Mexico - First and Second Trade Policy Project

Document of The World Bank FOR OFFICIAL USE ONLY MICROFICHE COPY Report No. 10386-ME Type: (PPR) Report No. 10386 WISE, G. / X31679 / T9 003/ OEDD2 PROGRAM PERFORMANCE AUDIT REPORT MEXICO TRADE POLICY LOANS I AND II (LOANS 2745-ME AND 2882-ME) FEBRUARY 28, 1992 Operations Evaluation Department IThis document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EOUIVALENTS (annual averages) Currency Unit - Peso (Mex$) On October 2, 1989, the exchange rat& in the controlled market was US$1 - Max $2,545.00; the free market exchange rate stood at US$1 - Max $2,560.00. ABBREVIATIONS AND ACRONY)IS AGSAL - Agricultural Sector Adjustment Loan BANCOMEIT - Bank for International Trade CHIE - Nationale Foreign Investment Commission COD - Country Operations Division DFI - Direct Foreign Investment EDP I - Export Development Loan I EDP II - Export Development Loan II FERTIMEX - Mexican Fertilizer Company FSL - Financial Sector Adjustment Loan GATT - General Agreement on Tariffs and Trade GDP - Gross Domestic Product GOM - Government of Mexico IDY - Industry and Development Finance Division IMF - International Monetary Fund ISPL - Industrial Sector Policy Loan NAFIN - Nacional Financiera, S.N.C. NTB - Non-Tariff Barrier OED - Operations Evaluation Department ORP - Official Reference Price PCR - Program Completion Report PEMEX - Mexican Oil Company PERL - Public Enterprise Reform Loan PPAR - Program Performance Audit Report PR - President's Report QR - Quantitative Restriction REER - Real Effective Exchange Rate SECOFI - Ministry of Trade and Industrial Development SIDERMEX - Mexican Steel Company SHCP - Secretariat of Finance and Public Credit SPP - Secretariat of Programming and Budgeting TFI - Trade, Industry and Finance Division TPL I - Trade Policy Loan I TPL II - Trade Policy Loan II FISCAL YEAR January 1 - December 31 FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Dwecto-Gewal Opratiom Evaluiaskm February 28, 1992 MEMORANDUM TO THE EXECUTIVE DTRECTORS AND THE PRESIDENT SUBJECT: Program Peformance Audit Report on Mexico - Trade Policy Loans I and II (Loans 2745-HE and 2882-ME) Attached, for information, is a copy of a report entitled "Program Peformance Audit Report on Mexico - Trade Policy Loans I and II (Loans 2745-ME and 2882-ME)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROGRAM PERFORMANCE AUDIT REPORT MIL TRADE POLICY LOANS I AND 11 (LOANS 2745-HE AND 2882-NE) TABLE OF CONTENTS PREFACE . .......... .. . . . . . . . . BASIC DATA SHEETS . . . . . . . . . . . . . . . . . . . . . . . . EVALUATION SUMHARY . . . . . . . . . . . . . . . . . . . . . . . vii PERFORMANCE AUDIT REPORT I. INTRODUCTION AND ECONOMIC BACKGROUND........ . . . .I II. THE REFORM PACKAGE OF JULY 1985. ......... . . . . 2 III. THE BANX-8 TRADE POLICY LOAMS........... . . . . . . 2 Desaign of TPL l &II . . . . . . . . . . . . . . . . . . . 5 Implementation of Trade Policy.Reform . . . . . . . . . . 6 IV. EVALUATION OF TPL I AND II. .......... . . . 7 Did the Bank Influence Trade Reform?. ........ . . . 7 Impact of Trade Reform. . ........... . . . . .8 Assistance in Mobilizing Financing. ......... . . .10 Another Objective - Technical Assistance..... . . . . .10 Evaluation of Administrative Arrangements. ..... . . .11 V. ISSUES AND LESSONS.. .............. . . . 12 Lessons. . ........... . . . . . . . . . . . 14 VI. . .CONCLUSIONS................. . . . . 15 PROJECT COMPLETION REPORT I. PROJECT IDENTITY........... ...... . . . .19 II.A ROUND . . . . . ...................19 A. Trade Policy Reform . . . . . . . . . . . . . . . . . 19 B. Protectionist Policies . . . . . . . . . . . . . . . . 20 C. Instruments of Protection . . . . . . . . . . . . . . 21 D. Interdependence Between Macro and Trade Policies . . . 22 E. Objectives of the Trade Reform . . . . . . . . . . . . 22 F. Linkage with Macro Objectives . . . . . . . . . . . . 23 This document has a restricted distribution and may be used by recipients only in the performance of their ufficial duties. Its contents may not otherwise be disclosed without World Bank authorization. TABlE OZ CP'dENT (cont'd.) Pate No. III. P BaMCTkgEoN . . . . . . . . . . . . . . . . . . . . . . 23 A. Background . . . . . . . . . . . . . . . . . . . . . . 23 B. World Bank Strategy . . . . . . . . . . . . . . . . . 24 TPL I . . . . . . . . . . . . . . . . . . . . . . . 24 TPL II . . .. . . . . . . . . . . . . . . . . . . . 26 C. Conditionality of the Loan Operations . . .. . . . 26 D. Credibility of the Trade Reform and Calendarization . 27 Z. Issues . . . . . .. . . . . . . . . . 28 Formulation of Conditionality ... . .. . . . . 28 IV. IMPLEMENTATION OF TRADE REFORM . o . . . . o . . . . . 29 A. Overview..................... . . . 29 B. Overall Evaluation of T-'. Operations . . . . . . . . . 29 Tariffs . . . . . ... . . . . . . . . .. . 32 Change of the Minimu irt Tariff to 10% 33 C. Export Promotion . ................... 33 D. Main Conclusions on Trade Reform to Date . . . . .. 34 V. MAJOR ISSUES .*. ......................... . . . . . . 34 A. Rnti-Dumping System . . . . . . . . . .. . . . . . . 34 B. Coordination of Trade Policy Reform with Accession to GATT . .. . . . . . . . . .. . . . . . . . . . 36 C. Sequencing of Reforms . . . . . . . . . . . . . . . o 36 D. Stabilization and Liberalization ... . . . . . . . . 37 E. Speed of the Reform . . . . . .. . . . . . . . . .o. 37 F. The Level and Changes in the Real Exchange Rate (REER) . . . . . . . . . . . . . . . . . . . . 38 G. Synchronization of Trade Reform with the Macro-Adjustment Program . .... . .. . . . . . . 38 Progress on Inflation Reduction . . . . . . . . . . 40 H. Domestic Regulations . . .. . . . . . . ... .. 40 Direct Foreign Investment . . . . . . . . . . ... .. 41 Public Procurement ... . . . . . . .. . . . o. .. 41 Subsector Programs . . . . . . . . . . . . . . . . . 42 Agriculture Sector Liberalization . . . . . . .. . 42 I. New Policy Agenda . . . . . . . .o. . . . . . . . . 43 J. Need for Remaining Reforms . ... . . .... . . . . 43 VI. IMPACT OF TRADE REFORM . . . . ....o . . . . 44 A. Introduction . . . . . . . . . o . . . . . . . 44 B. Exports and Imports . . . . . . . . . . . . . . . . . 45 C. Evolution of Total Exports . . . . . . .. . . . . . . 45 D. Evolution of Manufactured Exports . . . . . . . . . . 47 E. Evolution of Total Imports . . . . . . .. .. .. 47 F. Evolution of Manufactured Imports . . . . . . . . . . 49 G. Trade Ratios...... . . . . . . . . . . . o. .. . 49 TABLE OF CONTENTS (cont'd.) Page No. VI. IMPACT OF TRADE RUFORM (cont'd.) H. Private Soctor Perception of the Trade Reform . . . . 49 Perception of the Macro outlook . . . . . . . . . . 51 Lessons of Experience . . . . . . . . . . . . . 51 Impact of the Trade Reform on the Private Sector . . 52 I. Impact of Trade Reform on Public Sector Enterprises . 53 VII. SUSTAINABILITY OF TRADE REFORM PROGRAM . . . . . . . . . 54 VIII. WORLD BANK'S PERFORMANCE . . . . . ..... . . . . . 57 IX. PROJECT RELATIONSHIP. ................ . .. 58 X. OPERATIONAL ISSUES . . . . . . ............. . 58 A. World Bank-IMP Relations . . . .......... . .. 58 B. Disbursement Performance under the Loan Operations . . 59 C. Technical Assistance under TPL I . . . . . . . . . 59 D. Cofinancing......... ... . . . . .... .. 60 XI. CONCLUSIONS.................. . . . . . 60 ANNEXES I. Compliance Under TPL I & II Operations..... . . . . . .61 II. Trade Policy Loan I, Loan/Credit Data...... . . . . . . 63 III. Trade Policy Loan II, Loan/Credit Data . . . . . . . . . . 64 List of Tables I. Progress Under Trade Liberalization Program, 1985-1989 . . 30 II. QR Coverage . . . . . . . ... . . . . . . . . . . ... . 31 III. Remaining QRs by Sector, end-March 1989 . . . . . . . . . 32 List of Gralhs I. Inflation Rate . . . . . . . . .. . . . . . . . . . . . . . 25 II. Real Exchange Rate . . . . . . . . . . . . . . . . . . . . 39 III. Evolution of Exports . . . . . . . . . . . . . . . . . . . 46 IV. Structure of Exports . . . . . . . . . . . . . . . . . . . 46 V. Public and Private Exports . . . . . . . . . . . . . . . 46 VI. Manufacturing Exports . . . . . . . . . . . . . . . . . . 48 VII. Evolution of Manufacturing Exports . . . . . . . . . . . . 48 VIII. Structure of Total Imports . . . . . . . . . . . . . . . . 50 IX. Selected Import Groups . . . . . . . . . . . . . . . . . . 50 - L - PROGRAM PERFORMANCE AUDIT REPORT TRADE POLICY LOANS I & II (LOANs 2745-ME AND 2882-ME) PREFACE This is a Program Performance Audit Report (PPAR) on Trade Policy Loans I & II, the Bank's first and second loans in support of Mexico's trade liberalization program announced in July 1985. The first of these loans - TPL I (2745-ME) -- was approved at the end of July 1986, became effective in November 1986, wa suL--";tially disbursed within two months, and closed at the end of November 1991. The second -- TPL II (2882-MR) -- was approved in November 1987. became effective in January 1988 and was completed in August 1989. While there was no co-financing directly linked to these loans, disbursement of part of the second tranche of the first loan was made a condition for the drawdown of the first US$2.5 billion of a financing package by a group of commercial banks. The PPAR consists of the Program Performance Audit prepared by the Operations Evaluation Department (OED) and the Project Completion Report prepared by the Latin America Region. The PPAR is based on the attached PCR, the President's Report, 4he loan documents, on a study of Bank files, and on interviews of Mexican officials and Bank staff. An OED mission visited Mexico early in 1990, and discussed the effectiveness of the Bank's assistance with officials in the Ministry of Commerce and Industrial Development (SZOPI), Bancomext, and Banco de Mexico. Follow-up discussions were held in November 1990. The kind cooperation and valuable assistance of Mexican officials in the preparation of this report Js gratefully acknowledged. The PCR provides a good account and assessment of the program experience, and discusses in detail the Borrower's implementation of the trade reform. The PPAR elaborates on particular aspects such as the design and implementation of Bank support, the consistency with stabilization policies, and the performance of the technical assistance component. The PPAR was sent to the Borrower for comments. The Borrower offered no comments on the body o: the report, but requested correction of the basic data sheets to reflect that there had been partial repayment of the first loan. The sheets have been corrected accordingly. - iii - PROGRAM PERFORMANCE AUDIT REPORT TRADE POLICY LOAN I (LOAN 2745-ME) BASIC DATA SHEET LOAN POSITION (Amounts in US$ Million) As of Dec. 31. 1991 LAR Oiginal Disbursed Cancelled Renaid outstandina Loan 2745 500.0 494.7 - 83.34 413.35 CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS FY87 FY88 Y89 Appraisal Estimate (US$M) 494 5 1 Actual (US$M) 489.01 0 0.32 Actual as % of Appraisal (%) 98.9% - 32% Date of Final Disbursementt November 25, 1991 PROGRAM DATES original ActalA Initiating Memorandum 04/30/85 08/16/85 Letter of Development Policy 07/22/86 07/22/86 Negotiations 12/85 03/06/86 Board Approval 02/86 07/29/86 Signing 08/08/86 08/08/86 Effectiveness 11/06/86 11/18/86 Loan Closing 11/30/88 11/30/91 - iv - MISSION DATA No. of No. of Staff Date of Month/Year X2ka EqMWga ees Report .reparation 11&12/84 3.6 4 14.4 02/01/85 Preparation 4&5/85 2.0 5 10.0 05/06/85 Appraisal 09/85 3.0 8 24.0 10/85 Appraisal 11/85 3.6 8 28.8 04/86 Negotiations 04/86 1.0 4 4.0 12/18/85 Supervision I 11/86 2.0 6 12.0 12/86 Supervision II 09/87 1.0 2 2.0 09/87 Completion 1-8/89 2.0 1 2.0 08/89 OTHER PROGRAM DATA Borrower/Executing Agency: Bancomext Follow-on Overation(s): Project: Mexico Trade Policy Loan II Loan/Credit No.: Loan 2882-ME Amount: US$500 million Board Date: November 11. 1987 - v - PROGRAM PERFORMANCE AUDIT REPORT TRADE POLICY LOAm II (LOAN 2882-KE) BASIC DATA SHEET LOAN POSITION (Amounts in US$ Million) As of Dec. 31, 1991 k9An Original Disbursed Cancelled BeRai Outstandina Loan 2882 500.0 500.0 - 41.67 458.33 CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS FY88 EL89 Appraisal Estimate (US$M) 500 - Actual (US$M) 299.42 265.16 Actual as % of Appraisal (%) 59.8% - Date of Final Disbursements January 6, 1989 PROJECT DATES Oricinal Actual Initiating Memorandum 03/09/87 03/09/87 Letter of Development Policy 05/87 05/87 Negotiations 06/10/87 06/10/87 Board Approval 09/87 11/12/87 Signing 12/07/87 12/07/87 Effectiveness 12/87 01/20/88 Loan Closing 06/30/88 12/31/88 Actual Completion 08/89 08/89 - vi - OISSINQDATA No. of No. of Staff Date of Month/Year 192%_ Persons eekg Report Preparation 11/86 2.0 6 12.0 12/86 Appraisal 03/87 2.0 8 16.0 05/87 Post Appraisal 09/87 1.0 3 3.0 09/28/87 Supervision 1* 02/87 3. 11 30- 02/26/87 Supervision II 04/88 2.0 2 4.0 06/03/88 Supervision III 06/88 .8 3 2.4 07/28/88 Completion 01/89 2.0 1 2.0 08/89 OTHER PROGRAM DAT Borrower/Executing Agency: Bancomext Follow-on Operation(s): Project: Mexico Industrial Sector Policy Loan/Credit No.: Loan 3087-ME Amount: US$500 million Board Date: June 13, 1989 * These figures include mission t4me spent on preparation of the green cover Trade Impact Study Report. - vii - PROGRAM PERFORMANCE AUDIT REPORT TRADE POLICY LOANS I & II (LOANS 2745-ME AND 2882-ME) EVALUATION SUMMARY Introduction introduction of a duty drawback scheme, rebating duty on imports to 1. In the 1970s Mexico tried to be processed and re-exported, and stimulate its domestic market the issuance of a statement of com- through expansionary public expendi- mitment to the rationalization of ture, and restrictive trade policies protection provided the basis for in order to resuscitate the failing Bank support in the form of the import substitution strategy that Export Development Project Loan had been the basis of its economic (#2331-ME) and for the promise of growth during the preceding two support for further trade reforms. decades. The effects of expansion- The fiscal deficit was reduced and ary policies were initially masked the balance of payments returned to by the exploitation of newly dis- a position of slight surplus during covered oil in the latter half of 1983-84. the decade, but by 1982 the fiscal deficit reached 17.6% of GDP; the 3. In July 1985 the Government, rate of inflation shot up to 98% - faced with a pessimistic outlook for more than three times the average oil prices and with negative growth for the last four years; and the and persistent inflation, announced current account deficit in the for- a new stabilization package with a eign accounts as well as the ex- trade reform component aimed mainly ternal debt more than doubled be- at the liberalization of imports. tween 1977 and 1982. With the This component provided for the levelling off of oil exports in the replacement of quantitative restric- latter year, the supply of foreign tions with tariffs, the lowering of credit dried up and the net foreign the average level of tariffs, the reserves became negative. narrowing of the dispersion of tariffs rates, and the eventual 2. Respondi_g to this crisis the reduction in the coverage of of- incoming administration in December ficial reference prices. These trade 1982 reversed the expansionary fis- reform measures partly reflected the cal and monetary policies; imposed Bank/country dialogue during the exchange controls; suspended repay- preceding two years. ment of the principal on US$60 bil- lion of the external debt; and rein- Objectives stituted quantitative restrictions on nearly 100% of imports. After 4. The Bank offered to support some modifications, and a signifi- the trade reform provided Mexico cant nominal devaluation of the would make the process irreversible peso, the stabilization program by announcing a three year calendar attracted IMF support in 1983. The of liberalization measures, and - viii - would establish a sound macro-eco- pressures. An Economic Solidarity nomic framework and a realistic Pact with unions and businessmen was financing plan. By mid-1986 Mexico negotiated. It provided for a announced a calendar of measures freeze on prices and wages. with covering the period 1985-88; reached public sector pricing and the ex- an advanced stage of negotiations change rate serving as nominal an- with tLe IMF on a stabilization chors. and for the restraint of program which would qualify for a demand through higher interest rates Stand-by; and commenced talks with and curtailment of public invest- commercial banks on new financing. ment. Trade reform was accelerated and deepened to complement this IMplementation anti-inflationary program. The program proved effective in reducing 5. In July 1986 the Bank approved inflation from over 150% in 1987 to the first Trade Policy Loan of under 20% in 1989. With the infla- Us$500 million, to help defray the tion brought under control GDP cost of imports not subject to non- growth and exports showed strong tariff barriers in Mexico, and to signs of recovery in 1989. assist Mexico in mobilizing external financing. The loan became effec- Results tive in November 1986, and US$300 million was disbursed immediately. 8. The TPL operations were suc- The Bank accepted that the Standby cessful in helping to bring about a agreement reached between Mexico and considerable improvement in trade the Fund satisfied the macro-eco- policy environment and greater open- nomic conditionality attached to the ness of the economy; the quantita- second tranche, and that the vigor- tive objectives having been amply ous implementation of reforms jus- met by the measures implemented. By tified the release of the tranche at May 1988 only 23 percent of trade- the beginning of December. By the able output was subject to import end of December a total of over licensing compared with over 90 US$400 million had been disbursed, percent at June 1985, the average thereby allowing Mexico to draw the tariff had fallen from 28 percent to first $2.5 billion of new commercial 11 percent and the dispersion among bank financing that had been made rates had been reduced to a third of contingent on drawdown of the Bank what it was in 1985; the use of and IMF support. official reference prices had been practically discontinued. 