Agricultural adjustment and food policy reform in Mexico

NUMBER 117 * ED Precis U Operations Evaluation Department June 1996 Agricultural Adjustment and Food Policy Reform in Mexico The Bank's tzwo successfiul agricultural technology, achieve high yields, non-market mechanisms by mar- adjustmenit loans to Mexico w,ere un- compete in international markets, ket ones, while providing support usual in supporting a nuniber of key and earn high incomes. Small-scale for a poverty-oriented food assis- reforms to imlprove food security. Had farmers produce mainly rainfed tance program. The loan was pre- they igniored food security issues, like crops whose yields are well below pared before the Salinas govern- most comparable operations financed those with irrigation; they do not ment was elected in 1988, and the by the Bank, they would have had less generally compete in international Bank had done no in-depth sector impact on poverty and perlaps rmore markets, and earn subsistence in- work on Mexican agriculture for difficulty in accomplishing the struc- comes often supplemented by sea- five years. The idea was to go tural reforms needed for long-ruiii sonal work on commercial farms. ahead with a quick, relatively development. easy, operation, as precursor to a Before the reforms, government more demanding operation to be After the debt crisis of 1982 and agricultural policy sought to en- designed after the election. The several years of poor economic per- courage production by protecting AGSAL I program, some of which formance, Mexico began redefining the major food crops and cotton was implemented before the loan its basic economic strategy and through high guarantee prices, took effect, was designed to: growth policies. By the early 1990s, which were made feasible by im- fiscal adjustment, trade liberaliza- port restrictions on the crops in * reduce government interven- tion, and deregulation, as well as question. Since the same guarantee tion in agricultural markets, in- reforms in industry, agriculture, fi- prices applied throughout the coun- cluding replacing guarantee prices nance, and infrastructure, had re- try, producers near production cen- with a price band based on world placed the dirigiste, public produc- ters received lower subsidies than market prices, for some key tion and control model of the 1970s those far away. Simultaneously, commodities; with a market-oriented strategy food policy worked to keep urban that stresses integration into the consumer prices below farm prices * abolish quantitative restrictions world economy. The reforms were through heavy, untargeted subsi- on many agricultural imports; supported by sectoral adjustment dies and market regulations. These loans from the Bank (see also Precis consumer subsidies were expen- 97, Privatization and1 Deregulation sive, amounting to about 1 percent "P'rtor,nance audit report: in Mexico-ed.). Among these, the of GDP in 1986. They only covered A!le'.L: A,ricultur,I Sector first and second agricultural sector urban areas. In rural areas, where ,4d11,:t:qnent Loan, I and If." bit adjustment loans addressed struc- the poor are concentrated, consum- loh:n Baftt.s l-and fac1 A leemian. 1 tural issues in agriculture and food ers often paid more for food than totulhconnno,, Ju::te 1996. Pert01r- security. their urban counterparts. itinCYe a:1dit repo(irt s arc avz,7i- ,7a'le to, Bank A'C( iitwe direL tors In Mexico agriculture provides The loans MOzd s:aft ffrotn tlhe Internal more than a fourth of employment Dnocii,ents Unit an,l frotm, but contributes less than 10 percent The first agricultural adjustmttent Reton,al Intorntlrmtion Se; z,jce of GDP. Production is dualistic. loan, AGSAL I ($300 million, imple- Centel,s. Commercial farmers use modern mented 1988-90), sought to replace  ing the poverty of the most vulner- Box: Modern lech nology helps target Mexico's food subsidies able groups through better targeting of food programs. The loan package For the tortilla subsidy pro- bar code information is down- also contained measures to raise re- gram. a list of potential beneficia- loaded through a computer link, soucoductvit and to help ou- ries was selected through non- and the store is paid for the trans- source productvity and to help out- discretionary measures, including actions at the official price of torti- put and input markets function household surveys, and computer- llas. Payments are made by an better. In particular, it contained: ized. Criteria for including and independent specialized agency, excluding families were defined administered by the national cred- * trade and other policy reforms taking account of income, spend- it institution, Banco Nacional de to reduce further the government's ing, housing conditions, and fain- Comercio Interior, which works role in producing, planning, market- ily possessions, and the poverty on the basis of a fully automated ing, storage, and processing agricul- line was defined as families with system. tural products and inputs and to incomes of two minimum wages encourage more competition from or less. Calculations were made The list of beneficiaries is con- the private sector; for each family listed. tinually monitored and updated; random verifications are carried * tighter targeting of consumer Families who meet the eligibil- out each quarter in each city and food subsidies; and ity criteria each receive a plastic families who no longer qualify are card containing a unique bar code replaced. The program uses a geo- * changes in the role of the Minis- that identifies them as beneficia- graphic information system based ries of the program. In 1994, the on digitized mapping. In 1994, the try of Agriculture, away from at- sYstemn began using "intelligent" list of beneficiaries was consoli- tempted control of production and memory cards. Optical readers in dated with that of the milk subsi- toward policy formulation and the affiliated shops read the bar dy program, permitting economies regulatory functions. codes. W hen the beneficiaries of scale through the joint manage- present their cards the shop gives ment