.0-40.0 2.9-3.3 80s-lOOs DCH-32 (Average) SL 33.0-36.0 2.7-3.2 60s-80s Southern Andhra Pradesh Jayadhar/Suyodhar Jun-Oct Jan-May SM 21.0-23.0 4.6-5.6 Gaorani-6 Jun-Oct Jan-May SM 21.0-23.0 4.5-5.2 197/3 & Y-1 & Virnar Jun-Oct Jan-May M 22.0-24.0 4.0-5.3 26s AK-235/AK-277 Jun-Oct Jan-May SM 22.0-24.0 4.6-5.2 1007 Jun-Oct Jan-May L 25.0-27.0 3.5-4.5 MECH-1 Jun-Oct Jan-May SL 27.0-29.0 3.6-4.3 H-4 Jun-Oct Jan-May SL 27.0-29.0 5.0-5.7 40s-50s JKHY-1 Jun-Oct Jan-May SL 27.0-29.0 2.8-3.8 40s MCU-5 (Superior) Jun-Oct Jan-May SL 32.0-34.0 3.0-3.5 50s-60s MCU-5 (Average) Jun-Oct Jan-May SL 30.0-33.0 2.7-3.2 34s-50s Varalaxmi (Superior) Jun-Oct Jan-May SL 34.0-37.0 2.9-3.3 60s-80s Varalaxmi (Average) Jun-Oct Jan-May SL 32.0-35.0 2.8-3.3 44s-60s DCH-32 (Superior) Jun-Oct Jan-May SL 36.0-40.0 2.9-3.3 80s-lOOs DCH-32 (Average) Jun-Oct Jan-May SL 33.0-36.0 2.7-3.2 60s-80s Suvin Jun-Oct Jan-May SL 36.040.0 2.9-3.3 90s-120s Karnataka Jayadhar/Suyodhar May-Sept Oct-Apr SM 21.0-23.0 4.6-5.6 Vimar May-Sept Oct-Apr M 22.0-24.0 4.0-5.3 26s Varalaxmi (Superior) May-Sept Oct-Apr SL 34.0-37.0 2.9-3.3 60s-80s Varalaxmi (Average) SL 32.0-35.0 2.8-3.3 44s-60s DCH-32 (Superior) May-Sept Oct-Apr SL 36.040.0 2.9-3.3 80s-lOOs DCH-32 (Average) SL 33.0-36.0 2.7-3.2 60s-80s Tamil Nadu LRA-5166 L 25.0-26.0 3.54.0 40s MCU-5 (Superior) Aug-Sep SL 32.0-34.0 3.0-3.5 50s-60s MCU-5 (Average)_ SL 30.0-33.0 2.7-3.2 34s-50s Varalaxmi (Superior) Aug-Sep SL 34.0-37.0 2.9-3.3 60s-80s Varalaxmi (Average) SL 3.0-35.0 2.8-3.3 44s-60s DCH-32 (Superior) Oct-Nov SL 36.040.0 2.9-3.3 80s-2100s DCH-32 (Average) SL 33.0-36.0 2.7-3.2 60s-80s Suvin Aug-Sep SL 36.0-40.0 2.9-3.3 90s-120s Source: T. A. Bell and F.E.M. Gillham, 1989, 7he World of Cotton, Washington, D.C.: ContiCotton, EMR Annex 4 Page l9of 21 Table A4.4:1rrigated Area under Cotton Cultivation by State, 000 ha and Percentage Share of Total Cultivated Area (values in parenthesis) . States 1981-82 1982-83 1983-84 1984-85 1985-86 1986-87 1987-88 1988-89 1989-90 1990-91 AndhraPradesh 190.0 177.2 38.2 96.0 76.3 50.0 52.0 325.0 88.0 82.0 (40.1) (40.0) (8.0) (17.0) (12.3) (12.2) (9.2) (51.7) (13.5) (12.5) Gujarat 461.7 359.3 316.1 394.7 430.5 450.1 208.7 424.4 436.0 436.0 (30.2) (24.0) (22.6) (28.5) (30.7) (33.0) (29.0) (38.9) (36.7) (47.3) Haryana 310.0 379.0 390.0 288.0 343.0 379.0 413.0 430.0 469.0 489.0 (94.8) (95.5) (95.1) (97.6) (99.7) (100.0) (99.0) (99.3) (100.0) (99.7) Madhya Pradesh 54.2 53.9 51.5 66.2 60.0 89.0 79.0 131.5 136.0 147.0 (8.8) (9.0) (9.5) (13.0) (11.2) (17.0) (15.6) (23.4) (23.5) (24.1) Maharashtra 141.1 118.7 103.8 100.2 109.4 66.6 30.0 53.5 91.0 91.0 (5.2) (4.5) (3.9) (3.8) (4.0) (2.5) (1.2) (2.0) (3.4) (3.3) Karnataka 80.2 110.0 127.9 182.1 148.7 112.4 100.0 136.4 177.0 145.0 (7.7) (11.8) (14.1) (21.7) (22.0) (27.1) (21.0) (20.8) (25.3) (24.3) Punjab 683.0 724.0 635.0 464.0 553.0 562.0 616.0 754.0 728.0 696.0 (100.0) (100.0) (97.7) (98.2) (98.8) (99-1) (99.2) (99.5) (99.4) (99.2) Rajasthan 326.9 358.0 380.8 304.7 301.0 343.7 327.4 277.8 410.0 428.0 (86.1) (90.2) (91.5) (90.9) (90.4) (94.2) (95.2) (92.8) (94.4) (94.1) Tamil Nadu 96.1 73.2 84.7 129.4 138.0 75.2 108.6 100.3 111.0 77.0 (40.0) (45.0) (47.6) (51.1) (54.3) (40.4) (44.5) (41.2) (41.4) (32.1) Uttar Pradesh 25.2 28.1 22.7 15.8 23.9 16.8 18.2 18.3 17.0 14.0 (83.4) (79.4) (75.7) (79.0) (85.4) (76.4) (89.7) (87.6) (88.5) (89.1) Others 2.0 2.0 Neg. Neg. 2.1 2.0 2.0 2.0 1.0 1.0 AllIndia 12370.4 2383.4 2150.7 2041.1 2185.9 2146.8 1954.9 2653.2 2664.0 2606.0 (29.4) (30.4) (27.9) (28.1) (29.0) (30.9) (30.2) (36.1) (34.6) (35.0) Source: EICA India Cotton Annual, 1992-93 Annex 4 Page 20 of 21 Annex A4.5: Integrated pest management approaches for cotton in selected countries Crop Pest Country IPM Type IPM Method Results Reference Cotton Aphids China, Cultural Intercropping of cotton, rape and com with wheat in adjacent Local farmer's practice. In tests, the intercropping Zhou et al. 1986 southem fields. Com is planted twice during the year in two or three showed predator levels three to six times that of Hebei seedling bunches every 2 meters. monocropped fields. Pesticide applications were Province reduced by three to four sprays annually. Cotton Aphids; China Biological Release of the predator Coccinella septempuncta, which is C. Septempuncta is mass-reared on aphids or male BIRC 1991 Mites; collected at the end of the season from vegetable fields and pupae of honey bee, and recently on artificial diets, Cotton released in the next season. which allows cost-effective mass production. bollwomm Cotton Cotton China Biological Utilization of small "greenhouses" (bamboo and plastic) This technique is utilized on 6,000 ha of cotton and Liu and Qin aphids constructed in cotton fields for rearing the seven-spotted lady effectively reduces aphid numbers, generating savings 1987; Olkowski, beetle. of $61ha/yr. Zhang and Thiers 1991 Cotton Cotton China, Biological Field collection and release of the C7 lady beetle, Coccinella Well-established practice in Henan Province. In Olkowski, Zhang aphids Henan septempuncta Anyang County in 1974, 93.3% of all cotton fields and Thiers 1991 Province, (22,000 ha) utilized C7 beetles. Savings of about Anyang $309,000 in pesticide costs. County Cotton Cotton China Viral Application of nuclepolyhedrosis viruses (NPV). NPV for cotton has been mass-produced as "78-3" Zhang et al. 1981 bollworm insecticides and has been applied to about 100,000 ha, usually yielding 90% mortality after 11 days. Cotton Cotton Philippines Biological Locally produced Trichogramma australicum released The costs of one application of a synthetic pyrethroid Hasse et al. 1988 bollwonm preventatively at 40,000/ha twice a week equals the costs of Trichogramma releases for six or seven weeks. Cotton Cotton Philippines Chemical Peg-board for farmer's use. Simplified surveillance system. Reduction of pesticide use. Hasse et al. 1988 bollwormn Cotton Cotton Philippines Chemical Surveillance system for technicians. Threshold of 17-18 Reduced insecticide applications. Hasse et al. 1988 bollworm infested plants out of 20 plants considering predator population densities. Cotton Cotton Philippines Cultural Intercropping of 15-20 rows of cotton with 1 row of maize or Reduced sprays from 6.5 to 2.5 per season. Hasse et al. 1988 bollworm tobacco as trap crop (pest trap) with synchronized reproductive phase. Cotton Cotton China Biological Release of the predator lacewing Chrysopa spp. The predator has been mass-reared and released since BIRC 1991 bollworm; 1979. Chrysopa larvae are reared in eggs of rice Cotton moth, Corcyra cephalonica, and in artificial eggs aphid encapsulated by low-melting-point paraffin. Annex 4 Page 21of 21 Annex A4.5 cont'd: Integrated pest management approaches for cotton in selected countries Crop Pest Country IPM Type IPM Method Results Reference Cotton Cotton flower Philippines Cultural Cessation of irrigation for up to four weeks after the Forced boll maturation, increased pupal mortality (100% Hasse et al. 1988 weevil onset of flowering at 9% soil moisture), favorable environment for predator ant. Cotton Cotton flower Philippines Cultural Increase of planting density from 60,000 to 130,000 Increased yield security. Hasse et al. 1988 weevil plants/ha. Cotton Cotton flower Philippines Cultural Planting of all fields within the shortest time. Infestation remains below threshold if no migration Hasse et al. 1988 weevil from older fields occurs. Cotton Cotton flower Philippines Cultural Planting a trap crop of okra (Hibiscus esculentus L.) so Reduced early infestation with flower weevil Hasse et al. 1988. weevil cotton and okra flower at the same time. Cotton Cotton Philippines Chemical Chemical application at the threshold of 50% Yield increase is reported. Hasse et al. 1988 leafhopper defoliation Cotton Cotton Philippines Cultural Introduction of a hairy variety from Paraguay Reduction of early insecticide applications. Hasse et al. 1988 leafnopper Cotton Cotton pink China Biological Release of the braconid parastoid Bracon nigroruflum. Field parastism rates are highest (90%) for the first BIRC 1991 bollworm generation bollworms. 2nd and 3rd generation __________ _parasitism reaches 60%/o. Cotton Heliothis China Viral Application of the nuclear polyhedrosis virus (Ha Research started in 1975. It is now the first Guangyu 1991 armigera and insecticide SNPU) at 4.8 x 10ll PIB/ac by ground or air commercially available viral insecticide in China. H. assulta application. Sales volume increased from 600 ha ('86) to 2800 ha ('87) to 5700 ha ('88) and an estimated 10,000 ha in 1989. The price is equivalent to chemical insecticides ($5/ha). Cotton Hypomeces Indonesia Cultural Early planting to prevent infestation No information. Wirjosuhardjo, 1988. Cotton Mealybug China Biological Release of Cryptolaemus montrouzieri C. montrouzieri was introduced to China in 1955 and is BIRC 1991. mass produced on Aleurites molluccana trees. Cotton Old World China Biological Moving of overwintering queens of the predator wasp Increased infestation of bollworm and measuring worm Li and He 1988 bollworm, Polistes hebraeus into nests within controlled by 61% and 57% respectively. Suppression of the Small cotton temperature and humidity greenhouses. wheat armyworm and the wheat sawfly. measuring worm Cotton Spodoptera Indonesia Mechanical Handpicking of Spodoptera eggs No information. Hasse et al. 1988 Cotton, Cotton pink China, more Biological Local production and release of the pteromalid Mass production and release in more than 10 provinces BIRC 1991. storage bollworm than 10 ectoparasitoid Dibrachys cavus. Release in stored since 1960. Parasitism of cocoons reaches as high as provinces cotton to control overwintering larvae. 100%. Source: Uwe-Carsten Wiebers, 1993, "Integrated Pest Management and Pesticide Regulation in Developing Asia," World Bank Technical Paper No. 211. Structure & Performance nnex O of Cotton Markets A. Introduction A5.1 A competitive and efficient cotton marketing system would ensure that demand and supply signals are accurately transmitted between the ultimate consumers (textile industry) and producers (cotton farmers). The cotton marketing system in India is quite unique in that it is one of the few major cotton producing countries in the world where the marketing of cotton (seed cotton and cotton lint) is predominantly handled by the private sector (traders and cooperatives). Until 1996/97, it was also highly regulated by the GOI and State governments. Past interventions ranged from storage, movement, and credit controls to the fixing of ginning fees and restrictions on the scale of operations in the ginning sector to export controls. This Annex describes the structure of cotton trading in India and analyzes its performance in the context of a highly regulated environment. The following sections describe the structure and operations associated with trading seed cotton and cotton lint, ginning, cotton classification and grading, and cotton exports and imports. The last section investigates the performance of the cotton marketing system: processing margins, seasonal price efficiency and risk management. B. Marketing of Seed Cotton and Cotton Lint AS.2 The private sector has the lion share in cotton trade. There are three major players in the marketing of cotton: private traders, state level cooperatives and the Cotton Corporation of India (CCI). Private traders include owners of ginneries, local and terminal market merchants or their agents and textile mills. In the triennium ending Figure A5. 1: Private Traders Dominate Cotton Seed 1993/94, private traders handled about 77 percent of Procurement, TE 1993-94 marketed seed cotton and cotton lint (Figure A5.1) and Private their market share is expanding over time (Appendix Traders Table A5.1). Other major players in the cotton market 77% are the state cooperatives (15 percent) and the CCI (8 percent). In Maharashtra, all seed cotton is purchased by the state government through the Maharashtra State Cooperative Cotton Growers' Marketing Federation sta - (MSCCGMF) under its Monopoly Procurement 4% MSCCG 8% Scheme (Annex 2). In view of its procurement MF monopoly, MSCCGMF accounts for about 11 percent 11% of total market supply and is the single largest cotton trader in India. ~~~~~~~~~Source: EICA and CCI. trader in India. Seed Cotton Marketing A5.3 Regulated markets or mandis are the primary marketing channelfor seed cotton. Unlike most countries, farmers in India sell their cotton in the form of kapas or seed cotton, not as cotton lint. Cotton lint is, with cotton seed, the joint product of ginning seed cotton. Farmers sell their output in the large assembly markets, which maybe regulated or unregulated. The regulated markets or mandis were established under the State Agricultural Product Markets Act (Text Box A5. 1) and consist of a dense network of delivery points all over the country. About 75 percent of the seed cotton in India is marketed through over 1,000 of the 6,000 regulated markets in the Anne\ > PaLc 2 of .4 country. The regulated markets in the northern irrigated states (Madhya Pradesh, Karnataka. Maharashtra and Southern Gujarat) are important outlets for seed cotton. The exceptions are states such as Gujarat and Andhra Pradesh, where unregulated markets still function in manm areas. In other states, such as Punjab and Haryana, a large part of the sales occur outside tlle regulated markets-- at the village level by private negotiation. In the regulated markets, sales take place mostly through auctions and open outcry, in which the commission agent usually acts on behalf of the seller. In some markets, the seed cotton is sold by inviting tenders from buvers instead of open auction. A5.4 Commission agents generallv mediate transactions in the markets. Commission agents or adatias are prohibited by the rules of regulated markets from purchasing seed cotton onl their owIn account. If a commission agent desires to purchase from his own seller client, it is obligatory for him to make such purchases in the public auction or through tender in the market yard in the presence of market committee officials. The commission agents are also free to buy cotton sold through other agents in the market. However, very few adatias combine both trading and the commission agency business. Even when some do, their actual share in the total purchases of seed cotton in the assembly markets is small. A5.5 Commissions agents also provide other support services to Jarmers. Almost all transactions in these markets go through commission agents or adatias, although it is not obligatory for sellers to engage them. Cooperatives also act as commission agents for their member farmers. Usually cotton is brought by the farmer to the office of the commission agent, who exhibits and arranges for its sale in the auction. Besides providing direct mark-et related services, commission agents also offer cash advances to farmers for the purchase of seeds, pesticides, fertilizers and funds to cover non-production related expenditures. Commissions are fixed by market committees ranging from 0.4 percent in Gujarat to 2 percent in Punjab. Harvana and Karnataka. A5.6 Most of the kapas is sold duiring October to May, although the duration may vary according to regions. The marketing season in the irrigated areas of the north (Punjab. Haryana and Rajasthan) begins in October and virtually ends in March. In the central region (Madhva Pradesh, Gujarat) and Maharashtra, the marketing of cotton is mainly confined between December and March (extending into July for Gujarat). In the three southerm states of Karnataka, Andhra Pradesh and Tamil Nadu, marketing begins in February and ends in April-May. Most farners transport their seed cotton to the assembly markets by bullock cart. although larger farmers and cooperatives pooling member output are increasingly using trucks and tractor-trailers for transport. Text Box A5. 1: Agricultural Product Markets Act and Mandi Operations The Agricultural Product Markets Act (APMA) was legislated by the states for the purpose of regulating the marketing of agricultural produce, including oilseeds and oilseed productsm for the purpose of improving their efficiency and in the process ensure a more equitable distribution of gains from agricultural trade between consumers, traders and producers. The Uttar Pradesh Krishi Utpadan AMandi .Samitis Adhiniyam (1972) and the Punjab Agricultural Produce Markets Act (1961) are examples of this legislation. The act prescribes the setting up of mandis or regulated markets for agricultural produce and the establishment of Marketing Committees in each market area. 'The Marketing Committees are responsible for implementing and enforcing the provisions of the act and arc empowered to control and regulate admissions to the markets, charge fees (market and license fees), issue or renew license and suspend or cancel them. The Marketing Committees are supervised by the State Agricultural Produce Market Board, which is entrusted with the responsibility for state-wide market development. Revenues generated under the Act were to be allocated for the operation and development of tile respective markets, particularly (i) the operation and maintenance of the market yards and related facilities (Market Committee Funds) and (ii) the development and improvement of market facilities and related development works and activities (Market Committee Development Fund). Annex 5 Page 3 of 34 A5.7 lnfrastructure is inadequate and grading is non-existent in the mandis. Infrastructure facilities for weighing, handling, moving, and storing cotton in many regulated markets are inadequate, if not completely non-existent. In some markets, rable A5.l: Markets Cess on Kapas in inadequacy of space leads to cramped stacking and admixture of Selected States different lots. Moreover, poor handling practices lead to greater Statc Markct(Ce% s contamination of seed cotton. Grading is non-existent in most Punjab 3 25a states. Haryana 3.00 Madhya Pradesh 1.35 AS.8 Essential market information is poorly collected and Rajasthan 1.25 dlisseminated by the mandis. Market information on the Andhra Pradesh 1.00 transactions in the regulated markets, such as volume of arrivals, Karnataka 1.00 quantities sold, varieties delivered and prices realized, provide an Tamil Nadu 1.00 important basis for market participants to assess price trends, Gujarat 0.003 Note: a - includes market fees, rural supply conditions and variety-wise production in the short and development fund and marketcets long run. In some markets, proper markets records are not Source: Ministry of Textiles. "Report maintained on a regular basis. In other markets, although the of Working Group on Cotton Procurement. Marketing and Ginning information is collected, they are poorly disseminated. and Pressing Industry" A5.9 Market cess collected by the mandis is intended for improving/maintaining mandi facilities. As per the State Agricultural Product Markets Act, the mandi committees are permitted to charge an ad valorem market cess for the purpose of generating revenues for market improvement and development activities. The market fees charged vary from state to state, from as low as 0.003 percent in Gujarat to 3 percent in Haryana (Table A5. 1). The revenues generated through the cess is considerable. In 1994, the state of Rajasthan collected about Rs 900 million, while the state of Andhra Pradesh collected about Rsl billion. A5.10 A large share of mandi revenues is often diverted to non-market related purposes. Although intended primarily for financing market operations and improvements under the guidance of the market committees, the market cess has become a major Text Box A5.2. Expenditures Financed By Market Ccss Levied source of revenue for the state On Market Yards In The State Of Andhra Pradesh treasurv. For example in Andhra A I percent market cess is levied on products purchased on the 82 1 mandis in Pradesh, only about 45 percent of operation in the StateofAndhra Pradesh. Atotal ol'around Rsl billion. (LUS$32.2 the approximately Rs I billion million) was collected during 1994, ofwhich: market tax collection was re- . 30 perccnt (USS9.6 million) was usedto cover market and central administration invested in the markets--30 operatingcosts: * 10 percent (US$3.2 million) went into a "Central Market Ftund" which makes percent for operating and loans and grants to the mandis: administrative costs and 15 . 25 percent (USS8.0 million) was used tbr shoil-tenn credit for a maximum ol' percent for infrastructure three montis and 50,000 Rs. (USSI,600) to farmers at a low interest rate; improvement (Text Box A5.2). . 15 percent (USS4.8 million) was used to improve market infrastructure: and In Rajasthan, about 7,000 km of * 20 percent (USS6.4 million) went towards the mainterance and improvement of roads have been built using rural roads. revenues from the mandis. Note: Percentages are best estimates of order of magnitude. Limited financial authority of the Source: Field Estimates. respective marketing committees constrains access to the necessary capital to properly operate or upgrade the facilities of the agricultural markets. Weak financial authority and deficient accounting systems have further contributed to rent-seeking activities. A5.1 I There is a needfor increased decentralization offinancial and management auth)oritl in1 thie mclndis to market committees. The market committees should be given more authority and Annex 5 Page 4 of 34 accountability over the revenue generation and mobilization activities. The frequency at which fees are to be collected needs to be clarified and made more transparent. In order to inerease the efficiency and promote the development of individual agricultural markets, the financial rules need to be reformed to allow the marketing committees to retain a greater proportion of the revenue collected and greater authority in fund allocation. To improve the effectiveness and transparency of the fund allocation, the composition of the marketing committees should be improved to increase the representations of users--traders, farmers, and consumers and regular elections should be held. The State Agricultural Produce Marketing Boards should focus on monitoring and enforcement of the rules and regulations, while retaining financial control over state-wide market development initiatives. Where possible, the leasing out or full divestment of the mandis to user could be explored. Cotton Corporation of India A5.12 Central government interventions in cotton marketing are implemented through the Cotton Corporation of India (CCI). The establishment of the CCI ushered a new era of direct GOI intervention in cotton marketing in India. CCI was set up in 1970 to exclusively import cotton to bridge the domestic shortfall in cotton output. In 1972, it was charged with supporting cotton prices in all states except Maharashtra. As a result, in the early 1970s, CCI's main role was to purchase the surplus of seed cotton in assembly markets in several states at support prices. However, with prices above the minimum support price levels, CCI's role in the cotton market declined. Consequently, in 1975 its role was expanded to buying cotton for the public sector National Textile Corporation (NTC). Since the NTC mills are free to purchase their cotton from the cheapest sources, CCI could not depend solely on NTC mills from its survival. The Textile Policy announced in August 1978 therefore allowed CCI to enter into commercial transactions and operate a buffer stock, and purchase cotton not only for the NTC mills, but also for other mills in the private sector. The CCI also purchases cotton for other public sector mills, assisting in theory the weaker units among them. In order to relieve itself of mounting unsold stocks, CCI was appointed as the sole canalizing agency for the export of extra long staple (ELS) cotton. This policy was subsequently modified to permit state cooperative marketing federations to also export ELS cotton subject to annual quotas. A5.13 CCI purchases cotton in competition with private traders and local cooperatives. CCI has a network 200 purchasing centers operating in all states except Maharashtra. The corporation is also required to maintain minimum infrastructure in remote areas served by Primary Cooperative Marketing Societies although its last price support operation was in 1986. CCI also performs several other non-commercial functions including improvement in ginning, processing, and packaging of cotton bales, assistance in research work, assistance for increasing production of genetically pure high quality cotton seed for planting, and assistance to various state agencies to increase cotton production as well as assistance in pest management. A5.14 CCI's financial performance improved in the 1990s. Prior to 1988/89, CCI consistently incurred losses, which reached nearly Rs 290 million in 1981/82, owing to large accumulated stocks and escalating interest liabilities (Appendix Table A5.3). CCI's mounting losses were addressed by the GOI by raising its share capital from Rs 40 million in 1978/79 to Rs 230 million at present and partly through loans. Rising cotton prices and income from other sources (e.g. carrying charges for the private sector) in recent years contributed to its improved financial Annex 5 Page 5 of 34 performance.1 In 1996/97, CCI profits after deductions for interest charges and taxes reached Rs 491 million. A5.15 CCI's market share remains small and is declining over time. Between 1989/90 and 1993/94, CCI's cotton procurement declined from 1.2 million bales (170 kg) to 776,000 bales. By 1993-94, its market share was 6 percent, down from 9 percent in 1989-90. CCI's commercial operations are primarily centered in the states of Rajasthan, Gujarat, Madhya Pradesh and Andhra Pradesh (Appendix Table A5.4). The excess of purchases over domestic sales is exported, usually through private traders who act as agents for foreign buyers. CCI's domestic sales to the private sector are increasing, accounting for two thirds of sales in 1996/97. Due to the increased sickness of the NTC mills, the share of sales to NTC has declined from over 40 percent in 1991/92 to about 15 percent in 1996/97. MSCCGMF and the Maharashtra Cotton Procurement Scheme A5.16 The Government of Maharashtra launched its Cotton Monopoly Procurement Scheme in 1972/73. The objectives of the Scheme are: (i) to ensure fair and remunerative prices to the cotton growers in the state by eliminating middlemen, and (ii) to bring about stability and growth in the overall production of cotton in the state. This scheme assures cotton growers a guaranteed variety specific price for seed cotton fixed for the season. The Maharashtra State Cooperative Cotton Growers' Marketing Federation (MSCCGMF) is responsible for the enforcement of the procurement scheme (see Annex 2 for a more detailed discussion of the mechanics of the scheme). AS.17 The price support cum price stabilization system offered by the Scheme comes at considerable fiscal costs. By 1993/94, MSCCGMF's accumulated losses reached Rs. 1.4 billion ($40 million). Indeed, the Federation would have had to cease operations if not for a Rs 2.8 billion ($80 million) infusion by the State government.2 During the 1995-96 season, MSCCGMF was reported to have incurred additional losses amounting to Rs. 5 billion ($142 million), raising its accumulated losses to Rs 7.6 billion ($217 million).3 The MSCCGMF's financial losses eventually are subsidized by the Maharashtra state government. Based on available information, the state government has contributed Rs 514 million (US$ 43 million) in 1984/85, Rs 2.8 billion (US$ 229 million) in 1985/86, and Rs 2.8 billion (US$ 89 million) in 1993/94. 