6. During 1987 Mexico accelerated the implementation of reform mea- 9. However, the structural change sures to the point where it was well leading to greater efficiency and ahead of the agreed calendar. In improved international competitive- hovember 1987 the Bank approved TPL ness of Mexican production, which II, providing US$500 million of was the primary objective of Bank additional support for incremental support under TPL I and II, needed imports benefitting from the reduc- time to gain momentum. It takes tion of restrictions. time for the system to attain the necessary credibility with producers 7. Yn December 1987, Mexico, in a and for them to respond by making stronger balance of payments posi- the adjustments in output, in pro- tion and with enlarged international duction processes and In orientation reserves, shifted the focus of its to internal and external markets. macroeconomic management to the Reform is expected to have its im- abatement of resurgent inflationary pact over the longer term; and the - ix - speed and extent of the response to pursued under subsequent Bank loans trade reform will depend on other and through operations wholly policies affecting investment in financed by Mexico and there has plant and equipment and in the adop- been substantial improvement in tion of technological change. customs procedures and data collec- tion, the impression remains that 10. In 1988 the recovery of fixed the inclusion of thiu component as investment was adversely affected by early as in 1986 was premature. the appreciation of the exchange value of the peso against the dol- Findinas and Lessons lar, and by the fact that nominal and real interest rates remained 13. The main lesson from this high. Although nominal wages rose review is that in a country with more slowly than prices, the capable economic management and doubling of nominal wages coupled confidence to go beyond the tradi- with a freeze on the nominal ex- tional fiscal/monetary squeeze, it change rate led to a slowdown in the is possible to design and implement inflow of funds for direct invest- an effective stabilization strategy ment and to a rise in investment without sacrificing progress toward outflows. This may have slowed trade liberalization. The stabili- structural change in response to zation package reflected in the trade reform. After 1988 the ex- December 1987 Pact was undertaken change rates were adjusted at a pre- while continuing to reduce the announced rate in an effort to avoid coverage of import licensing, to real appreciation of the peso. This abolish reference prices, and to effort only partly prevented a real lower tariffs. The trade reform was appreciation. Domestic interest seen as one of the means of keeping rates were reduced after mid-1989. domestic prices from being buoyed up Since then investment has been by inflationary expectations. While gaining momentum, which augurs well some stabilization measures may have for the impact of trade reform. delayed the response to import liberalization, the effectiveness of 11. TPL I undoubtedly met the the stabilization effort in bringing objective of assisting Mexico in inflation rapidly under control may mobilizing new commercial bank have contributed substantially to financing. TPL II, extended when the resurgence of investment en- Mexico was in a stronger foreign abling a positive response to trade exchange position, was never in- reform after 1988. tended to be a trigger for commer- cial financing. 14. Another lesson is that the formal announcement of a reform 12. The technical assistance com- calendar can do much to reduce the ponent of TPL I was only partly ambiguity and to enhance the sus- successful. It's implementation was tainability of the reform process. slowed because of budgetary con- Besides, the discussion of the straints during the tight fiscal calendar can provide an excellent stance of 1987-89. Up to late 1990 opportunity for the Bank to influ- some studies had not been completed. ence its content and timing without Some equipment was purchased only resort to overt conditionality. after considerable delay. About half of the original provision re- 15. The success of the Mexican mained undisbursed at the end of May trade reform should be interpreted 1991. Although the objectives of with caution by Dank staff in ad- this component of TPL I have been vising on trade reform in other -x - countries. Even though its fiscal situation was weak. Mexico could reduce tariffs because revenues from trade were a small share of total revenues; Mexico's system of pro- tection depended in large measure on non-tariff barriers. In many other countries, protection may depend on high tariffs, and revenues from trade may constitute an important share of total revenues; which would make the pace of trade reform a more delicate issue, not to be pressed without a careful analysis of the fiscal situation since other polit- ically acceptable and administra- tively feasible ways of replacing the lost revenue from trade taxes will have to be found. 16. Another lesson concerns tech- nical assistance. The experience with TPL I suggests that the Bank should re-examine the merit and manner of including a technical assistance component in quick dis- bursing operations. Technical as- sistance operations require careful preparatory work to ensure that adequate priority is given to the objectives to be pursued, that the government is ready and willing to provide the counterpart funds, and that administrative and implementa- tion arrangements have been made and are acceptable to all parties; in short the technical assistance project, and specifically the focus and terms of reference of studies, should be well identified before Bank funds are committed. In the absence of this preparatory work, delays are likely and disbursement of the loan will be slowed. PROGRAM PERFORMANCE AUDIT REPORT TRADE POLICY LOANS I & II (LOANS 2745-ME AND 2882-ME) I. INTRODUCTION AND ECONOMIC BACKGROUND 1. After more than two decades of sustained growth Mexico's strategy of import substitution failed to provide the impetus for continued expansion into the 1970s due to slow growth of the domestic market. In the 1970s the Government sought to enlarge the market through the expansion of public expenditure accompanied by restrictive trade policies. The predictable resuLt -- increasing public sector deficits and a continuing deterioration in the external trade balance -- were masked during the second half of the decade by the discovery and exploitation of vast oil resources. 2. Oil euphoria fostered the perception that high growth without politically costly reforms was feasible. The value of the peso in exchange against other currencies was allowed to appreciate; restrictions on private borrowing from abroad were lifted; and public spending was allowed to increase even more than before. As a result between 1977 and 1981 there was an upsurge in private foreign debt; growth of imports outstripped that of exports such that the deficits in the trade and current accounts of the balance of payments tripled; and capital flight soared as people grew pessimistic about the country's ability to hold the exchange rate. 3. With the levelling off of oil exports in 1982 the supply of foreign credit dried up and net foreign reserves of the Central Bank became negative. The fiscal deficit which had by then reached 17.6% of GDP had to be covered in large measure by monetary expansion, and the rate of inflation shot up to over 98% in 1982 compared with the under 30% of the preceding two years. 4. In December 1982 the incoming administration of President Miguel de La Madrid initiated a stabilization program and policy measures responding to the dire shortage of foreign exchange. These measures included the return to quantitative restrictions covering nearly 100% of imports, the forcible conversion of dollar deposits held by nationals, and the suspension of payments of principal on US$60 billion of the external debt. After some modifications in the stabilization approach and a rethinking of trade policy components, a program calling for the reversal of expansionary fiscal and monetary policies, the increase in prices of public sector services, and the first large scale devaluation of the peso attracted IMF support in 1983. 5. Significantly, the agreement with the Fund and policy statements by the government suggest that there was a clear commitment to the rationalization of protection. Although there was little progress in the reduction in the coverage of import licensing, in the use of official reference prices, or - 2 - change in tariff structure for the next two years, a free trade regime for exporters abolishing QRs and instituting a duty drawback scheme involving the refunding of duty on imports to be processed and re-exported was introduced. The World Bank's first Export Development Loan supported the export expansion objectives of this reform by financing a revolving fund to finance the imported inputs needed by exporters and by financing medium and long term investment sub-loans to export industries. 6. This program resulted in the reduction of the operational deficit of the government in 1983/84 and in the turnaround of the current account of the oalance of payments from a deficit of US$5 billion in 1982 to a surplus of the same amount in 1983 and 1184. However, the counter-inflationary impact was less than expected, and GDP growth was a negative 5% instead of the zero percent that had been expected. These results coupled with the sluggish performance of non-oil exports and the pessimistic outlook for oil prices led the government to take a fresh look at its stabilization/adjustment program and to announce a new reform package in July 1985. II. TH8 REFORM PACKAGE OF JULY 1985 7. The July 1985 Reform Package included a stabilization component designed to strengthen the balance of payments by curtailing demand for imports while giving a greater incentive to exports. It provided for cutting public expenditure, increasing public sector prices, increasing interest rates, and a nominal devaluation of the peso. The package also contained a trade reform component to reduce the level of protection through the liberalization of imports so as to induce Mexican producers to undertake longer term structural changes leading toward greater efficiency in production. The reform was to start by eliminating import licenses on 3600 items and reducing the coverage of QRs from 92% to 47% of tradeable production. 8. There were some measures to soften the initial impact of the trade reforms, e.g. slight increases in tariffs for half the items freed of QRs, and the addition of 400 items to the official reference prices (ORP) system.' The Government also announced its intention to introduce anti-dumping rules to prevent other exporting countries from flooding the Mexican market with subsidized goods and surpluses; an announcement which may have been interpreted as continuing protection under another guise. Notwithstanding these measures, the commitmeni to trade reform was firm, and by the end of the year the government announced its decision to negotiate Mexico's membership in GATT. III. THE BANK'S TRADE POLICY LOANS 9. In response to Mexico's 1985 trade liberalization package, which was substantially in line with Bank/Mexico dialogue, the Bank decided to negotiate loan support provided that: IPresident's Report on TPL I; Annex VII, paras. 25-28. - 3 - a. Mexico would design and announce a three year calendar of trade liberalization; b. liberalization would be backed by a sound macro-economic framework; c. there was a realistic financing plan. 10. Partly because of impending negotiations with GATT the government wat initially reluctant to issue a statement of its longer term intentions in foreign trade policy, as a condition for Bank support. The Bank insisted on the announcement of the policy because it regarded the trade reform process as lacking in transparency, conducted on a case by case basis and subject to substantial bureaucratic discretion. As such the process seemed imminently reversible. The Bank sought and received assurances that QRs were no longer being regarded as an important or permanent feature of the trade regime. Within a few months the government announced a tariff reduction calendar scheduling the reduction in tariff positions from 10 to 5, the narrowing of tariff dispersion, and the setting of a maximum tariff of 30%. These tariff reductions were to be achieved in four rounds, one every ten months, beginning in April 1986. 11. Whi)e accepting the Bank's need for assurances on the macro-policy framework, Mexico suggested that macro-policy issues and by implication macro- conditionality should be kept in the background as far as possible, given the pending negotiations with the IMF. The Bank, satisfied that the 1986 Budget and other parameters of the proposed economic program would contract domestic absorption and assist the liberalization program, if supported by appropriate exchange rate and interest rate management, "chose to conclude negotiations in the absence of a full agreement on the macropolicy program in view of renewed efforts by the Government to negotiate a program with the IMF".2 The President's Report on TPL I did not include an analysis of the macroeconomic program; but in reply to Board concern staff said that a program satisfactory to the Bank would be a condition of effectiveness. 12. Regarding the realistic financing plan, "it was jointly perceived by the authorities and the Bank that Bank financial and technical support for trade reform in Mexico could provide the fulcrum for Mexico's broader external financing strategy beginning in 1986".s Instead of a realistic financing plan being a condition for Bank support, Bank/IMF support and specifically the disbursement of a substantial part of the Bank loan became a drawdown condition 2Memo d/d July 2, 1986. President's Report d/d July 14, 1986 page 20, para. 40. Memo d/d July 28, 1986 - Questions from Executive Directors: "The Trade Policy Loan's (TPL) President's Report (PR) makes clear how critical the macropolicy framework (paras. 38-41) and the external financing plan (paras. 72-73) are, yet no details are provided on these subjects". s Back to Office Report d/d Dec. 18, 1985 - Senior Econ., LC1MX to Acting Chief, LC1MX: Mexico - Trade Policy Mission, November 7-25, 1985, Back-to-Office Report. -4- for the first tranche of new financing by commercial banks to Mexico. Thus financial urgency was perceived as such as to produce a favorable recommenda- tion of the loan to the Board prior to the conclusion of an agreement between Mexico and the IMF and between Mexico and the consortium of commercial banks. 13. TPL I was submitted to the Board and approved in July 1986. The amount of the loan US$500 million was justified on the ground that "given Mexico's need for external financing in the next four years, there is a critical mass of Bank financing which is perceived by the Mexicans as crucial to the liberalization process".' Of the total amount US$489 million was to help to pay for additional imports resulting from the removal of NTBU;' and was to be made available in two tranches of US$300 million and US$189 million. The first tranche was essentially to be retroactive financing of specified imports as from August 8, 1986, while the second tranche was to be disbursed on evidence of continlued progress in the implementation of the announced trade reform program, the publication of internationally acceptable anti-dumping regulations, and consistency between macro-policy and the trade reform program. Of the remaining US$11 million some US$8 million$ was linked to the estab- lishing of an anti-dumping system, i.e. for equipment and technical assistance to improve the quality, availability and timeliness of trade data, and US$3 million was for studies relating to the reform. 14. The objectives of TPL I were stated to be the following: a. Support the government's long term objective of gradually opening the economy through trade reforms as a means of improving domestic efficiency and international competitiveness. Bank resources were to finance part of the increase in imports resulting from liberalization; b. to be part of a larger program of Bank financial and technical support for Mexice's 1986-88 trade liberalization program; c. assist Mexico in obtaining external financing. "Commercial lenders have since late 1985 indicated to the Government and the Bank that major new lending in 1986 would not be possible without a new IMF program and structural reforms supported by the Bank, particularly in the area of trade liberal- ization". 15. TPL I became effective in November 1986. By early December the Bank staff informed the Board that the conditions for release of the second tranche had been met. Mexico was ahead of the agreed calendar for reform measures; and Memo on Loan Committee Proceedings, Sept. 27, 1985. 7 President's Report on TPL I; Loan and Project Summary - "The loan would finance part of the additional imports resulting from the removal of NTBs." S Schedule I of the Loan Agreement: "2(b) establishment of anti-dumping system under sub-paragraph (ii) $8,000,000". 