The Mexican sugar industry : problems and prospects

Pu's,, Room oh l% ExmdhzeAfllmfre 1 VVOHKING P~APLR.S in|neUo.mi Tr J International Economics Department The World Baik February 1991 WPS 596 The Mexican Sugar Industry Problems and Prospects Brent Borrell The economic costs of the extensive government regulation of the Mexican sugarindustry are quantifiedandpolicy changes are suggested. TePdicy.Reuch. and External Affairs Complex distribute PRE WorkingPapens to disseminate the finding of work in progress and to encourage the exchange of ideas among Bank stff and aU othen interated in development issues. The papen carry the names or the authors, ztflect oly their views, and should be used and cited accordingly. The rindings, interpreutiocs, and canclunions are the auithos own. They should not be attibuted to the World Bank, its Board of Directors, its managemrnt, or any of its maeber countrieg.  Pet, n_#-kan tnm li ilulml WPS 596 Ihis paper- a product of the Intemational Trade Division, Intemational Economics Department- is part of a larger effort in PRE to understand the implications for world commodity markets of changes in developing countries' trade policies and to assist developing countries in designing good trade and industry policies. Copies are available free from the World Bank, 1818 H Street NW, Washington DC 20433. Please contact Pauline Kokila, room S7-040, extension 33716 (60 pages). The Mexican sugar industry has been subject to econometric model of the Mexican sugar indus- extensive government controls over land owner- try. This was linked to a global model of the ship, cultivation, harvesting, milling, marketing, world sugar industry. Stochastic simulations distribution, and pricing. The many objectives projecdng the Mexican sugar industry under of these intervendons include protecting produc- these policies show consumption increasing ers and consumers from world price variability, faster than production and Mexico increasing its ensuring self-sufficiency, guaranteeing employ- sugar imports. It appears unlikely that under ment and social welfare, providing cheap milling such policies Mexico would return to being an services, and protecting domestic soft drink exporter of sugar. manufacturers. in simulations of a sugar industry operating Borrell's calculations show that although the under essentially free trade conditions, Mexico interventions have helped stabilize industry becomes a significant sugar exporter. Produc- rcturns to some degree, the estimated effective tion, trade, and stocks are more variable, and rate of assistance points to a high degree of consumption growth is curtailed. But welfare in resource distortion. In years when world prices the economy ac a whole would be increased were low, such as in 1985-88, the effective rate substantially. The main beneficiaries would be of assistance is estimated in the range of 70-390 sugar producers, and the losers consumers - but percent. the loss to consumers would average less than US$3.20 per person per year. To analyze the impact of changes in Mexi- can sugar policies, Borrell constructed an The PRE Working Paper Series disseminates the findings of work under way in the Bank's Policy, Research, and External Affairs Complex. Anobjective of the series is to getthese findings out quickly, even if presentations are less than fully polished. The findings, interpretations, and conclusions in these papers do not necess . ly represent official Bank policy. Produced by the PRE Dissemination Centcr  The Mexican Sugar Industry Problems and Prospects by Brent Borrell Table of Contents I. Introduction 1 I. Background 3 The Structure of the Industry 3 (i) Production 3 (ii) Consumption 6 Government Intervention in the Sugar Industry 7 III. The Economic Effects of Intervention 11 The Static Costs of Existing Price-Fixing Arrangements 11 The Effective Level of Assistance 14 The Effects of Intervention on Industry Costs 17 Overall Assessment of Economic Costs of Existing Policies 19 Recent Changes to Regulations and Controls 22 IV. Prospects for the Mexican Sugar Industry Under Different Policies 23 Illustration of the Effects of a Continuation of Existing Policies to 2004 23 The Effects of a Change to Free Trade 29 The Effects of a Guaranteed Minimum Price Scheme 35 Discussion of Simulation Results 36 V. Conclusions and Recommendations 37 Suggestions for Policy Reform 38 References 40 Appendix 1: The Nature of the Present Cost Burdens 41 Appendix 2: Model Specification, Estimation, and Limitations 52  I. INTRODUCTIO* 1. The Mexican sugar industry operates under strict government controls. The sugar parastatal, AZUCAR, and other statt agencies govern virtually all aspects of pricing and, until recently, AZUCAR controlled virtually all aspects of marketing. State agencies also affect the cultivation and milling of cane and influence the operations of the sweetener- using sector. Market forces appear to play only a minor role in the allocation of resources within the industry and between the industry and the rest of the economy. On theoretical ground,, and based on empirical evidence from other countries, there is a strong prima facie case for believing that the economic performance of the industry is highly constrained as a result of the interventions and that these imoOse significant costs on the wider economy. 