4 A5.18 Several factors contribute to the high risks and poor commercial viability of the scheme. First, the Maharashtra Federation commits to a guaranteed price, thus exposing itself to considerable price risks on very large volumes. The exposure to price risks is further amplified by the absence of price risk management mechanisms, such as hedging on futures markets. Although in principle the guaranteed price is to be set at the GOI's MSP--in general significantly lower than free market price, thus potentially reducing the exposure to price risks-- in practice the announced guaranteed prices could be much higher.5 The absence of a price discovery mechanism --which cotton futures markets could provide-- increases the difficulty of coming to a ' CCI income from carrying charges ranged from Rs 235 to580 million between 1992/93 and 1995/96 (CCI, 1997). 2 A. Gulati, S. Bhide, S. Bhagat, and S. Shroff, 1996, "Economic Reforms and Agricultural Parastatals: The Case of Cotton Corporation of India and Maharashtra Federation," IRIS-India Working Paper No. 19 Center for Institutional Reform and the Informal Sector, University of Maryland. 3 Cotton Statistics and News, No. 35, November 26, 1996. 4 A. Gulati, S. Bhide, S. Bhagat, and S. Shroff, 1996, "op.cit. 5 For example, although the MSP for the 1996-97 range from Rs 1150 to 1480 per qt., it is reported that the State is committed to paying Rs 2,100/qt.Cotton News and Statistics, No. 35, November 26, 1996. Annex 5 Page 6 of 34 commercially defensible guaranteed price. It also exposes the price setting process to political influence. During the 1995/96 season, an election year, the MSCCGMF was again reported to have incurred additional losses amounting to Rs 5 billion ($142 million). A5. 19 Second, porous borders and competition with neighboring states reduce the enforceability of the monopoly procurement scheme. It also severely limits the ability of the MSGCCMF to set a first payment low enough to reduce the Federation's exposure to risk. Instead, high guaranteed prices attract cotton from other states, raising the Federation's exposure to risk and the subsidy costs if final prices are set too high. A5.20 Third, improper grading of cotton contribute to significant losses andfurther contributes to the Federation's exposure to risks. Unlike other traders, MSCCGMF cannot adjust its prices within a season. According to the High Level Committee on Monopoly Cotton Procurement Scheme (April 1987), the grading of "inferior" cotton as "superior quality" was the major cause of the Scheme's losses, especially during the bumper harvest years of 1984-85 and 1985-86. This serious problem Table A 52:; Percent Distribution8 ade Cotton by MSCCGMF. persists as over 50 percent 3. _ .w of cotton proured ___8___- 57.1% 38.4% 4.2% 0O.3% 10.0 of cotton procured 1990-91 47.3% 45.5% 6.7% 0.5% 0.1% between 1989/90 and 1991-92 59.3% 37.9% 2.5% 0.2% 0.1% 1993/94 was graded as 1992-93 41.3% 50.0% 8.4%° 0.2% 0.0% superior and about 95 1993-94 50.8%° 46.1% 2.9%1 0.2%, 0.0% average 51.2%1 43.6% 4.9%1 0.3% 0.0% percent as FAQ or above Source: MSCCGMF. (Table A5.2). A5.21. Fourth, the Price Fluctuation Fund provision rules are insufficient. Under the scheme, only 25 percent of the annual profits generated by the Federation can be credited to the Price Fluctuation Fund and cover for the accumulated or future losses, while the 75 percent balance is remitted to farmers as a bonus payment. This is clearly insufficient to cover for the risks faced by the Federation, and the state government has to come in and cover for the net losses incurred. Even with a 100 percent replenishment rule, there are compelling reasons to believe that the Price Fluctuation Fund can never become financially viable. Recent empirical work on commodity prices (Deaton, 1992) shows that most commodity prices do revert eventually to their mean --a requirement for a stabilization fund to be viable-- but only very slowly, with an average reversal time measured in years, not months. Because of this, reserves have to be very large to be effective. This calls either for a very large contribution by the Maharashtra state government, an option which is neither politically possible nor fiscally tenable. Alternatively, it calls for a very low first guaranteed payment which, for the reasons outlined above, is neither politically nor practically acceptable. A5.22 Finally, the absence of competition under the Monopoly Scheme reduces the incentives for improving the efficiency of marketing operations and unnecessarily raise operating costs. A5.23 The Maharashtra Monopoly Procurement scheme is not sustainable because of its inherent large fiscal costs andfinancial risks. The central government appears to have reached a similar conclusion. For the last few years, the central government has been pressing the Maharashtra state government to devise a better alternative to the monopoly procurement scheme. In the absence of an alternative solution, the central government grants extensions to the Maharashtra monopoly procurement scheme, the most recent two-year extension expiring on July I st 1998. Better alternatives to the monopoly procurement can be designed which are outlined in the final section of this Annex. Annex 5 Page 7 of 34 Cooperatives Marketing in Other States A5.24 Cooperatives account for only a small share (4 percent) of marketed cotton. Cooperative cotton marketing generally follows two methods:: * pooling of farmer members' cotton and selling the pooled produce in the assembling markets; and, * outright commercial purchases from farmers for the sale to cooperative or public sector mills. A5.25 This system is best exemplified by the cooperatives in Gujarat. Under the Gujarat cooperative system, members hand over their seed cotton to the societies for processing and sale of pressed bales. Advance payment of 65-70 percent of the estimated value of the seed cotton is made at this stage to farmer members. After processing and sale at the end of the season, the average variety-wise price is worked out and the difference between final and advance prices is paid to the farmers. A5.26 The cooperative marketing structure in Gujarat is a three tier system. It consists of the apex Federation at the top, 7 regional marketing unions at the middle level, and over 500 primary level societies at the village level. A total of about 200,000 cotton farmers are members of the Gujarat State Cooperative Cotton Federation (GUJCOT). GUJCOT has established a High Volume Instrument Testing Laboratory (HVITL) to facilitate quick and reliable quality evaluation of farmer output based on technical standards. Cooperative marketing in Gujarat has acquired added strength in recent years owing to the allotment of export quotas. A5.27 The cooperative marketing arrangement has been beneficial to members in several respects. One recent study found that farmers garnered 93 to 96 percent of the final price of cotton.6 In addition, the cooperatives ensure adequate and timely supply of credit, given partly in kind and partly in cash. These loans are recovered from the sales proceeds after marketing the cotton. This arrangement has also helped in the dissemination of improved production practices among farmers. A5.28 Besides Gujarat, cotton cooperatives are also developed in Punjab, Haryana and Karnataka. These cooperatives also receive export quotas when unsold stocks accumulate. C. Cotton Ginning and Pressing Industry A5.29 Cotton ginning is largely a private sector activity. Ginning involves separating the cotton lint from the seed to which it is attached. On average, cotton lint accounts for two-thirds the weight of kapas and cotton seed accounts for the remainder. After ginning, the cotton lint is pressed into bales of standard sizes (170 kg) in pressing factories for ease of transport and storage. Currently, there are 3,237 ginning and 795 pressing factories distributed across nine states employing about 250,000 workers. Gujarat has the highest number of ginning (1,002) and pressing (189) factories (Table A5.3). Almost 90 percent of these factories are privately owned with the rest belonging to the cooperative sector. Besides ginning their own seed cotton purchases, most of these factories undertake ginning and pressing of cotton for other merchants and mills at fixed charges. Until 1996, these charges were fixed by the state governments under the Cotton Ginning and Pressing Factories Act. The charges varied from state to state and are revised yearly (See Appendix Table 5.5 for a listing of ginning charges). Ginning and pressing charges amount to nearly 3 to 4 percent of the combined sales returns from cotton lint and cottonseed. 6 K.S. Parmar and K. Ramchandran (1992): Cotton Growing & Marketing in India. Tecoya Trend. pp. 141-142. Annex 5 Page 8 of 34 A5.3 0 Most of the gins are of the Table A5. 3CQotton Ginnng and Pressing Factories In India, 1995 roller type. The adoption of modem roller gins, for which pre- and post iUbi cleaning and pneumatic handling could 2,52 14 1 be integrated has been slow. The large 1ayana 37 2,552 114 115 54 Punjab 137 3,008 834 240 106 investments involved and the labor Rajasthan 220 1,701 506 100 67 displacing nature of the technology Gujarat 1,002 15,810 372 10 189 against the backdrop of low labor costs Madhya Pradesh 246 2,412 1,387 64 and the notified fixed conversion Maharashtra 387 3,010 4,477 12 151 and the notified fixed conversion Andhra Pradesh 446 5,785 79 charges by weight, militates against the Karnataka 368 2,653 60 more widespread adoption of this type Tamil Nadu 394 3,034 25 of technology. Saw gins which are Total 3,237 28,493 19,162 477 795 suitable for short staple cottons are Source: Report of the Committee on Modernization of Cooperative being adopted in the northem states Ginning and Pressing Sector and East India Association, Bombay. where higher volumes of production per unit (irrigated ) area make them more economically viable. A5.31 The ginning and pressing sectors are characterized by inefficient and outdated technologies. Almost half of the gins operating in India is over 75 years old and some of them are about a century old. Most of the ginning factories have no pre-cleaning facilities, and handling activities are done manually. Because of their age, most gins suffer from low productivity, high costs and high percentages of trash in the pressed bales. While this may be acceptable for older spinning mills, cotton contamination poses serious problems for the modern high speed and automated spinning mills, both in terms of production disruption and inferior product quality which is ultimately reflected in defective quality of the final product. A5.32 Cotton lint Table A5.4 Degree of Cotton Contamination in Seiected Countries contamination remains a ;; = 1 D * serious problem, 13 22o~tt 6S 1fth ultimately lowering the Shankar 4/6 25 9 19 72 quality of cotton products. DCH-32 21 7 21 72 In 1991, the International MCU-5 18 9 16 75 Textile Manufacturers China Xinjiang 16 4 14 82 Federation conducted a Sudan Shambat 10 1 16 83 Federation conducted a USA Acala 9 1 14 85 world-wide survey of 201 Texas High Plains 45 5 9 86 mills and 1,465 samples of Rio Grande Valley 27 5 9 86 cotton covering 81 cotton Cameroon Cameroon 23 2 13 85 types to identify the order Note: Based on 1993 ITMF survey covering 1575 samples of 82 cotton descriptions from 235 mills in 30 countries. of contamination (Table Source: C.H. Mirani and K.S. Parmar, 1995, Baling the Gold, Bombay: Tecoya, Table 6. A5.4). For India, 30 samples consisting of 5 different cotton types (Sankar 4/6, H-4, DCH, MCU and others) were analyzed. Sixteen contaminant categories were identified and the seriousness of the impact on the production process was grouped as "serious", "moderate", and "insignificant". Of the 81 cotton types ranked in terms of decreasing contamination, India ranked 1, 2, 3, 5 and 8. In terms of "stickiness", the five Indian cotton types ranked 6, 18, 25, 27 and 32. In terms of the presence of seed coat fragments, India placed at number 3, 5, 6, 9 and 17.7 The main sources of contamination observed were strings and fabrics made of jute/hessian and woven plastic, organic 7 S. Chaudhuri, 1994, "Cotton Yarn Spinning in India, The Competitive Advantage and Prospects," ICRA Sector Focus Series #1, ICRA, New Delhi. Annex 5 Page 9 of 34 matter like leaves, feathers, paper, and leather, inorganic matter like sand, dust, rust, and metal wire, oily substances and chemicals like grease, oil, stamp color and tar. The level of contamination in India is much higher relative to other cotton producing countries, such as China, Sudan, Cameroon and the USA. A5.33 Improving the efficiency of the ginning and processing sector requires greater attention to handling. While it is true that outdated ginning technology in the sector raises operating costs, there is a greater loss in the value of cotton lint output due to high levels of contamination from poor handling. Moreover, upgrading the ginning technology, such as the replacement of single roller units by double roller units,8 and improved maintenance require only slight modifications/investments, compared to the wholesale replacement of the existing ginneries. The more serious problem, however, lies with the poor of handling of seed cotton and lint. Infrastructure facilities for proper handling are inadequate or lacking in most ginneries resulting in high contamination levels. These include inadequate transport infrastructure, storage facilities, cleaning and handling equipment at ginneries, primary and secondary markets, and poor training and supervision of workers. Improved handling to reduce contamination will require several measures: * improving transportation infrastructure, including roads, etc, * improving storage and handling at primary and secondary markets, * storing of seed cotton on clean, concrete foundations and use of labor to remove contamination, etc., * installing of chalnis to remove immature and offgrade cotton, * installing of conveyor systems for transporting kapas from heaps to gins and from gins to presses, * better methods of removing trash and foreign objects from both seed and lint cotton (contrary to Indian wisdom, these does not necessarily include seed cotton and lint cotton cleaners), * installing automatic seed cotton feeders, * storing of lint cotton prior to pressing in good lint bags on platforms or in houses dependent on weather, and * supervision of the work to include the realization of the importance of clean cotton. A5.34 Several factors discouraged efficiency improvements in the ginning and pressing factories. First, until 1996, the fixing of ginning and pressing rates by state governments, which are not remunerative for the factories, and the small scale reservation through licensing of ginning discourage productivity improving investments. Second, the storage (lifted in 1997) and Selective Credit Controls (lifted in October 1996), increase the costs of doing business further reducing the profitability of ginning. Third, weak implementation of a grading system for lint cotton that would reward or penalize the ginner for the high trash content in the ginned bales discourage investments to improve quality management. Textile mills have to be willing to pay for improved cotton quality. A5.35 After ginning, cotton lint is brought to the terminal markets, where it is sold to either the textile mills or exporters. There are few of these types of markets located in major textile industry centers. Bombay, Ahmedabad, Coimbatore, Madras, Nagpur, Indore, Sholapur, and Kanpur, are some of the important terminal markets. However, as the cotton textile industry is now widely dispersed in the producing areas, the importance of many terminal markets has 8 Double roller units are more efficient than the single roller units Annex 5 Page 10 of 34 declined. Several textile mills buy directly in the adjoining markets, though many still prefer to employ the services of merchants and commission agents to ensure quality and better bargains. D. Cotton Grading and Classification A5.36 Grading is an important aspect of cotton marketing because it establishes the spinning utility and hence the market value of cotton. Spinners depend on accurate grading and classing information to facilitate purchase of the appropriate quality of cotton required for the efficient manufacture of the type of yarn demanded by their customers. In turn, quality premia associated with quality differences provide a signaling mechanism to agents in the cotton marketing system on the quality characteristics demanded. These premia encourage farmers, traders and ginners to adopt appropriate methods of picking, storage, packing, transportation, and ginning to deliver and preserve quality characteristics desired. A5.37 Grading also assumes considerable significance, because of the sensitivity of cotton output to a multitude offactors which affect quality. Few other industrial crops show as much variation in critical characteristics as cotton does. There are numerous characteristics of cotton like staple length, fineness, strength, maturity, feel, color, luster, trash and moisture content, each of which has a bearing on yarn quality and the final textile good produced. This problem is further confounded by the large number of varieties in India and the divergent conditions under which they are grown. Besides inter-seasonal variations, the same variety displays variations from location to location within the same season. Apart from the genetic base, the quality of cotton is also influenced by factors such as soil, climate, soil and atmospheric moisture, nutrient application, pest and disease infestation, and picking time and methods. The quality and development of lint are affected by these factors when it is still in the boll and also when the boll opens and it is exposed to natural elements. Since the bulk of cotton in India is marketed as kapas, varietal admixture prior to ginning is also a problem. In view of the multiple sources of variation, grading therefore becomes critical in determining actual quality of cotton traded in the market. Minimal Formal Grading of Kapas AS.38 Seed cotton grading standards exist. The quality of raw cotton is affected by a number of physical properties: color, luster, fineness, trash content, level of stained and immature fiber, feel and moisture content. The color of the fiber affects the dye absorption capacity and finishing. Stains in cotton affect dyeing uniformity. Fineness and maturity affects the strength and length of the fiber. Luster measures its ability to reflect light. Trash results in lower quality output. Moisture affects fiber strength. These qualities influence the type and quality of the final product produced from the raw materials. The Directorate of Marketing and Inspection of the Government of India has laid down specific procedures for grading kapas based on these characteristics. These procedures cover the drawing of samples, steps in grading, and the grade specifications of different varieties. Grade standards for each variety in each market are to be established based on consultation with local trade, the State Agricultural Department and the Directorate of Marketing and Inspection of the GOI (Table A5.5). In addition to grading, staple length and ginning percentages are also determined. A grading officer determines the grade and staple length of the kapas, while the ginning percentage is determined by a grader in charge of ginning. Annex 5 Page 11 of 34 A5.39 Seed cotton grading Table A5.5: Typical Set of Specifications of Different Grades of a Variety. standards are used infrequently. The3=i _ Special Shall be white in color and silky in feel (characteristic of the above described kapas grading variety); clean and free from stained or immature kapas; procedure, however, is hardly shall be free from added moisture followed in any market, except a few A Shall be white in color and silky in feel (characteristic of variety); slightly leafy; free from stained and immature centers like Hubli in Karnataka and kapas; shall be free from added moisture the procurement centers in B Slightly dull in color, soft in feel, slightly free from leaf, Maharashtra. CCI, the marketing stained or immature kapas, shall be free from added moisture. federations and private trade all C Slightly dull in color, soft in feel; moderately free from recognize the need for grading. A leaves, stained or immature kapas, shall be free from added centrally sponsored scheme launched D moisture cetrll sDull in color, good in feel, contains stained and immature in the 1 970s included the kapas; shall be free from added moisture. establishment of about a dozen x Any kapas that does not fall into any of the above grades. grading centers in different markets to Source: K.S. Parmar and R. Ramchandran, 1992, Cotton Growing and promote grading, but when Central Marketing in India, Bombay: TECOYA Trend. funding came to an end, the grading centers also became inoperative. A5.40 Maharashtra is the only state, which implements grading on a large scale. When procuring seed cotton, MSCCGMF follows a simpler system with five grades (super, fair average quality [faq], fair, X-1, and X-2) for each variety.9 A grader first identifies the variety of kapas (with reference to the characteristics of different varieties in the prescribed schedule), then classified the kapas according to grades. However, as noted earlier, the poor implementation of cotton classification under the monopoly procurement scheme contributes to the financial woes of the MSCCGMF. Lint Classification and Grading A5.41 Unlike kapas, lint classification and grading are done on a more extensive scale. Trading of cotton bales is carried on the basis of grade and staple. The East Indian Cotton Association (EICA) maintains official cotton standards for each of the commercially grown varieties (with different staple lengths) as per the schedule in each season. The standards are prepared after a survey of the crop is undertaken by the Sworn Surveyors for length and uniformity. Representative samples are evaluated in the laboratories of EICA and the Central Institute for Research on Cotton Technology for staple length and uniformity ratio. These readings are compared with those of the Surveyors and are then selected for preparing standards. After they are agreed to by the Standards Committee of EICA, they become official standards and the trade is subsequently informned of the same. A5.42 Each variety has six grades: extra superfine, superfine, fine, fully good, good, and fully good and good. The basic grade of each variety is 'fine' as over 50 percent of a given variety would be equal to this grade. Lint grading is still done manually. Samples are drawn from fully pressed bales for determination of grade and staple length. Normally, a 2 kg sample is drawn from one bale out of every lot of 50 bales (or from 2 to 3 bales from lots of 100 bales). Grade classification is based on appearance which is judged visually, taking into account color, leaf, and preparation. Staple characteristic is judged by sight and touch and by measuring a typical portion of fiber sample pulled out by hand and compared with that of the official standard. 9 Disputes on grades are handled under the provisions of the Agricultural Produce Market Committee Act - Maharashtra Agricultural Marketing Rules, 1963 - Section 10 with no opportunity for appeal at the present time although it has been recommended to appoint a district level appellate committee. The dispute settlement committee is composed of 5 member (the vice-chairman of APMC who acts as chairman, 2 grower members, I trader member, and I APMC expert member). Annex 5 Page 12 of 34 A5.43 In practice, the bulk of the cotton lint traded in India is graded on the basis of variety or staple length only, instead of standards described above. Other important fiber attributes --trash content, fiber strength, uniformity, maturity-- are not scientifically included in determining the value of the cotton fiber. For example, the generic terms J-34 and H-4 represent more a differentiation according to staple length than seed variety of the numerous (and mixed) seed varieties of so-called America cotton grown in the Northern region. The practice of grading according to variety is becoming increasingly inconsistent with the technological and quality requirements of a textile industry that is modernizing and increasingly oriented towards exports markets where quality requirements are exacting. These inadequate grading practices result in considerable under-spinning by the textile industry or inefficient utilization of raw cotton by the textile industry, and the yarn quality suffers. For example, in the higher yarn counts, technological change in the Figure A5.2: Quality Premium Between J-34 Superior Grade Purchased By An EOU And The spinning industry is resulting in the replacement of 1 0,000- Average Market Price 12,000 RPM ring frames with high speed 16,000-18,000 RPM units. Mills are forced to shift to longer staples to 70.0 avoid breakage. The low strength associated with the staple 60.0a varieties is due to the number of immature fibers which Al w should be under 25 percent for normal spinning needs. In s the medium counts, an increasing percentage is being spun , 40.0 - with open-ended spinning with very high RPM as 30.0 compared to spindle speeds of older conventional ring 20.0 V frames. This requires cottons with greater strength, so that o11111 II staple length is no longer the most important fiber attribute. High speed looms also require increased yarn strength. 9 9 A5.44 Export-oriented spinners do pay a premium for BOU -Average better grade cottons. Export-oriented units need to remain -_S_ Caudury_ 196__ "ottn_ Scto .