of the two programs. * A pilot program of food assis- them a kilogram of tortillas free of tance to the poor was introduced in charge. Subsequently the store Source: Government comments on selected rural areas. The program employee takes the optical reader the project completion report for operated successfully and the Bank to an exchange center where the AGSAL II. is currently discussing with the gov- ernment the possibilities for its rep- lication in a number of poor states. * reduce the role of agricultural on output and exports went further parastatals, including selling and than loan conditions required. It re- The Bank and borrower had more closing some enterprises and limit- duced the share of agricultural and difficulty agreeing on the consump- ing the government's coverage of agro-industrial output protected by tion assistance than on the growth parastatal operating losses; quantitative restrictions from 18 and productivity measures. The percent in 1986 to 13 percent in government suggested breaking the * liberalize domestic agricultural 1988. The agricultural economy AGSAL II program into two opera- trade; moved closer to a market-pricing tions: one to deal with growth and system. Arbitrary and costly trade productivity, the other with food * eliminate subsidies on farm restrictions were reduced. consumption. The Bank argued inputs; against this because the funding for Food vouchers (tortibonos) were the targeted food consumption pro- * increase the efficiency and introduced to safeguard the food grams would depend on the elimi- volume of public investment in consumption of the urban poor, nation of general subsidies, and the agriculture; with purchase prices that were to impetus for reform would decline increase with inflation. This pro- considerably if the components * decentralize and reduce staff of gram reached a significant fraction were separated. The Bank argued the agricultural ministry; of the nutritionally at-risk popula- for an expensive countrywide nu- tion at an acceptable cost. The value trition assistance program. As fi- * establish a more targeted food of urban food assistance reached nally agreed, the loan provided subsidy program with adequate $456 million in 1988, compared with for a pilot nutrition project in four funding. the target of $250 million. states, whose results could be used to guide Mexico's Nutrition and The first adjustment loan was The second agricultural adjustment Health Program. successful overall. The parastatals loan, AGSAL II ($400 million, imple- were divested on schedule. The mented 1991-93), sought to increase Another source of contention reform of quantitative restrictions agricultural growth while alleviat- was the continuing protection of Junte 1996  maize, wheat, and sugar. The First, the tortibonos program sub- The Bank contributed to this Bank accepted the government's stituted in part for Mexico's gen- success story because of the heavy position that these crops were vital eral subsidies on food. Tortibonos emphasis it put on food security elements in the negotiation of the were vouchers distributed to urban issues during the preparation and North American Free Trade Agree- consumers who used them to buy implementation of both adjust- ment (NAFTA) and that any addi- tortillas at below-market prices. ment loans. For the Bank this em- tional conditionality on the Bank The vouchers, whose value to con- phasis was unusual. More than 85 loan, to encourage speedier liberal- sumers was to keep pace with in- percent of the Bank's agricultural ization, would undermine the gov- flation, were accepted at affiliated adjustment loans have completely ernment's NAFTA negotiating tortilla shops, which subsequently ignored food security. Other Bank position. cashed them for amounts reflecting operations dealing primarily with the market price of tortillas. The agricultural policy reforms would Overall, the second adjustment program was not targeted in that do well to build in similar, sub- loan was successfully implemen- it had no instruments to monitor stantial attention to this need. ted. Remaining export restrictions who was benefiting. on key food commodities were Farm income support eliminated, as was the Ministry of Second, under the second adjust- Agriculture's system of imposing ment loan, the tortibonos program Mexico is also exceptional in in- crop production targets. Prices of was replaced in 1991 by a targeted troducing a system of agricultural important food commodities were subsidy on tortillas. The tortilla sub- subsidies that does not distort re- liberalized, as were those of pesti- sidy program is designed to trans- source allocation. PROCAMPO cides. The agricultural investment fer income to poor urban families is an income support scheme budget was increased. The govern- through provision of a staple food. that is designed eventually to re- ment met most of its contractual Participating families receive one place agreed prices. The scheme obligations. kilogram of free tortillas daily. is consistent with NAFTA require- The amount of the transfer is the ments for gradually liberalizing However, backtracking on the important element, not the nutri- agricultural pricing and marketing. earlier price rationalization, the tional content or contribution to Farmers receive subsidies based government gave grain producers the beneficiaries' diet. By 1993, the not on what they produce, but on marketing support as part of the tortilla subsidy program was oper- the amount of land that they have precios de concertaci6n scheme (i.e., ating in more than 200 cities, had cultivated for specific crops. Be- agreed prices between producers, more than 2 million beneficiary cause payments are based on consumers, and the government). families-60 percent of the target past cultivation practices, they Along with delays caused by families-and 13,000 affiliated do not affect prices, production Mexico's NAFTA negotiations, tortilla shops. The use of leading levels, or any other requirement these subsidies, which were at odds edge technology makes possible of performance, and thus