2. The purpose of this study is to make transparent the main economic effects of existing sugar policies. Three broad measures are used to estimate the resource misallocation effects of intervention: the nominal rate of protection, the effective rate of assistance and the net subsidy equivalent. Theoretical arguments are also used to demonstrate other potential inefficiencies in resource use. To estimate the effects of efficiency- improving policies, an economic model of Mexican production, demand, stock * Brent Borrell completed this study when he was employed as a long-term consultant with the International Commodity Markets Division; at the time he was on leave from the Australian Bureau of Agricultural and Resource Economics. Presently, Brent is Chief Market Economist with The Centre for International Economics in Canberra, Australia.  - 2 - demand and cane-pricing arrangements is constructed. This model is linked to a model of the world sugar market to evaluate the trade and other economic opportunities which should arise from policy reforms. 3. While the methods used to evaluate the current and potential performance of the industry have shortcomings, taken together they provide a broad indication of the costs and benefits of the different policies. Broad policy reccmmendations are made on the basis of the analytical results.  -3- 1I. BACKGEOUND The Structure of the Industry 4. Mexico is a large producer and consumer of sugar in world terms but has switched from being a significant exporter to an importer during recent years; presently it is roughly self-sufficient in sugar. For the 20-year period to the mid-seventies, Mexico exported between 20X and 40% of its domestic production (see Table 1). Despite strong growth in sugar consumption, export performance was largely maintained due to reasonably steady advances in production. However, during the oil boom years from 1975 to 1982, production stagnated. Consumption, meanwhile, continued to increase strongly. From 1977 to 1986, Mexico was mostly a net importer of sugar. Only recently has Mexico returned to the world market as an exporter. 5. Production growth resumed after 1982, following policy changes (the :canero decree of 1979) which were intended to maintain real cane prices in the face of spiraling inflation. In 1987 and 1988, sizable quantities of sugar were exported, albeit due to some extent to a rundown in stocks. For 1989, net exports are estimated to be small and for 1990 imports will likely be necessary to meet the growth in consumption and a shortfall in production. (i) Production 6. Production is widely dispersed throughout 9 main geographical regions and 14 states (see Table 2). There are 70 sugar mills employing about 40,000  - 4 - Table 1: luppiy-Deamnd Balance Sheet for Mexican Sugar Net End-Year Exports/ Production Consumption Exports Stocks Production --------------------('000 tons)-------------------- 1960 1,530 1,121 462 404 0.30 1961 1,487 1,156 586 149 0.39 1962 1,531 1,249 365 66 0.23 1963 1,735 1,312 393 96 0.22 1964 1,933 1,424 491 114 0.25 1965 2,107 1,478 555 201 0.26 1966 2,266 1,555 513 399 0.22 1967 2,412 1,650 572 589 0.23 1968 2,338 1,767 676 482 0.28 1969 2,564 1,875 625 554 0.24 1970 2,402 1,992 612 351 0.25 1971 2,489 1,920 551 369 0.22 1972 2,587 2,075 598 282 0.23 1973 2,810 2.298 586 207 0.20 1974 2,838 2,344 496 206 0.17 1975 2,636 2,526 217 1on 0.08 1976 2,710 2,675 13 121 0 1977 2,790 2,677 0 234 0 1978 3,131 2,934 74 363 0.02 1979 3,095 3,059 30 506 0 1980 2,457 3,152 -760 650 -0.30 1981 2,642 3,261 -673 705 -0.25 1982 2,739 3,514 -522 452 -0.19 1983 3,076 3,241 -819 1,106 -0.26 1984 ',308 3,343 -273 1,349 -0.08 1985 3,492 3,547 66 1,227 0.01 1986 4,068 3,451 219 1,625 0.05 1987 4,060 3,657 518 1,510 0.12 1988 3,908 4,070 1,014 334 0.25 SourceS International Sugar Organization, London.  Table 28 Mexican Sugar Procaction, by Region and State 1985 Sugar Production (Base estandar, tons) Region state Private Mills Public Mills Alto Veracruz Veracruz 326,214 418,516 Oaxaca Bajo Veracruz Veracruz 422,014 Balsam Michoacan 143,655 Centro Morelos 280,179 Puebla Huastecas San Luis Potosi 65,907 358,324 Veracruz Tamaulipas Noroeste Sinaloa 46,822 312,593 Nayarit Occident. Jalisco 107,837 401,402 Colima Pacifico Sur Oaxaca 144,388 Chiapas Sureste Tabasco 39,611 158,464 Quintana Roo Source: Latin America & Caribbean Country Department II, Agriculture Operations Division, World Bank.  -6- workers and around 130,000 cane suppliers and 90,000 cara eutters. Average sugar production per mill, is around 50,000 tons per year, which is roughly equivalent to the average in Cuba but only about one-half the Brazilian average. Cane production is predominantly small-scale. The average area of cane harvested per supplier is 4.3 hectares. Most cane is harvested by hand, compared to Cuba and Brazil, where over one-half of the cane is mechanically harvested. About 40% of cane area is irrigated. Cane is usually grown in an eight-year cycle and is not grown in rotation with other crops. Other crops could be grown on cane lands; the two most likely are maize and beans. 