~~~ol maktan.a Source: S.Chaudhury, 1996, "Cotton Sector cost and quality competitive on the world market and can Study, World, ankbackgroundworkingpaper ill afford to rely on inadequate grading practices which continue to prevail in the cotton market. Increasingly, spinning and textile mills are introducing new technologies (such as High Volume Instrument Testing Laboratory) and practices with their own cotton traders to overcome the uncertainty and risks associated with the shortcomings of the current grading system. Quality conscious mills are willing to pay significant premiums and incur significant additional costs to acquire good quality cotton. Industry sources indicate that individual mills and traders invest considerable resources in improving their market intelligence and knowledge about quality conditions of the crop, and in identifying good quality cotton. Industry sources estimate that about 20 to 25 percent of cotton lint is now subject to more systematic and precise grading practices. Figure A5.2 shows the average quoted price in the market for J-34 relative to the price paid by an export-oriented spinner superior grade J-34. As indicated by Figure A5.2, the quality premium ranges between 10 and 20 percent of the average market price under normal market conditions. Supply and demand factors, however, influence strongly the magnitude of the quality premium. The production shortfall during the 1994/95 season practically eroded the quality premiums. A5.45 Poor grading practices are intimately related to poor seed management identified in Annex 4. As indicated in Annex 4, at least 40 percent of India's seed requirements are mixed seeds. Consequently, cotton maturity and uniformity are major problems. Immaturity coupled with poor ginning results in the presence of a significant amount of seed coat fragments in the ginned bale and eventually an inferior quality of fabric. Immaturity causes uneven dyeing due to differential dye take-up. This is less of a problem in the manufacture of denims (counts 1-10) Annex 5 Page 13 of 34 which uses the natural dye (indigo), but is for the remaining manufacture of textile which utilizes chemical dyes. The multiplicity of varieties has resulted in about 80 to 100 varieties being grown in India, of which about 20 constitute the major part of production. The production of newly introduced varieties has not remained genetically stable, except for the cotton grown in the Northern zone, some areas of Gujarat and the Guntur tract of Andhra Pradesh. Variety quality has gradually deteriorated with the passage of time. Examples are numerous. When Varalaxmi was introduced in the Raichur tract, the cotton was capable of spinning for 80s yarn counts. As this variety spread to Maharashtra, Gujarat and Madhya Pradesh, quality deterioration resulted in cotton suitable for spinning 40s yarn count. Examples of other varieties include H-4 and JKHY- 1. A5.46 Inadequate grading (and seed management) practices are the natural response to a historically protected and inward-looking textile industry. Until recently, the inward-looking and highly protective textile policies produced very few market incentives for managing quality throughout the cotton and textile industry and cotton marketing system. Market conditions are however changing rapidly with the textile industry becoming increasingly export-oriented and gradually subjected to international competition at home. This calls for rapid and concerted effort at introducing premia for cotton fiber attributes into the current marketing system. If the Indian textile industry is to capture the market opportunities created by the dismantling of the MFA (see Annex 6 for more detailed analysis), price signals must flow freely from textile producers to cotton growers; the ginning industry must be rewarded for preserving the value of fiber attributes; cotton growers must be informed through transparent and consistent grades about the value of fiber attributes, and rewarded for producing the quality of cotton demanded by the textile industry. This would require adjusting the role of government agencies and regulations from an interventionist and punitive approach towards a new supportive role in the standardization of cotton and contracts by the private sector, in the dissemination of market information, and in training. AS.47 International experience attests to the complexity and difficulties in successfully implementing grading systems for cotton and cotton lint. Different countries have adopted different methods of grading and classifying cotton and lint with varying degrees of success. Their experiences highlight several lessons (Text Box A5.3). First is the importance of transmission to all market participants of information on the quality characteristics demanded by downstream consumers--textile mills. Second, the system of premiums and discounts need to be market driven, otherwise it will not be adopted by traders. Third, the establishment of reliable systems is critical to prevent rent-seeking activities. E. International Marketing of Cotton Cotton Exports Are Restricted A5.48 India has a long tradition of exporting cotton lint. As early as 1764, India is known to have exported 10,000 bales of cotton lint to Great Britain. Before the great depression in 1929, India's cotton exports peaked annually at 3.8 million bales (638,000 mt). Cotton exports declined thereafter, following the growth of the domestic textile industry. With the partition of the country, exports were cut drastically as 40 percent of the country's annual cotton production was lost to Pakistan. Since then, cotton export policies have largely restricted exports to non-staple varieties like Bengal Deshi and Assam Comilla. Only in the 1970s, when India regained self-sufficiency in cotton were exports of staple resumed (Figure A5.3). Annecx J 'ige 14 o- ,4 Text Box AS.3 Cotton Grading in Other Cotton Producing Countries Pakistan, The Pakistan Cotton Standards Institute established the grades and stsndards for seed cotton and cotton lint currently applied in a limited scale, pending formalization of the system. Prescribed prcmiums and discounts are applied and anmound by the government, based on two main charactertcs-non-lint content and degree of reflectance. Grading is done by giers. Price diffarentials for quality, however, have not been extensively adopted. Ginners do not always follow the announced premia and discounts, but rather follow their own standards depending on what they expect to receive from textile mills or exporters. Even if they do adopt the grading system, since alimost 95 percent of themer output of raw cotton is.sold through intermediaries, it is the intcrmediaries who directly receive the benefits from grading. The poor dissemination of price information in turn limits farmer's ability to negotiate better deals with the intermediaries and ginners. USA. The USA has the most scientific and systematic grading system in the world, wherein practically every bale of cottonis being evalateI and daased by the U.S. Dep ittnnt of Agriculture. Unlike other countries, however, all cotton is processed and baled byproducers and; sold: as tint. Although classiication is not mandatory, farmers consider it essential for marketing their crop and for participating in the government price support program. There are 18 cotton classing centers sategially located across the country. Thescenters make use of the most modern testing equipment for high volume instrument (liv!) classification. Classification ks based on fiber length, length unitbrmiy, strtngth, micronaire, color, preparation, leave and extraneous mater. The Hl system compltes the tests in mter of seconds and the results on the fiber tests are immediately available to producers or their authorized agents tugh th intemet, diskettes, staples, and printouts. In several countries, namely Egypt, Brazil, China, Tanzania, Mali, and Uzbekistan, formal classification is implemented by the state or parastatal agencies. The grading systems of some of these countries are described below. Egypt. In Egypt, each variety is classified into several grdes on the basis of fiber quality ranging from good' to ,fully good'. Prices for grades within each variety are deternined by thle Ministry of Economy and Foreign Trade. Farmers are pd80 percent of the value of the cotton after the first evaluation at the collection center, and the balance was given after the final evaluation of the ginner. Recent liberalization enables any licensed ginner to perfbrm the grading procedure. Tanzania. In Tanzania, seed cotton from farmers is classified into two grades: white clean cotton (AR) and dull stained cotton (BR), After hand-picking, farmers grade their cotton before delivering it to the society store. Grade inspection at th buying post is done by primary society commitee members on duty. Cotton inspection in the society store however, has deteriorated. Problems of mixing high and low quality cotton lowered the AR grade of cotton delivered toginneries. Competition at the society level, corruption, fear of members, and a general lack of appreciation of the importance of quality has undermined the grading system. Mali. In Mali. the seed cotton is divided into thrwe grades based on color and trash. Producers, who are organized into village associations and big individual producers transport their seed cotton directly ttginneries and are paid on the basis of their fiber grade. In viilages where farmers are not organized, the seed otton is ciassified by the govcrnmentparastatal Compnie Malienne pour Ie Developpement des Textiles (CMDT), China. In China, seed cotton like lint is classified into 7 classes and a sub-standard class. Each class has three grades except the sub-standard class with two grades. The classification of the cotton according to quality, staple length, moisture content. trash content, and ginning outturn is performed byginneries operated by the Cotton and Jute Company, an agency of the Ministry of Internal Trade, who also makes the payments. Source: F. Gillham, et. at., 1999, i Cotton Production Prospects for the Next Decade, ' World Bank T echnical Paper No. 287 and M.B. Lal and K.S. Parmar. 1995, Cotton Marketing, Bombay: TECOYA. A5.49 Currentl v, Indias csotion mu7 exports aIre liim7itedi and highlv erratic. Over the past II years, cotton lint exports fluctuiated considerably (Figure A5.4). They ranged from a low of 7,000 mt in 1987/88 to over 200,000 mt in 1989/90, 1990/91 and 1992/93. These large fluctuations are primarily the result of the (GOI's export policy whichi sets quotas oii an anilual basis (See Aninex 2). Exports quotas change considerably from vcar to vear. and ratige frotn 7.000 to 3(00,000 tnt of staple cotton dependinig on the supply/demanld situation in any particular year.. The principal purpose of these export controls is to assure cottoin supplies at low prices to the textile industry. The quotas seldom constitute more thani a small percentage of productioll, for example, the 1994/95 quota Was set at 5)00.000 bales, whichi is equivalent to about 5 percenit of total production . Annex 5 Page 15 of 34 A5.50 Arbitrary implementation of Figure A5.3: Indian Cotton Exports in Volume and Value, quota system undermines cotton 500 export operations and credibility. Even when the total volume of the a400I export quota has been decided, it is 300 released in ad-hoc installments at v 200 different intervals depending on the EI 9 ~~0 100 government's assessment of the o cotton supply situation. This creates 0 considerable uncertainty for y o m exporters. In 1994-95, quotas were > '1 _ Z Z suddenly suspended and shipments Volume 000 mt Value $ million stopped temporarily for contracts entered into with overseas buyers. Source: FAO Trade Yearbook and USDA/FAS 1995 Annual Cotton Report. These discretionary and abrupt suspensions undermine the credibility of Indian exporters with respect to delivery and shipment Figure A5.4: Cotton Exports Are Generally Below shexpoters Export Quotas schedules. 350 300 A5.51 Indian cotton lint exports suffer from large 250 discounts in the world market. Several factors E 200 contribute to low export prices received in the world 3 150 market. First, the perceived lower and lack of consistent 100 quality of Indian cotton result in discounts in cotton 50 prices. Second, because of poor quality management, oY the cotton has to be regraded by the importer. Third, the - state agencies, who dominate cotton exports, have little expertise in international cotton marketing and, until Exports o Quota recently, any incentive for profit was masked by the MIEP. Consequently, most CCI and state agency export Source: Office of the Textile Commissioner, Bombay. sales are made on a FOB basis to international cotton merchants, who use Indian cotton merchants as their agents. Fourth, the arbitrary implementation of the export quota system increases the uncertainty and risk Table A5. 6 Agency-wise pattern of cotton exports, 00 bales, 1988/89 to 1993/94 of buying cotton from India. ttoc X X < CCI 9 614 246 2 601 146 A5.52 The cotton lint export State Cooperatives, 0 547 450 1.3 652 166 policy discriminated against the including NAFED non-cooperative private sector. Private Traders 1 100 390 1 CCI and state agencies till 1994- Total 20 1261 1086 3.3 1253 312 II. Non-staple Cotton 95 season exclusively CCI - 7 monopolized exports (Table Private Traders 56 103 106 74 124 77 A5.6). Private trader exports Total 57 110 106 74 124 77 were restricted to the non- Source: Office of the Textile Commissioner, Bombay. spinnable short and non-staple varieties like Bengal Deshi, Yellow Pickings, and Assam Comillas. It was only during the 1995-96 season that cotton exports were opened to private trade. Exports are now awarded through auction. Annex 5 Page 16 of 34 Cotton Export Policy Depresses Cotton Lint Prices A5.53 India's cotton export policy taxes cotton lint. The Figure A5.5: Nominal Protection Coefficients for MCU-5, S-4, and restrictions on cotton lint exports J-34 Cotton Lint (Exportable Hypothesis) are depressing domestic cotton lint prices below international prices, 2.00 l notably for the ELS cotton grown 1.80 in mostly rainfed areas. The 1.60 nominal protection coefficients 1.20 - (NPCs) for the MCU-5, S-4, and J- 1.00 34 varieties have been used as 0.80 indicators of the degree of 0.60 protection or disprotection for 0.40 ELS, medium and short cotton 0.20 . . I,,,,,,,,, . ,. lints, respectively. Cotton lint has 71 73 75 77 79 81 83 85 87 89 91 93 95 been implicitly taxed since the late __+MCU-5_a S-4 J-34 1970s (Figure A5.5). Of the staple groups, extra long staple cotton Source: G. Pursell and A. Gupta, , "Background Statistics, Protection and Incentive lint is the most heavily taxed with Indicator, 1965-95, World Bank. NPCs hovering around 0.65 since 1980. Short staple cotton lint prices (J-34) have been about 15% below international prices on average since 1980. Medium staple cotton lint (S-4) has been at export parity level, except in those years when domestic production was in short supply. The low cotton lint prices represent a significant source of implicit subsidization to the domestic textile industry. The price differential between domestic and international prices has narrowed somewhat since 1991. Two factors explain this reduction in the gap between domestic and international prices. First, the textile boom fueled by liberalization and devaluation is exerting strong demand pressures on the domestic cotton market which are driving upward the domestic cotton lint prices. Second, pest problems and associated production shortfalls further exacerbate the upward pressure on domestic cotton lint prices. Oilseed Policies Compensated Cotton Growers for the Low Cotton Lint Prices Figure A5.6: Percentage Share of Cotton Seed Oil in Edible Oil Market, 1990-1994 A5.54 Oilseed policies spillover to the cotton Other Edible Oils sector via their impact on cotton seeds. Ginning 87% of kapas produces two joint products: cotton seed and cotton lint. In turn, cotton seed produces two joint products, cotton meal and cotton seed oil. Cotton seed oil, an edible oil, accounts for about 16 percent of the weight of oil o v b cotton seed. Over the last 15 years, inadequate 5% Seed Oil domestic production prompted the GOI to 8% pursue an import substitution strategy, which involved very high non-tariff protection till Source: FAO Production Yearbook and A. Gulati, A. Sharma, & D.S. Kohli, 1996, "Self-sufficiency and Allocative Efficiency, Case of Oilseeds in India." Annex 5 Page 17 of 34 1994/95.10 In the early 90s, edible oil trade restrictions resulted in nominal protection levels for the major edible oils (groundnut, mustardseed, Figure A5.7: Nominal Protection Coefficients for Cotton Seed soybean, sunflower, and coconut oil) of 74 (Importable Hypothesis) percent to 233 percent. The high domestic 35 edible oil prices spilled over to cotton seed oil, which accounts for about 8 percent of 3.0 . domestic edible oils supply (Figure A5.6). In 2.5 the case of cotton, cotton seed oil accounts for 2X0 about 14 percent of the value of J-34 seed 2.0 cotton, and 3.5 percent of the value of MCU-5 1.5x seed cotton. This results from the fact that the .7 cotton seed accounts for 25 percent of the 1.0 value J-34 seed cotton, and 6 percent for 0.5 MCU-5. 71 73 75 77 79 81 83 85 87 89 91 93 95 97 A5.55 Oilseed policies gave a high and positive protection to cotton seeds. As shown in Figure A5.7, the .PCs for cotton eed, Source: G. Pursell and A. Gupta, , "Background Statistics, In Figure A5.7, the NPCs for cotton seeu Protection and Incentive Indicator, 1965-95, World Bank. remained over 1.5 for most of the 1980s, reaching 2.9 in 1988. NPCs for cotton seed declined dramatically since 1990 as the combined result of near self-sufficiency in edible oils, exchange rate devaluation, free edible oil imports since 1994/95 and lower tariff protection. The renewed domestic protection observed in the last two to three years reflects production shortfalls in cotton as well as a shift to other oilseeds (e.g., soyabean). Figure A5.8: Nominal Protection Coefficients for Seed Cotton (Exportable Hypothesis) A5.56 High cotton seed prices compensated cotton growers for the 31 low cotton lint prices, but more so for 2.5 - farmers growing shorter staples than longer staples. The high degree of 2 X disprotection of cotton lint observed 1.5 earlier (Figure A5.5) has been partly 1 compensated by the high protection afforded to cotton seeds as a result of 0.5 the oilseed import substitution policy. o _--_-_-____ In effect, edible oils consumers cross- 71 73 75 77 79 81 83 85 87 89 91 93 95 97 subsidized cotton growers through the joint-product of cotton lint production. |*+MCU-5 n S-4 -34 1 The equivalent kapas or seed cotton Source: G. Pursell and A. Gupta, "Background Statistics, Protection and NPCs for short (J-34) and medium (S- Incentive Indicator, 1965-95, World Bank. 4) staples generally exceeded 1.0 during the 80s, although their cotton lint NPCs were generally less than 1.0 (Figure A5.8). The benefits from cotton seed protection, however, did not provide as much buffer to farmers producing extra-long staples for two reasons. First, ELS cotton lint was more heavily taxed than the other cotton varieties. Second, the relative value of cotton seed in the total value of ELS seed cotton is considerably lower than the others. As a result, ELS seed cotton NPCs were consistently below 1.0. 10 See World Bank, 1996, "The Indian Oilseed Complex: Capturing Market Opportunities, " Report No. 15677-IN, Vol. I & II, Washington, D.C.: World Bank for more detailed discussion on the GOI's oilseed sector policies. Annex 5 Page 18 of 34 A5.57 The overall protection of seed cotton helped maintain the relative attractiveness of cotton cultivation compared to other major crops. Appendix Figure A5.1 presents the ratio of the wholesale price indexes of MCU-5, S-4 and J-34 seed cotton to the wholesale price indexes of oilseeds, cereals, pulses and sugar, khandasari and gur. Seed cotton prices in general moved favorably relative to all other crops between 1981/82 and 1993/94, except pulses. A5.58 The recent oilseed policy reforms will have a depressing effect on seed cotton prices. As of July 1996, the import of most major edible oilsl1 is under Open General License (OGL) and tariffs have been brought down to 20 percent. The resulting reduction in protection will not only affect oilseeds directly, but also affect the cotton and textile industry indirectly through its impact on cotton seed prices. The protection of cotton seed which was important in providing a buffer against the disprotection of cotton lint is likely to disappear. A decline in cotton seed prices could have a notable impact on cotton production patterns, especially the northern regions growing shorter staples. A5.59 The impact of oilseed policy reforms will vary across cotton producing regions. Farmers in the Northern Region will be more greatly affected by the expected decline in edible oil prices Figure A5.9: Cotton Imports, Volume and Value that those in the Central or 300 __________________________ Southern Regions. Farmers in 30250 - the northern regions, because ~ 200 l they grow the shorter cotton 4 200 staples, will be more affected 150 by the expected decline in S 100 edible oil prices, since cotton i 100 l t l ll . L ll seed accounts for a larger. 0 ~~~~~~~~~~~~~~~share of the kapas. 3 3> 33333333 3^ 3z Cotton Imports _Import Volume ~Import Value l_____________________Imort_Volue _--*-Imort_Valu A5.60 India, until the 1970s, was a major importer of Source: FAO Trade Yearbook and USDA/FAS 1995 Annual Cotton Report. cotton. Excess domestic demand for long staple cotton necessitated continuous imports prior to the 70s. In the 1940s, India's combined imports from Egypt, Jordan, East Africa, and the USA reached 600,000 bales, increasing to about 800,000 bales by the 1960s. In 1970, CCI was established and imports were canalized through CCI. Imports were still subject to quota restrictions and tariff barriers as existed earlier. Domestic allocations of imported cotton were assigned by the Textile Commissioner to mills and trading of allocations was banned. But as India approached self- sufficiency in the late 1970s, especially in long-staple cotton, imports declined (Figure A5.9). During the 1980s, cotton imports were quite negligible. A5.61 Low domestic prices limit cotton imports. Imports by textile mills were permitted under the Advance Licensing Scheme starting in 1986-87. However, imports by mills remained small 11 Coconut oil, RBD palm oil and palm stearin remain canalized. A subtle distinction has been introduced by the 1995 Export-Import Policy guidelines. The recently published guidelines classify all major vegetable oils in crude form -- the standard quality on the oil world trade-- as canalized by the STC, and only "edible grade" as freely importable. This distinction introduced by the Export-Import Policy may diminish the economic significance of the liberalization in the oilseed complex. Annex 5 Page 19 of 34 (172,000 bales in 1987-88 and about 400,000 bales in 1988-89). Since domestic cotton prices were generally lower than world prices, mills preferred to rely mostly on domestic cotton. A5.62 The lifting of import restrictions and the production shortfall in 1993/94 resulted in a surge in imports. In 1994, cotton lint imports were placed on OGL, with a zero import duty, mainly prompted by the supply crisis-- output in 1994 was considerably below normal and domestic prices rose sharply.12 Not surprisingly, the 1993-94 and the 1994-95 cotton seasons witnessed a sharp rise in imports. Generally, Indian millers import cotton when domestic prices are at least 90 percent of comparable landed foreign prices. At 90 percent, mills can often justify the more expensive foreign cotton because it is cleaner and not subject to local taxes.13 F. Cotton Market Performance Seed Cotton Processing Margins A5.63 Cotton processing margins display a wide range but generally provided for adequate coverage of ginning and pressing costs. The gross processing margins (GPM) in real terms (1990 rupees) were estimated for two short staple varieties --J-34 and F-414 (Figure A5.10) using average monthly prices from November 1990 to February 1995 in Punjab.14 Data limitations confined the analysis to these two varieties. The GPM for J-34 covered a narrower and slightly lower range, from Rs 86 to Rs 204 per quintal, compared to F-414, Rs 100 to Rs254 per quintal. Nonetheless, these margins provide adequate coverage of the ginning and pressing fees set by the state. In Punjab during this period, ginning fees in real terms (1990 rupees) averaged Rs 53 per qtl, while pressing fees averaged Rs 35 per qtl. 15 Figure A5. 1O: Gross Processing Margins for J-34 and F-414 10_ 12 9. 10- 87 9 7 7 5- 6- ~~~~~~~~~~~4, 4. 3=- 34 2, 2 A 1 ~~~~~~~~~~~~~~0 r , N (N (N O v v O 0 - 0O CO - s O0 ( CO zl - : 0 N 0 ('7 - ~ C- (N ( ax t s o:1C1 n F-414, Rs/qtl J-34, Rs/qtl Source: computed. 12 G. Pursell and A. Sharma, 1995, "Indian Trade Policies Since the 1991/92 Reforms," mimeo. 13 USDA/FAS, 1995, Cotton Annual Report. 14 Gross processing margin= [(0.34* cotton lint price) + (0.66*cotton seed price)]-seed cotton price. Seed cotton and cotton seed prices were from the Bhatinda market in Punjab, while the cotton lint prices used were the average upcountry spot prices (Punjab, Rajasthan, and Haryana). 15 The fees are normally expressed in bales, and ginning fees average Rs9O per bale and pressing fees are Rs6O per bale during the period 1990-91 and 1993-94 (1990 Rupees). Annex 5 Page 20 of 34 Farmer's Share of Cotton Sale's Price A5.64 Private traders offer better prices to growers than CCI and Maharashtra monopoly procurement scheme. In the case of CCI and the Maharashtra cotton marketing federation, the farmer's share of the final cotton sales price was estimated directly from the cost of purchases of all seed cotton as a proportion of the agency's gross revenue from cotton sales. For the private trade, due to data limitations, the analysis was confined to two varieties, J-34 and F-414 in Punjab using average monthly prices. The farmer's share was estimated by the ratio of seed cotton prices to the weighted average returns from the sale of the equivalent cotton seed and cotton lint. Although the results are more indicative than precise due to the slightly different methodologies uses, it appears that private trade still display the lowest margins, and therefore allow a higher share of the cotton sale price for farmers (Table A5.7). CCI, which directly competes with private traders, show higher marketing margins than private traders. The Maharashtra Monopoly Procurement Scheme does not perform much better than the CCI. Moreover, the relative farmer share of the final price does not take into account the opportunity cost of the "payment by installment system" under MMPS, in contrast to private trade which generally conducts all transactions in cash. We have no information on the marketing performance of cotton cooperatives in other states, except for the study of the Gujarat State Cooperative Cotton Federation quoted in Parmar & Ramchandran (see footnote 3) which quoted that growers received as much as 93 to 96 percent of the final price of cotton. Although it is difficult to generalize from only one study, this may indicate that well managed cooperatives may perform better than private traders and offer better prices to growers. Table A5.71Farner's Share of the Final Price of Cotton Lint and Cotton Seed under Different Marketing Systems I. CCI 84 79 86 80 78 ____ H. Monopoly Scheme 83 81 80 80 85 III. Private Trade l (a) J-34 87 89 88 89 89 (b) F-414 87 88 87 87 87 Source: CCI, Bombay; Maharashtra State Cooperative Growers' Marketing Federation Ltd., Bombay; North India Cotton Association, Bhatinda; East India Cotton Association, Bombay. A5.65 The performance of private traders and capacity to offer better cotton are burdened by the numerous rules and regulations imposed upon them (see Annex 2 for an inventory of these domestic regulations). The Selective Credit Controls (lifted in October 1996) together with the Cotton Control Order(s) have been powerful instruments in reducing competition and raising marketing costs in cotton trading. In particular, the limited access to trade finance has compelled private cotton traders to rely mostly on their own financing, thereby reducing entry and competition in the trade. Similarly, the absence of forward and futures trading raises risk premiums in marketing margins by forcing traders to assume speculative positions. This further reduces competition and entry in the cotton trade. Seasonal Price Efficiency A5.66 Shorter staple cottons display a higher degree of seasonality. Seasonal price indexes for 5 selected cotton varieties, namely J-34, F-414, H-4, Shankar 6, and DCH-32 were estimated.16 These 5 varieties are broadly representative of the patterns of production in India. J-34 and F- 16 The seasonal price indexes were calculated using monthly average spot prices from March 1984 to February 1995. Annex 5 Page 21 of 34 414 represent the northern region, Shankar 6 and H-4 the central Table A5.8: Percentage Seasonal Pice Increase of Selected Cotton region and DCH 32 the southern region. J-34 is a superior medium Pr 'Varieties staple cotton, F-414 is a long staple cotton, Shankar 6 and H-4 are mostly extra-long staple cottons and DCH-32 represents the longest J-34 200 staple cotton of the group. The prices of the shorter staples, in F-414 20.4 particular J-34, F-4 14, and Shankar 6, display strong seasonal price Shankar 6 17.7 trends (Figure A5.1 1). Prices generally reach their lowest point in H4 5.4 November and reach their peak in August. The longer staples, H-4 DCH-32 8.7 and DCH 32 displayed a flatter seasonal price pattern. Source: Compute A5.67 These price patterns parallel the differences in the market arrival patterns. In the north and center, there is a shorter and more distinct marketing season: October to March in the North and December to March in the Central Region. In contrast, the relatively more evenly distributed pattern of arrivals in the south explain the flatter seasonal price pattern of DCH-32 and H-4 prices. A5.68 Seasonal price increases provide Figure A5.1 1: Seasonal Price Indexes of Selected Cotton Varieties adequate financial incentives for storage. 