9 President's Report on TPL I. - 5 - had reached agreement with the IMF and agreement in principle with commercial banks on the provision of additional external financing; and the Regulations for the Anti-Dumping Law of November 1985 had been published. 16. Work on TPL II began almost immediately after the release of the second tranche of TPL I. The initiating memorandum revealed general satisfac- tion by the Bank regarding the appropriateness of the macroeconomic framework since mid-1985. It was particularly satisfied that the difference between the controlled exchange value of the peso and the free market rate had declined from 50% at mid-1985 to between 2 and 3% during wost of 1986; that the real exchange rate had depreciated by 55% by the end of 1986; that the rate of monetary expansion had fallen; and that the public sector operational deficit as a percentage of GDP had been reduced.10 17. Noting that in spite of the rapid growth in non-oil exports GDP had failed to grow in 1986, and that there had been a sharp rise in inflation, the Bank was concerned about the risk this situation would pose for the continua- tion of trade reform. But Mexico's implementation of trade reform measures continued to outpace the calendar that had been announced, thereby providing no reason for delay in Bank processing of TPL II. 18. Trade Policy Loan II was appraised in March/April 1987, negotiated in June-September, and received Board approval in November of the same year. The amount of the loan was US$500 million, a figure consistent with financing estimates which were the basis for negotiations with commercial banks. This figure was maintained even though during the year the balance of payments performance was extremely good, and Mexico's foreign exchange reserves reached a record high of US$13.7 billion in gross terms. The loan was in two tranches with conditionality for both being the continued implementation of the reform calendar and consistency of macroeconomic policy with the reform program. It should be noted that the quantitative targets for reduction in NTBs, tariff reductions, and phasing down in the use of ORPs to meet conditions for effectiveness were already substantially met by the time of Board approval. 19. The objectives of TPL II were to continue the support of the Trade Policy Program's actions of 1987. Although not a formal condition, TPL II also supported the phasing out of the 5% import surcharge introduced for fiscal reasons the previous year. The surcharge was eliminated the month after Board approval, actually before the loan became effective. Design of TPL I & II 20. Both loans were designed to be quick disbursing. The time between effectiveness and release of the second tranche of TPL I was barely two weeks. Disbursement was linked to eligible imports which excluded categories subject to non-tariff barriers (NTBs); the rationale ostensibly being to gain leverage over the pace of reform by rewarding categories that had been liberalized. Given the fact that the targets for reform had been substantially achieved on the basis of actions already taken when each loan was submitted for Board approval, disbursement was not likely to be delayed by conditionality. 10 Memo d/d Feb. 11, 1987 - Second TPL Initiating Memorandum. -6- 21. In general both TPLe were designed with minimum conditionality; significant tightening of the reform program having been achieved during negotiation of the Mexican statement on trade policy which preceded the loan, and significant commitment to its implementation having been demonstrated by Mexico. Essentially, the requirements were progress with the announced reform program especially the reduccion in the coverage of NTBs, consistency between macro-stabilization and trade reform programs, and the publication of internationally acceptable anti-dumping regulations. These conditions were the minimum required for the improvement in efficiency and export competitiveness of Mexican products which was the ultimate objective of trade liberalization. 22. Although consistency between macro-stabilization and trade reform was recognized as necessary, TPL I & II did not include explicit macroeconomic conditionality. The Bank deferred to the Fund in the area of the macro-policy framework and specifically the stabilization program which was to be at the heart of the Fund/Mexico program. Implementation of Trade Policy Reform 23. Mexican trade policy reform occurred in two phases; the first between July 1985 and December 1987 and the second after December 1987. In the first phase the coverage of import licensing was reduced from 92.2% of the production of tradeables to 25.4%; that of official reference prices was reduced from 18.7% to 0.5% also of production of tradeables. At the same time the maximum tariffs were reduced from 100% to 40%, and the average level of tariffs fell from 28.5% to 23% excluding a temporary 5% surcharge during 1987. 24. This liberalization was accompanied by significant real devaluation of the peso. The real exchange rate index (trade weighted foreign prices relative to the domestic price level) rose from 100.0 in June 1985 to 157.4 in December 1986, falling to 147.2 in December 1987; implying a real devaluation of the peso by about 50%. This was brought about mainly by a 22% nominal devaluation shortly after the reform was announced in 1985 and scheduled mini- devaluations enough to slightly more than offset increases in domestic prices. 25. The second phase of trade policy reform began in December 1987, when in response to mounting inflationary pressures Mexico announced an Economic Solidarity Pact involving a reorientation of stabilization and trade policies which it had agreed with the trade unions and the representatives of entre- preneurs in the major business sectors. Primary goal of the pact was to reduce inflation through a price freeze followed by forward indexation of key prices, and through an incomes policy involving smaller increases in wages; with the exchange rate and public sector prices serving as nominal anchors for the program throughout 1988. Aggregate demand was restrained through continued tight monetary policy and higher interest rates. The trade reform program was deepened and accelerated to complement the anti-inflationary stabilization program. Maximum tariff and the average tariff were reduced to 20% and 13%, respectively; the coverage of import licensing was reduced to 22%; and the system of official reference prices was abolished. 26. Adjustments in the nominal exchange rate during 1988 were too small to prevent some appreciation in the exchange value of the peso against the US dollar. An extension of the pact was negotiated by the incoming administration in December 1988 to last through to the end of July 1989; was subsequently extended to 1991, and is still in force. The value of the peso was initially -7- reduced daily at an average annual rate of about 15% and later at about 10%, in an effort to avoid any appreciation in the real exchange value of the peso. 27. The distinction between the two phases lay not in the content of reform or in the vigor of its implementation, but in its relative priority among economic goals and in the orientation and content of macroeconomic policies. In December 1987 the curtailment of sharply rising inflationary pressures became the highest priority objective. Liberalization of imports took on added importance as it assisted in moderating inflation; the avail- ability of cheap imports tending to restrain price increases by domestic producers. The peso was allowed to become overvalued because this also served to help to hold down prices and make the incomes restraint acceptable, but this came at the expense of a stronger balance of payments position as the incentive to export was reduced and the incentive to import was increased. Pricing of public services was also affected by the Pact; and tight monetary policy and higher interest rates became the main toole for demand management. These high interest rates most likely discouraged investment necessary to the restruc- turing of enterprises to achieve greater efficiency in the face of lower levels of protection. IV. EVALUATION OF TPL I & II 28. The evaluation of TPL I & II must take as its point of departure the two main objectives which were being pursued through these operations. The first objective of trade policy support was to assist Mexico in taking measures designed to strengthen its external account by improving the efficiency and hence the international competitiveness of its tradeable goods sector. The second was to assist Mexico's effort to mobilize external financing from the intirnational commercial banking community. 29. The performance of Mexico in the implementation of trade reform was marked by an unswerving commitment throughout the period between July 1985 and the May 1988. By the latter date all the targets whith Mexico had agreed in its reform calendar had been met or exceeded. Thus from the point of view of opening up the economy the quantitative policy objectives of TPL I and II were amply met. Using the readily available measure of openness - the sum of merchandise imports and exports as a percentage of GDP - the degree of openness increased from 20.7% in 1984 (the year before the major import liberalization) to 27.9% in 1987. Did the Bank Influence Trade Reform? 30. An evaluation of the efficacy of the Bank's Trade Policy Loans in achieving or at least furthering this objective must determine whether the Bank's operations influenced the content and timing of trade reform, and determine the impact of the measures taken. In the case of TPL I, although it was appraised by staff some months after the July 1985 trade policy reform was announced, there is reason to believe that its content and probably its timing were influenced by the Bank/Mexico dialogue. 31. As early as 1982 after the incoming administration announced a multi- pronged strategy to relieve the pressure on the balance of payments and to put the economy back on a sustainable growth path, (and reached agreement with the IMP on an EFF), the Bank offered to provide some urgently needed foreign - 8 - exchange through its Special Action Program. Because a SAL was not considered the appropriate vehicle at the time, it was agreed that the Bank would provide quick-disbursing support for discrete facets of the Government's strategy. As part of this support the First Export Development loan, intended as the first step in a series to assist the longer term program of trade reform, was appr,jved in July 1983. 32. After a joint review of the Bank's program for Mexico, the Government invited the Bank to send a Trade Policy Mission in Novamber-December of 1984. T At mission proposed a multi-year strategy of gradual reduction of QRs and tariff rationalization. The Bank sent a follow-up trade policy mission in April 1985. and assisted in the organization and financing of a major trade policy seminar in Mexico in July of the same year. Bank records suggest that the TPL I proposal grew out of this earlier dialogue with Mexico on trade policy reform. While the extent to which the dialogue influenced the content of the Government's July 1985 announcement is impossible to determine with any precision, it seems clear that there had been a "meeting of minds" between Mexico and the Bank in this matter. The content of the Government's Statement on Foreign Trade Policy was certainly influenced by the Bank; four successive drafts of the statement were discussed during appraisal and negotiations of the first loan." 33. The conclusion to wh;ch one must tend, albeit on the basis of the limited evidence available, is that TPLe (especially TPL I) influenced the content of the trade policy reform. Bank operations, especially the associated policy dialogue, were influential in the application of imrroved quantitative criteria, specifically the use of production weights instead of import weights for coverage of licenses and reduction in tariffs. Import weights tend to ignore the protection to industries producing items which are not or are no longer imported. The use of production weights in Bank conditionality therefore tended to put pressure to speed up liberalization. Impact of Trade Reform 34. While the implementation of trade reform by Mexico produced dramatic changes in the policy environment within a short time, it may be too early to judge the effectiveness of reform in achieving the ultimate objective of structural adjustment of the tradeables sectors to make them more efficient and hence internationally competitive. This is so Iartly because of the relatively long gestation period of the investments requirod to improve efficiency; and partly because of the lag with which data becomes available to allow the effect of trade reform to be assessed and separated out from that of other policy initiatives and economic circumstances. Ideally, the effects of trade reform should be gauged by changes in production functions and total factor efficiency at the sector and subsector levels. In the absence of data at this level of detail one must resort to the more superficial analysis of whether the amount and structure of trade has responded to trade reform, and whether the level of investment has grown to bring about the improvements in technology, plant and equipment, changes in product mix and orientation to new markets. 1Fax d/d Feb 28, 1989; Bank Res. Rep. to LA2CO: Mexico Trade Policy Loan/ Some Preliminary Observations - "The role of the Bank in the July 1985 reform has been in the background". - 9 - 35. While the quantity of exports could be one indicator of international competitiveness of the Mexican economy after import liberalization, so many other factors may have influenced the growth of exports since 1985 that it is difficult to attribute the growth of exports to the trade reform supported by TPL I & II, which was essentially import liberalisation. There were other policy measures (supported by the Bank's Export Development Project Loan) favoring exports, implemented as far back as 1983, which influenced the behavior of exports after mid-1985. There were also some policies, e.g. significant devaluation of the peso, which provided the incentive for more intensive use of capacity in export industries.n Preliminary results from a small sample survey of firms to determine perception and impact of trade policy reforms suggest that these have been the dominant influences on export growth, rather than the reforms beginning in 1985 leading to new exports and increased efficiency of production of tradeables.1 Real growth of non-oil exports was negative in 1988 but recovered in 1989 to a small but positive figure. The difference in performance of non-oil exports before and after the policy reform of December 1987 was probably due partly to the difference in the exchange rate policy. Whereas the 1985 reform was followed by a large nominal devaluation resulting in a fall in the real value of the peso vis a via a trade weighted basket of currencies, the nominal exchange rate immediately after the December 1987 Pact was fixed to serve counter-inflationary objectives and resulted in an appreciation of the peso at least against the US dollar, and in erosion of the competitiveness of Mexican exports especially in the US market. 36. The behavior of merchandise imports between 1985 and 1987 would suggest that the response to the liberalization of imports initially may have been masked or offset by other influences. After mid-1985 imports would have been affected positively by the reductions in tariff and the lowering of NTBs and negatively by the relative devaluation of the peso as well as by the 5% import surcharge imposed during 1987. After December 1987 the surcharge was no longer applied but the real exchange value of the peso was allowed to creep upward relative to the US dollar. The behavior of total value of merchandise imports, falling in 1986-87 below the 1985 level and recovering strongly to surpass it in 1988, suggests that the exchange rate and the surcharge were the dominant influences determining the responsiveness of the economy as a whole to import liberalization. Although exchange rate policy after 1988 moved toward correction of the overvaluation of the peso, imports have grown much more rapidly than has GDP. To the extent that this has been due to changes in the sources of intermediate goods it may be have been an effect of trade reform. 