7. Mills generally produce one or two of three grades of sugar: raw, estandar (mill white) and retined. About 16X of sugar output is in raw form, a little less than 35% is refined and the remainder (about 50X) is estandar. By-products include molasses, bagasse and a small amount of alcohol. About 25% of the bagasse (the cane fiber) is used as pulp in paper making (paper sales make up only a minor proportion of mill revenues). The remainder is burned as fuel in the mills. (ii) Consumption 8. Over the past decade the annual growth in consumption has averaged a little over 3%. Mostly this reflects Mexico's strong population growth; but relatively strong growth in disposable incomes during the late 1970s and early 1980s also stimulated growth. By international standards, pei capita consumption is high at around 40 kg. About 50% of sugar consumption is in the form of processed focts with the remainder being consumed directly. In 1988,  - 7 - 56Z of sugar purchased by the food-processing sector was used in soft drinks. Alternative sweeteners hold a very small share of the Mexican sweetener market. Government Intervention in the Sugar Industry 9. Government intervention is pervasive. It almost completely eliminates the normal workings of commercial markets in the allocation of resources. O Legal restrictions on the sale and renting of land largely influence who can grow cane and greatly influence the scale of operation as well as the methods of production. O Regulations governing the pricing and delivery terms for cane establish the incentives which affect the quality of cane, the proportion of sugar and molasses produced from the cane, the technology used in mills, the narvesting and milling season length, the utilization of mills, the cane ratooning pattern (i.e., the number of years the cane plant is harvested before being replanted), the scheduling of h'arvesting and mill delivery, and many cultivation practices. o Government ownership of many of the l.sills and centralized control of milling operations greatly affect their output, technology, investment, location, size and level of employment. o The monopoly acquisition and marketing powers of the sugar parastatal "AZUCAR" mean that patterns of storage, distribution, handling, exporting and importing are centrallv  - 8 - controlled--though recently changes have been implem4nted which reduce AZUCAR's monopoly powers. o Price fixing, price discriminetion and trade barriers determine the volume and pattern of production, consumption, trade and stocks, as well as the availability and range of sugar products. 10. The objectives of Mexican sugar policies are not explicitly intended to circumvent the workings of commercial markets nor to impose a system of (virtual) central planning on the industry. Intervention has a very long history. It was introduced to protect the industry from the volatility of world prices as early as 1856 and maybe before then. At that time prices were fixed and imports were controlled. Current policLes are the result of a complex series of political and economic responses to ptoblems and unintended consequences of the initial controls; rather than a deliberate, internally- consistent policy. For instance, fixed consumer and producer prices (introduced with the objective of protecting both groups) created an tnintended cost-price squeeze which left many mills insolvent. Government buy outs and ownership of mills were the result even though there was not a deliberate policy of nationalization. 11. The accumulation of ad hoc responses lef .n its wake a wide variety of implicit objectives rather than a prioritized set of consistent goals. Over the last two decades the main purposes of intervention appear to have been:  - 9 - o To protect the cane-growing sector during intermittent periods of very low world prices and to tax it heavily during short periods of extremelv high world prices (see Figure 1). o To provide a level of social welfare to cane suppliers and mill workers in excess of that generally available to other rural groups. O To retain the small-scale cane production system to maintain rural employment. 3 To ensure a continuing supply of cheap milling services to the cane-growing sector and to expand employment in mills, irrespective of the implicit tax on private sector mills in general and the direct bud.3etary costs of st bsidizing public mills. o To shiels ronsumers from variability in world prices, providing v--y high subsidies at times and extracting relatively small taxes at others (see Figure 1). o To protect domestically-owned soft drink manufacturers from foreign-owned competition through discriminatory pricing. o To maintain a high degree of self-sufficiency in sugar production, and to encourage production through the provision of subsidized inputs.  Figure 1: MEXICAN CANE AND DOMESTIC SUGAR PRICES AND WORLD SUGAR PRICES 60 - 50 40 - 30 10 66 67 6B 69 70 71 72 73 74 75 76 77 7B 79 50 B1 02 03 04 05 B5 07 CWIMC O 0 CANE PRICE -I WORLD PRICB