110 The seasonal movements in cotton prices 105 - offer adequate incentives for storage, since 100 : they provide ample room to cover the cost 95 _. of working capital (Table A5.8). Although 90 ___' the percentage seasonal price increase for 80 the longer staples (H-4 and DCH-32) is . * , c low, the longer marketing season and more 0 ° 0 8 U $ i even distribution of market arrivals reduce the need for longer storage. . J-34 F-414 A5.69 Cotton export uncertainty may 110 also contribute to the larger price 105 seasonality observed for the early cotton 100 varieties. Cotton export policies are highly 95 uncertain, and export volumes highly volatile. This uncertainty may help 90 l depress prices of those cotton varieties , et which are marketed early in the marketing X c s season, until export quotas are actually K4 to DCH32 x, S-6 released during the marketing season. Source: East India Cotton Association, Bombay. Cotton Price Volatility A5.70 Farmers and cotton traders are exposed to significant inter-year price variability. To analyze the degree of variability of cotton prices, the prices of three cotton varieties grown in the three major cotton regions were selected. They are DCH-32 for the southern region, H-4 for the central region and J-34 for the northern region. With the exception of the southern variety DCH- 32, the inter-year price volatility of the other cotton varieties (northern and central) during 1983-94 was estimated at 23 percent (Figure A5.12). This, however, parallels the international price, the Cotlook "A" Index, volatility during the same period. Annex 5 Page 22 of 34 Figure A5.12: Inter-Annual Price Volatility for Selected Cotton Varieties, A5.71 Farmers and cotton 1983-94 traders are also exposed to significant intra-year price 0.3' 'I __ variability. Cotton prices also 0.25-/ display significant seasonal 0.2- variation. Figure A5.13 shows the 0.15- highest and lowest prices during the 0.0- _ _ _ _crop year (expressed as a 0.051 percentage of the average crop-year J-34 F0414 H Shan DCH- price) and the range of prices k-6 32 (between the lowest and highest value). The coefficients of Source: P. Varangis, IECCP, The World Bank variation (CV)17 of the detrended Note: Volatility is estimated by the coefficient of variation. price series during the period 1984/85 and 1993/94 (using monthly averages) indicate that this variation could be as high as 25 Figure A5.13: Seasonal Variation for Selected Cotton Varieties percent within a year. However, (% of the average crop-year price) the CV of cotton prices were less than 10 percent for several years. 100- ~~~~~~~~~~G. Risk Management in the 80- Cotton Sector 40 MHighest Price Risk Management 20 iN# ;7 L L L L URa Instruments . _ _ . ,. . d l A5.72 Prior to February 1997, I j; E = only NTSD contracts in raw cotton were allowed, and regulated by Source: The East India Cotton Association eight recognized associations spread over the country. The NTSD cotton contract is a type of forward contract between two parties in which a commodity of specific grade, has to be delivered to a specified location during a pre-determined time frame at a predetermined price. Neither buyer or seller can transfer the contract to another party and financial settlement is not allowed. Grade, location and delivery dates cannot be renegotiated after the contract has been signed. Currently, only NTSD contracts in cotton are allowed. At present, there are eight associations]8 in the country that are recognized and allowed to trade NTSD contracts by the Forward Markets Commission, which implements the FC(R) Act (Table A5.9). A5.73 The turnover of cotton NTSD contracts was small, highly seasonal and trade only for afew months. The total volume of NTSD contracts reportedly traded at all recognized exchanges has remained more or less stable since 1990, at an average of 1.8 million bales (each of 170 kg) a year. 17 The coefficient of variation is estimated by dividing the standard deviation by the mean of a series. 18 An association has to apply for recognition from the central government. Taking into account public and the industry's interests, the government may or may not approve the application. In general, recognition is granted for only a short period at a time --six months to three years. The central government also retains the right to withdraw its recognition at any time. For more details about commodity exchanges and commodity forward and futures trading, see World Bank 1996, "Managing Price Risks in India's Liberalized Agriculture: Can Futures Markets Help?," Report No. 15453-IN. Annex 5 Page 23 of 34 This is equivalent to about 15 percent of total production, although there are reports that more cotton is sold using forward contracts than is officially reported. Table A5.9: Cotton Associations Permitted to Trade Cotton A5.74 Both in Bhatinda (in the Punjab, in NTSD Contracts, 1995 the heart of one of India's cotton producing L I ---+ e centers) and Bombay (the traditional transit Ahmedabad Cotton Merchants Association Ahmedabad point for both domestic and external trade), Andhra Pradesh Cotton Association Guntur trade in NTSD contracts is highly seasonal. Central Gujarat Cotton Dealers Association Vadodara In Bhatinda, during the period 1990 to 1993, Central India CoKton Association Ujian on average 52 percent of the year's trade is Cotton Association Indore in only two of the annual six contracts, East India Cotton Association Ltd. (EICA) Bombay namely the contracts for November- South India Cotton Association Coimbatore December and January-February delivery. In Southern Gujarat Cotton Dealers Association Surat Bombay for the same period, the two delivery Source: World Bank, 1996, "Managing Price Risks in India's periods plus March-April, account for 64 Liberalized Agriculture: Can Futures Markets Help?" percent of total volume. Trade in Coimbatore's exchange, serving the sizable textile industry in and around this southern Indian town, is better distributed, with only the November-December contract being minimally traded. A5.75 Trading is limited to near-by months. NTSD contracts allow delivery up to six months out. However, most trade is in nearby delivery months, with the vast majority being no more than three months forward. For example, in the most liquid months, up to 50-60 percent of NTSD contracts are in the nearby months. In contrast, only 10-15 percent of NTSD contracts are entered into more than two months before the start of the contracts' delivery period. A5.76 NTSD contracts are highly imperfect risk management tools. The stringent limitations imposed on NTSD contracts by the FC(R)A make them highly imperfect risk management tools. This explains the small share --15 percent-- of physical trade covered by NTSD cotton contracts, their highly seasonal nature, and the fact that most contracts are only for up to three months. NTSD contracts turn out to be rather risky forms of forward contracts under present regulations. Once a company has entered into a NTSD contract, say for delivery in six months time, it cannot get out of the contract or even renegotiate its specifications. Cotton ginning mills, for instance, often sell their lint forward in the expectation that they will be able to buy the seed cotton at an attractive price. If the harvest turns out to be lower than expected, the mills who entered into forward contracts have no choice but wait until the contracts' due dates. The only exception to this is when a general production crisis would cause a general default, a situation in which the commodity exchange board can decide to forcibly close out and settle all outstanding contracts, at a negotiated price. Such a forced closure of contracts --experienced by the Northern India Cotton Association in Bathinda in late 1994-- causes significant distress and results in large financial losses. It also reduces public and government confidence in futures markets. The Northern India Cotton Association in Bathinda has since lost its trading privileges in NTSD cotton contracts. A5.77 Such a default crisis would not occur under the standard, international futures contracts definition. As soon as the news of a disappointing harvest is known, futures prices increase. Those who sold contracts will have to close them or pay extra margins. The commodity exchange clearing house guarantees contract performance and secures itself through the margin payments it receives. Default will be avoided automatically: speculative mills would be forced out of the market relatively fast and will be unable to stay with an open position until the situation becomes unbearable. Annex 5 Page 24 of 34 A5.78 The poor suitability of NTSD contracts prompts an active illegal futures trading In response to the legitimate hedging needs of cotton gins, a significant illegal trade in seed cotton and raw cotton futures is reported to have developed with its main center in Surendranagarin, where V-797 cotton futures are traded primarily around harvest time, but also in other centers. Commodity futures contracts are forward contracts that are tradable and standardized in their obligation to make or take delivery of a fixed quantity and quality of a commodity at a specific location, on a specific future date or time. Unlike the NTSD contracts, futures contracts are backed by margin money which insures against default. This illegal trade appears to be relatively well-organized: participants pay margins in case their positions move against them, and default rates are reportedly quite low. A5.79 Given the recent lifting of the ban on trading of cotton futures contracts, the government would need to play a major supportive role in the revival of cotton futures markets. Improvements in the regulatory and institutional environment governing cotton futures trading operations both at the Forward Markets Commission, and among the cotton commodity exchanges are needed to ensure their orderly development. On the regulatory front, the FMC would need to curb its discretionary interventions and revert to the original intent of the three tier regulation model provided by the Forward Contracts (Regulation) Act. Under this model, the FMC would still approve exchanges and set the general legal and regulatory framework. Cotton commodity exchanges would need to up-grade their rules and regulations (trading procedures, delivery system, supervision), clearing operations, promotional and development activities, and their implementation and monitoring capacities. These topics are analyzed and developed in greater detail in our companion report on India futures markets mentioned above. A5.80 The prospects for cotton lint and seed cotton futures are good. In the case of cotton lint, several cotton futures contracts have been traded in India in the past, and the presence of illegal trading in cotton lint futures in some areas attests to the demand for such contracts. More detailed analysis, however, will be required in determining for which cotton varieties could cotton lint contract be introduced.19 Simple correlation analysis of monthly average spot prices between the main cotton varieties show a high degree of correlation between varieties, opening possibilities for using just a few contracts to hedge all the major varieties. (Table A5. 10). In the case of seed cotton, the potential for a futures contract is also quite good. This stems largely from its attractiveness to both ginners and cotton growers. A5.81 To succeed, the cotton lint and seed cotton futures contracts would need to carefully develop appropriate delivery, grading and warehousing systems, which balances the trade-offs between liquidity (the number of transactions) and the basis risks (the difference between the futures and the market price) among and within the large number of cotton varieties produced across India. Available evidence suggests that the cotton policy reforms aimed at improving the performance of the physical markets for cotton and their stable integration with the world markets, by systematizing and standardizing cotton grading practices, by improving contract enforcement procedures, by promoting the developing of warehousing receipts system would go a long way towards alleviating the current trade-offs between liquidity and basis risks. A5.82 The re-introduction of cotton futures contracts would be greatly facilitated by India's strong legal and regulatory system for commodity exchanges and experience. Our companion study on futures markets (see footnote 16) indicates that India, unlike most other developing 19 For more detailed discussion of the prospects of introducing futures contracts for seed cotton, cotton lint and cotton yam, see World Bank, 1996, "India, Managing Price Risks in India's Liberalized Agriculture: Can Futures Markets Help?", Report No. 15453-IN. Annex 5 Page 25 of 34 Table A5. 10:The Correlation Of Monthly Aege Spot Prices For Main Cotton Varieties, 1984-1995 [ ___ J-34 F-414 HSjDH32[6__ IF-414 1S-6 ] HA4 DCH-32 J-34 1 10.99 0.99 0.978 0.94 J-34 1 0.95 0.83 0803 1 0.85 F-414 1 1 0.99 0.98 0.92 F-414 -1 0.87 0,86 0.380 S-6 ~~1 0.991 0.97 S-6 1 0.871 -0.60 H4 ~~ ~~~110.95 HA4 1 10.58 DCH-32 ~~~~~~I DCH-32 - 1 ______ - = ~~1986-87 1987-898 =____ I ~ J-34J F414 IS6IH4 tDCH-321 _ __IJ-34 [F-414 I S-6 H-4 DCH-32 J-34 1I 0.78 0.86 0.207 0.949 J-34 -- 1 f0.95 f0.39 0.89 0.60 1F14 1 1 1 0.61 0.328 0.50 FA414 -[ 1 [0.59 0.82 0.42 S-6 ~~~~~~1 0.119 0.57 *S-6 -[ 1 0.19 -0.467 H4 ~~~1 0.30 H14 { 1 f0.72 DCH-32 I~~~~ DCH-32 T__ ______ ___ 1 _____ - _____ - 1988-891 19891-990 =f J-34___ F414 IS6IH4 ~DCH-321[ I__ J-34 jFA414 I S-6 J HA [ DCH-32] [J-34] 1 0.998 0.986 0.98 0.497 J-34 1 J0.88 0.37 -0.07 0.61 [F-4141 1 06 0.98 0.98 FA414 I 0.33 -0.13 0.49 S-6 I~~~~ 0.97 0.96 S-6 1 0.68 -0.17 H-4 1~ ~~~~~~~~ 0.99 HA [ 1 0.31 DCH-32 ~~~~I DCH-32 I - ~~~~~~1990-9193 -_______ - 1991-92 ____ I ___ J-34[ F414 H I S- -4 DH-32I ____ _34_ F414 I S-6 [ HA DCH-32J __34__ 0.989 0.91 0.91 0.92 J-4 1 10.99 0.99 -0.98 0.96 IF-414 1 1 10.98 0.88 0.91 FA414 1 1 0.99 -0.98 0.195 S-6 ~~~1 0.85 *0.96 S-6 1 0.97 0967 H -41 1 0.82 HA4 1 0.98 DCH-32 I~ ~~~~~ DCH-32 1 Source: World Bank, 1996,"nda MaaigPie9ik9nIdi'3ieaizArclue:CnFtrs9 akt4ep?,Rpr o 1-4F5453-N.H4 DH-2F14IS- - DH3 Source Wntrndatina196 experie,Mncein suggeRstsi th dat'utrs miearkzets Arcouldue open nuuew aktep" op Rtunitie iNo fcooredst examined lnew aproachesionpoiigicm and pxeinewt uue raicei suportton aginculuralg oprioducersdn. This aledso being driveIndby haeps to rel-eeinilovernmentexenes and rgltyssthem desre teofproueratos tof havefriedityerchangs-ad moefexibledsppr maretchaniiesms,uadche requiwremeusntst costmply whith inentonal travde thagreements.e Govdernydveomentsr alof ctrying ftonures thatkenew Approachtesrmationtain supeitable safeggests fort producersbtwtotitreigwt markets ouforcne ppruntes.i goenmtiscntext,sr commodity deivtieonsrumens suh0a futureens, optionsou one futres, saps bend prdces This andethefollo ing paarahdrien borrowedmfro P. Vraingis & gD.eLars nt"D eal ngswit Commditye Pricre Uncertainty", Policy Research Working Paper No. 1667, October 1996 The World Bank. Annex 5 Page 26 of 34 commodity bonds are being examined and tested to determine their viability in providing suitable level of price protection and income stabilization that is cost effective and trade neutral.- A5.84 A number of experiments are taking place worldwide. The U.S. Department of Agriculture (USDA) and Agriculture and Agri-Food Canada have implemented pilot projects that use commodity-derivative based instruments to provide participating producers with an alternative to traditional farm income support programs. The USDA program provides participants with option on futures contracts and small subsidies to cover brokerage commissions. The program is available to a small subset of regular feedgrain and wheat program participants. The option pilot program in Canada provides all cattle producers with market access to a specialized put option, comprised of Canadian dollar-converted US live cattle futures. The put option is roughly one-fifth the size of the regular futures contract in Chicago. These two pilot projects are example how governments are using financial instruments to provide risk management instrument mechanisms to producers. A5.85 Other governments have introduced financial instruments to re-insure their own risks directly. Such experiences could be provide useful venues for covering the risks of government- sponsored cotton programs in India or Maharashtra state, or the risks assumed by cooperatives on behalf of their members. For example, the Mexican government has used financial instruments to manage their commodity price risk exposure in offering price protection to cotton growers (Text Box A5.4). A5.86 Other cotton producing countries have recognized the need to offer risk management tools to private operators in liberal cotton market environment. Turkey, for example, is setting the stage for the introduction of cotton lint futures contract in Izmir and Adana, with the assistance of the World Bank and UNCTAD. The Government of Turkey is planning to complete already existing forward cotton contracts with the introduction of futures contracts. Turkey found it expedient to offer access to a domestic cotton contract since existing international contracts were not suited to manage the risk management needs of its cotton community (basis risk with a New York cotton contract would be too large). It is likely that a Turkish futures market, when established, will attract considerable interest from neighboring cotton countries such as Egypt and Uzbekistan. These countries produce similar types of cotton, and their cotton liberalization efforts are currently hampered by lack of access to market-based risk management tools. A Turkish futures contract may also have some interest for India, and in particular its ELS cotton. Annex 5 Page 27 of34 Box A5.4. Mexico's Cotton Price Support Scheme Because of the recent liberalization of agricultural trade and intemal marketing systems in Mexico, farm-level prices of several agricultural products are now determined mainly by intenaational markets, and farmers have had to cope with price uncertainty between planting and harvest times to a degree unknown before. The Government of Mexico has responded with a program designed to guarantee a minimum price to cotton growers. Through ASERCA, a government organization providing support services for agricultural commercializationr Mexican cotton producers are able to manage their price risks during the harvest period and are guaranteed a minimum price during the planting season. Although programs guaranteeing minimum prices are common, most programs simply transfer price risks from producers to the government budget through the floor price mechanism, and most programs fail -sometimes spectacularly- when sudden price changes overstrain the government budget necessity is the mother of intervention, and ASERCA. lacking the budget to assume the price risk direetly, designed a sustainable program to transfer price risks from growers to international markets. During the planting season, ASERCA offers farmers the chance to participate in a program guaranteeing a minimum cotton price for a fixed fee. The minimum price is fixed using the New York cotton futures exchange. For a fee, ASERCA offers a guaranteed price (in US dollars) and hedges its own risk by using the fee to purchase a put option on the exchange for future delivery at harvest time. (The put option gives ASERCA the right to sell cotton on a specific future date at a prescribed price, known as the strike price). Should prices subsequently fall, ASERCA pays farmers the difference between the New York price at harvest and the minimum price. The difference is exactly equal to the payoff value of the put option. If prices rise instead. ASERCA makes no payments to farmers. By paying a fee and participating in the program, a famer in effect purchases insurance against a drop in prices below a certain level --in fact, the program refers to the fee as a "premium". As with insurance, payouts do not always occur. so the program is not without costs to farmers. Private brokers could offer similar programs, although the private sector has had litle experience in providing such services directly to growers. Since ASERCA's program is inexpensive to administer and demand driven, however, ASERCA can readily reduce its presence should a market for private brokers develop. Source: P. Varangis, D. Larson (October 1996): "Dealing with Commodity Price Uncertainty", Policy Research Working Paper No. 1667, World Bank. Page 28 of 34 Annex 5 H. Appendix Tables & Figures Appendix Table A5.1: Cotton Procurement by Different Agencies, 10,000 bales of 100 kg. Agency 1989-90 I 1990-91 1991-92 1992-93 1993-94 Cotton Corporation of India 1240 1019 1001 1186 776 Maharashtra Monopoly Procurement Scheme 2080 1350 1063 1990 1336 State Cooperatives Haryana 65 68 106 123 78 Punjab 223 156 191 141 101 Rajasthan 12 22 24 36 41 Gujarat 188 215 191 134 195 Other States 55 41 28 69 30 Subtotal 543 502 540 503 445 Private Trade 9712 8829 9474 10321 9768 Total 13575 11700 12078 14000 12325 Note: (1) Excludes Maharashtra; (2) Figures in parentheses denotes percentages to total Source: EICA and CCI. Appendix Table A5.2 Minimum Support Price of Seed Cotton of Fair Average Quality, 1990-91 to 1994-95, Rs per quintal Basic Staple Micronaire length in Value millimeters 1990-91 1991-92 1992-93 1993-94 1994-95 (2.5% Span Variety length) Assam Comilla 7.0-8.0 - 600 690 780 870 Bengal Deshi 6.8-7.2 525 600 690 780 870 Wagad 20 5.5-6.0 550 625 720 810 900 V-797 22 4.24.8 585 660 760 855 950 Jayadhar 22-23 5.0-5.6 585 660 760 855 950 J-341Bikaneri 23 3.6-4.4 605 680 785 885 985 Narma G.Cot 12 23.5 4.2-5.0 605 680 785 885 985 AK/Y-I 24 4.8-5.2 615 690 795 895 995 F414/H- 24 -25 3.84.2 620 695 800 900 1000 7771Agatti Digvijay 'A' (Guj) 25 4.04.5 630 700 790 885 975 AHH-468 25 -26 3.8-4.2 675 760 875 985 1090 1007 27 3.74.0 695 780 900 1000 1110 LRA-5166 27 4.04.5 695 780 900 1010 1120 JHKY-1/MECH 30 3.84.2 750 840 950 1050 1200 H4 30 3.6-4.2 750 840 950 1050 1200 Shankar-6(Sau.) 28 -29 3.7-4.2 750 840 955 1055 1205 Shankar-6(Guj.) 30 3.7-4.2 765 860 975 1080 1230 MCU-5(South) 33 3.0-3.5 770 865 980 1085 1235 DCH-32(M.P.) 33 -34.5 3.0-3.3 775 870 985 1090 1245 DCH-32(Mah.) 34.5 3.0-3.2 775 870 985 1090 1245 DCH(South) 39 3.0-3.5 820 920 1040 1150 1300 Suvin 40 3.2-3.6 1140 1280 1450 1600 1830 Source: Office of the Textile Commissioner, Govemment of India. Page 29 of 34 Annex 5 Appendix Table A5.2 contd: Minimum Export Prices for Selected Varieties of Raw Cotton; 1990-91 to 1994-95, (US cents per kg F.O.B.) Cotton Variety 1990 - 91 1991-92 From From 10.10.1990 23.10.1990 I 19.02.1991 08.11.1991 18.11,1991 Extra Long (34.5 mm & above) DCH-32 (AP/Kar,TN) 200 190 Long (28 mm to 34 mm) DCH-32 (AP/Kar/TN) 170 170 MCU-5 (AP/Kar/TN) 170 170 Shankar-6 163 163 H-4/MECH-II (Mah) 162 162 ". (MP) 160 160 Superior Medium & Medium (24.5 mm to 27.5 mm) 161 155 LRA 5166 150 150 F-414 150 150 H-777/Agatti 145 145 J-34 (SG) Bengal Deshi Extra Superior Fine/Choice 120 128 120 120(CF) 112 Super Fine 115 123 115 115(CF) 107 Fine 110 118 110 IIO(CF) 102 Assam Comilla Hand ginned 160 168 160 Machine ginned 135 143 135 Appendix Table A5.2 contd: Minimum Export Prices for Selected Varieties of Raw Cotton; 1990-91 to 1994-95, (US cents per kg F.O.B.) Cotton Variety 1992-93 From 30.10.92 2.12.92 1 1.1.293 15.2.93 2.3.93 27.4.93 July'93 20.8.S93 Extra Long (34.5 mm & above) DCH-32 (AP/Kar/TN) 160 160 160 165 165 165 Long (28 mm to 34 mm) DCH-32 (AP/Kar/TN) 130 135 135 140 140 140 MCU-5 (AP/KarJlN) 120 123 123 123 123 110 Shankar-6 113 122 122 122 122 109 H-4/MECH-11 (Mah) 110 111 119 119 119 119 119 (MP) 108 109 117 117 117 117 117 Superior Medium & Medium _ _ _ (24.5 mm to 27.5 mm) LRA5166 103 106 115 115 115 115 102 F-414 110 111 119 119 121 121 121 H-777/Agatti 107 108 118 118 120 120 120 J-34 (SG) 102 106 117 117 117 117 104 Bengal Deshi Extra Superior 92 92 92 92 92 92 92 Fine/Choice 89 89 89 89 89 89 89 Super Fine 86 86 86 86 86 86 86 Fine _ _ Assam Comilla Hand ginned 112 112 112 112 112 112 Machine ginned ' 92 1 1 92 92 92 92 92 Page 30 of 34 Annex 5 Appendix Table A5.2 contd: Minimum Export Prices for Selected Varieties of Raw Cotton; 1990-91 to 1994-95, (US cents per kg F.O.B.) Cotton Variety 1993-94 1994-95 From From 13.5. 94 26.10.94 14.2.'95 13.6.'95 4.7.'95 25.7.'95 25.8.'95 28.9.'95 Extra Long (34.5 mm & above) DCH-32 (AP/Kar/lN) 262 300 300 275 Long (28 mm to 34 mm) DCH-32 (AP/Kar/TN) 237 MCU-5 (APIKarJTN) 227 245 245 215 Shankar-6 205 222 222 200 199 H-4/MECH-II (Mah) 192 216 216 (MP) 192 Superior Medium & Medium (24.5 mm to 27.5 mm) LRA 5166 188 220 210 210 185 F-414 198 - 212 215 187 H-777/Agatti 198 J-34 (SG) 190 227 205 207 185 Bengal .Deshi Extra Superior Fine/Choice 136 Super Fine 132 Fine 128 Assam Comilla Hand ginned Machine ginned Notes: AP = Andhra Pradesh; Kar = Karnataka; TN = Tamil Nadu; Mah = Maharashtra; MP = Madhya Pradesh Source: The East India Cotton Association, Bombay. Appendix Table A5.3: Profits and Losses of CCI, 1978-79 to 1993-94 (Rs. million) Year Tumover Profits before Profits after Profits after ____________ _________________ interest interest before tax interest and tax 1977-78 1805.6 174.2 (-)53.5 (-)53.5 1978-79 1395.1 172.0 69.6 (-) 69.6 1979-80 2975.2 237.1 (-) 136.5 (-) 136.5 1980-81 3622.2 416.8 10.6 10.6 1981-82 2238.5 144.2 (-) 286.5 (-) 286.5 1982-83 3040.8 206.6 (-) 251.4 (-) 251.4 1983-84 2518.9 181.4 (-) 142.2 (-) 142.2 1984-85 1813.2 141.5 (-) 106.8 (-) 106.8 1985-86 3017.4 170.2 (-) 186.7 (-) 186.7 1986-87 3656.1 312.2 (-) 2.2 (-) 1.7 1987-88 3310.4 186.2 (-) 30.6 (-) 30.6 1988-89 1382.3 195.2 98.4 85.6 1989-90 3763.5 528.8 282.4 232.4 1990-91 7346.1 912.5 688.5 628.5 1991-92 7641.0 754.6 519.8 279.8 1992-93 5968.2 901.4 601.1 306.1 1993-94 9176.5 788.0 562.1 362.1 1994-95 7878.2 647.2 540.8 283.3 1995-96 9083.0 685.4 472.0 252.0 1996-97 13096.5 1067.5 800.0 491.2 Note : (-) indicates loss Source: The Cotton Corporation of India, Bombay Page 31 of 34 Annex 5 Appendix Table A5.4: CCI Statewise Purchases and Buyer-wise Sales (Bales of 170 kg) States 1989-90 1990-91 1991-92 1992-93 1993-94 Purchases Punjab 285,967 131,240 207,369 328,043 69,043 Haryana 142,218 79,353 109,857 137,737 47,014 Rajasthan 121,040 108,045 149,389 162,892 130,153 Gujarat 216,409 204,851 154,365 226,602 163,999 Madhya Pradesh 185,375 213,646 126,484 101,618 168,684 Andhra Pradesh 231,249 253,012 171,619 157,827 162,062 Kamataka 38,312 21,314 74,017 66,019 31,063 Tamil Nadu 21,471 7,679 7,147 5,310 4,334 Others 822 97 434 550 0 Total 1,242,863 1,019,237 1,000,681 1,186,598 776,352 Sales I NTC 392,527 352,100 406,943 225,999 305,008 SSTC/Cooperatives 162,948 318,918 396,601 135,178 171,786 Privates 43,836 66,347 155,463 133,708 241,259 Total 599,311 737,365 959,007 494,885 718,053 Source: CCI Appendix Table AS.Sa: Ginning Fees in Selected States, Rs/ bale of 170 kg. State I Centre 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 Punjab: All centres 69.00 69.00 72.00 72.00 72.00 87.00 109.00 115.00 117.00 Madhya Pradesh: Khandwa 98.60 102.00 102.00 102.00 108.80 121.55 144.50 146.20 154.70 Khargone 100.30 103.70 103.70 103.70 105.40 121.55 144.50 147.05 149.60 Gujarat: __ ___ Mehsana 78.20 83.80 92.48 98.26 109.82 121.38 135.88 144.50 167.62 Surendranagar 72.25 80.92 80.92 --- ... 