37. Data on the structure of imports suggest that producers did respond to the opening up of the Mexican economy at least in deciding on the source of their inputs. The share of intermediate inputs in merchandise Jmports Memo d/d May 15, 1990; Consultant (Research Institute of overseas Investment, The Export-Import Bank of Japan) to Chief, OEDD2: Draft Report, Section C - Micro Level Perceptions and Responses. "Automobiles and auto parts accounted for a significant share of the growth in exports during this period; but the performance of this sector cannot be ascribed to trade reform measures under TPL I. Non-tariff barriers for autos continued, and autos were an exception to the phasing out of QRs under TPL I. (See TPL I President's Report page 33 and Annex VII, page 85, para. 24.) - 10 - increased from 67.9% in 1985 to 72.2% in 1987. Meanwhile, the shares of consumer and capital goods declined. In contrast to consumer goods, the share of which recovered strongly in 1988 when the peso was allowed to appreciate against the dollar, that of capital goods imports did not.1' The sluggish behavior of capital goods imports very likely reflected the impact of high domestic interest rates on Domestic Investment and of the exchange rate on direct foreign investment inflows, which fell sharply in 1988 after having doubled in nominal terms between 1985 and 1987.1s 38. Gross Domestic Investment in 1986 and 1987 was 22% below the 1985 level, and although it grew in 1988 was still more than 10% below that of 1985; reflecting the contractionary effects of tight monetary and fiscal stabiliza- tion policies. While the slowdown of investment would imply that stabilization efforts may have slowed some of the structural response to trade reform, the resurgence of investment in 1989 as inflation was brought under control suggests that no long term setback to the benefits of reform need result from the simultaneous attention to macroeconomic stability. Assis&ance in Mobiligina Financing 39. TPL I was effective in assisting Mexico in mobilizing external financing, both through direct Bank lending and by satisfying the conditions under which commercial banks were willing to lend. A review of the disburse- ment experience is relevant in this connection. Disbursement of TPL I proceeded as rapidly as planned, the actual being 99% of the planned during the first year. This was ensured by initially providing for up to $300 million to be disbursed as retroactive financing for qualified imports during the three months prior to effectiveness of the loan, and by later allowing retroactivity to eleven months to ensure that at least $400 million was disbursed as soon as possible to enable Mexico to meet the conditions for drawdown of the commercial bank financing. 40. In contrast, disbursement of TPL II was slow, with the actual being only half the planned level during the first year after effectiveness. The PCR suggests that the factors responsible for this difference in disbursement performance were the reduction in the number of eligible import categories, the inadequacy of documentation and processing arrangements, and the simultaneous existence of other policy-based Bank operations. Another Objective - Technical Assistance 41. The performance under the technical assistance component of TPL I was the least satisfactory among components in a generally successful program. Budgetary resources were not provided by Mexico due to the tight fiscal situation; and with the change in administration there was some time lost in implementing this component. As of late 1990 some studies had not been done, 16Source: Banco de Mexico. isSource: National Foreign Investment Commission, Bureau of Foreign Investment. - 11 - and some equipment originally envisaged for the project had not been bought.1 However, in the meantime Mexico pursued many of the improvements in customs data handling with its own resources instead of using those prce ided under TPL I. The undisbursed balance stood at Us$5.37 million at May 31, 1991 and the closing date has been extended to November 30, 1991. Some of the original objectives of the technical assistance component are now being pursued in the context of the Bank's Export Sector loan approved in February 1991. 42. Several facts point to inadequate design and follow-up of this component. The studies were not precisely defined at the outset.1' In a letter dated September 1989 Bancomext proposed to undertake six studies at a *otal cost of US$1.1 million. In March 1990 the Bank informed Bancomext that it was still awaiting information requested the previous year regarding terms of reference, coordination and personnel who would do the studies. In response Bancomext suggested cancellation of the balance for studies. Regarding equipment for data management, the Bank did provide fairly detailed recommenda- tions, and demonstrated flexibility by eventually accepting proposals which were not in line with its initial views. However, the arrangements for coordination among beneficiary agencies and for counterpart local cost financing were left undefined. Disbursement of the technical assistance provision was delayed even though implementing agencies submitted proposals for approval as far back as May 1987. Only SECOFI submitted " follow-up proposal, one substantially exceeding its original allocation. The Bank approved part of this; approval of the remainder was left pending proposals from other agencies which apparently lost interest after not receiving budgetary allocations for this purpose. Evaluation of Administrative Arranaements 43. Arrangements for the local auditing of the accounts for these operations apparently have not worked well. This is clearly the case with those for auditing the use of loan proceeds. Recent review in the Bank suggests that the Bank audit requirements regarding Statements of Expenditure and examination of documents and the specified imports have not been complied "Memo d/d May 30, 1990; LA2TP Loan summary - "In order to permit further commitments and disbursements under the technical assistance component of the Loan, the closing date has been extended to November 30, 1990. Out of US$11.0 million allocated for T.A., us$7.03 million has been committed and US$3.77 million disbursed. The closing date will most likely have to be extended again." "Memo d/d June 3. 1988; terms of reference of Supervision Mission included "discuss with Mexican authorities the design of data collection system". "Memo d/d Oct 11, 1989; Consultant to LA2TF - noted that on review INEGI proposals did not fully meet Bank recommendations, especially as regards the use of "dumb" terminals instead of PCs. - 12 - with.1' In other words the Bank cannot be sure that disbursements were made against imports which qualif!.ed in being incremental imports of classes affected by trade liberalization measures. This restriction had been made in order to ensure that Mexico would only have got the financing if it did liberalize imports. As it turned out Mexico did take the measures ahead of the agreed schedule, so that the safeguard the Bank had built in was unnecessary. However, it is now obvious that arrangements were not in place to ensure that the safeguard would have been effective had it been necessary. 44. The failure to adequately follow up the technical assistance component seems to have been the result of administrative arrangements within the Bank, under which the TPL operations revolved around one officer and therefore were adversely affected by his untimely death. Specifically, the proposals for drawdown of resources under the TA component apparently were not promptly attended to because the Bank was left an incomplete record of the dialogue with the implementing agencies. 45. The incompleteness of the record has also made ex-post evaluation of the operation difficult. In this case the inability of the Management Information System data to provide the basis for differentiating between time spent on supervision and on other activities has been an additional source of difficulty. V. ISSUES-AND LESSONS 46. The review of these TPL operations suggests a number of issues in the promotion of trade reform. One concerns the feasibility of structural reform in a climate of macroeconomic instability. This issue frequently centers around the question of whether to stabilize and then reform or to attempt both together, i.e. the issue of sequencing. The fact is that in the real world a country usually does not have the luxury of choosing the ideal sequence of stabilization and adjustment approaches. Macroeconomic crises have to be attended to whenever they arise even if. it means temporarily relegating adjustment to a lower priority. The experience under TPL showed that opportunities for making trade reform and stabilization policies mutually reinforcing can be created, and that the attention to stability need not stop the forward thrust of trade reform. However, it must be conceded that the success of the heterodox stabilization strategy introduced in December 1987, and the fact that it was designed to take advantage of accelerated trade reform, must be attributed to the willingness and ability of Mexican economic managers to go beyond the traditional Bank/Fund demand management and exchange rate devaluation approach and attempt a more comprehensive approach which 19 Memo d/d April 17, 1990; LATPS to LAC - "There is no indication of the work done to review the SOEs supporting the imports neither if the current conditions to accept the S0Es were checked; if the imports took place within the period provided in the Loan Agreement, if they were eligible according to such agreement, if the customs documentations was (sic) checked to verify that the goods effectively (sic) arrived and if the goods were paid." 'Memo d/d Aug.8, 1989; Consultant to LA2TF & Memo d/d Nov. 22, 1989: Economist, LA2CO through Chief LA2TF to Chief OEDD2 re Trade Policy Loans ... Project Documentation Issues. - 13 - addressed the problem of wage inflation and its dependence on the prices of imports and on the exchange rate. 47. Economic theory provides no unambiguous guide as to what constitutes consistency betweLA macroeconomic strategy and trade reform. This makes it important for the Bank either to: (a) state clearly the criteria for determining consistency of macroeconomic strategy with trade reform when this consistency is made part of the conditionality attached to the operation; or (b) simply make access to Bank support subject to the conclusion of a standby agreement between the client country and the IMF. In the case of TPL I the Bank provided no guidance as to what mix of fiscal deficit, interest rates and exchange rates would have constituted a consistent economic framework. There is some evidence that the Bank had designed the operations within a macro- economic framework; but this framework was not the basis for a transparent set of conditions to ensure consistency. Instead, the macroeconomic policy mix was left to be negotiated and monitored by the IMP within the context of a standby arrangement, apparently without discernible Bank input. 48. Another issue concerns the extent to which the importance of taxes on trade as a source of revenue may affect the feasibility and pace of trade reform. In Mexico the relative unimportance of trade taxes in total revenues was an initial circumstance favorable to trade reform. This was due partly to the dependence on quantitative restrictions rather than high tariffs as the primary means of industrial protection, and partly due to the high reliance on oil revenues. Between 1983 and 1988 taxes on trade accounted for 1.4 to 2.6 percent of total revenue of the government; the high of 2.6 reached in 1986 with the imposition of a temporary import surcharge. What this meant was that the replacement of QRs with tariffs and the eventual reduction of tariffs did not result in erosion of government revenues. Indeed the shift from QRs had a salutary effect in 1985 and 1986. It also meant that it was politically feasible to resort to the temporary expedient of an import surcharge in 1986. Thus although a weak fiscal situation was part of the Mexican stabilization problem, due mainly to the dependence on oil revenues, trade reform was not perceived as a threat to fiscal stability. This situation is probably not representative of many developing countries and the replication elsewhere of the Mexican trade reform approach should be approached with caution. Care should be taken to ascertain the feasibility and ease of devising substitute revenue sources. 49. Still another issue concerns the sensitivity of reform to the fact that industries need time to adjust to new rules. Reform should be designed in such a manner as to maximize the survivability of firms which can become efficient and internationally competitive. Mexico's approach to trade reform paid due regard to the need to avoid undue hardship by giving producers time to adjust. This was reflected in some measures to soften the initial impact, e.g., by slightly raising tariffs on many items freed of QRs and by temporarily adding to the items under the official reference prices system. In retrospect these actions may have enhanced the acceptability of the program in that once entrepreneurs believe that the government is sensitive to the problems the reform poses for them they are unlikely to mount an effective opposition to it. This approach did not appear to slow down its implementation eignificantly. - 14 - Lessons 50. From the above the following lessons for future Bank operations are worth noting: a. In a country with capable economic management it is possible to design and implement an effective stabilization strategy while undertaking trade reform. b. Formal announcement of a reform calendar can do much to reinforce the certainty and credibility of trade reform, and to enhance the sustainability of the reform process. The discussion of the calendar can provide an excellent oppor- tunity for the Bank to influence its content without resort to overt conditionality. The Bank's performance in this regard was a very positive aspect of the Trade Policy Operations. c. The relative unimportance of taxes on trade as a sot.rce of government revenue in Mexico was a favorable initial Aircum- stance which may limit the replicability of the TPL type operation in other countries. Bank staff should be careful to identify sources of revenue to replace lost revenue from taxes on trade in countries where these are an important component of total revenue. d. The use of domestic production as the basis for judging the progress of trade liberalization and the linkage of disburse- ment to the imports of goods in liberalized categories are design features worthy of consideration for similar opera- tions in othev countries. Given the extensive use of retroactivity in order to make TPL I fast disbursing, these features may not have been particularly effective in pushing trade reform in Mexico, but the approach could be invaluable where the commitment to reform is more tentative than it was in the Mexican case. e. In general, the Bank should try to avoid linking technical assistance to an operation where the provision of fast disbursing financing is one of the main objectives. TA does not usually lend itself to fast disbursement. Furthermore, TA is usually a small part of the loan and typically will not attract adequate attention from the Bank or from the client. In the case of TPL I the Bank did not pin down the design nor did it push for the expeditious implementation of the component in the early period after loan effectiveness. f. If TA is considered sufficiently important to the structural change that it must be included in a fast disbursing policy loan, then appropriate conditionality should link the use of TA to disbursement of the last tranche. If the objectives of the TA are achieved without use of Bank financing, then the provision for TA should be cancelled. Successive extension of the closing date in the hope of finding some other use for the funds should be avoided as this intrudes - 15 - on the prerogative of the Board to approve loans in full awareness of the purposes to which they will be put. g. TA should be substantially designed before loan approval and the Bank should plan to monitor its implementation. Notwith- standing the smaller size of technical assistance components the Bank should provide for their adequate supervision. Even where only small amounts of equipment are required, the Bank should still Insist that procurement be according to Bank rules. This will foster a general rather than a selective respect for Bank procedures, and ensure expected benefits from this component will in fact materialize. VI. CONCLUSIONS 51. The Bank's Trade Policy operations played an important supporting role in what has turned out to be a significant opening up of the Mexican economy, by influencing the content of Mexico's trade reform through the economic dialogue and by providing financing for additional imports and assisting in mobilizing commercial bank resources during a period characterized initially by weakness in the balance of payments. Clearly the success of these operations has been due mainly to Mexico's willingness to take brave measures to restructure the policy environment. - 17 - PROGRAM COMPLETION REPORT MEXICO TRADE POLICY LOANS I AND II (LOANS 2745-ME AND 2882-ME) November 16, 1989 Trade, Finance and Industry Division Country Department II Latin America and the Caribbean Regional Office - 19 - PROGRAM COMPLETION REPORT MEXICO TRADE POLICY LOANS I AND II (LOANS 2745-ME AND 2882-ME) I. PROJECT IDENTITY 1. Project Names: Trade Policy Loans I and II Loan Numbers : 2745-ME and 2882-ME RVP Unit : Latin America and Caribbean Country : Mexico Sector : Trade II. BACKGROUND 2. Mexico experienced three decades of high and stable growth from the early 1940s to the end of the 1960s, based on an inward-looking growth strategy. By 1970, however, Mexico had largely exhausted the easy possi- bilities for import-substitution, and growth started to slow down. The Government attempted to stimulate growth through an expansion of public sector expenditures and the encouragement of high-cost, capital-intensive import-substitution. This strategy failed to sustain sound economic growth, and in 1976, Mexico experienced a serious financial and economic crisis. The exchange rate was devalued, and an International Monetary Fund (IMF) stand-by program was introduced. Revenues from new oil discoveries in 1977 resulted in a quick economic recovery and greatly enlarged Mexico's access to foreign borrowing. However, discovery of oil in Mexico led to the overvaluation of the currency and reduced protection to domestic indus- tries. By 1980, rapidly rising public expenditures, unmatched by gains in revenue, were generating unsustainably large public sector deficits and although domestic inflation rose rapidly, no significant pressure was felt to adjust the economy because of rising oil export earnings and abundant foreign financing. At the same time, imports rose rapidly and non-oil exports declined. Growing fears of a major devaluation of the then fixed exchange rate, combined with domestic interest rates, which were low rela- tive to rising international rates, encouraged substantial capital flight. However, the increasing precariousness of the balance of payments was masked temporarily by massive public sector borrowing, which nearly doubled the level of public external debt in just two years from US$40 billion in 1979 to US$78 billion in 1981. A. Trade Policy Reform 3. Based on the realization that import-substitution and public sector-led growth had reached their limits, the authorities began to intro- duce reforms in the trade regime to provide a new stimulus to growth. - 20 - Since mid-1985, Mexico has undertaken a fast and far-reaching liberaliza- tion of the trade regime, aimed at expanding the tradeables sector, opening the economy to international competition and encouraging efficiency in both exports and import-substitution activities. In 1986, Mexico joined the GATT. The pace of liberalization was accelerated in late 1987 and early 1988 as part of the introduction of the macro-stabilization program aimed at drastically reducing domestic inflation, which had accelerated rapidly during 1987. 