their with the reform program, delayed more accurate targeting and effect on production and trade is the release of the second tranche of greater operational efficiency neutral. PROCAMPO restricts the loan. The Bank finally accepted (see box). payments to the highest income the government's position that the new subsidies were soon to be re- placed by a new income support Annual growth rates scheme for farmers, PROCAMPO. (percent) Food security 6 The reforms supported by the two agricultural adjustment loans 4 - led Mexico to reduce the expensive general food subsidy and introduce 2 - a program of targeted subsidies. The targeted program costs less o _ __ and has improved the govern- ment's fiscal position. -2 A key element of Mexico's food -4 security program is the subsidiza- 84-87 1988 1989 1990 1991 1992 1993 1994 tion of tortillas for the urban poor. -D GDP - Agriculture This subsidy evolved in two stages. OED Pr&cis  farmers by imposing a benefits from 0.2 percent in 1984-87 to 2.2 program will stimulate overall agri- ceiling. percent in 1991-94. cultural growth. In the Mexican case, the failure of the reforms Initially the Bank resisted The expected faster growth to induce a supply response has PROCAMPO and advocated em- of agriculture had been based on caused confusion and may have ployment-generating and other the assumption that Mexico's agri- reduced support for the reform infrastructure in rural areas, culture was heavily taxed in the program. especially in the poor southern years immediately before the re- states, rather than a high cost sub- forms, 1984-87. But it now appears Supporting analysis sidy to benefit all farmers, rich that the discrimination against agri- and poor. But finally it accepted culture was more moderate. Fur- Agricultural policy reform in- PROCAMPO as a way to facilitate ther, Mexico's two main crops, volves substantial learning, but commodity price liberalization and maize and beans, accounting for it is better to begin with sector to deal with the political need to more than half the value of crop analysis rather than attempting support agriculture while mini- production, were essentially left to provide the knowledge base mizing the negative allocative out of the reforms and remained and the reform actions almost effects. Although initially expen- heavily protected by quantitative simultaneously. When the first sive, PROCAMPO payments are restrictions on imports. Production agricultural adjustment loan to planned to decrease and then of wheat, also an important crop, Mexico was conceived, the Bank cease after some years. was protected throughout the re- had limited knowledge of the form period. Mexican institutions most impor- The scheme has been reduced tant for policy reform. Policy and in scope because of its administra- Real producer prices for crops institutional issues were analyzed tive complexity and high cost. nevertheless fell, on average, during collaboratively by Bank and bor- Nontheless, it is a considerable the reform period. Not only was the rower. But because of the lack of improvement over agricultural peso appreciating in 1991-94, hold- background sector work from be- subsidy programs, such as those ing down the value of export crops, fore the operations, knowledge of OECD countries, that tie sub- but world market price trends for about the consequences of policy sidy payments to the production Mexican agricultural commodities changes was slow in coming. of specific commodities, leading to were also falling throughout the substantial overproduction of pro- decade 1984-94. Institutional development tected commodities and excessive use of labor and capital in pro- Meanwhile, on balance the The adjustment loans did not tected subsectors. agricultural adjustment program have explicit institutional develop- probably raised input prices. Agri- ment goals. Even so, the cumulative Supply response cultural input subsidies fell substan- institutional development resulting tially on average. In 1988, they had from the policy changes they sup- The second agricultural adjust- amounted to about $2 billion-a ported will be enormous. Wide- ment loan sought to increase the large figure in relation to the ranging privatization of production rate of growth of agriculture by $22 billion of value added in agri- and commerce stimulates the devel- eliminating policy and institu- culture. In 1991-94 most of these opment of new production and tional distortions. But, through subsidies, notably on credit, seed, market institutions. Opening up ex- 1994 at least, Mexico's agricultural power, and fertilizers, were elimi- port markets leads to institutional output hardly expanded. Agricul- nated. Both the elimination of input change as farmers and other pro- tural GDP during 1991-94 grew subsidies and the exchange rate ducers respond by learning about at 0.7 percent a year, close to the appreciation would have reduced new markets, developing new busi- 1984-87 average of 0.8 percent. By profitability. This output-input ness relationships, and applying comparison, the growth of agricul- price squeeze explains the lack new technology. In the public sec- tural exports rose from 1.8 percent of supply response. tor, there is now a new role for the a year in 1984-87 to 2.7 percent a agricultural authorities, and new year in 1991-94, while the annual It is important to be clear at the organizations and rules for support- growth of total GDP increased outset on whether or not a reform ing the poor have been developed. OED Pr_is is produced by the Operations Evaluation Department of the World Bank to help disseminate recent evaluation findings to development professionals within and outside the World Bank. The views here are those of the Operations Evaluation staff and should not be attributed to the World Bank or its affiliated organizations. This and other OED publications can be found on the Internet, at http:// www.worldbank.org/html/oed. Please address comments and enquiries to the managing editor, Rachel Weaving, G-7137, World Bank, telephone 473-1719. Internet: rweaving@worldbank.org June 1996