118.58 130.00 150.28 170.50 Maharashtra: Selected centres 101.15 106.33 109.62 123.59 129.00 146.45 166.60 195.50 210.80 Andhra Pradesh: Adilabad 93.50 93.50 93.50 93.50 119.00 130.90 153.00 142.80 195.50 Source: East India Cotton Association, Bombay. Appendix Table A5.5b: Pressing Fees in Selected States, Rs/ bale of 170 kg. State i Centre 85-86 86-87 87-88 88-89 89-90 90-91 91-92 92-93 93-94 Punjab: All centres 44.00 44.00 46.00 46.00 50.00 60.00 75.00 74.00 75.00 Madhya Pradesh: Khandwa 61.00 63.00 64.00 66.00 74.00 90.00 100.00 102.00 110.00 Khargone 62.00 64.00 65.00 66.00 70.00 89.00 85.00 103.00 109.00 Gujarat: II I_ _ _ _ Mehsana 50.00 52.00 54.0( 56.00 63.00 70.00 80.00 105.00 108.00 Surendranagar 49.00 50.50 50.50 --- --- 68.00 78.00 102.00 110.00 Maharashtra: Selected centres 58.25 53.80 57.63 61.10 72.00 83.05 97.00 105.00 112.00 Andhra Pradesh: Adilabad 58.0 8.0 60.00 60.00 80.00 82.00 95.00 91.00 111.00 Source: East India Cotton Association, Bombay. Page 32 of 34 Annex 5 Appendix Table A5. 6Farmers' Share in Gross Income Earned by Cotton Corporation of India and Maharashtra Monopoly Procurement Scheme, Rs million. COTTON CORPORATION OF INDIA MAHARASHTRA MONOPOLY PROCUREMENT SCHEME 1989-90 1990-91 1991-92 1992-93 1993-94 1989-94 1989-90 1990-91 1991-92 1992-93 1993-94 1989-94 _______________________ __________ ~~~~~~~ ~~Average __ _ _ _ _ _ _ _ _Average I Gross Income 5416.9 6932.0 8951.9 6987.3 8302.9 7318.2 9498.9 7666.3 7128.9 11012.3 11049.8 9271.2 (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) (100.00) 2. Payments made to farmers 4549.6 5457.5 7726.2 5611.9 6497.7 5968.6 7907.1 * 6178.6 5705.8 8857.5 9366.7 7603.1 (83.99) (78.73) (86.31) (80.31) (78.26) (81.56) (83.25) (80.59) (80.03) (80.43) (84.77) (82.01) 3. Total Expenses _ _ i Ginning and 291.8 387.7 432.2 395.2 460.7 393.5 531.8 401.4 373.0 744.1 576.0 525.3 processing charges (including (5.39) (5.59) (4.85) (5.65) (5.55) (5.38) (5.60) (5.24) (5.23) (6.76) (5.21) (5.66) overheads) ii Interest on bank 186.8 224.1 234.8 300.3 225.9 234.4 442.6 146.3 244.3 487.6 179.4 300.2 loans (3.44) (3.23) (2.62) (4.30) (2.72) (3.20) (4.66) (1.91) (3.43) (4.43) (1.62) (3.24) iii Selling, 28.5 214.6 46.7 101.7 229.2 124.1 352.0 327.7 413.7 566.5 465.5 425.1 distribution and forwarding (0.53) (3.10) (0.52) (1.45) (2.76) (1.70) (3.71) (4.28) (5.80) (5.15) (4.21) (4.58) -charges iv. Miscellenous 2.3 8.18 15.2 8.2 313.9 84.3 140.8 97.7 120.5 203.9 145.7 141.7 expenses (0.04) (0.12) (0.17) (0.12) (3.78) (1.15) (1.49) (1.27) (1.70) (1.85) (1.32) (1.53) 4. Provision for Taxation 50.0 11.42 217.1 263.8 213.3 171.7 (0.92) (0.16) (2.42) (3.39) (2.57) (2.35) Total 559.4 846.0 946.0 1069.2 1443.0 972.7 1467.2 973.1 1151.5 2002.1 1366.6 1392.2 (10.33) (12.20) (10.57) (15.30) (17.38) (13.29) (15.46) (12.70) (16.16) (18.19) (12.36) (15.02) 5 Net Profit 307.9 628.6 279.8 306.1 362.1 376.9 121.3** 514.5 271.6 152.6 316.5 275.4 (5.68) (9.07) (3.12) (4.39) (4.36) (5.15) (1.28) (6.71) (3.81) (1.38) (2.87) (2.97 Notes : 1) Figures in brackets are percentages to gross income * Payments made to farmers under Monopoly Procurement Scheme is inclusive 'bonus' Break up of payments received by the cotton growers under **Figures in this column refer to amount transferred to Price Fluctuation Fund. Monopoly Procurement Scheme. Sources: Cotton Corporation of India Maharashtra State Cooperative Cotton Growers' Marketing Federation,Bombay. Page 33 of 34 Annex 5 Appendix Figure A5.1: Ratio of the Wholesale Price Index of Seed Cotton to the Wholesale Price Index of Oilseeds, Cotton, Pulses, and Sugar, Khandasari and Gur. MCU-5 1.80 1.60 1.40 X 1.20 j4IXi~Z~ 1.00 0.40 ii 0.60 0.40 00 00 00 0 N N O m u oi 00 00 00 00 00 O (ON | Oiliseeds z Cereals A Pulses > sugar, kg S-4 2.40 1.90 ___ 1.40 0.90 0.40- r- -1 1 1 1 1 1- 00 00 0 00 ON ON O -Z w '~ r_- ON - ' 00 00 00 00 00 ON N | 0ilseeds r Cereals L = Pulses sugar, k&g J-34 1..80 1.40 1.20 0 80 0.40 ______ _____ ______ _____ 00 o0 00 00 ON ON O ...Oilseeds .n Cereals _lses X sugar, k g Source: Reserve Bank of India Bulletins and G. Pursell and A. Gupta, "Background Statistics, Protection and Incentive Indicator, 1965-95, World Bank Page 34 of 34 Annex 5 Appendix Figure A5.2: All India Oilcake and Cotton Oil Cake Exports and Domestic Cotton Oil Cake Production. 3500 3000 2500 E 2000 9/ 0 1500 O 1000 1 500 e ' CD 0 cv t C 0 X CV CD a) XD 0) 0) CD a: X a) a) Ca Cotton Cake o Total Oilcake Cotton Cake Prodn Source: FAO Trade Yearbook and Ministry of Agriculture, GOI. MFA Phase-out and Annex 6 Cotton and Textile Policies A. Introduction' A6.1 Recent developments in the world arena create major market opportunities, but at the same time exposes India to more intense market competition. The agreement to phase-out the MFA beginning in 1995 creates a new environment for world trade in fibers, textiles and apparel. The abolition of the MFA will vastly increase the scope for countries to expand their exports and create employment in this sector, but the competition will be fierce. The growth potential of India's textile industry is particularly important in light of the reforms in textiles and wearing apparel trade, foreseen under the Uruguay Round agreement of the GATT and the ambitious export-led growth and liberalization programs undertaken by the Indian Government since 1991. However, the implementation of GATT agreements and recent developments in the global economy will also expose India's textile and clothing sectors to more intense competition, both at home and abroad, from other major exporters such as China and other newly industrialized countries. This challenge is especially acute in light of India's textile policies. Traditionally, the cotton spinning, weaving and clothing industries have tended to be viewed as part of integrated cotton sector. Policies have provided incentives for cotton to be processed in India and have consequently discriminated against cotton production. In spite of recent and significant progress, policy interventions remain heavy in the spinning and weaving industries, attempting to reserve parts of each sector for small-scale industry and to maintain employment even at the expense of sharply lowering productivity. High import barriers and domestic taxes have discriminated strongly against man-made fibers, increasing India's reliance on cotton-based products. A6.2 To rise to the challenge India would need to reform its textile policies to take advantage of these new opportunities. Expanding export opportunities contributing to the expansion of -domestic cotton demand from the textile sectors outstripping the supply, may make export controls on cotton redundant. If, on the other hand, the industry shifts strongly to man-made fibers, the same quotas might become impossibly restrictive, greatly inhibiting the production of cotton. Certainly, the liberalization of world trade in textiles and apparel will require reforms to allow increases in the efficiency of the processing industries. Raising the productivity of the industry seems a much more reliable path to job creation and better-paying jobs, than policies designed to preserve employment using traditional techniques in these industries. A6.3 This annex presents the results of a study that investigates the impact of a series of policy reforms that would encourage the textile industry to become more efficient. The impact of these reforms are examined under full and partial abolition of the MFA. B. Overview of the Indian Cotton Textile Industry A6.4 India is one of the largest cotton producers in the world, but remains a small player in the world market. India is the third largest cotton producer (after the USA and China) with 25% of the world cotton acreage and 13% of the world cotton output. India also ranks 4th in the world 1 This Annex is drawn from A. Elbehri, T. Hertel and W. Martin, 1997, "Estimating the Impact of Trade Reforms on the Indian Cotton and Textile Sectors: A General Equilibrium Approach," Background working paper. Annex 6 Page 2 of 20 in terms of staple fiber output and 6th in the production of filament yarn producers. Cotton-based garments dominate India's exports of apparel. A6.5 India's export share has historically been small but began to grow significantly after 1980, reaching a total share of 2.3% of world apparel exports. Indian apparel is exported to over 120 countries, the most significant are the Multi-Fiber Agreement (MFA)-importing countries with a total share of 67.5% compared to 32.5% of exports to non-restricted markets. The MFA provides a framework under which developed countries impose quotas on exports of yarn textiles and apparel from developing countries. The most important export destinations for Indian apparel are the U.S. and Western Europe, which account for 61.2% of total exports.2 By 1986 India's exports hit the quota ceilings imposed by MFA-importing countries. The industry's remarkable growth in the 1980s reflects the inherent advantage which this sector enjoys. By contrast, world trade in textiles and apparel grew at an average annual rate of 12% between 1988 and 1992. In 1992 world imports of apparel were estimated at US$125 billion of which 80% was imported by MFA-importing countries. The major exporters include Hong Kong, China, Korea, Taiwan, and Thailand which share among themselves 40% of world exports. A6.6 In addition to its sheer size, the cotton textile sector has a dualistic manufacturing structure dominated by a fast expanding decentralized small-scale manufacturing segment and a declining vertically integrated, large-scale composite mill segment. For example, the apparel industry consists of 27,700 domestic manufacturers and over 48,000 fabricators, and 1,000 manufacturer exporters. The vast majority (80%) of the manufacturers are small operations (11- 20 sewing machines) and 99% of the domestic manufacturers are of proprietorship/partnership type. The technologies for processing cotton textiles and apparel in India also cover a broad range of techniological sophistication--from hand-operated to sophisticated automated technology.3 Given this structure and the extensive governmental regulations, the overall picture of the textile and apparel sectors in India is one of great potential but under-performance. A6.7 Over the lastfour decades, the Government of India's cotton textile policies centered on two major objectives: assuring an adequate supply of reasonably priced cotton to the textile industry, while promoting growth in employment. Cotton export quotas ensured that a major share of cotton output was processed domestically, which in turn discriminated against cotton production. The spinning and weaving industries continue to operate under a variety of government-imposed restrictions such as yarn export quotas, hank yarn obligation, and barriers to entry and exit, labor utilization and importation of production materials. C. MFA and its Implications for India A6.8 The MFA has been a major force shaping global textile trade. The MFA controls developing country textile product exports to the industrialized nations. The MFA quotas are typically applied on a bilateral basis, under the threat of unilateral restraints to be imposed by the importing country. The quotas that emerge from bilateral negotiations are specific to particular product categories, as defined by fiber and function. This has allowed discrimination not only against specific fibers and products but also among exporting countries.4 The MFA export quotas are administered by the exporting country government, which allocates them based on 2 R. Jaikumar and ICICI, 1995 3 R. Jaikumar and ICICI, 1995, "Beyond the Multi-Fiber Agreement: Strategies for the Indian Apparel Industry," The Industrial Credit & Investment Corporation of India, Ltd. and Harvard Business School. 4 W. Martin, 1996, "The Abolition of the Multi-fibre Arrangement and its Implications for Fiber Markets," International Trade Division, World Bank. Paper for presentation to Conference on the WTO and the Uruguay Round Agreement: Implications for South Asian Agriculture, Kathmandu, April 1996. Annex 6 Page 3 of 20 predetermined criteria. When these quotas are restrictive, there is a substantial quota rent generated. This accrues to the exporting country government, and sometimes the rents are also shared by importing firms when they wield large market shares in the importing countries.5 A6.9 Bilateral quotas are generally more binding for wearing apparel compared to textiles. Using export tax equivalents for 1992 as a measure of the restrictiveness of the quotas, Hertel, et al.6 reported the highest rates for Indonesia with 46% and 48% on wearing apparel exports to North America and Western Europe, respectively. The rates for China and South Asia followed with 36% and 40% to these markets while S. Korea, Taiwan and Hong Kong face relatively lower export taxes for wearing apparel, resulting from high quota rates compared to their productive capacity. Another way to assess the competitiveness among exporting countries is to look at the share of exports to MFA restricted markets. Countries with a small quota relative to their production, export higher share to the unrestricted non-MFA markets. This share was shown to be an important factor in determining whether an exporting country will benefit or not from the elimination of MFA quotas. A6. 10 The MFA is a very costly form of protection for the developed countries for several reasons. First, higher costs to suppliers translate into higher prices for textiles and clothing. Second, MFA-importing countries do not receive quota rents or import duty revenues. Third, bilateral quota agreements create an inefficient system that does not benefit from the flexibility of sourcing of imports across supplying countries. For exporting countries, the impact of the MFA on welfare derives from two sources. On the one hand, suppliers to the restricted MFA-importing countries benefit from higher prices, but on the other hand these same suppliers face lower prices in unrestricted markets. More importantly there is a loss of efficiency to the exporting economy as suppliers are forced to scale back production in a sector for which they have a comparative advantage. Tightly restricted exporters like India are more likely to be net losers under the MFA than less restricted (Bangladesh) or more mature markets (S. Korea, Taiwan, and Hong Kong) who benefit from large quotas relative to their export levels.7 A6. 11 One of the most significant accomplishments of the Uruguay Round Agreement was the Agreement on Textiles and Clothing (ATC) which calls for MFA quotas to be progressively phased out during a 10 year transition period. Leaving aside the potential political pitfalls in the final implementation of the agreement, MFA abolition does offer great opportunities for exporting countries, particularly in South and Southeast Asia, to expand textile and clothing exports and stimulate demand for fibers. A significant benefit will come from a positive impact on employment in these labor-intensive industries. Under the ATC agreement, quota growth rates will increase in three stages: 16% in the first three years, 25% in the next four years, and finally 27% in the final three years. A6.12 During the Uruguay Round, India agreed to a package of tariff reforms that will greatly reduce tariff levels on import competing industries. This will greatly reduce the burden on efficient, export oriented industries like the production of cotton, textiles and apparel both by 5 K.Krishna, R. Erzan, and L. Tan. 1994. "Rent-sharing in the Multi-Fibre Arrangement: Theory and Evidence from US Apparel Imports from Hong Kong," Review of International Economics 2(1):62-73 6 T.W.Hertel, W. Martin, K. Yanagishima, and B. Dimaranan. 1995. "Liberalizing Manufactures Trade in a Changing World Economy," Chapter 4 in W. Martin and L.A. Winters (eds.), The Uruguay Round and the Developing Economies, World Bank Discussion Papers No. 307, Washington, DC (also forthcoming Cambridge University Press). 7 Yang, Y., W. Martin, and K. Yanagishima. 1996. "Evaluating the Benefits of Abolishing the MFA in the Uruguay Round Package," in T.W. Hertel, (ed.) Global Trade Analysis: Modeling and Applications, Cambridge University Press, Cambridge. Annex 6 Page 4 of 20 reducing the costs of the inputs needed for export production, and by reducing the prices of non- traded goods and factors entering into the production process. In addition to its commitments under the Uruguay Round, India unilaterally lowered its tariff protection on a wide range of products, including dramatic reductions in tariffs on imports of man-made fibers. These products were further liberalized under bilateral agreements concluded shortly after the Uruguay Round between India and the USA and India and the EU. A6.13 Harrison, et al. (1995)8 found that elimination of MFA provides particularly large benefits to the industrial country importers that impose this protection. Among the exporters, the Republic of Korea that already benefit from preferential access to the restricted markets will suffer from the agreement because they currently enjoy large quotas compared with emerging exporters such as China and India. One implication is that India could greatly benefit from the abolition of MFA if it is able to improve its competitive position with respect to its main rivals: China, Hong Kong and Taiwan. India will also likely gain more than some other textile and apparel exporters from eliminating the MFA since its has been shown that these quotas tend to discriminate against cotton-based fibers which tend to be more labor-intensive than other fibers. For example, Martin9 finds that, on average, the Hong Kong export tax equivalents are 20% higher on cotton products than on man-made fiber. Since India has a natural comparative advantage in cotton and cotton-based fiber, the abolition of MFA will clearly benefit India's cotton industry as well as the cotton-based textiles sector and the farmers producing this raw material. A6.14 Overall, for India to fully benefit from the liberalized trade regime and the abolition of the MFA quotas, complementary domestic policies that encourage exports and enhance productivity and efficiency are critical to increase the competitiveness of the domestic industries. Among the major distortions that need to be addressed are: a) export quotas on cotton and cotton- based textiles, b) restrictions on the firm activities such as the hank yarn obligation, c) restrictions on imports of textiles and other materials for the production of exports, and d) restrictions on imports of textiles and apparel. India has already embarked on a series of policy reforms to correct some of these restrictions. A6.15 The analysis of the policy reforms under consideration in this paper considers not only the impact of competition between the cotton industry and other sectors within India, but also competition between India and other dynamic suppliers such as China, Indonesia and Thailand in fiercely competitive markets for fibers and their products. Taking these interactions into account requires a global general equilibrium model. D. Methodology A6.16 Overview of GTAP. The impact of the MFA phase-out and domestic regulatory reforms on the Indian cotton and textile sectors is examined using a modified version of an applied general equilibrium model, known as GTAP (Global Trade Analysis Project).10 Appendix A6.1 describes in greater detail the model used in this analysis. The GTAP model is a relatively standard, multi-region model which is currently in use by over one hundred researchers in 30 countries on five continents. The data base builds on contributions from many of these individuals, as well as the twelve national and international agencies represented on the GTAP 8 Harrison, G., T. Rutherford, and D. Tarr. 1995. 'Quantifying the Uruguay Round," Paper presented to the Conference on the Uruguay Round and the Developing Countries, Washington, DC: The World Bank. 9 W. Martin, 1996, op.cit. 10 Hertel, T.W. and M.E. Tsigas. 1996. "Structure of GTAP," Chapter 2 in T.W. Hertel (ed.), Global Trade Analysis: Modeling and Applications, New York: Cambridge University Press. Annex 6 Page 5 of 20 advisory board. Unlike most such models, GTAP utilizes a sophisticated representation of consumer demands which allows for differences in both the price and income responsiveness of demand in different regions depending upon both the level of development of the region and the particular consumption patterns observed in that region. In the simulations presented below, India, and many of the East Asian economies are projected to continue to experience extremely rapid growth rates, so that the income elasticities of demand play an important role in the model. A6.17 On the supply-side, differences in relative rates of factor accumulation interact with different sectoral factor intensities to drive changes in the sectoral composition of output. The GTAP production system distinguishes sectors by their intensities in four primary factors of production: agricultural land, labor, physical capital, and human capital. Thus in a region where physical capital is accumulating rapidly, relative to other factors, it is expected the capital intensive sectors to expand at the expense of labor intensive sectors such as agriculture in East Asia. These "Rybczynski effects" are not present in the partial equilibrium analyses of the cotton sector provided to date. However, they have been found to be important determinants of structural change in rapidly developing economies.1I A6.18 Since it is a general equilibrium framework, GTAP is built on a complete set of economic accounts for each of the 30 regional economies represented. This is based on the year 1992 and incorporates an exhaustive description of inter-industry linkages at the 37 sector level. (In this paper we work with a 13 sector/15 region aggregation of the GTAP data base; see Table A6.1 below.) In addition to differences in intermediate input intensities, import intensities are also permitted to vary across uses. Since much of the global trade in textiles and wearing apparel comprises intermediate inputs, the distinction between sales to final consumers and sales to other firms can be quite important. Lowering the cost of imported wearing apparel to consumers is quite different from lowering the cost of intermnediate imports used by domestic firms which in turn may be competing with consumer-oriented imports in the product market. Finally, the GTAP model employs the common "Armington" specification in which products are differentiated by origin of production. Thus Indian textiles are treated as distinct from (but highly substitutable for) Chinese textiles, etc. This permits the tracking of bilateral trade flows. Table A6.1 Commodity Aggregation in the GTAP Model Rice paddy rice, processed rice Cottext Cotton textiles (yarn and fabric) Wheat wheat Ncottext Non-cotton textile (yam and fabric) Cotton Cotton Cotwapp Cotton wearing apparels Crsgrns other grains Ncotwapp Non-cotton wearing apparels OthCrops non-grain crops LightMnfc leather, etc., lumber, fabricated metal products, other manufacturing LstkProd meat products, milk products, other TMEq transport industries, machinery and equipment livestock products, wool ProcFood other food products, beverages and HeavyMnfc chemicals, rubber, and plastic, primary ferrous tobacco, fisheries metals, nonferrous metals, fabricated metal products, pulp paper, etc., petroleum and coal, nonmetallic minerals NatRes forestry, coal, oil, gas, other minerals Services electricity water and gas, construction, trade and transport, other services (private)., other services (govemment), ownership of I dwellings Source: Authors' model disaggregation of each category. 