4. Mexico's trade reform has been carried out in a hostile external environment. Between 1982-88, Mexico's terms of trade fell by more than 35?, implying a cumulative income loss of around US$40 billion. Real interest rates in international markets remained high and continued to entail a significant burden on the economy. Net resource transfers to Mexico shifted from a historically positive position to a negative 6Z of GDP on average during 1982-88. The debt overhang created a climate of macroeconomic uncertainty which inhibited investment and placed enormous pressures on domestic financial markets. Despite these adverse external shocks, increasing exports enabled Mexico to improve its debt servicing capacity, with its total stock of debt falling in real terms over the period. 5. Before the 1985 crisis, the World Bank had made only one trade- related loan to Mexico, a 1983 Export Development Loan operation (EDP I, Loan 2331-ME) that achieved mixed success. The loan package negotiated between Mexico and its creditors in 1985 thrust the Bank into a central role in the policy dialogue, since the release of much of the commercial bank money was linked to various policy-based Bank loans then under discus- sion with the authorities. Between late 1986 and early 1988, the World Bank Board approved lending operations for policy-based loans (including hybrid operations) in the agriculture (Loan 2918-ME), fertilizer (Loan 2919-ME), and steel (Loan 2916-ME) sectors, Trade Policy Loans I and II (Loan 2745-ME and Loan 2882-ME), and another Export Development Loan (EDP II, Loan 2777-ME) all of which supported, to varying degrees, changes in the import regime. TPL I and II provided an overall framework for the rest of the lending program, with gross disbursements totalling over US$3.3 billion during 1986-88. B. Protectionist Policies 6. For three decades, Mexico sought to achieve the development of its industrial sector through the rapid expansion of its domestic market behind protective barriers. Industrial growth took place in the setting of a largely closed economy. Compared to Brazil and Argentina, protection in Mexico was not high. Resources were shifted to industry directly through public sector investments and indirectly through the incentive system for the private sector by means of quantitative import restrictions (QRs), tariffs, fiscal incentives and industrial regulations. Domestic industry was also protected by numerous barriers to entry and other regulations affecting factor and service markets. Industrial sector policies, based on - 21 - import-substitution and widespread controls, resulted in poor performance of overall productivity and the international competitiveness of Mexican industries. 7. Mexico's present economic difficulties date back to the beginning of the early 1970s, when increases in public consumption contributed to domestic inflation and to the appreciation of the domestic currency in real terms under a fixed exchange rate policy as discussed in paragraph 6. The overvaluation of the exchange rate, in turn, increased the application of protectionist measures that permitted the establishment of inefficient activities oriented towards the domestic market while discriminating against exports. 8. Protection in Mexico raised the costs of production with unfavor- able repercussions for exports. It created scarcity margins for the recip- ients of licenses, and reduced the transparency of the trade regime, espe- cially in the granting of import licenses. As a result, domestic produc- tion became much less efficient. Because of high protection, the cost of production became particularly high for iron and steel, capital goods, automobiles, and consumer durables. For the manufacturing sector as a whole, total factor productivity fell by 0.7% a year between 1970-82. 9. In the past, protectionist policies were defe-aded on the assump- tion that Mexico's large domestic market could support practically the full range of industries. This argument conflicts with the experience of both developed and developing countries. In fact, even countries such as France and Germany, whose market size for industrial products is much larger than Mexico's, rely on international specialization to ensure efficient produc- tion. C. Instruments of Protection 10. The following paragraphs give an overview of the protection devices used in Mexico before the introduction of the trade reform in 1985. 11. Tariffs (ORs) were high with a maximum rate of 1002 in 1982. 12. Quantitative Import Restrictions (QRs) were used as the main vehicle of protection. 13. Temporary Import Surcharges. In addition to the base tariffs, the Government used the option of applying surcharges of up to 502 of the base to provide temporary protection to industries, whose non-tariff barriers (NTBs) were eliminated. According to the agreement with GATT, the duration of these surcharges could be extended up to eight years. However, in early 1987, the authorities announced that temporary surcharges would be applied on a limited basis. The authorities also declared that, in practice, these surcharges would apply for no longer than five years, except in special cases. - 22 - 14. Import Tax. A uniform general import tax of 52 was applied for fiscal purposes. 15. Rule 8 (Regla Octava) granted the right to import inputs at a preferential ad-valorem tariff rate of 1OZ to industries included in special industrial programs. By increasing the spread between tariffs on inputs and tariffs on outputs, this mechanism raised effective protection for industries producing for the domestic market. 16. Official Reference Prices (ORPs) were administratively-determined prices that the Government used in place of actual CIF values to calculate import duties. ORPs were often significantly higher than CIF values. However, ORPs provided more transparent protection than QRs. 17. Public Procurement Polici. Public agencies oriented their pro- curement to domestic firms irrespective of price and quality considera- tions, increasing protection to domestic industries. 18. In addition, NTBs in the form of domestic content requirements, sanitary restrictions and product quality norms provided additional protec- tion to domestic industries. D. Interdependence between Macro and Trade Policies 19. In Mexico, trade policies have shown a significant interdependence with macro economic policies. QRs have tended to rise fastest during major balance of payments crises. As a result, Mexican policy makers have tradi- tionally used the import regime as an instrument to counteract balance of payments problems. The introduction of the licensing system in 1947, and the sharp increases in coverage rates in 1954, 1976 and 1982 coincide with Mexico's successive foreign exchange crises. Similarly, when foreign exchange constraints were not present, such as during the oil boom of the late 1970s, the import regime tended to become looser as a rule. In the 1950s, QRs covered about one-tenth of domestic production. There was a charp tightening of the import regime during 1982, and by September of that year virtually all imports were subject to licensing. E. Objectives of the Trade Reform 20. At the heart of the reform begun in 1985 is the Government's desire to increase the efficiency of Mexican industries and expand exports through greater exposure to international competition, to reduce the econo- my's dependence on oil exports and foreign borrowing, and to create a new stimulus for the resumption of growth. 21. The Mexican authorities believed that the liberalization of imports would provide an inducement for productivity improvements and for product specialization through increased import competitiveness and also reduce the anti-export bias of the system of incentives. - 23 - F. Linkage with Macro Objectives 22. As Mexico's domestic inflation rate soared in 1987 (Graph I), trade liberalization came to be seen by policy makers not only as an instrument for improving the long-term resource efficiency of the economy, but also as a supportive element in an anti-inflationary "shock" strategy which relied not only on fiscal/monetary contraction, but also on a freeze of nominal wages, key prices, and the exchange rate. To reduce domestic inflation, the authorities introduced a macro-stabilization program in late 1987. The key elements of this program included: (i) an increase in the fiscal effort of about 3Z of GDP; (ii) an immediate 22Z nominal devalua- tion; (iii) a lowering of the maximum import tariff from 45Z to 20Z; (iv) up front adjustments in public sector prices (64%), the minimum wage (382), and basic goods; (v) followed by a wage and price freeze; and (vi) a forward-looking monthly adjustment of wages and prices tied to a basket of basic goods and services. This program has been extended with some modifi- cations several times, most recently until March 31, 1990. III. PROJECT DESIGN A. Background 23. Once it became evident that the 1982 crisis would not be transi- tory, Mexico was forced to look for a more permanent solution to its devel- opment problems. At that time, the only clear viable alternative for the country was seen to be trade liberalization, which could help integrate the domestic economy to the world economy and provide a growth stimulus. Internal discussions within the World Bank (the Bank) supported the hypoth- esis that the country could return to sustained growth only through trade reform. It was also reasoned that trade liberalization would go a long way in addressing the fiscal issues by removing protection for the goods pro- duced by parastatals and in this way force them to become more competitive. Similarly, fiscal reforms in the absence of a complementary trade reform would not be expected to lower the public sector deficit. Furthermore, there was a politically strong lobby for trade liberalization in Mexico. 24. In this context, the Bank planned three loan operations based on a multi-year reduction in restrictions on imports. In 1984, the Bank sent a letter to the Government of Mexico proposing the change of direction in the Bank's lending strategy. A major discussion reportedly took place within the Mexican administration on the direction and importance of the Bank role in Mexico. In the end, the authorities agreed that they wanted the Bank to provide quick disbursing policy-based loans to Mexico and that a trade loan would be the first such operation. The authorities also requested that the Bank assist in their negotiations with the commercial banks to obtain reschedulings and new money. 25. The design of the loan packages was based on the notion that quick disbursing loans are anchored in a country's medium-term balance of pay- ments problems. While at the time both loans were prepared Mexico produced - 24 - a trade surplus, the external debt burden was still a major issue. It was also believed that the trade reform would increase the efficiency of the economy and help reallocate resources to the competitive tradeable sectors, and also help reduce domestic inflation. B. World Bank Strategy 26. Since the Mexican Government did not wish to receive a Structural Adjustment Loan (SAL) for political reasons, given the nature of the struc- tural issues the country faced, the Bank's strategy in helping Mexico address its long-term structural issues was through the formulation of a program consisting of a series of loans, which taken together included major elements of a SAL program. The trade policy loans were designed as the backbone of this strategy in a multi-year program with progress on trade reform being measured by a year-to-year reduction in trade barriers. The second loan for export development (EDP II), and the agriculture, steel and fertilizer sector adjustment loans formed part of the total package of reforms to reduce trade barriers for these subsectors and to improve the administrative framework for exporters. The strategy was based on the premise that these loans would reinforce each other within a well-designed package. 27. The Bank's judgement at the time was that trade liberalization was the key to a successful adjustment, since trade barriers were constraining the growth of the economy. In this context, TPL I started the reform process, and TPL II aimed at speeding up the momentum gained under TPL I. 28. Trade policy reform was the core of Mexico's adjustment program. Trade liberalization was also expected to strengthen external confidence in policy management. In this context, the TPL program provided a basis on which the Government mobilized external financing from commercial sources: TPL II was a trigger for drawdown on the financing package signed in March 1987. Commercial banks made some Bank loans, including TPL II, link loans that would allow the Mexican Government to draw down the commercial bank loans. This strengthened the hands of the reform-minded elements in the Mexican administration in pursuing a successful implementation of the trade reform program. TPL I 29. TPL I was conceived as the first in a series of fast-disbursing repeater loan operations over a three-year period (1986-88) in support of the Government's trade policy reform program. Specifically, TPL I was designed to support the reduction of NTBs and the simplification of and reduction in dispersion of tariffs. The gradual and programmed reduction of tariffs and NTBs was preannounced so as to induce the desired effect on expectations and new investment decisions. Trade liberalization, however, was expected to be phased over a period of several years given its far- reaching consequences throughout the economy and the time needed for adjustments at the firm and sectoral levels. Graph .1: Inflation Rate (Monthly Index) 16 14 12 - 10- 8- 6- 4 2 - Jai-80 Jan-81 Jan-82 Jan-83 Jan-84 Jan-85 Jan-86 Jan-87 Jan-88 Inflation Rate - 26 - 30. TPL I was designed to finance part of the increase in imports resulting from the removal of NTBs. After an adjustment period, during which further import liberalization measures were to be introduced, non-oil export growth resulting from the trade policy reforms was expected to com- pensate for an expansion in imports. The benefits of the program were assumed to derive from further improvements in the balance of payments, given the need to service large external debt obligations, the gradual recovery of sustainable growth and employment based on productivity and efficiency improvements; and the restoration of external confidence and creditworthiness. Repeater TPL operations were planned to support contin- ued reduction of QRs, further tariff reform, public sector procurement, and anti-dumping policies. TPL I was also designed as the first step in the process of trade liberalization which could be reinforced by more conven- tional operations and a large-scale cofinancing package. TPL II 31. TPL II was designed to continue and deepen trade liberalization in the period from 1987 to mid-1988, with a possible TPL III expected to follow in 1988 to support continuation of the liberalization program to the end of the sexenio (six year governmental period) in November 1988. This possible third operation would have continued the focus on the elimination of NTBs and the reform of import tariffs. Related issues to complement these reforms were identified as the progressive freeing of domestic prices from controls, the removal of export controls, duties and of distortions created by special subsector programs, as well as by ORPs, and the liberal- ization of government procurement policies. The Bank decided to address these issues in the context of other sector adjustment loans under prepara- tion at the time. C. Conditionality of the Loan Operations 32. To formulate the targets of the TPL loans in a quantitative way, a system of production weights was estimaLed for import tariff positions. A significant part of the conditionality of TPL I and II was expressed in terms of quantitative targets for the reduction in the coverage of licenses and in ORPs calculated with production weights. 