11 A. Krueger, 1977, 'Growth, Distortions and Patterns of Trade Among Many Countries" Princeton Studies of International Finance, Princeton; W. Martin and Warr P.G. 1993. "Explaining the Relative Decline of Agriculture: A Supply-Side Analysis for Indonesia," World Bank Economic Review 7(3):381-401. Annex 6 Page 6 of 20 E. Policy Reform Simulations A6.19 Five experiments (El-E5) using the GTAP model were conducted to capture the economic impact on the cotton and textile sectors of various policy reform scenarios in the context of a complete or partial phase-out of the Multi-Fiber Agreement. These include the liberalization of cotton and yarn exports, the elimination of the hank yarn obligation, and the abolition of restrictions on foreign investment and on industry structure in the apparel sector. These experiments are summarized in Table A6.2. Table A6.2 Experimental Design and Specification of the Shocks El Groi projectlons 1992-2005; Uruguay Round tariff and export subsidy cuts; GTAP database 1992 Indo-US and Indo-EU agreement: Abolition of MFA quotas E2 Elimination of export tax on cotton and cotton-based textile; Increased productivity of Post-El 2005 database cotton textile sector (.2%/Jyear); Increased labor productivity by 67% (to China level) E3 Growth projections 1992-2005; ATC quota growth projections; Uruguay Round tariff GTAP database 1992 and export subsidy cuts; Reduction in non-cotton textile import tariffs (from 195% to 30%/6); (MFA quotas in place) E4 Elimination of export tax on cotton and cotton-based textile; Increased productivity of Post-E3 2005 database cotton textile sector (.2°/Wlyear); Increased labor productivity by 67% (to China level)(vfFA quotas in place) E5 Indo-US and Indo-EU agreement: (MFA quotas in place) Post-E3 2005 database A6.20 The size of the domestic textile policy reform shocks is specified in the following manner: - Cotton export quotas: The effects of these quotas were represented by an export tax equivalent on cotton. The average export tax equivalent of these quotas was estimated to be 15 percent (see Annex 5).12 * Cotton-based textile export quota: The effects of the yarn export quotas were represented by an export tax equivalent, estimated to be about 15 percent in 1992 (see Annex 5). * Hank yarn abolition: The hank yarn obligation raises the costs of the domestic cotton spinning industry by requiring the use of a technology that would not otherwise be chosen by spinners. The cost of this obligation is minimized by allowing trade in the hank yarn obligation. The ability to observe trade in this obligation allows an estimate of the magnitude of the associated cost. While the nominal obligation is 50 percent, exemptions for export oriented firms and new entrants reduce the overall obligation to roughly 35 percent of total yarn. Applying an average estimated transfer price of Rs 0.5 per kilogram to 35 percent of total spinning implies a productivity cost of roughly 0.2 percent of the output of the spinning sector. The same technical inefficiency parameter was applied to the weaving component of the overall cotton textile sector (see Annex 3) * Productivity improvements in the apparel sector: Productivity in the apparel sector appears to be well below that prevailing in other exporting countries, and to have declined rapidly relative to China and other suppliers. Trela and Whalley (1990)13 used data for 1982 and 1984 and their productivity measures 12 G. Pursell and A. Gupta, 1995,"Background Statistics, Protection and Incentive Indicator 1965-1995," Worldbank 13 Trela, I. and Whalley, J. 1990. "Global Effects of Developed Country Trade Restrictions on Textiles and Apparel." Economic Journal 100:1190-1205. Annex 6 Page 7 of 20 imply that work that took 1 hour in Hong Kong required 5.4 hours in China and 6.5 hours in India. Economist Intelligence Unit (1996)14 presents data for 1993 and 1995 implying that work that requires 1 hour in Hong Kong requires 8.7 hours in China and 14.4 hours in India. Since it seems reasonable to suppose that the level of labor productivity achieved in China could be achieved in India if India followed similar policies, a 67 percent increase designed to take India's labor productivity to China's level was specified. Experiment El: Growth Projections to 2005 and Abolition of MFA Quotas A6.21 Since liberalization in the textile and clothingsectors is phased in over a 10year period, culminating in the abolition of quotas in 2005, this experiment incorporates growth projections from 1992 to 2005. These projections are based on a relatively small number of exogenous shocks, including projections of regional endowments of agricultural land, physical capital, human capital, the state of technology, population and labor force (see Table A6.3). These are derived from a combination of historical data and World Bank projections for the growth of population, labor force, and real GDP and investment.15 Experiment El also models the Uruguay Round import tariff cuts and export subsidies as agreed on in the GATT agreement. In addition, El incorporates the Indo-US and Indo-EU agreements. In these agreements, the US and the EU committed to early abolition of selected MFA quotas while India committed to a phased removal of quantitative restrictions on imports of fiber, yarn, fabric, and apparel as well as lowering import tariffs on textiles and clothing. Experiment E2: Domestic Reforms in Textile Policies with MFA abolition A6.22 This experiment builds on El. It takes as its starting point the equilibrium data base generated by running El and implements several reforms and structural changes in the textile and clothing sectors. The simulated changes in E2 are: a) elimination of cotton and cotton yarn export quotas, captured in the model by the elimination of the 15% export tax equivalent on cotton and cotton-based textiles, b) increased total factor productivity (0.2%) in the cotton textile sector due to the abolition of the hank yarn obligation, and c) abolition of restrictions on foreign investment and on industry structure in the apparel sector, so that labor productivity rises (by 67%) to the level currently enjoyed by China. The results from this experiment may be interpreted as the incremental change in selected variables, in the year 2005, if the Uruguay Round agreement were accompanied by further domestic reforms in the Indian cotton and apparel sector. A6.23 The two experiments El and E2 complete the set of policy reforms and productivity changes investigated in this study. In addition, key components of the policy reforms have been examined individually, through a series of carefully constructed experiments: E3 to E5. The key difference between these experiments and El-E2 is that with E3-E5, the impact of India's reforms in the absence of complete MFA quota abolition is modeled separately. Experiment E3: Incomplete Abolition of MFA A6.24 Experiment E3 differs from El in two respects. First, the MFA quotas are not abolished at the end of the transition period. Projections for ATC quota growth following the UR agreement 14 Economist Intelligence Unit (1996), Textile Outlook International. Economist Intelligence Unit, London. 15 For more details on the methodology used to carry out these projections see Hertel, et al., 1995, op. cit. and K. Anderson, B. Dimaranan, T. Hertel, and W. Martin. 'Asia-Pacific Food Markets and Trade in 2005: A Global, Economy-Wide Perspective," Paper Commissioned for the International General Meeting of the Pacific Basin Economic Council Washington, DC, 20-22 May 1996. Annex 6 Page 8 of 20 Table A6.3 Assumptions Used in the Projections: Cumulative Percentage Growth Rates Over the Period (2.2) (I.6) (11.3) (3.5) (748) Indonesia 129 30 152 242 129 (1.3) _ (2.0) (7.4) (9-9) (6.6) Philippines 33 40 92 109 576 (2.2) (2.6) (5-1) (5.8) (4.4) Thailand 1 8 26 265 150 173 (1.3) (1.8) (10.5) (7.3) (8.0) Malaysia 28 41 214 257 174 (1.9) (2.7) (9.2) (10.3) (8.1) Korea I11 12 209 119 131 (0.8) (0.9) (9.1) (6.2) (6.7) Taiwanl 11 18 211 119 115 (0.8) (1.3) (9.1) (6.2) (6.1) Hong 11 13 158 83 117 Kong/Singapore (0.8) (0.9) (7.6) (4.8) (6.1) Japan 3 -2 69 81 44 (0.2) (-0.2) (4.1) (4.7) (2.8) Australia/New 15 16 46 155 54 Zealand (1.1) (1.1) (3.0) (7.5) (3.4) NAFTA 15 18 56 95 40 (1.1) (1.3) (3.5) (5.3) (2.6) Western Europe 2 1 36 217 37 (0.2) (0.1) (2.4) (9.3) (2.5) Former Soviet 6 8 10 10 10 Union (0.4) (0.6) (0.7) (0.7) (0.7) India 24 31 98 107 94 l = l _________ (1.7) (2.1) (5.4) (5.6) (5.2) Rest of the 32 28 50 133 57 World (2.2) (2.6) (3.2) (6.7) (3.5) Source: Authors' modifications of World Bank projections. were used to capture the resulting expansion of quota access. The second difference from El is that additional cuts in Indian textiles and apparel tariffs from the Indo-US and Indo-EU agreement were not included since these agreements include provisions for withdrawing India's concessions on these goods should the developing countries fail to abolish the MFA quotas. However, E3 implemented the reduction in the import tariff on man-made fibers (non-cotton textiles) from 195% in 1990 to 30% (starting in 1995). Experiment E4: Domestic Reforms with Incomplete Abolition of MFA A6.25 This experiment builds on E3 and uses the post-E3 2005 database to simulate selected policy and structural changes. E4 is similar to E2 except that the ATC quotas remaining in 2005 are not abolished. This experiment was designed to measure the interaction between India's domestic reforms and the UR agreement -- specifically abolition of the quotas on textiles and apparel exports. A6.26 Experiment E5: Indo-US and Indo-EU Agreement. This experiment also builds on E3 and uses the post-E3 2005 database to simulate the effects of additional cuts in Indian textiles and apparel tariffs from the Indo-US and Indo-EU agreement. The agreement also calls for selected abolition of the quotas from the US and EU. Annex 6 Page 9 of 20 F. Economic Impact of Policy Reforms Abolition of the MFA (El) A6.27 The combination of the sectoral growth, the Uruguay Round liberalization measures, MFA phase-out and complete abolition, and the US-EU agreement contribute to the decline in agriculture's share in GDP (first seven entries in column 1) (Table A6. 4). This is in line with historical evidence as development occurs. The sharpest declines are for the grains sector for which the income elasticity of demand is typically relatively low. This decline is much smaller for cotton, which experiences a significant increase in derived demand due to abolition of the MFA quotas. Cotton output would increase by an estimated 72% by 2005, equivalent to about 1.3 million mt of cotton. This is a remarkable increase in cotton production to be achieved over a 13 years period. By comparison, it took India the last 15 years to increase its cotton production by about 500,000 mt (a 40% increase, then). The magnitude of the projected increase in cotton production gives a measure of the future demand challenges and growth opportunities facing cotton growers. Other crops, meat and livestock, and processed food outputs also increase their output by more than grains due to the higher income elasticity of demand for these products. Notably, where sectoral output (column two) increases more rapidly than economy-wide GDP (97.9%), the sectoral share in GDP rises. When the rate of output increase is below average, the share of GDP generated declines in that sector. Table A6.4 Change in Composition of GDP, Exports, Import, and the Trade Balance for India; 1992-2005 Rice -25.1 48.2 219.3 72.5 1069.2 Wheat -21.1 56.2 317.9 16.2 -66.3 Coarse grains -31.6 35.4 104.4 4.6 4.7 Cotton -13.1 72.0 114.9 127.3 -161.5 Other crops -10.9 76.2 149.9 167.7 1359.7 Meat-livestock -13.0 72.2 1091.9 119.2 1422.8 Processed food -13.7 70.7 141.7 702.6 243.8 Natural resources 5.4 108.6 92.6 103.2 -6357.1 Cotton textile 1.5 100.9 199.4 333.9 2378.5 Non-cotton textile -4.8 88.4 196.3 868.3 527.6 Cotton clothing 98.3 292.4 622.7 278.2 8875.9 Non-cotton clothing 103.1 302.1 745.9 267.9 3542.7 Light manufacturing 8.7 115.2 157.5 58.8 3324.2 Transportation/equipment -5.9 86.2 149.7 121.5 -4113.6 Heavy manufacturing 1.1 100.1 127.1 150.6 -7748.3 Tei7vices 5.1 108.1 127.1 79.0 1615.9 All industries 0.0 97.9 196.3 125.3 5918.3 Source: Authors' simulation results. A6.28 In contrast to food and agriculture, the share of GDP generated in highly labor-intensive industries --wearing apparel, light manufacturing and services-- increases significantly. A quadrupling of apparel industry output is fueled by abolition of the MFA quotas and tariff reductions under the Uruguay Round agreement. The apparel industry would become the fastest growing sector in India, increasing its GDP share by an estimated 100%. The employment opportunities generated in this highly labor-intensive industry would be enormous, with a potentially significant impact on poverty reduction since the labor skill requirements in the apparel industry are not as demanding as in other industrial sector and also tend to favor women' s employment. In other words, the MFA phase-out offers tremendous potential for a more participatory growth, which further and better investments in basic education and health would Annex 6 Page 10 of 20 help capture. India's comparative advantage also shows up in light manufacturing. The increase in the share of GDP generated by services is indicative of strong domestic demand and an income elasticity of demand in excess of unity. A6.29 India becomes a net importer of cotton. In spite of the remarkable projected increase in domestic cotton production, India is projected to become a net importer of cotton to fuel the demand of its cotton-based textile industry. The projected net trade deficit in cotton is estimated at (1992) US$ 166 million by the year 2005.16 This result suggests that cotton export quotas may become counterproductive by depressing rather than promoting domestic cotton production. With cotton export quotas maintained (assuming a constant 15% rate of export taxation over this period), exports are projected to increase by 115% while imports rise by 127%. This suggests that while it is beneficial for India to export some of its cotton varieties and import others, cotton export quotas and their implementation may stand in the way of capitalizing on such arbitrage opportunities. A6.30 The largest improvements in commodity-specific trade balances are for cotton textiles and apparel. Indeed, the combined increase in net exports (exports less imports) of textile and apparel products is $US 15,323 million. This is fueled by massive increases in apparel exports (623% and 746% in cotton and non cotton, respectively). Textile imports rise much more rapidly than exports. The rapid increase in textile imports is because of the very large increase in derived demand fueled by much higher apparel exports in a post-Uruguay Round setting. In the case of non-cotton textiles, where imports increase by more than 800%, this propensity to increase imports is further fueled by the very substantial reductions in tariffs begun under recent trade reforms, and extended under the agreements with the US and the EU. However, despite these dramatic increases in gross imports, India is projected to remain a significant net exporter of textile products in 2005. This is a reflection of the current surplus of exports over imports. Although exports are projected to grow more slowly over this period, they are growing from a much larger base. Domestic Reforms in Textile Policies with abolition of the MFA A6.31 Table A6. 5 reports the incremental impact on the Indian economy in 2005 of the package of reforms identified in experiment E2. This reform package scenario is explored in the context of the MFA quotas fully abolished in 2005, so that unfettered trade in textiles and wearing apparel is achieved. A6.32 As shown by experiment E2, the elimination of the export taxes on cotton results in substantially (67%) higher exports in the year 2005 than would have been the case in the absence of these reforms. The elimination of the cotton export quotas raises cotton prices by 3% in real terms, and further boosts cotton production by a projected 5%, and cotton exports by 67%. This has a positive effect on the trade balance for cotton, but not enough to outweigh the change envisioned under El. The cotton output increase is dampened somewhat by the tendency of domestic textile producers to substitute man-made fibers for higher priced domestic cotton. This follows from abolition of the 15% export tax equivalent, which raises cotton prices, leading to substitution away from cotton fiber in the textile industries. On the other hand, the productivity increases in the cotton textile and wearing apparel sectors lead to expansion of these sectors and hence higher demand for cotton inputs. The net result of these two opposing effects is an increase 16 Since the initial trade balance for cotton is negative $US 5 million, the projected increase in the difference in the value of fob exports and cif imports under full implementation of the Uruguay Round agreement is projected to be negative $US 161 million, implying a net trade deficit in cotton of $166 million in 2005. Annex 6 Page 11 of 20 in cotton output, implying that the expansion in derived demand fueled by the downstream productivity demands dominates the inter-fiber substitution effect. A6.33 In the case of cotton textiles, elimination of the export restrictions combines with Table A6.5 Impact of Domestic Reforms on Indian Output, Exports, and Imports in the Year 2005. Rice 0.4 -4 -82.4 4 -0.1 -2.4 1.4 -310 Wheat -0.2 -7.5 4.3 -24.5 0.0 -2.6 1.5 -8.0 Coarse grains 0.0 -8.2 4.7 -0.6 0.0 -2.8 1.6 -0.2 Cotton 4.8 66.9 8.1 14.5 3.2 77.5 3.5 47.5 Other crops -0.2 -7.3 4.7 -244.3 -0.1 -2.4 15 -86.5 Meat-livestock -0.1 -4.0 5.4 -90.3 0.0 -1.3 1.9 -28.3 Processed food -1.2 -8.0 4.4 -381.7 -0.4 -2.6 1 .4 -134.7 Natural resources -1.6 -4.1 0.6 -424.2 -0.5 -1.3 0.2 -154.3 Cotton textile 10.1 60.9 74 1948.3 5.1 36.7 2.2 984.2 Non-cotton textile -0.2 -6.3 4.9 -158.5 0.2 -0.8 1.4 -22.2 Cotton clothing 15.1 20.1 -7.3 1881.2 5.9 8.3 -14.0 339.0 Non-cotton clothing 10.8 14.5 -7.7 534.1 5.7 8.8 -14.1 122.4 Light manufacturing -1.9 -8.6 4.0 -501.2 -0.7 -2.8 1.3 -188.4 Transportation/Equip. -2.4 -9.9 3.9 -762.1 -0.8 -3.3 1.3 -263.9 Heavy manufacturing -0.7 -4.9 2.4 -707.8 -0.2 -1.5 0.8 -233.1 Services 0.1 -5.8 4.2 -704.6 0.1 -2.0 1.4 -240.5 All industries 0.4 3.5 2.6 296.0 0.2 1.2 0.9 102.0 Source: Authors' simulation results abolition of the hankyarn obligation to make this sector more competitive internationally as well. Cotton textile output in 2005 increases by 10%, relative to the level achieved under El, and there is another $2 billion added to the trade surplus in this commodity group. A6.34 The elimination of investment restrictions in the apparel industries, which is assumed to raise labor productivity standards to the level of China, has a substantial impact on output (Table A6. 5). The largest increase is in cotton apparel, where output rises by 15% and exports by 20%. Non-cotton apparel follows with increases of 11% and 15%, respectively. When combined with reductions in imports, these developments generate an additional $2.4 billion trade surplus in apparel in the year 2005. Domestic Reforms with Incomplete Abolition of MFA A6.35 In this case, we consider what happens if the bilateral quotas are accelerated under the ATC, but complete abolition of the quotas fails in 2005. Thus, Indian exports to the US and EU markets remain constrained, albeit at a higher level than at present. The non-abolition of the remaining ATC/MFA quotas in 2005 would imply a slightly smaller increase in cotton and (cotton and non-cotton) textile output compared to the complete MFA abolition scenario. The incomplete abolition of the ATC/MFA quota, however, would have large implications for the clothing industry, since it would cut by an estimated two-third the increase in the value of output (in constant dollar terms) in that industry; the value of cotton and non-cotton clothing industry would still rise by an estimated 80%, compared to an estimated 270% when the MFA/ATC quotas are fully eliminated. Continued restricted access to the US and EU markets would therefore have significant implications for India in terms of foregone growth, employment and poverty reduction opportunities in the clothing industry. Nevertheless, it is still the case that the size of the clothing industry would increase significantly. Annex 6 Page 12 of 20 A6.36 Focusing on the impact of domestic reforms (scenario E4), the difference in the results between this scenario (E4) and the earlier reform scenario (E2) captures the interaction of domestic reforms with MFA quota abolition (Table A6.5). We hypothesize that the domestic reforms will be more valuable if the ATC quotas are fully abolished. When the same policy change is undertaken where the ATC quota constraints remain, the results are not as encouraging. In terms of the trade balance in these products, the change in net apparel exports is now $450 million, as opposed to the $2.4 billion noted above. The increases in cotton textile and non- cotton apparel output are also only half as large, and the cotton apparel output growth under E4 is only one-third as large as E2. As a result of the much smaller increase in derived demand for cotton from the textile and apparel industries, the net trade deficit in cotton diminishes by about $ 50 million. In the next section, we analyze the effects of the individual components of the domestic reform package on output and trade. Individual Domestic Reforms with or without Abolition of the MFA A6.37 The abolition of the export quotas on cotton alone would have a larger impact on cotton Table A6.6: Impact of Removing Cotton Export Tax on Indian Output, Exports, and Imports in the Year 2005 Wheat - 0.0 -0.1 0.05 -0.31 0.0 -0.16 0.08 -0.46 Coarse grains 0.0 -0.11 0.06 0.0 0.0 -0.18 0.09 -0.01 Cotton 0.81 86.45 0.22 52.43 1.08 85.08 0.3 62.17 Other crops -0.02 -0.09 0.04 -3.02 -0.02 -0.14 0.07 -5.1 Meat-livestock 0.0 -0.06 0.07 -1.26 0.0 -0.09 0.11 -1.83 Processed food -0.02 -0.08 0.03 -3.51 -0.02 -0.13 0.06 -6.55 Natural resources -0.01 -0.01 0.0 -1.07 -0.02 -0.03 0.0 -2.73 Cotton textile -0.04 -0.13 0.05 -5.27 -0.02 -0.1 0.1 -3.34 Non-cotton textile -0.02 -0.05 0.01 -1.09 -0.02 -0.04 0.04 -0.99 Cotton clothing -0.1 -0.14 0.06 -13.69 -0.04 -0.06 0.13 -2.74 Non-cotton clothing -0.08 -0.11 0.04 -4.27 -0.03 -0.05 0.1 -0.76 Light manufacturing -0.02 -0.05 0.02 -3.08 -0.03 -0.1 0.04 -7.07 Tmeq -0.01 -0.06 0.02 4.14 -0.03 -0.11 0.04 -9.02 Heavy 0.0 -0.03 0.01 4.81 -0.01 -0.06 0.03 -8.49 manufacturing l Services 0.0 | -0.03 0.01 -3.3 0.0 -0.07 0.03 -7.41 Source: Authors simulation results. exports, but a smaller impact on cotton production, than does the complete package of reforms (Table A6.6). Clearly, the improvements in productivity included in the complete package of domestic reforms increased dramatically the domestic demand for (and production of) cotton, hence reducing exports. Without such increase in domestic demand for cotton, cotton production increases little. The abolition of the cotton export quotas with the ATC quotas in place gives rise to a larger increase in cotton exports because the reduction in domestic use of cotton following the increase in the cotton price is reduced by an induced increase in the price of cotton textiles. A6.38 The removal of India's export quotas on cotton textiles could result in a dramatic expansion in the exports of these goods, particularly in the absence of the ATC quotas. In the presence of the ATC quotas, exports of textiles still grow by 36 percent (Table A6.7), but they can only expand into non-quota markets. Consequently, India's terms of trade for these goods deteriorate more rapidly than in the absence of these restrictions. Removing cotton textile export Annex 6 Page 13 of 20 quotas has a much larger impact on cotton production than the elimination of the cotton export quota alone; the impact is much lower when ATC quotas remain in place. A6.39 The abolition of the hank yarn obligation directly results in an increase in exports of cotton textiles (Table A6.8). The reduction in the price of cotton textiles stimulates the cotton clothing industry and results in an expansion in output and exports of these goods. The increase Table A6.7: Impact of Removing Cotton Textile Export Tax on Indian Output, Exports, and Imports in the Year 2005 Rice -0.17 -24Z I.31 -27.94 -0.12 -1.46 0.8 -19.13 Wheat -0.11 -2.57 1.35 -7.8 -0.06 -1.58 0.85 -4,69 Coarse grains -0.03 -2.83 1.52 -0.21 -0.01 -1.75 0.94 -0.14 Cotton 3.67 -5.59 4.72 -19.71 1.96 -3.26 2.56 -10.24 Other crops -0.46 -2.21 0.98 -68.39 -0.21 -1.4 0.71 48.54 Meat-livestock -0.1 -1.27 1.71 -27.74 -0.05 -0.82 1.07 -16.81 Processed food -0.5 -2.76 1.32 -126.14 -0.37 -1.73 0.85 -87.26 Natural resources -0.62 -1.36 0.08 -119.14 4.54 -0.83 0.04 -75.86 Cotton textile 8.62 65.25 3.32 2113.4 -0.21 36.27 2.03 974.64 Non-cotton textile -0.93 -2.42 1.03 -52.35 -0.17 -0.73 0.89 -16.86 Cotton clothing -3.33 -5.21 1.09 -508.61 0.01 -0.92 2.61 -39.23 Non-cotton clothing -2.98 -4.9 3.69 -184.67 -0.54 -0.79 2.43 -11.31 Light manufacturing -0.78 -3.19 1.35 -182.96 -0.56 -1.97 0.83 -132.41 Tmeq -0.89 1 -3.69 1.35 -272.26 -0.21 -2.28 0.84 -179.65 Heavy manufacturing -0.32 T-1.78 0.78 -241.93 -0.03 -1.09 0.46 -146.02 Services 1.33 -241.77 0.08 -1.34 0.83 -156.59 Source: Authors' simulation results Table A6.8: Impact of Removing the Hank Yarn Obligation on Indian Output, Exports, and Imports in the Year 2005 i~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~-r Rice --0.6 0.0 -0.04 0 -0.0 0.03 0.02 -0.34 Wheat 0.0 -0.09 0.05 -0.31 0.0 -0.03 0.02 -0.12 Coarse grains 0.0 -0.09 0.05 0.0 0.0 -0.03 0.02 0.0 Cotton -0.04 -0.06 0.0 -0.04 -0.07 0.03 -0.05 0.1 Other crops 0.0 -0.0S 0.06 -2.92 0.0 -0.03 0.03 -1I.17 Meat-livestock 0.0 -0.04 0.06 -0.99 0.0 -0.01 0.02 -0.31 Processed food 0.0 -0.08 0.05 4.24 0.0 -0.02 0.02 -1.24 Natural resources -0.01 -0.06 0.01 -6.8 0.0 -0.02 0.01 -3.94 Cotton textile 0.13 0.85 -0.46 33.6 0.04 0.44 -0.53 14.57 Non-cotton textile -0.01 -0.06 0.04 -1.57 0.0 0.0 0.01 -0.19 Cotton clothing 0.18 0.25 -0.13 24.49 0.05 0.1 -0.23 4.09 Non-cotton -0.1 -0.15 0.08 -5.71 -0.01 -0.01 0.01 -0.14 clothing Light manufg -0.01 -0.09 0.04 -5.21 0.0 -0.02 0.01 -1.21 Tmeq -0.02 -0.11 0.04 -8.2 0.0 -0.03 0.01 -2.25 Heavy manufg 0.0 -0.04 0.02 -6.6 0.0 0.0 0.0 -1.36 Services 0.0 -0.06 0.05 -7.83 0.0 -0.01 0.02 -2.39 Source: Authors' simulation results Annex 6 Page 14 of 20 in exports of cotton textiles is much lower in the presence of the ATC quotas, and the increase in exports of cotton clothing is virtually eliminated in this case. A6.40 The liberalization and consequent productivity improvement in the apparel industry has a strong impact on the production and exports of clothing. Increase in output of cotton clothing generates a substantial demand for cotton textiles, and causes a substantial reduction in net exports of cotton (Table A6.9). Clearly, this policy reform will reduce the likelihood of export quotas being a binding constraint on exports. The increase in exports of apparel, and the induced reduction in raw cotton exports, are very substantially reduced when the ATC quotas remain. Table A6.9: Impact of Increased Productivity of Apparel on Indian Output, Exports, and Imports in the Year 2005 (, I 01 W1 01 I g i1 1|1 I 10 | 61 5.0 g : Wheat -0.04 -4.76 1 2.76 1-15.57 10.03 r o 84 1 0.52 -2.76 1 Coarse grains 0.0 -5.21 2.95 -0.38 0.02 -0.91 0.52 -0.08 Cotton 0.62 -5.45 3.09 -14.16 0.24 -1.05 0.68 -2.88 Other crops 0.26 -4.86 3.49 -167.52 0.1 -0.87 0.68 -32.7 Meat-livestock 0.0 -2.58 3.45 -58.27 0.05 -0.43 0.64 -9.53 Processed food -0.68 -5.13 2.83 -244.18 -0.05 -0.77 0.48 -41.1 Natural resources -0.92 -2.66 0.47 -292.7 -0.09 -0.46 0.18 -72.54 Cotton textile 1.66 -3.75 4.55 -169.41 0.47 -0.09 0.59 -5.3 Non-cotton textile 0.42 -3.69 3.47 -99.06 0.43 -0.04 0.5 -4.36 Cotton clothing 19.33 26.64 -10.82 2491.6 6.29 9.51 -16.1 387.85 Non-cotton 10.41 14.47 -11.34 535.42 5.88 9.97 -16.17 138.31 clothing l_ll__l__ll_ Light manfg -1.08 -5.34 2.44 -308.83 -0.11 -0.74 0.36 -50.06 Tmeq -1.4 -6.13 2.37 -468.72 -0.18 -0.91 0.37 -75.46 .Heavy manufg -0.34 r -2.97 1.52 -444.03 0.04 -0.4 0.31 -78.67 IServices 0.211 -3.62 2.66 -442.85 0.11 -0.56 0.5 -76.14 Source: Authors' simulation results Welfare Gains from Domestic Reforms in Textile Policies A6.41 Domestic reforms in the absence of ATC quotas result in overall gains total $1,928 million. These gains derive from terms of trade improvements (about 15%), allocative efficiency gains from eliminating export tax distortions (30%), and improved efficiency in the cotton textile and apparel sectors (Table A6.10). Dividing the impacts up between the shocks considered in Table A6.6 to A6.9 makes it clear that the largest source of this welfare gain is the increase in productivity in apparel. The allocative efficiency gains result primarily from the removal of the cotton textile export tax, and from the increase in productivity in apparel, both of which expand activities that are currently below their optimal size. The terms of trade gain is attributable to the increase in productivity in the apparel industry. The deterioration in India's terms of trade in apparel is more than offset by improvements resulting from falling exports of other goods whose export prices are more sensitive to India's export levels. A6.42 Welfare gains from domestic reforms remain substantial even in the absence of a full MFA phase-out. If ATC quotas are not eliminated by 2005, welfare gains are reduced by more than half, but the welfare gains remain substantial at about $ 800 million. The second half of Table A6.8 shows the impact of these reforms in 2005 with binding ATC quotas still in place. The overall increase in GDP (0.4% under E2) is cut in half. Annex 6 Page 15 of 20 A6.43 Overall, the domestic reforms are significantly more valuable in the context offree trade in textiles and apparel. The gains in Indian real income from the domestic reforms are more than twice as large under E2 than E4 (Table A6. 10). In terms of the individual components of welfare, the largest reductions are in the terms of trade, where the benefits of increased apparel productivity fall dramatically as India's major export markets become quota constrained. Impact of Agreements with the US and the EU A6.44 The net impact of the US-EU agreement on the Indian economy in the year 2005 is about a $100 million improvement in the total trade balance for textiles and apparel. Overall, however, the impact of the US-EU agreement on the Indian economy is much smaller than the domestic reform package discussed above. This is reflected in the smaller welfare gain to India, $239 million (Table A6.11). Relaxation of the quotas on textiles and apparel gives a boost to exports-especially of apparel. However, the tariff reduction in Indian textiles and apparel also generate a very large increase in imports. Table A6. 10 Impact of Domestic reforms on India's Welfare in Year 2005 (1992$US mill.) QUOTAS ABO)LISBED Removing cotton export tax -0.47 -3.46 2.988 na na Removing cotton-textile export tax 119.45 -201.78 321.24 na na Removing Hank Yarn Obligation 83.65 -0.35 2.59 81.41 na Raising productivity on apparel 1700.5 476.13 254.67 na 969.68 All reforms combined (E2)* 1928.2 292.2 592.3 85.6 958.1 QUOTAS NOT-ELIMINATED _ Removing cotton export tax 4.28 -0.69 4.98 na na Removing cotton-textile export tax 163.7 -10.2 173.9 na na Removing Hank Yarn Obligation 66.65 -2.45 -2.47 71.57 na Raising productivity on apparel 577.41 61.91 73.54 na 441.9 All reforms combined (E4)* 809.2 44.8 249.4 73.3 441.7 Note: *The sum of individual reforms do not exactly add up to the combined effects due to rounding errors, and because of interactions between the reformn options. Source: Authors' simulation results G. Summary and Conclusions A6.45 This study sought to evaluate the potential gains from domestic reforms in India's cotton and textile sectors, in light of projected developments in the world economy in the coming decade. Emphasis is placed on the impact of the Uruguay Round reforms on trade in textiles and apparel and on the environment they create for complementary reforms in India's policies towards the sector. The domestic reforms considered include: abolition of export restrictions on cotton and cotton textiles, abolition of the hank yarn obligation, and elimination of restrictions on foreign and domestic investments in the apparel industry. We find that the gains from this reform package are quite significant, amounting to as much as $1.9 billion in the year 2005. The size of these gains is heavily dependent on elimination of the MFA/ATC quotas, scheduled to be abolished in phases between now and the year 2005. However, even in the absence of quota abolition the gains from reform total more than $800 million/year in 2005. Annex 6 Page 16 of 20 Table A6.11 Impact of Indo-US and Indo-EU Agreement on Indian Output, Exports, and Imports in the Year 2005. [Experiment E5] Rice 0.01 -0.17 0.11 -2.28 Wheat r 0.02 -0.2 0.13 -0.71 Coarse grains 0.01 -0.21 0.12 -0.01 Cotton -0.34 0.07 -0.27 0.87 Other crops 0.04 -0.21 0.19 -7.99 Meat-livestock 0.02 -0.09 0.14 -2.01 Processed food 0.01 -0.17 0.13 -9.89 Natural resources -0.03 -0.13 0.04 -18.11 Cotton textile -0.06 2.85 62.05 -145.16 Non-cotton textile -0.75 1.74 13.17 -73.17 Cotton clothing 2.93 7.85 195.51 306.14 Non-cotton clothing 0.41 2.63 193.97 27.34 Light manufacturing -0.03 -0.17 0.09 -12.14 Tmeq _ -0.04 -0.22 0.09 -18.11 Heavy manufacturing _-0.01 -0.08 0.04 -12.51 IServices 0.02 -0.12 0.11 -16.44 JAll industries 0.07 0.71 0.61 15.79 Source: Authors' simulation results Annex 6 Page 17 of 20 H. Appendix - Structural Specifications of the GTAP Model A6.46 The impact of MFA phase out and domestic reforms on the Indian cotton and textile sectors was investigated using a modified version of an applied general equilibrium model, known as GTAP (Global Trade Analysis Project). The GTAP model is built on contributions made by many of researchers, as well as the twelve national and international agencies represented on the GTAP advisory board. It is based on a complete set of economic accounts for each of the 30 regional economies represented. This model used is based on the year 1992 and incorporates an exhaustive description of inter-industry linkages at the 37 sector level. For this study, a 13 sector/15 region aggregation of the GTAP data base is used (see Appendix Table A6.1) A6.47 Throughout the analysis, the standard GTAP parameters as documented in chapter four of Hertel (ed.) (1996)17 is used, with two exceptions. First, the income elasticities of demand for selected products have been updated. Secondly, larger values were used for the Armington elasticities of substitution used to specify the extent to which similar products from different countries substitute for one another. In his analysis of changing trade shares in the Asia-Pacific region over the decade of the 1980, Gehlhar (1995)18 found that the standard GTAP trade elasticities were too small to accurately predict changes in trade shares over this 10 year period. By increasing these parameters he obtained a better prediction of historical changes in export shares in the region. Accordingly, the standard parameter values in the 13 year projections reported here have been doubled. A6.48 A final, methodological point relates to the treatment of the bilateral export quota constraints. These have been modeled explicitly. Traditionally quota restrictions have been modeled indirectly in GTAP by endogenizing the export tax rate and exogenizing the quantity variable for the bilateral flow of textiles and clothing. While this method is appealing for its simplicity it is inadequate in cases when the quota changes status during the course of simulation (i.e. from binding to non-binding or vice versa). An improved method of explicitly modeling quotas developed by Bach and Pearson (1996) called GEMPACK19 is therefore used in application. A6.49 Region and Sectoral Aggregation. Since it is cumbersome to conduct projections with the full 37-sector, 30-region GTAP data base, aggregation was performed up to a level which highlights sectors and countries of interest for this particular study. The 30 regions in GTAP were aggregated into 15 regions allowing us to identify India individually, as well as the restricted textile and wearing apparel markets in North America and Europe. The regional aggregation also separates out most of the key export competitors in these markets. A6.50 Table A6.1 displays the 16 aggregated sectors employed for this analysis, and their relationship to the 37 sectors available in the GTAP database. The agricultural sector is fairly disaggregated due to the relative importance of this sector to India, and owing to its importance in terms of the Uruguay Round. Within agriculture, cotton is treated separately because it is a major 17 T.W.Hertel,1996, Global Trade Analysis: Modeling and Applications, New York: Cambridge University Press. 18 M. Gehlhar, 1994, "Economic Growth and Trade in the Pacific Rim: An Analysis of Trade Patterns," Unpublished Ph.D. dissertation. 19 F.C. Bach and K.R. Pearson. 1996. "Implementing Quotas in GTAP Using GEMPACK - or How to Linearize an Inequality," GTAP Technical Paper No. 4. Annex 6 Page 18 of 20 focus of this study. Textiles and wearing apparel are also an important part of the study, since they represent the predominant end-use of cotton. Here, four separate sectors were included: cotton textile, non-cotton textile, cotton apparel, and non-cotton apparel. The light and heavy manufacturing, which have very different primary factor intensities, were also disaggregated. Transportation equipment and machinery represent about 40% of global trade, and this commodity category is an especially important component of rapid growth in the Asia-Pacific region. When combined with services, these sectors exhaust all economic activity. A6.5 1 Modifications to the Data Base. The version 3 GTAP data base aggregates all non-grain crops into a single category. Likewise, textile products and wearing apparel are also not disaggregated between cotton and non-cotton based products. In order to obtain the level of commodity detail shown in Table A6.1, it was necessary to break these three sectors into two parts: that which is cotton-based, and that which is not. In order to do so, a two stage strategy was followed. Stage I involves getting the right overall size and exportldomestic use for each of the disaggregated sectors in the regions represented in Table A6.1. Stage II involves implementing the appropriate inter-industry mapping of usage among these sectors. A6.52 The most difficult part of breaking out sectors in a global, general equilibrium data base stems from the need to retain overall balance between supply and demand, expenditures and revenues, income and outlays, etc. This problem is vastly complicated by the existence of full bilateral detail in the data base, as well as the presence of inter-industry sales and the sourcing of imports by use. The best approach is to implement share-based splits of existing sectors. To do this, a special purpose DAGG software (1996 version) developed by Mark Horridge was used. Furthermore, the splits must be done in a particular sequence in order to guarantee, for example, that exports will equal the difference between production and domestic use, and that, when summed over sources, they equal the total demand for imports in any given region. A6.53 The first step in disaggregation involves splitting the cotton-based sectors' exports out of their respective aggregates. The assumed shares and sources used are reported in Appendix Table A6. 1. Of particular note is the high intensity of cotton-based products in total textile and apparel exports (67% and 75%, respectively, relative to a world average of 51%). On the other hand, raw cotton exports are very small, reflecting the attempts of the Indian government to encourage domestic use, via export taxes and periodic bans. By comparison, the share of cotton in non-grain production is almost three times as high as the 2.6% export share. The export splits from each source are implemented uniformly across destination regions. Once these disaggregated flows are aggregated by destination, over all sources, we have an estimate of how imports must be split apart in the destination economy. In light of the data in Appendix Table A6. 1, the larger the share of Indian imports in a region's total textile or apparel imports, the higher the post-split intensity of cotton-based products. A6.54 Step II involves estimating the cotton-intensity of domestic demands and using this to split out intermediate and final demands for cotton, textiles and wearing apparel in the GTAP data base. The second part of Appendix Table A6.1 displays the shares and sources for the domestic use splits. For textiles and wearing apparel, these are based on fiber content in all products, and the splits are therefore identical in both the textiles and wearing apparel sectors. For cotton, an estimate of direct cotton consumption is obtained from the FAO data base, and this is compared with estimated total, non-grains crops use in the GTAP data base, to obtain a share split. This split varies by country. For India, the estimate is 10%. Annex 6 Page 19 of 20 Appendix Table A6. 1: Shares Used in Splitting the Cotton/Textile/Wearing Apparel Factors Exports o6f Cotton apparel from India 75% total apparel Ramakrisha, Annex III Cotton apparel from Other 51% total apparel Ramakrishna, p. 1 Cotton textiles from India 67% total textiles Ramnakrishna, Annex ItI Cotton textiles from Other 51% total textiles Ramakrishna, p. I Cotton from India 2.6% non-grain crops Pursell, App. Table 8 Cotton from Other varies by country FAO Domestic use of Cotton apparel India 70% of all textile fibers E. Thigpen App. (last page) and textiles USA 44.8% total Chaudhuri, Appendix I EC 46.7% total Chaudhuri, Appendix I JPN 45. 1% total Chaudhuri, Appendix I Other developed 46.0% total Chaudhuri, Appendix I Developing 59.7% total Chaudhuri, Appendix I Cotton All countries varies by country FAO A6.55 Once the estimates of exports and domestic usage were obtained, by region, the required output, by sector and region is estimated. This is done using a special purpose program, BALDATI.TAB. The output of this program are the weights to be used in disaggregating factor demands, and hence total output, by industry. Step I is completed by checking the global data base in order to ascertain that it is still fully balanced. This is a very good check on the accuracy with which Step I has been implemented, since any mistakes will automatically result in an unbalanced data set. A6.56 Step II of the disaggregation procedures involves specifying the desired mapping of inter-industry flows in the data base. In particular we strive to achieve the following: 1. Own-use of cotton in the aggregated non-grain crops sector is assigned to the disaggregated cotton sector. Similarly for the own-use of cotton textiles and cotton wearing apparel. 2. Correct the preliminary allocation of non-grain crops and cotton flowing to final demand and processed food after Step I by shifting cotton usage to the textiles sector and shifting non-grain crops usage to offset this shift. 3. Reallocate cotton intermediate sales resulting from Step I so that the predominant flow is to the cotton-based textile sector. 4. Reallocate cotton textile sales to apparel so that the predominant flow is into the cotton-based apparel sector. These changes to the data base are implemented using a second program, BALDAT2.TAB, which also checks the basic accounting relations and writes the final data base. Modifications to the Model A6.57 Over the past 15 years, the fiber content of Indian textiles has been changing. In particular, the share of man-made fibers (MMF) has been increasing. This is a logical response to a continuing decline in the domestic price of synthetic fiber, relative to cotton, resulting from: (a) the lowering of customs and regulatory duties during the reforms of the 1980's and early 90's, and (b) periodic high prices for cotton during seasons of low cotton output. In order to capture the potential for future fiber substitution two new equations to the GTAP model are introduced, which is implemented in linearized form, but solved via non-linear methods: Annex 6 Page 20 of 20 EQUATION FIBERPRICE ! Price of fiber composite in each sector/region. This is a new addition.! (all,j,PROD_COMM)(all,r,REG) pfb(j,r) = sum(k,FIBR_COMM, SFIBER(kj,r) * [pf(k,j,r) - afb(kj,r)]); EQUATION FIBERDEMAND ! Demands for fiber commodities. This is a new equation.! (all,i,FIBR_COMM)(all,j,PROD_COMM)(all,r,REG) qf(ij,r) - afb(i,j,r) + qfb(j,r) - ESUBFB() * [pf(ij,r) - afb(i,j,r) - pfb(j,r)]; A6.58 The first of these, FIBERPRICE, computes the percent change in an average fiber price index for each sector using cotton and synthetics: pfb(,r). This is just a cost share-weighted sum of the change in the prices of individual fiber products. SFIBER(k,j,r) refers to the cost share of fiber type k in sector j of region r. This share is updated over the course of a simulation, to reflect the change in cost share which might result from inter-fiber substitution. Variablepf(kj,r) is the percent change in the corresponding fiber price. The term afb(kj,r) capture the effect of exogenous fiber-augmenting technical change (assumed to equal zero here). A6.59 The second equation, FIBERDEMAND, describes the percent change in derived demand for fiber type i, by sector j in region r. This is a function of the ratio of the price of that particular fiber type to the average fiber price. In percent change form, this corresponds to the difference between the percent changes in each of these prices. The difference is premultiplied by the constant elasticity of substitution, ESUBF(j), among fiber types. This is set equal to 2.5 for the textile and apparel sectors and zero for all other j. In addition to this substitution effect, there is an expansion effect captured by qfb(j,r). That is, if relative prices are unchanged, individual fiber demand will increase in proportion to the total demand for fibers, which is, in turn, proportionate to the increase in total sectoral output. A6.60 Now consider a thought experiment in which the cotton export tax is removed in India. This results in the diversion of some domestic cotton to the export markets, and a subsequent rise in the domestic price of cotton, i.e. pf(cotton,j,India) > 0. This causes a general rise in the price index of fibers. However, since SFIBER(cottonj,India) < 1, this rise is smaller than the cotton price hike. Consequently, some substitution away from cotton, towards man-made fibers will occur. In addition to this substitution effect, there will be an output effect, driven by the increase in the price of aggregate fibers. Here, a higher price will reduce the overall demand for fibers, thereby adding to the reduced demand for cotton. tA7L7 Policy Options for Liberalizing nnex , the Cotton Textile Industry A. Introduction A7.1 India stands to benefit greatly from domestic reforms that encourage productivity growth and an efficient allocation of resources in its cotton textile industry. Productivity growth in the highly labor-intensive cotton textile industry is a powerful policy instrument for creating new and better paying jobs, because it raises the industry's competitiveness in an increasingly competitive world market. Not surprisingly, we have found that the more opportunities the world market offers and the more competitive the industry becomes, the more India would gain. Annex 6 projects that the gains from broad-based domestic reforms, including eliminating cotton and cotton yarn quotas, the hank yarn obligation, and other policy and regulatory barriers to productivity growth in the apparel industry, could be significant, potentially amounting to as much as $800 million/year by 2005 if the MFA/ATC quotas are not fully eliminated (or more than double this amount if the MFA/ATC quotas are fully eliminated). A7.2 The policy reforms listed above, if taken individually, will have income distribution implications. A better understanding of how the costs and benefits of individual textile policy actions are distributed among economic agents --e.g., cotton growers, spinning mills, handloom and powerloom operators-- in the short run and how they interact with one another could offer useful insights in designing a sustainable reform package. The global model used in Annex 6 is not adequately equipped to answer some of these questions. Consequently, the objective of this annex is to investigate the short run distributional implications of, and interactions between, selected changes in textile policies, particularly cotton and cotton yarn export liberalization and the elimination of the hank yarn obligation on various participants in the textile industryl B. Methodology Description of the Model A7.3 A partial equilibrium multi-market model is used to measure the impact of three key policy changes on the income of cotton farmers, cotton yam spinners, and cotton textile manufacturers (handlooms, powerlooms, mills, and knitting). The policy changes investigated include: the liberalization of cotton and cotton yarn exports and the removal of the hank yarn obligation. A7.4 The multi-market model consists of six different markets: cotton markets (India and rest of world), cotton yarn markets (India and rest of world), and cotton textiles market (India and rest of world). The cotton yarn market in India is further divided into two kinds of yarn - cone and hank yarn. The cotton textiles sector in India is divided into two subsectors; the first subsector consists of powerlooms, textiles mill and knitting firms, and the secondsubsector consists of handlooms. 1 This Annex is based on the working paper by Masanori Kondo, 1995,"Liberalization Policy for the Cotton Sector in India," South Asia II Agriculture and Water Division, World Bank. Annex 7 Page 2 of 14 A7.5 The model captures the essential structural andpolicyfeatures of the cotton textile industry represented schematically in Figure A 7.1 (See Annex 4 and 3 for a detailed description- of cotton marketing and the textile industry, respectively). After raw cotton is harvested, it is ginned to separate the cotton lint from the cotton seed. The lint is sold to spinning mills or vertically integrated composite mills (includes spinning, weaving, finishing) and processed into cone or hank yarn. Hank yarn is consumed by the handloom industry, while cone yarn is consumed by the powerlooms, composite mills and knitting industries. Fabrics produced by handlooms,powerlooms and composite mills are either processed into apparel and other textile products or directly exported as fabrics. It is assumed that handlooms directly compete with powerloom and mills on the domestic or export market for fabrics. Figure A7. 1: Simplified Commodity Flows in the Cotton-Textiles Industry |Cotton Farmers Hank Yr OCotton Lint | Spinin |Exports Export Quota Cone Yarn Hank Yarn |Powerlooms, | |Handlooms| |Composite Mills A : l | parel Textl Product EJApparel & Textile~ s r ~~~~~~~~~~Product Exports - Nn Table A7. 1: Irnitial Value for Simulation Cotton 200 1,89 10 1656 1,735I56 Cotton Yarn 1,547 1,318 229 16,605 16,834 91 97 109 Cone Yam 1,177 948 229 93 100 HankYamr 370 370 0 83 89 Cotton Textiles 1,364 969 395 9,613 10,008 217 221 221 Powerloom 926 561 365 216 Handloom 438 408 30 219 Sources: EICA (1995), ICAC (1994), Office of the Textile Commissioner (1995) etc. Annex 7 Page 3 of 14 A7.6 The model is solved for five successive periods. Demand and supply of cotton, yarn and textile are assumed to become more responsive to changes in prices over time. This standard assumption is that in the very short run producers and consumers have a somewhat limited capacity to adjust to changes in relative prices. Over time, consumers and producers are in a better position through reallocation of resources and investment decisions to adjust and respond better to changes in relative prices (Table A7.2). A7.7 Cotton export quotas Table A7.2: Price Elasticities restrict India's participation in the world cotton market. Although the g annual export quota is usually set at INDIA: Cotton Supply with respect to cotton producer price 0.10 0.20 500,000 bales (170 kg each), actual Demand with respect to cotton consumer price -0.09 -0.18 quota levels are announced each Demand with respect toyam producer price 0.15 0.30 year and vary accordingly (see INDIA: Yarn Supply with respect to cotton consumer price -0.06 -0.12 Annex 1). As a result, India is only Supply with respect to yarn producer price 0.12 0.24 a minor player in the world cotton Demand with respect to yam consumer price -0.09 -0.18 lint market. In 1993/94, India Demand with respect to textiles producer price 0.15 0.30 lint market. In1993/94, India NDIA: Textile exported 66,000 MT of cotton or Supply with respect to yam consumer price -0.06 -0.12 about 1 per cent of the total world Supply with respect to textile producer price 0.12 0.24 Demand with respect to textiles consumer price -0.09 -0.18 trade volume. In contrast, India WORLD: Cotton accounts for about 25 per cent of Supply with respect to cotton producer price 0.07 0.13 total world acreage and 13 per cent Demand with respect to cotton consumer price -0.06 -0.12 Demand with respect to yarn producer price 0.10 0.20 of world production (see Annex 5). WORLD: Yarn In the model, the cotton export Supply with respect to cotton consumer price -0.04 -0.08 quota is captured as an export tax Supply with respect to yarn producer price 0.08 0.16 Demand with respect to yam consumer price -0.06 -0.12 equivalent. Based on the evidence Demand with respect to textiles producer price 0.10 0.20 presented in Annex 5, it is assumed WORLD: Textile that the export quota creates a Supply with respect to yam consumer price -0.04 -0.08 Supply with respect to textiles producer price 0.08 0.16 wedge between domestic and Demand with respect to textiles consumer price -0.06 -0.12 international prices of cotton, which depresses domestic prices below world market prices by an estimated 15 percent. A7.8 GOI annually fixes the ceiling for exports of cotton cone yarn ranging I to 40 counts (coarse- and medium counts). Yarn quota levels are announced each year and varies from year to year. As of January 1996, the ceiling on cotton yarn exports was set at 80 million kg. The exports of cone yarn above 41 counts are exempted from the quota. The export of hank yarn up to 60 counts (most of the hank yarn), which account for a quarter of the total cotton production, is also prohibited. Domestic cotton spinning mills are protected from imports by an import duty of 25 per cent. India's share in the world yarn market remains small (see Annex 1). The yarn export quota is captured as an export tax equivalent in the multi-market model. Based on the evidence presented in Annex 3, the yarn export tax equivalent on spinning mills is set at 15 percent. A7.9 The cotton spinning industry is subject to the Hank Yarn Obligation, as well as other government regulations. The hank yam obligation (HIYO) requires mills to reserve half of their total domestic cotton yarn supply for hank, which is exclusively processed by handlooms (see Annex 1). In practice, only about 25% of the total yarn production is produced in hank form as a result of an increasing number of exemptions and lax implementation of the HYO (see Annex 3). Other regulations protect workers in the organized sector (including spinning and composite mills); these include minimum wage laws, minimum employment requirement laws, and the laws which require Annex 7 Page 4 of 14 mills to obtain permission from state governments to close unprofitable mills or sell off the assets. These regulations are not captured in the model, but the Hank Yarn Obligation is. A7. 