33. Production weights were considered superior to the eadily avail- able import weights, because the latter would not incorporate the coverage of prohibitive restrictions whose protective effects could be enormous. The Bank concluded that the production weights system offered a reasonable compromise between the theoretically desirable and the practically possible and relied on the Mexican Government to produce quarterly statistics of progress on trade reform. Although this effort was clearly very useful in quantifying the progress of the reform, undue attention paid to interna- tional price comparisons--which apparently erroneously showed the effect of protection to be insignificant--may have been excessive, given the dubious validity of these comparisons. - 27 - 34. The advantages of the quantitative targets are apparent. They are clear and transparent and measurable both by the Bank and the Government in the same way, and hence they are unambiguous. Definition of the targets in a multi-year program allows for room to maneuver, while keeping the program on track, although it becomes progressively more difficult to achieve QR reduction per dollar of loan proceeds, given that liberalization of domes- tically produced consumer goods is more difficult than imported inputs and capital goods. This indicates that further operations on trade might aim at a more limited reduction in import controls than that achieved under the TPL II operation. D. Credibility of the Trade Reform and Calendarization 35. For successful implementation of trade liberalization, it is cru- cial to explain the goals and continuity of the reform process to the pri- vate sector and to convince them that the trade liberalization process is irreversible. An element intended to convince the private sector of the permanency of the liberalization was the Government's announcement in early 1986 of its intention to reduce the highest tariff rates of 50Z at that time to 302 in three years. 36. The TPL I operation was very important and any evaluation of it has to consider the historical context in which the operation was prepared and implemented. The initiation of the trade reform process, which the TPL I operation supported, was arduous and it required intensive and very sen- sitive negotiations with the Mexican authorities. In fact, before 1985, even a slight modification of the trade regime was virtually unthinkable. In this context, TPL I proved to be valuable in starting a very successful process of reducing nominal and effective protection in Mexico. It also paved the way for subsequent World Bank policy-based lending operations for Mexico. 37. The announcement of the trade reform program by the Government was also very helpful in sending an unambiguous signal on the future direction of commercial policies. This also committed the authorities to the reform program and contributed to strengthening the credibility of the program. It has to be noted, however, that the Bank initially met resistance from the authorities for insisting on an official announcement of the program and it took sometime before the authorities accepted this principle. It is recommended that future policy-based lending operations on trade should follow the Mexican example and require a formal government announcement on the future direction of commercial policies. 38. Four rounds of tariff reductions were scheduled, with the first one in April 1986 and subsequent ones every 10 months, so that the last one would take place in October 1988. The authorities also announced their intention to gradually abolish all ORPs for imports by the end of 1987. This was also part of a commitment under the accession to GATT, particular- ly with the subscription to the Customs Valuation Code, which does not allow ORPs as a customs valuation device. However, no calendar was estab- lished for QR reduction. - 28 - 39. The calendarization of the tariff reduction and the timely announcement of other liberalize'icn measures reinforced the perceived commitment of the Government to the reform, and gave the private sector time to prepare itself for the announced changes. However, in the context of the introduction of the macro-stabilization program, the tariff reduc- tions were accelerated in December 1987--at the same time that the third round of tariff reductions was to be carried out--with tariff reductions exceeding what had been previously announced in both rounds three and four combined. E. Issues 40. Use of Official Reference Pries. During the reform process, one main unfavorable development was the use of ORPs. The Government expanded the scope of ORPs providing additional protection to domestic industries. The net effect of tariff reductions and increases in reference prices actually increased effective protection for some domestically produced products. While the action was defended on anti-dumping grounds, this would not explain the use of ORPs for 1,171 tariff positions, covering over one-sixth of domestin production. Therefore, one of the lessons of the Mexican experience is that attention has to be paid to all forms of protec- tion in the formulation of conditionality to ensure that reductions in tariffs and QRs are not offset by other actions. 41. Temporary Increase in Effective Protection. Effective protection rose during 1985 due to the introduction of ORPs. Effective protection was further increased for final consumer goods since the first stage of the reform was targeted to liberalize imports of inputs. 42. Overperformance. The fact that the Government went beyond its initial announcements of tariff reduction in 1987 raises an issue to be discussed within the Bank. Should the Bank reward overperformance if reform goes beyond the announced schedule. This report is not the right forum to resolve this issue, but suggests that on balance overperformance compared to initially agreed trade reform targets be considered favorably. First, overperformance underscores the Government's commitment to the trade liberalization program. Second, reducing domestic inflation was the overriding goal of the macro-stabilization program in 1987, and accelerated trade reform was one of the tools used. On the negative side, however, deviations from announced targets may reduce credibility of the reform program and increase uncertainty for the private sector in undertaking new investments. Formulati.on of Conditionality 43. The conditionality under TPL II was met in the context of the actions taken by the authorities to introduce the macro-stabilization pro- gran. In this respect, it may prove to be difficult to replicate the quan- titative achievements under both operations for other countries. It is also important to note that major trade actions--both in 1985 and in 1987--took place outside the established calendar for tariff reform under TPL I and II. - 29 - 44. Although both operations included monitoring of macroeconomic performance, neither operation incorporated any quantitative macro perfor- mance targets. For future policy-based lending operations, consideration may also be given to link quantitative performance targets with macro tar- gets. Obviously, this is a difficult task and would require great skill and knowledge of country circumstances to formulate these targets. More- over, given the high frequency of external shocks of large magnitudes, macro targets could be defined in ranges. In this way, performance could also be measured in terms of a successful economic outcome--vital for the sustainability of the trade reform. 45. Regarding the conditionality on the macro-consistency, neither operation incorporated any explicit model or criteria for success. Al- though both operations were designed within a macro framework, the Region decided not to use them later and relied on IMF stabilization programs for ensuring macro consistency. This is another issue which may deserve fur- ther discussions outside this report, especially, since the World Bank Board recommended on September 13, 1988 the development of an explicit quantitative macroeconomic framework to analyze adjustment programs. IV. IMPLEMENTATION OF TRADE REFORM A. Overview 46. The trade reform consisted of two main stages. In the first, undertaken in July 1985, a number of tariff positions, mostly intermediate goods and machinery, were liberalized. Then, after some modest further progress, a second quantum leap in the liberalization process was made in November/December 1987. In the second stage, many finished consumer goods were also liberalized. B. Overall Evaluation of TPL Operations 47. When measuring the success of the TPL operations in terms of the liberalization of the import regime, there is little doubt about their success. The conditionality of both of those operations were satisfactori- ly met. Annex I presents the conditions under each Loan operation and the compliance record. From mid-1985 to the present, licensing coverage on domestic production hLs been decreased from well over 90Z to less than 23Z. The Mexican trade reform has been far-reaching. In 1982, virtually all imports were subject to non-tariff barriers. Today less than 20Z of the value of imports is subject to any form of quantitative restrictions. In fact, present QR coverage is about 10 percentage points below the TPL II target. Tariffs have also been reduced drastically from a range of 0? - 1002 with 12 different rates to a range of 10Z - 20? with only 5 rates. The maximum tariff is 10 percentage points below the TPL II target. More- over, all ORPs have been eliminated. However, the present anti-dumping system may have introduced some protectionist bias to the trade regime. - 30 - Given the short time the system has had since its inception, this may be understandable, but efforts should be made to remove the protectionist elements from the system. 48. It may be too early to judge the economic performance, especially since economic activity stagnated during the period of TPL I and II imple- mentation. On average, output growth averaged about zero percent over the past seven years, while real per capita income and real wages fell sharply. Adverse external shocks mentioned in paragraph 8 and large resource trans- fers abroad also contributed to this unfavorable outcome. The key remain- ing question, whether and how quickly the economy will respond to the changes in the trade regime, will largely depend on keeping domestic infla- tion under control as well as on the credibility of the reforms and the ease with which resources can flow in the economy from contracting to expanding sectors. 49. TPL I. Under the TPL I program, implementation was good. The Government lowered the maximum tariff rate from 100% to 50%, announced a three-year plan to narrow the range of import tariff rates from 0% - 100? in 1986 to 0% - 30% by October 1988, and committed itself to eliminate all ORPs by the end of 1987 (in combination with the introduction of a modern anti-dumping system compatible with GATT procedures), and to further reduce QR coverage by the equivalent of 5? of domestic production by no later than the end of 1986. 50. TPL II. Implementation under TPL II was also good. As part of its commitments, the Government lowered the maximum tariff rate from 45? to 40? in February 1987, and during the succeeding six-month period reduced tariffs on some 4,900 tariff positions. Tariffs were raised in August 1987 on some items freed from NTBs, but they stayed within the 0% - 40? range. In late 1987, the Government exceeded its commitment, and accelerated its program sharply by halving the maximum tariff rate to 20Z. The authorities further reduced QR coverage from 49.8? of 1983 domestic production to 43.1?. Table I shows some quantitative indicators of liberalization during the trade reform program: Table I: PROGRESS UNDER TRADE LIBERALIZATION PROGRAM, 1985-1989 June 85 Dec 85 June 86 Dec 88 June 87 Dec 87 May98 March 89 4/ Import Licensing 1/ 92.2 47.1 48.9 89.8 35.8 26.4 28.2 22.8 Reference Prices j/ 18.7 26.4 19.8 18.7 13.4 0.5 0.0 0.0 Tariffs-Maximum 100.0 100.0 45.0 45.0 40.0 20.0 20.0 20.0 -Average V 28.6 28.5 24.0 24.6 22.7 11.8 11.0 12.6 Real Exchange Rate 8/ 100.0 127.1 148.6 157.4 164.8 147.2 124.8 122.4 S/ I/ Percentage coverage of production of tradeables, using 1986 weights. 2/ Weighted by production of tradeables in 1988; excludes 5% surcharge. 8/ Trade-weighted foreign prices relative to the domestic price level. 4/ Bank Staff estimates. 5/ As of December 1988. - 31 - 51. The following paragraphs show, in detail, progress towards program goals. 52. NTBs on Imports. As of August 1988, less than one quarter of domestic production of tradeables was protected by quantitative restric- tions on competing imports based on 1986 weights. Three hundred and twenty-five items (under the newly introduced Harmonized Tariff System) are still controllc4, accounting for less than 32 of all items and about 162 of imports (1987 values). 53. Table II shows the situation as of March 1989 as well as TPL I and II target positions on the basis of both 1983 and 1986 weights. The original targets were expressed in terms of 1983 production weights. The reweighting is approximate. Table II: QR COVERAGE 1/ (2 of domestic production) Situation TPL II targets TPL I targets March 1989 21 end-1987 end-1987 1986 1983 1986 1983 1983 weights weights weights weights weights All tradeables 22.3 34.0 33.0 43.1 49.8 3/ Oil & Der. 5.5 20.5 5.5 20.5 20.5 Other 16.8 13.5 27.5 22.6 29.3 1/ Coverage of domestic production of competing tradeables: 1983 or 1986 weights. 2/ Preliminary estimates. 3/ Plus 11.3% of coverage of ORPs, resulting in a total net NTB coverage of 61.12 of tradeables. 54. QR coverage was reduced from the TPL II target of 43.12 for end- 1987 to roughly 342 in March 1989, using 1983 weights. Based on the 1986 weights, the comparable reduction is from about 33Z to 22.3Z in March 1989. According to the TPL II President's Report, QR coverage was expected to range between 252 and 302 of tradeables production (1983 weights) by end- 1988. Current coverage at 342 (1983 weights) remains above this expecta- tion, largely due to the existence of remaining QRs on agroindustrial products. 55. The remaining QRs are applied to various sectors. These include petroleum and petroleum refining (5.52), agricultural sector (7.7%), and agroindustry, including grains, oil seeds, tobacco, milk products, eggs, sugar, and coffee (6.1Z), as well as industrial sectors with sectoral pro- grams (3.3Z). The latter group includes autos, auto parts, pharmaceuti- cals, textiles, electric and electronic equipment. Through its operations, both planned and ongoing, the Bank is supporting the Government to phase out some of these QRs in agriculture and subsector progi..,s. The following table presents the remaining QRs by major subsectors. - 32 - TABLE III: REMAINING qRs BY SECTOR, END MARCH 1989 Sectoral Contribution to Sector Coverage Production Coverage Petroleum 100.00 8.69 Derivatives 37.20 1.54 Sub-total Oil 5.53 Agriculture 68.00 6.18 Livestock 17.50 1.09 Fishery 68.80 0.49 Sub-total Aariculture 7 78 Meat and Dairy 28.50 1.64 Coffee processing 99.70 1.66 Sugar proee*ing 90.00 0.97 Vegetables olls and fate 67.00 0.74 Tobacco products 100.00 0.81 Other 0.81 Sub-total Aar.i3ndustry 6.18 Autos * Autoparts 62.10 8.16 Pharmaceuticals 12.00 0.14 Sub-total Sectoral Programs 8.80 Other 0.50 All Tradeables Goods: 28.20 Source: SECOFI, Bank Staff Estimates. Tariffs 56. With tbe halving of almost all non-zero tariff rates in late 1987 and the elimination of the 5% temporary import surcharge, the tariff range was reduced to 0% - 20Z. The tariff system shows the typical escalation by stage of processing; most raw materials and intermediate goods have rates of 0, 5%, and 10?, most capital goods 5?, 10 or 15? and most consumer goods 15? or 202. 57. A major progress was achieved in tariff reduction in 1987. The second round of the calendarized scheme was implemented in March 1987, which reduced the maximum tariff rate from 45% to 40? and the average tariff rate from 24.1? to 20.9?. The third round was scheduled for - 33 - December 1987. However, a much more ambitious reduction was carried out instead, going further than rounds 3 and 4 combined, with the average tariff rate decreasing from 20.9Z in March to 10% in December 1987. 58. Official Reference Prices. From 1985 to 1988, the official refer- ence price system was phased out, so that the ORP coverage of 21.5% of domestic production in December 1985 was reduced to zero in mid-1988. The largest changes in ORPs took place in the second semester of 1987 when coverage was reduced from 11.1% to 0.5? of production. 59. Part of the reason for the vigorous pursuit of liberalization was to support and assist measures to reduce domestic inflation in order to encourage lower priced imports which would help moderate domestic infla- tion. During 1988 and early 1989, the policies introduced in December 1987 were followed and selective tariff reductions were implemented on products whose domestic prices kept increasing such as beef, paper and paper prod- ucts and glass products. Also, the elimination of import licensing for apparel in April 1988 was part of the inflation control measures introduced under the macro-stabilization program. Change of the Minimum Import Tariff to 10? 