10 The hankyarn obligation undermines the performance of mills. The cost of producing hank yarn is higher than that of cone yarn by about Rs 4/kg 2, while its selling price is lower than that of cone yarn by an average of about 10 percent. This additional cost of producing hank yarn is captured in the model. As a result of the Hank Yarn Obligation, the enforced production of unprofitable hank yarn reduces the profitability of spinning. The current 50 per cent hank yarn obligation has also contributed to the oversupply of hank yarn, pushing down the prices of hank yarn further (see Annex 3). Limitations of the Model. A7.11 The present model is intentionally kept simple and has a number of limitations which are important to keep in mind when interpreting its results. A7.12 First, the model does not incorporate changes in the world market arising from the MFA phase-out. As a result, the output growth and employment opportunities in the domestic cotton textile industry which the MFA phase-out will create are not captured by this model. From Annex 6, we know that these impacts could be quite large, whether or not the MFA is fully eliminated in 2005. A7.13 Second, the model is a partial equilibrium model and as such does not capture the interaction of the cotton textile industry with the rest of the economy. As a rule of thumb, this generally tends to over-estimate the predictions of a partial equilibrium model. A7.14 Third, unlike the general equilibrium model presented in Annex 6, the present model does not incorporate any productivity change (encouraged by supporting textile policy reforms) in the textile industry. As shown in Annex 6, productivity changes would be beneficial to the Indian cotton textile industry by improving its competitiveness and its ability to absorb higher input costs, for example, raw cotton for the spinning industry, yarn for the weaving (and knitting) industry, or gray cloth for the clothing industry. A7.15 Fourth, the parameters of the multi-market model are not derived from econometric estimations. Instead, the elasticities used in the multi-market model are assumed and reflect the authors' best guess based on available indicators and literature in India and elsewhere. The robustness of the multi-market model was assessed by conducting sensitivity analysis of the results (see Appendix). Similarly, the relative competitiveness of the spinning industry --i.e., its competitiveness when paying world prices for its raw material-- was assumed, based on the analysis presented in Annex 3. From the available evidence, it is assumed that the spinning industry, as a group, will remain competitive when faced with world cotton and yarn prices. Textile Policy Reform Simulations A7.16 The model estimates the impact on net incomes3 and export revenues of specific policy reform packages on cotton farmers, cotton yarn spinners, and cotton textiles producers. Six different policy scenarios examine the impact of eliminating (Table A7.3): * Scenario 1: cotton export quotas; * Scenario 2: hank yarn obligation and cotton export quotas; * Scenario 3: cotton and yarn export quotas; 2 Office of the Textiles Commissioner, Bombay 1993/94 3 Net income means gross margin (i.e. total revenue minus raw materials cost). Annex 7 Page 5 of 14 * Scenario 4: cotton and yarn export quotas and hank yarn obligation; * Scenario 5: cotton and yarn export quotas and hank yarn obligation, and introducing an explicit input cost subsidy (10%)4 to handloom weavers. Table A7.3: Summary of Policy Reform Scenarios. ,:~~~~~~. Base Quota Quota Yes No I Free Quota Yes No 2 Free Quota No No 3 Free Free Yes No 4 Free Free No No 5 Free Free No Yes A7.17 Scenario 1 investigates the income distribution effects of removing the cotton export quota on the cotton textile industry. Scenario 2 through 4 assess whether and how selected complementary textile policy reforms interact with each other and how they will affect the income and competitiveness of the cotton textile industry. Namely, the model investigates whether the elimination of the hank yarn obligation and the removal of the yarn export quota--individually or together--could compensate the spinning industry. Because the income and competitiveness of the handloom weavers will be negatively affected by higher cotton yarn prices and the elimination of the hank yarn obligation, the model also investigates the fiscal and income distribution implications of introducing a direct input subsidy targeted to the handloom industry. C. Income Distribution Impacts of Cotton Textile Liberalization Cotton Export Liberalization. A7.18 Cotton growers would benefit greatly from the liberalization of cotton exports (scenario 1) with their gross income rising by an estimated 13% (Table A7.4). The income gains accruing to cotton growers would come from the spinning industry which would stand to loose significantly from free cotton exports. The net of income of the spinning industry would fall by an estimated 39%. Yarn prices would rise by 4% which is smaller than the 17% increase in cotton prices triggered by the liberalization of cotton exports. Yarn export quotas prevent the spinning industry from raising its prices more and passing its higher costs of production to the textile industry. Less efficient mills with dated technology would suffer the most as well as those mills for which the yam export quota is binding --e.g., those producing below 40 counts, for which India seemed to have a larger comparative advantage (see Annex 3). As a result of their cost of production increase, less yarn is exported. The textile industry as a group would experience a small and insignificant loss of income (1%). Cotton Export Liberalization & Elimination of the Hank Yarn Obligation A7. 19 The elimination of the hank yarn obligation could bring some relief to the spinning industry, but insufficient to fully compensate its income loss caused by freeing cotton exports (scenario 2). The removal of the Hank Yarn Obligation imposes a greater loss of income to the handloom weavers (4%), since they would no longer be shielded from their low level of productivity. With the decline in handloom production, less textile products would be exported to satisfy domestic demand. Less cotton could be exported since the removal of the Hank Yarn Obligation helps 4 The 10 percent handloom production cost subsidy is equivalent to about Rs 10/ kg. Annex 7 Page 6 of 14 improve the productivity and competitiveness of the spinning industry. More cotton is therefore processed domestically, and yarn exports fall by a smaller amount than previously. Table A7.4: Changes in Net Income & Export Values Policy Changes Export Policy Cotton Free : Free -Free Yam Quota Quota Free HYO 50% None 5 00/0 10% subsidy to handloom No No No Net Income* Cotton 13% 13% 14% _ Changes Yam -39% -29% i % Textile -1% -2% -10% Mill/Powerloom -1% -1% -8% Handloom -2% -4% -12% Total -1% -1% 0% Export Value Cotton 46% 39% 13% Changes Yam -6% 4% I 10/0 Textile %0% -2 Total 2% 1% 2 Note: * Net income is measured by the gross margin, defined as total revenue minus raw materials cost. For cotton growers, it corresponds to gross income.AIl percentage changes are computed from the base case, at the end of year 5. Source: Computed. Free Cotton and Cotton Yarn Exports A7.20 Lifting both cotton and yarn export restrictions together (scenario 3) could bring significant income relief to the Indian spinning industry. In fact, freeing cotton yarn exports would virtually compensate the spinning industry from the increase in cotton prices resulting from liberalizing cotton exports. Cotton growers are unaffected by the liberalization of yarn exports; in fact, they gain a little since the yarn industry is now in a position to actually raise slightly its production and export volumes. Cotton export revenues would increase much less (13%), compared to scenarios 1 and 2 (46% and 39%, respectively). Less cotton would be exported, since more cotton could be processed domestically by the spinning industry whose level of competitiveness improves with the liberalization of yarn exports. Indeed, with domestic yarn prices allowed to rise to export parity levels, the spinning industry could register a slight gain in income (1%). Accordingly, yarn export revenues increase by 10 percent. The wide diversity in financial performance of Indian spinning mills (See Annex 3), however, suggests that financially weaker firns (poorly managed and/or dated mills with obsolete technologies) will be adversely affected. A7.21 The income transfer to cotton growers (14%) would come from the weaving industry. Rising production costs contribute to a decline in exports and in turn net incomes of the weaving sector. This is critical for handloom weavers, who are more sensitive to increases in raw material prices, because of their low initial productivity level. They are projected to incur a larger loss of income (12%) than the millpowerloom sectors (8%). Assuming the estimated number of handloom units is about 4 million, the income loss per handloom unit as a result of export liberalization is about Rs 1,894 ($54). The income loss experienced by handloom units turns out to be quite similar in magnitude to that under policy scenario 2 (removal of the hank yarn obligation and free cotton exports). Free Cotton and Cotton Yarn Exports, Removal of the Hank Yarn Obligation. A7.22 The elimination of both export quotas and the hank yarn obligation (scenario 4) results in a further redistribution of income from textile weavers to the spinning mills, with handloom weavers bearing most of the income loss. The elimination of the hank yarn obligation (when combined with free yarn exports) more than compensates spinning mills for the higher cost of cotton: the spinning sector's net income increases by an estimated 10 percentage points over the base scenario (Table Annex 7 Page 7 of 14 A7.5). However, the handloom industry incurs singificant additional income losess. The drop in hank yarn production and therefore rising prices of hank yarn further reduces the net income of handloom weavers by an additional 3 percentage points, so that overall their incomes drop by 15 percent (Rs 2,305 or $66/unit) from the base scenario. Overall, the total net income of the cotton textile industry remains largely unchanged. Compensatory Scheme to Handloom Weavers A7.23 A 10 percent direct subsidy targeted to handloom weavers is introduced to address the adverse impact of export liberalization and removal of the hank yarn obligation. Being economically disadvantaged, a compensatory package of an additional 10 percent production cost subsidy (Rs 10/kg of fabric) targeted to handloom units is examined. It is assumed that the subsidy would be supported by the government budget, and would amount to an estimated Rs 3.6 billion per year (US$ 103 million or US$ 26 per handloom unit). The 10 percent input subsidy would reduce the Table A7.5: Changes in Net Income & Export Values Policy Changes Export Policy Cotton Free Free Free Yarn Free Free Free HYO 50% Abolished Abolished 10% subsidy to handloom No No Yes Net Income* Cotton 14% 14% 14% Changes Yam 1% 10% 10% Textile -10% -11% -9% Mill/Powerloom -8% -9% -9% Handloom -12% -15% -9% Total 0% 0% 1% Export Value Cotton 13% 6% 6% Changes Yam 10% 12% %12 Textile -2 -2% -2% Total 2 1% 1% Note: * Net income is measured by the gross margin, defined as total revenue minus raw materials cost. For cotton growers, it corresponds to gross income. All percentage changes are computed from the base case, at the end of year S. Source: Computed. income loss to handloom weavers by an estimated 40%, and bring their income loss in line with other segments of the weaving industry. In effect, a 10 percent subsidy would be sufficient to compensate for the lower productivity level of handlooms relative to other textile producers. It would not be sufficient, however, to compensate the handloom weavers for the loss of competitiveness caused by the liberalization of cotton and yarn exports. Obviously, higher levels of input subsidy targeted to handloom weavers would compensate them more. D. Summary and Conclusions A7.24 The objective of this annex was to investigate the short run income distribution implications of liberalizing cotton and yarn exports and removing the hank yarn obligation, and the interactions among these three individual reforrns. It was motivated by the belief that a better understanding of these relationships may offer some useful insights and guidance in designing a socially acceptable implementation package of reforms. It is not the intention of the present note to provide final and exact numbers, nor to make finn policy recommendations. Instead, the present note and supporting model should be thought of a potentially useful tool for assisting in the possible design of textile policy reforms on which to build upon. In the following paragraphs, we conclude by drawing some of the main lessons from this exercise. Annex 7 Page 8 of 14 A7.25 First, liberalizing cotton yarn exports is clearly essential for supporting the profitability of the cotton spinning industry in the face of either rising cotton prices or free cotton exports. It is essential in two complementary ways; first, by raising yarn prices and second by boosting the spinning industry's incentives for investments in modernization and productivity improvements. Today, there are strong upward pressures on cotton prices stemming from cotton production problems on the one hand, and fast rising demand for cotton fueled by rapidly increasing domestic and foreign market opportunities on the other. The results of the analysis presented in Annex 6 also suggest that the upward pressures on cotton prices are likely to increase further over time. Recent government policies are making yarn export quotas less binding by increasing the number of exemptions. These policies should be continued and expanded further with the view to eliminate the implicit discrimination against those spinning firms that cannot access those export quota exemptions (e.g., non export oriented units producing yarns below 40s count). Giving similar market incentives and opportunities to both domestic-oriented and export-oriented spinning mills would be a powerful instrument for encouraging and accelerating the much needed modernization of the entire spinning industry, in particular of those domestic-oriented units where modernization needs are more acute. A7.26 Removing the Hank Yarn Obligation is a second policy instrument to promote the increased efficiency of the spinning industry. It would improve the competitiveness of the spinning industry by eliminating the implicit yarn export and productivity tax arising from the hank yarn obligation. It has, however, important and negative income distribution effects on the handloom industry which would need to be addressed. A7.27 The above analysis suggests that it is possible to think about alternative options to compensate the handloom industry for its loss of competitiveness brought about by higher yarn and cotton prices. An explicit budgeted input subsidy targeted to the handloom weavers would represent one option. An explicit input subsidy could avoid some of the shortcomings of the current hank yarn obligation. First and foremost, an explicit subsidy would have the advantage of transparency, making it readily apparent to society and taxpayers the price of social protection. Second, besides avoiding the efficiency costs associated with the hank yarn obligation, a demand- based subsidy targeted to handloom weavers could avoid some of the weaknesses inherent with the current Obligation, including moving away from a quantity-based to a demand-based regime, which would minimize the incentive problems of over-supply of hank yarn, leakages to other segments of the industry, mis-reporting, and uncertainty about actual number of --and yarn quantity demanded by-- the handloom units.5 A7.28 Additional measures, such as handloom conversion and quality improvement programs, may offer more viable and effective alternatives for assisting handloom operators in increasing their level of productivity. Tamil Nadu offers an example of a successful handloom conversion program which assists and encourages handloom owners to process man-made yarn. Such a program makes handlooms less dependent on cotton, and at the same time opens new market opportunities in cotton blend textile products. Another conversion program could encourage the shift from handloom to powerloom units. Training programs could also assist handloom operators in up-grading their skills to seek alternative employment opportunities in the booming and highly promising garmenting industry. Training programs could assist handloom operators up-grade the quality and design of handloom products for niche, high-value markets, domestic or foreign for which the need for special protection would be reduced. 5A demand-based regime could be designed by, for example, providing a per-unit subsidy to handloom units for the purchase of hank yarn , or for the production of handloom fabric. Annex 7 Page 9 of 14 A7.29 More generally, the model confirms the policy changes needed to facilitate and accelerate productivity-enhancing investments in the weaving and clothing industries, since it would help compensate for the loss of income and competitiveness the higher yarn and cotton prices would impose upon them. A7.30 In terms of phasing of policy reforms, the model results underscore the need for pro-active textile policies. Pro-active textile policies would aim at taking early measures to facilitate and encourage productivity-enhancing investments in the spinning, weaving, knitting and clothing segments of the textile industry. Pro-active textile policies would anticipate on the large expected increase in cotton prices fueled by the exogenous boom in cotton textile demand, and set in place the necessary conditions for the phased liberalization of cotton exports. Annex 7 Page 10 of 14 E. Appendix Simulation Model for the Cotton Sector Liberalization in India Structure of the Model A7.31 A comparative static partial equilibrium multi-market model is used in this analysis to measure the impact of two key policy changes on the income of cotton farmers, cotton yarn spinners, and cotton textile manufacturers (powerlooms, mills, knitting and handlooms). The two major policy factors simulated are the liberalization of cotton and cotton yarn exports and the removal of the hank yarn obligation. A7.32 The multi-market model consists of six different markets, including cotton markets (India and rest of world), cotton yarn markets (India and rest of world), and cotton textiles market (India and rest of world). The cotton yarn market in India is further divided into two kinds of yarn - cone yarn and hank yarn. The cotton textiles sector in India is subdivided into two subsectors: powerloom, textiles mill and knitting sector and handloom sector. Assumptions A7.33 Each market is assumed to have its own supply and demand function, and the equilibrium depends on the initial quantities, initial prices, and price elasticities in the market. For instance, the supply function of cotton yarn in India is a function of consumer price of cotton and producer price of yarn. The percentage change in the total yarn supply is equal to the supply elasticity of cotton with respect to cotton consumer price multiplied by the change in cotton consumer price plus the supply elasticity of cotton with respect to the cotton yam producer price multiplied by the change in the cotton yarn producer price. A7.34 The price changes in India are the result of the changes in various tax, incentive or subsidy rates. In the previous case, for example, the cotton consumer price will change when the government changes the cotton export incentive rate or cotton consumer tax rate. Cotton consumer yarn prices will change when the government changes the cotton yam export incentive rate or cotton yarn producer tax rate. In each market in India, price changes are the only factors which will affect the values of production and consumption. Therefore, domestic supply is independent of domestic demand for each commodity. The balance between supply and demand is assumed to be exported for each commodity, since it is assumed that there are no stocks in the simulation. A7.35 While changes in Indian prices depend only on tax, incentive and subsidy rates, price changes in the rest of the world depend on the demand and supply in the rest of the world and India's exports. The Indian market equilibrium is reached by changing the volume of exports; while the world market equilibrium adjusts to world price changes. This model follows a "large country assumption," such that changes in India's export volume could affect the world price for cotton and yarn. However, the simulation results indicate that the change is fairly small, since India is a minor exporter even after liberalizing the exports of cotton and yarn. A7.36 Although there were initially six equations for six markets, exports from India are a common variable in both the Indian and world market equations for each commodity. Plugging these into the world market equilibrium conditions reduces the number of equations to three: the world cotton market, world cotton yarn market and world cotton textiles market equations with three unknown variables - the world cotton price, world yarn price and world textiles price. Solving these Annex 7 Page 11 of 14 three equations simultaneously by inverse matrices provides estimates of how much the income of each sector - cotton production sector, cotton spinning sector, and cotton textiles sector -- increases or decreases as a result of specific policy changes. Production of Hank Yarn with and without Obligation A7.37 In recent years, domestic supply of hank yarn has not exceeded one third of the total supply of cone and hank yarn. Despite the 50 per cent hank yarn obligation, some mills do not obey the regulation. It is therefore assumed that even if the 50 per cent hank yarn obligation is maintained, mills will produce at the current rate of 28 per cent of their domestic supply of cotton yarn in hank form. Since hank yarn is much less profitable, it is assumed that without hank yarn obligation, mills will supply only cone yarn and hank yarn will not be available in the market. This assumption is reasonable because the production process of hank yam is exactly same as cone yarn except for the final process - winding as cone yarn or winding as hank yarn, and it is easy for mills to switch from hank yarn to cone yarn production. A7.38 The ratio of production capacities between these two sub-sectors are set exogenously in this simulation. It is also assumed that the cotton textiles manufactured by these two kinds of looms are identical in terns of quality. The main difference between the two subsectors is that the former processes only cone yam while the latter exclusively processes hank yarn. If the hank yarn obligation is phased out, hank yarn will not be available in the market and handlooms will have to buy cone yarn in the market and reel it in hank from by paying the additional cost (Rs 4/kg). Market Clearing Conditions and Identities T. Cotton Market Indian Market World Market Sc (Ppc) = Dc (Pcc, Ppy) + Xc Swc (Pwc) + Xc = Dwc (Pwc, Pwy) s.t. Ppc = Pwc /txc *tpc Pcc = Pwc / txc * tcc Ppy = Pwy / txy * tpy II.Yarn Market Indian Market World Market Sy (Pcc, Ppy) = Dy (Pcy, Ppt) + Xy Swy (Pwc, Pwy) + Xy Dwy (Pwy, Pwt) s.t. Pcc = Pwc ltxc *tcc Ppy = Pwy I txy * tpy Pcy = Pwy Itxy * tcy Ppt = Pwt / txt * tpt HI. Textiles Market Indian Market World Market St (Pcy, Ppt) = Dt (Pct) + Xt Swt (Pwy, Pwt) + Xt = Dwt (Pwt) s.t. Pcy = Pwy / txy * tcy Ppt = Pwt / txt * tpt Pct = Pwt / txt * tct Annex 7 Page 12 of 14 IV. Endogenous Variables: Sc = Supply of cotton, India Swt = Supply of textiles, Rest of the world Dc = Demand of cotton, India Dwt = Demand of textiles, Rest of the world Swc Supply of cotton, Rest of the world Ppc Producer price of cotton, India Dwc Demand of cotton, Rest of the world Pcc Consumer price of cotton, India Xc = Export of cotton, India Pwc = World price of cotton Sy = Supply of yarn, India Pwy = World price of yarn Dy = Demand of yarn, India Ppy = Producer price of yarn, India Xy = Export of yam, India Pcy = Consumer price of yarn, India Swy = Supply of yam, Rest of the world Pwt = World price of textiles Dwy = Demnand of yarn, Rest of the world Ppt = Producer price of textiles, India St = Supply of textiles, India Pct = Consumer price of textiles, India Dt = Demand of textiles, India Xt = Export of textiles, India V. Exogenous Variables: txc = Cotton export tax rate (or quota equivalent), India tpc = Cotton producer tax / subsidy rate, India tcc = Cotton consumer tax / subsidy rate, India txy = Yarn export tax rate (or quota equivalent), India tpy = Yarn producer tax I subsidy rate, hidia tcy = Yarn consumer tax / subsidy rate, India txt = Textiles export tax rate (or quota equivalent), India tpt = Textiles producer tax / subsidy rate, India tct = Textiles consumer tax / subsidy rate, India VI. Integrating Markets Sc (Pwc, txc, tpc) + Swc (Pwc) = Dc (Pwc, txc, tcc, Pwy, txy, tpy) + Dwc (Pwc, Pwy) Sy (Pwc, txc, tcc, Pwy, txy, tpy) + Swy (Pwc, Pwy) = Dy (Pwy, txy, tcy, Pwt, txt, tpt) + Dwy (Pwy, Pwt) St (Pwy, txy, tcy, Pwt, txt, tpt) + Swt (Pwy, Pwt) = Dt (Pwt, tt, tct) + Dwt (Pwt) Ppc = Pwc / txc *tpc, Pcc = Pwc / txc *tcc, Ppy = Pwy / txy * tpy Pcy = Pwy I txy *tcy, Ppt = Pwt I txt *tpt, Pct = Pwt I txt * tet Data A7.39 The data for cotton, yarn and textile supply, demand exports for India and the rest of the world and domestic and world prices are based on the official data of the past five years shown in Office of the Textiles Commissioner (1994), EICA (1995), and International Cotton Advisory Council (1994). The Cotton Advisory Board figures for cotton production in India is used in order to make them consistent with other values. A7.40 The producer price is the price which the producer receives after paying taxes (or receiving subsidies) while the consumer price is the price consumers (in the case of cotton or cotton mill, in the case of cotton yarn, powerlooms or handlooms) have to pay including the sales tax. As for the prices of cotton textiles, a reasonable adjustment is made. Cotton textiles consumer price and world price are set equal since there is no export restriction. All the initial values are in shown in Appendix Table A7. 1. Annex 7 Page 13 of 14 Appendix Table A7. 1: Initial Value Table for Simulation ---1'~~~~~~~~~~~~~~~~~- Cotton 2,000 1,893 107 16,566 16,673 53 55 61 Cotton Yam 1,547 1,318 229 16,605 16,834 91 97 109 Cone Yam 1,177 948 93 100 Hank Yam 83 89 Cotton Textiles 1,364 969 395 9,613 10,008 217 221 221 Powerloom 926 561 365 216 Handloom 438 408 30 219 Sources: EICA (1995), ICAC (1994), Office of the Textile Commissioner (1995) etc. Elasticities A7.4 1 This model has twenty price elasticities. The mathematical notations of all these elasticities are given below (Appendix Table A7.2). In the multi-market model analysis, the ratios among elasticities are critical, and therefore consistency in the methodology for estimating the elasticities is important. However, although some studies have estimated some of the elasticities, no comprehensive econometric study has estimated all the twenty elasticities. Therefore, rather than relying on the results of different econometric studies, it is assumed that the elasticities range from - 0.18 to 0.3 in 1-year period and from -0.16 to 0.6 for the 5-year period.6 These elasticity figures seem reasonable for most commodities, and are not very different from other commodities. Sensitivity analysis using different sets of elasticities were examined (see Appendix table A7.3), but the results were not very different from those presented in this paper. A7.42 The ratios among the elasticities are computed based on the cost structure of the products. For example, the ratio between the supply elasticity of yam with respect to cotton and the supply elasticity of yam with respect to yarn price is assumed to be -1/2, because the cotton constitutes roughly half the total cost of yarn production. Since long term elasticities are larger than short term values, it is assumed that the absolute values of elasticities increase from year 1 to year 3. Finally, all the elasticities in the rest of the world are set to be 2/3 of the equivalent Indian elasticities, because cotton, cotton yarn and textiles are produced not only in developing countries but also in high-income countries like the United States and member countries of the European Union Appendix Table A7.2: Elasticity Table for Base Simulation Case ,, It