60. The authorities raised the minimum import tariff rate from 0? and 5% to 10? in early 1989 to reduce dispersion in nominal and in effective protection across importables for a significant part of commodities. This adjustment not only helped reduce the dispersion in effective protection, but also helped to generate tariff revenues needed to decrease the public sector borrowing requirements. On the other hand, the increase has raised effective protection for input and capital producing industries and since the increase was not publicly announced earlier, it has tended to reduce the credibility of the trade reform program. Moreover, this increase could raise the cost of production at a time when priorities are clearly pointing towards a reduction in domestic inflation. C. Export Promotion 61. The TPL operations did not specifically address administrative reforms for export promotion. The preparation of the EDP I operation led to the policy dialogue on the TPL I operation. The EDP II operation was designed to introduce administrative reforms to complement the TPL I opera- tion in the reform of the trade regime. Since a reduction in import barri- ers is a key to promoting export growth and administrative improvements alone can not guarantee export growtb, TPL operations did not burden the programs with administrative reforms in export promotion. Measures were introduced to establish a "free trade" system for exporters by eliminating controls on temporary imports, giving exporters access to foreign exchange, and providing in-bond facilities. The authorities also made efforts to simplify administrative procedures for exporters. Other measures included the elimination of export taxes which had been imposed on certain largely agricultural products, and a reduction in the number of products subject to export licenses. Domestic letters of credit now permit exporters and their - 34 - suppliers to obtain access to export financing and foreign exchange needed for imports. Under an expanded temporary import admission scheme, export- ers signing multi-year agreements with the Government are also allowed to import equipment and intermediate inputs with reduced guarantee require- ments and rebates of duties and indirect taxes. Various export financing credit lines were consolidated, and export insurance and guarantee schemes simplified (for a detailed review of the experience under EDP I and EDP II operations, see the combined PCR on these operations). Administrative steps were also simplified, and some export taxes--notably in the mining sector--removed. Addressing the issues in the administrative framework for exporting remains one of the main areas for further improvement to fully benefit from the trade liberalization program. D. Main Conclusions on Trade Reform to Date 62. By the standards of most developing countries, both nominal and effective rates of protection rates are now low in Mexico. At the end of March 1989, QR coverage (other than on oil and derivatives) equalled 16.8Z of 1986 domestic production of tradeables based on 1986 weights. TPL III operation did not materialize mainly since the authorities asked for wide- ranging reforms to be supported by three World Bank-financed sector adjust- ment operations for the financial and industrial sectors and for public enterprise reform, which were made in June 1989. The next stages of the reform should aim at consolidating the gains achieved thus far and deepen- ing the reform of domestic regulations and institutions with the aim of removing or lowering domestic barriers to mob!.lity. V. MAJOR ISSUES 63. This section is devoted to a discussion of major issues identified in the design and implementation of the trade reform program supported by TPL I and II operations, including: the anti-dumping system, coordination between trade policy reforms with accession to GATT, the sequencing of the reform, the relationship between stabilization and liberalization, the desired speed of trade reform, the impact of the level and fluctuations in the real exchange rate on the sustainability of the trade reform, synchro- nization of trade reform with the macro-adjustment program, and the influ- ence of domestic regulations. This report also points out some issues for which no answers are provided, including the different pace of liberaliza- tion in industry versus agriculture (and services), the specific relation- ship of the trade policy loans to other loans in the lending program and formulation of conditionality for trade policy loans. A. Anti-Dumping System 64. The Anti-Dumping Law and regulations were introduced in 1985 and in 1986, respectively. In 1988, Mexico signed the GATT Anti-Dumping Code. While the experience with the new system is limited and SECOFI (the Minis- try of Commerce and Industrial Development) under the present Government - 35 - has taken a more liberal approach compared to the stance under the previous administration, there is some indication of protectionist tendencies incon- sistent with the trade opening. First, the number of anti-dumping cases initiated is increasing more rapidly than appears consistent with trade liberalization. Second, the "injury test" does not play the dominant role in evaluating anti-dumping cases as it should. Third, the protective mea- sures imposed under anti-dumping procedures generally involve imposition of high duties which have correspondingly high welfare costs for the economy. Fourth, in 1988, the first "ex-officio" case was initiated, covering 12 countries and a wide range of products. And lastly, the anti-dumping cases have offered the potential for collusive action by oligopolistic firms. Overall, current anti-dumping procedures have revealed some protectionist tendencies, leading to the notion among some enterprises that the anti- dumping system could replace, to a significant extent, the protection lost by the removal of QRs and the reduction of tariffs. 65. In the recent past, most of the accusing firms happened to be subsidiaries of transnational or public companies with a complete or virtu- ally complete monopoly position in the domestic market. This suggests some attempts to abuse the system. 66. Without careful design and checks, administrative steps could easily become a form of protection to replace previous trade barriers. In practice, though, there is considerable leeway to use anti-dumping proce- dures as a protective device, particularly since legal procedures are extremely complex. The implementation of anti-dumping procedures on sound principles, therefore, depends very much on the neutrality of the Tariff Commission in balancing the interests of producers and users of imported products. 67. Part of the problem has been that the trade opening is a new phe- nomenon, and the anti-dumping cases reflect a teething problem for Mexico. Industrial countries have had open and transparent trade policies over 20 to 25 years, creating a built-in stabilizer. Hence, it should be expected that Mexico will only resolve these issues over time. 68. This report suggests that the authorities continue to monitor and review the experience of the anti-dumping system with regard to consistency with the trade liberalization process, transparency of procedures, and the costs and benefits provided. In particular, it is recommended that the authorities: - staff a fully independent Ta:iff Commission with strong consumer representation; the Tariff Commission should pro- vide an annual report on the costs and benefits of anti- dumping actions on the domestic economy; - strengthen the role of the "material-injury" test; and - introduce a "sunset" clause for anti-dumping measures, with no reopening of cases for three years. - 36 - B. Coordination of Trade Policy Reform with Accession to GATT 69. Mexico's accession to GATT became effective in August 1986, en- abling Mexico to participate as a full GATT member in the opening session of the Uruguay Round in September 1986. The accession incorporated a pack- age of reforms, covering tariff reductions, the removal of import prohibi- tions, and the relaxation of quotas on 373 tariff positions, comprising 15.9Z of imports (1985 values). Many negotiated tariff rates were higher than the rates in force, leaving the Government some leeway to increase tariffs. By July 1987, Mexico had already signed the GATT's codes dealing with anti-dumping, customs valuation and technical norms. 70. Mexico's accession to GATT did not imply a deepening of its import liberalization program which was already well under way when Mexico's membership became a fact. Also, some Mexican officials apparently felt that negotiations with the Bank on Trade Policy Loans reduced the country's bargaining power with GATT, since the concessions granted by Mexico with its joining GATT were already incorporated in its trade liberalization program. In fact, the only real concession to GATT could be the elimina- tion of the official reference prices for imports in a period of one and a half years. Mexico's accession to GATT was important. It underscored the seriousness of the Government's intention to carry out the trade liberali- zation further, and thus increased the credibility of the reform program. 71. GATT membership provides various benefits to Mexico. First, bi- lateral negotiations with the United States are partially replaced by the. application of multilateral rules. Second, in cooperation with other developing countries, Mexico can now influence the development of new rules in GATT. In so doing, Mexico could join other developing countries that have become active in GATT affairs. Third, under GATT rules, countervail- ing duties cannot be imposed unless injury to domestic industry is proven. C. Sequencing of Reforms 72. The timing of the trade reform was well-conceived, with restric- tions on exports being reduced in advance of the import side, thereby avoiding the possible short-term adverse balance of trade consequences that may occur if imports respond more quickly than exports to trade liberaliza- tion. Moreover, a sharp real devaluation during the implementation of the trade reform was a key factor in encouraging rapid export growth. 73. The process was carried out in stages to provide opportunities for adequate adjustment. First, tariffs replaced non-tariff protection to provide transparency, and then this transparent protection was monitored and later reduced. 74. The elimination of import licensing began with intermediate inputs and capital goods, increasing effective protection for final consumer goods. Luxury consumer goods were left to the final stage, with their importation being liberalized as the authorities increased tax rates on the sale of luxury comodities. - 37 - D. Stabilization and Liberalization 75. Some of the lessons from the Mexican experience of the 1980s are of general relevance to development policy, while others are of primary interest in the context of the Bank's relations with borrowers during a trade reform process. Conventional wisdom argues that when inflation rates are relatively high, priority needs to be given to stabilization, with structural adjustment policies (i.e. including trade reform) postponed until the economy has stabilized. 76. Mexico's experience underscored the difficulty of stabilizing an economy before a highly distorted incentive structure is improved. The ongoing mRcro-stabilization effort, begun in late 1987, has been instrumen- tal in reducing domestic inflation, and this effort has been aided by reductions in the distortion of incentives brought about by the trade poli- cy reform. In this respect, the Mexican experience has cast doubt on the general validity of the conventional wisdom. However, the final conclusion on this issue should be drawn only after domestic inflation has been perma- nently reduced. E. Speed of the Reform 77. There is no benchmark or guideline on the appropriate time frame for undertaking a trade liberalization program. This depends on the ini- tial conditions, and accompanying macro measures, such as the implementa- tion of a stabilization program. 78. During the reform process, there was a great deal of debate within the Mexican Government on the proper speed of reform. SECOFI officials argued for a more gradualist approach, whereas the Ministry of Finance and Public Credit (SHCP) and Ministry of Programming and Budgeting (SPP) was in favor of a shock approach. 79. There are a number of arguments in favor of a smooth path, there are other arguments in favor of a shock approach. Given that the restruc- turing of the economy is a long-term process, a period of 3 to 5 years to carry out the reforms could be considered reasonable. Moreover, the grad- ual approach to import liberalization tends to reduce disruption of the productive sectors, allowing time for the troubled industries to adjust to the new economic conditions. In this context, it is important to announce the trade liberalization measures in advance so that firms can prepare for it. From a political point of view, however, it may be desirable to under- take the trade reform quickly, reducing the possibility that interest groups can get organized and form a strong lobby against reform. In fact, some Mexican policy makers who introduced the reform argued that if the reform had been slower, the initial force of the reform would have been lost and it would likely have been reversed by the political pressures likely to be lodged by entrepreneurs. In the final analysis, the speed was determined by the desire to conclude the core of the reform process before the end of the Miguel de la Madrid Administration and by the short-term macroeconomic objective of inflation control begun in late 1987. - 38 - F. The Level and Changes in the Real Exchange Rate (REER) 80. The exchange rate policy contributed to the success of the trAde reform as real devaluations encouraged a sharp increase in non-petroleum exports and discouraged import growth during the implementation of the tariff reform. However, drastic fluctuations in the real exchange rate increased uncertainty and adversely affeLted long-term planning (Graph II). When the first stage of the trade reform was introduced in 1985, the nomi- nal exchange rate was devalued by 222 -- a compensatory real devaluation to avoid a surge in imports. The dramatic changes in Mexico's terms of trade and massive capital flows led to the large swings in the real exchange rate. At the same time, the increased public sector deficit (in 1980-82, and again in 1986) and the consequent deterioration in the net resource balance have added to the pressures on the real exchange rate. 81. After a major devaluation, the authorities froze the nominal exchange rate in December 1987 to provide a nominal anchor for inflation control as part of a macro-stabilization program, including significant fiscal adjustment. Since early 1987, the real exchange rate has appreci- ated and the export growth rate has declined. To ensure the success of the trade reform, exchange rate policy has to be consistent with macro economic fundamentals. Moreover, temporary and unexpected fluctuations in the terms of trade for Mexico underscore the critical importance of anti-cyclical macro-management. Furthermore, a widely fluctuating real exchange rate is not conducive for improving resource allocation since it tends to delay investment decisions. G. Synchronization of Trade Reform with the Macro-Adjustment Program 82. To control domestic inflation, the Government introduced a macro- stabilization program at the end of 1987, initially for a three month period, and it has extended the program since then at three or six month intervals (para. 26). The program was based on: (i) raising the primary surplus of the public sector; (ii) freezing the exchange rate, wages, and public utility prices as nominal anchors; (iii) further liberalizing trade; and (iv) pursuing a vigorous process of concertation with the various sectors of the economy to control domestic inflationary pressures. 83. The Government implemented demand-management policies to reduce domestic absorption with supportive actions aimed at breaking the inertial component of domestic inflation through the macro-stabilization program. The authorities came to the belief that any stabilization program with a hope of averting major output losses and unwanted income redistribution needed to deal directly with the inertial component of inflation (e.g. by adjusting major macroeconomic variables in accordance with a forward-look- ing schedule within the framework of a freeze). Graph II: Real Exchange Rate (Monthly Index) 170- 150- 130- 110 90- 70 - Jan-80 Jan-81 Jan-82 Jan-83 Jan-84 Jan-85 Jan-86 Jan-87 Jan-88 Real Exchange Rate - 40 - Progress on Inflation Reduction 84. Domestic inflation has to be brought down to low levels on a last- ing basis to make trade reform stick. In a highly inflationary environ- ment, relative prices among sectors are so variable that they will give uncertain signals about where investment should occur. Inflation adds "noise" to relative price signals, and could alter patterns of shifts fol- lowing changes in trade policy. Economic agents may be confused on what sectors will gain or lose from trade reform. Therefore, the Government aimed at reducing high and variable rates of domestic inflation through the implementation of the macro-stabilization program since it considers the success of the stabilization effort a prerequisite for a lasting recovery. The macro-stabilization program has been a success so far, resulting in an increase in the primary fiscal surplus from 5Z of GDP in 1987 to 5.8% in 1988 and a projected 7.8? in 1989 and a dramatic reduction in domestic inflation from 160% in 1987 to 52Z in 1988 and to a projected rate of only 20% in 1989. In spite of the high primary surplus, however, continued reductions in the public sector borrowing requirement is complicated by high real domestic interest rates, reflecting expectations that the under- lying level of domestic inflation may not yet have been abetted. There- fore, further fiscal adjustment would help consolidate the gains achieved thus far and would help maintain price stability when the price controls under the macro-stabilization program are eventually phased out. H. Domestic Regulations 85. In addition to international trade controls, the Mexican economy has been subject to various domestic regulatory controls. These include restrictions on direct foreign investment, public procurement policies, subsector programs, controls on the agriculture sector, domestic price controls, and factor market rigidities. 86. Domestic regulations have not been adequately addressed during the implementation of the trade reform. Hence, the possibility exists for discretionary use of these instruments in an attempt to reverse the trade liberalization process, or more likely, to reduce the impact of the trade liberalization on domestic production and investment decisions, should things go wrong. This risk includes possible use of instruments to limit import competition, including the use of the anti-dumping regulations to protect domestic industries, the use of administrative controls (health regulations, safety and technical standards) and the discretionary use of public procurement policies. Moreover, domestic regulations significantly reduce the benefits which could be realized from the trade reform. Recent- ly, the authorities began to address some of those issues in a concerted way. The Bank is supporting the Government in addressing important regula- tory framework issues through an IndustrIal Sector Policy Loan (ISPL), approved by the World Bank Board in June 1989. - 41 - Direct Foreign Investment 87. TPL operations did not address issues related to the regulatory framework on direct foreign investment. Rules for direct foreign invest- ment (DFI) required a case-by-case approval of investments with greater than 49Z foreign participation. Moreover, investment in a number of impor- tant sectors was restricted to public, Mexican and majority Mexican owner- ship. In addition, the withholding tax on dividends exceeded the rates credited abroad which also discouraged DFI. The Mission's interviews with some foreign-owned companies indicated that the following factors acted as deterrents to DFI: (i) the complexity of the foreign investment law; (ii) the discretionality shown in the implementation of the regulations; (iii) majority ownership requirementG; (iv) domestic content and other perfor- mance requirements (i.e. balance of payments effect, employment, minimum capital investment); (v) the existence of price controls; and (vi) an uncertain outlook for business conditions. 88. Since 1982, and especially in recent years, the authorities have introduced a number of steps intended to mitigate to some extent the prob- lems caused by the strict rules governing foreign investment and relax the rules and administrative procedures. They clarified the basic legislation regulating foreign investment, identifying those sectors where the share of foreign ownership in local companies is unrestricted, partially restricted, or prohibited. They simplified the approval proce4a and expanded the num- ber of sectors where foreign ownership is permitted, which now includes nine major sectors, and virtually any business that promises significant exports. 89. Considering the importance of DFI in the financing of a recovery of investment in Mexico, the authorities introduced new rules in early 1989 to introduce flexibility to the exist!ig DFI law by reducing discretionality and improving transparency. The measures included: (i) elimination of deductibility of dividends and of the withholding tax on remittances to foreign companies beginning in 1989; (ii) publication of the regulations for the DFI Law containing the following measures: (a) allowing automatic 10O% DFI participation in activities comprising 60Z of GDP; (b) allowing increased participation of the private sector (domestic and for- eign) in areas subject to specific regimes; (c) increasing the transparency and speed of CNIE (National Foreign Investment Commission) decision-making; and (d) simplifying and deregulating registration requirements, establish- ing automatic approval of registries in the CNIE. Additionally, the Gov- ernment plans to implement a comprehensive DFI promotion program to posi- tion Mexico as a country with significant comparative advantages and will- ingness to compete in the international market for capital. In the medium- term, there is a need for modifying the foreign investment law to further liberalize the framework for foreign investments. Public Procurement 90. While the Government seeks to purchase goods and services on a competitive basis, there has been a tendency to help local contractors. - 42 - The importance of these rules is seen in the large share of tvtal purchases (30%) and imports (25-332) accounted for by the public sector. These pur- chases created a strong anti-import bias and provided additional protection in industries producing for the public sector. Although the Bank did not include liberalization of public procurement in TPL operations in order not to burden the loan packages, this issue has been addressed in the Public Enterpiise Reform Loan (PERL), approved in June 1989, which includes as a second tranche condition liberalizing the existing regulations for public procurement. Subsector Programs 91. The subsectoral programs were developed in the early 1980s primar- ily to ease balance of payment problems and develop cars, autoparts, phar- maceuticals, petrochemicals, computers, and textiles. The programs provide protection from imports, tax incentives, entry restrictions and price con- trols. In this respect, these programs run counter to the spirit of trade liberalization. While the programs appear to have met some of their tar- gets, such as helping to improve the subsectoral balance of payments (although subsequent policy reforms after the establishment of these pro- grams make it difficult to attribute these improvements to the subsectoral programs), the extensiv2 distortions created have led to efficiency losses and a reduction in competition. Domestic content requirements increase protection, while the fulfillment of export targets may substantially raise costs for the domestic user. The Government has announced that it does not intend to renew these programs beyond the dates committed under contractual agreements and it does not intend to introduce new programs in the future either. The ISPL operation provides support to phasing out these programs. Agriculture Sector Liberalization 92. TPL operations focused on liberalization of the industrial sector and had very limited impact on the liberalization of the agricultural sector. Both domestic and external trade of agricultural commodities is highly regulated by the Government. Mexico's agricultural sector is char- acterized by state truding. The public sector manages the trading of about 16 major crops (i.e. including grains, oil seeds, sugar, coffee, and tobacco). As a result, tariffs are redundant as an instrument to manage trade. Mexico is still a net importer of basic grains despite efforts to achieve self-sufficiency. Trading of imports of most agricultural products is subject to licensing. Moreover, quantitative import restrictions are extensive and official reference prices are applied to cattle exports. 93. The authorities have taken some actions to liberalize the agricul- tural sector in the past. They liberalized imports of some inputs (i.e. machinery and equipment and agrochemicals) used in the sector, as well as, exports of high quality beef. Clearly, much more needs to be done to liberalize the sector. The Government, with Bank assistance, is in the process of defining a program to liberalize the agricultural sector. - 43 - 94. The Agricultural Sector Loan (AGSAL, Loan No. 2918-ME) did not have any cross-conditionality with the TPLs. However, both operations included in their conditionality the reduction of NTBs; while the AGSAL specified the goods for liberalization, the TPLs were general on the goods to be liberalized. Conditionality under both types of operations rein- forced each other and actions taken under agriculture were credited to both operations. I. New Policy Agenda 95. With Bank support, the Government defined a new policy agenda to complete the trade reform and to define an industrial policy that would encourage growth based on an attractive investment environment and competi- tive goods and factor markets. The authorities are introducing reforms to deregulate the industrial sector. The objective is to achieve an industri- al structure based on competitive markets which is responsive to the need for technological change and modernization. The program involves the phas- ing out of the sectoral approach to industrial development, the easing of price restrictions, bureaucratic procedures, barriers to entry, increasing factor mobility (especially direct foreign investment, technology, and to some extent labor), and improving key services to industry. Deregulation measures would allow the economy to benefit from the trade reform by allow- ing for a clear transmission of international price movements into the domestic economy as well as to encourage investments in competitive trade- able sectors. 96. The Bank is supporting the Government through the ISPL to intro- duce the above-noted deregulation measures. The ISPL will help implement initial and in many cases fundamental changes. It also includes studies to identify further policy or legislative changes needed to improve the regu- latory framework. Deregulation measures supported by this Loan operation follow up on other measures already introduced by the authorities. How- ever, as pointed out earlier, given the extent of domestic regulations, much more needs to be accomplished to deregulate the economy. J. Need for Remaining Reforms 97. Despite the introduction of deregulation measures and other mea- sures which are planned to be introduced, the regul.ory framework might continue to obstruct an efficient restructuring of the economy. Sectoral liberalization and deepening in the domestic deregulation effort should be the primary focus of further liberalization efforts, which could be sup- ported by subsequent World Bank lending operations. 98. First, the reform process has so far by-passed agricultural imports to a large extent, with virtually all basic food imports still under direct and indirect government control. Second, it is not clear that the anti-dumping system established in Mexico is consistent with GATT rules. Third, non-tariff barriers on certain health regulations may be used as non-transparent protective devices. Fourth, export restrictions (export taxes, export controls) affecting mostly agricultural goods were - 44 - relaxed somewhat, but the extent of this liberalization process may have been less dramatic compared to industrial goods. Fifth, the cost of doing business in Mexico can still be considered high, despite recent government measures. 99. Another area which deserves attention is the administrative frame- work for trade, including customs administration. The liberalization mea- sures have not yet been matched in efforts to improve the administrative framework. For example, the Mission encountered complaints from the pri- vate sector regarding difficulties and "hidden charges" at customs, which act to frustrate trade liberalization. The planned Export Adjustment Loan aims at improving the administration of customs and address other remaining trade policy reform issues. Vi. IMPACT OF TRADE REFORM A. Introduction 100. The trade opening process has been significant since 1985. The sharp depreciation of the currency helped the trade opening process by contributing to the export growth and a slowdown in import growth. With an appreciating real exchange rate and further trade liberalization since the end of 1987, domestic firms started to face stiff import competition in late 1988 and in 1989. The private sector which has been used to operating under high protection, and with cheap credit and fiscal subsidies, now faces the disciplining effect of the trade opening together with the recently announced financial liberalization measures and the significantly scaled-down fiscal and financial subsidies. These measures have led the private sector to introduce efficiency measures aimed at becoming interna- tionally competitive. However, given the lags in resource flows from con- tracting to expanding sectors, the lasting impact of the reform will likely take time. 101. Judging the differential impact of the trade opening on the econo- my is difficult, given that price incentives have altered the prices of importables to a significant degree. The introduction of the macro-stabi- lization package together with the recently announced financial liberaliza- tion measures (i.e. freeing of interest rates) and the expectations con- cerning external debt negotiations with external creditors has had an impact on resource aIlocation decisions during the trade reform process. Hence, the conclusions re,ched on the outcome of the trade reform should be regarded as tentative at best. Given these caveats, some very preliminary conclusions emerged based on mission interviews with the private sector in Mexico. First, some reconversion began to take place. The maquila sector has continued to expand. Prices of some tradeable products fell in real terms, reflecting the impact of the macro-stabilization program and import liberalization. Second, P group of private entrepreneurs with a vested interest in the t 'rde op!c-oing began to emorge and became vocal in domestic politics. Third. de,: i.s 'hi fnct that mjor changes were introduced under - 45 - the trade reform and that protection was granted over a forty year period, no major bankruptcies occurred. This has helped sustain the trade reform. 102. The impact of the Mexican trade reform is described in great detail in the green cover Trade Impact Study, which was prepared and made available to the Mexican authorities in 1988. As explained in that report, a serious review of the impact of the trade reform could only be carried out after a sufficient lapse of time after the introduction of major changes that took place in 1987. This project completion report discusses developments in exports and imports and presents preliminary conclusions on the impact of the trade reform based on a limited enterprise survey con- ducted by the Mission in the field. 103. The Mission's enterprise survey indicated that the beneficial effects of trade reform should occur over time as it will take some time before resources move into the internationally competitive tradeable sectors. Clearly, the rate at which the resources move would be acceler- ated within the context of stable macro framework and the faster adoption of domestic deregulation measures. B. Exports and Imports 104. This section describes the behavior of exports and imports since 1980. This report does not attempt to isolate the effects of the changes in the trade regime on the performance of exports and imports, but suggests that this analysis be carried out. 105. Manufactured exports are concentrated in a limited number of firms and products. In fact, exports of cars and autoparts make up the largest single manufactured export group. Most firms are not yet export-oriented. Even those that export sell a large share of their output in the domestic market. Particularly, small scale industry is not generally oriented towards the export market. 106. Manufactured exports now account for more than half of total exports, while oil represents less than one-third compared to three-fourths in 1982. Following the introduction of the trade reform, some sectors and industries have been modernized, and the Mexican economy has become more competitive and export-oriented. C. Evolution of Total Exports 107. Total exports reached a peak in 1984 at approximately US$24 billion and oil exports played a prominent role in this growth (Graph III). The share of oil in total exports fell from almost three-fifths in 1982 to about about one-third in 1988 because of a sharp decline in oil export revenues--in particular since 1986--as well as the surge in manufactured exports (Graphs IV and V). The increase in manufactured exports has been particularly marked since 1986, when they grew by 30% on two occasions. Agricultural exports, however, have lost their dynamism, with their share in the total falling to 8% in recent years, except in 1984 and 1986 (Graphs VI and VII). Graph IV: Structure of Exports % Point Change 1979-1988 Graph Ill: Evolution of Exports Billions of US Dollars -20 -10 0 10 20 30 197e o ae 20 -mos m uin.n.. ClN.-.u m.e..~ mau...s..,s Fitpree Oar ms arO booo on J"m-Nov. 158 101 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 20- Total Exports E Oil E Agriculture j Manufacturing 1 Figures for 1988 are for Jan-Nov. .Pbt Exot MPriva8314te E1988 =Public Exports M Private Exportz Fiqie ow 1964 moB tow ian-Now. 4 7 108. The weakness of the domestic economy in recent years has been partly offset by a strong export performance. The increase in non-oil exports was particularly notable during 1987, reaching a level of US$12.0 billion, compared to US$9.7 and US$7.1 billion in 1985 and 1986, respec- tively. Oil exports also rose to US$8.6 billion during 1987, but they remained substantially below the level of non-oil exports. Manufactured exports, in particular, continued to expand sharply, increasing by 352 and 392 in nominal terms during 1986 and 1987, respectively. 109. During 1988, however, export growth slowed down as the nominal exchange rate was frozen under the macro-stabilization program. The level- ing off of export growth was generalized among sectors. Growth of manufac- turing exports declined from 39% in 1987 to 17% in 1988; of food, beverages and tobacco exports, from 40% to 4%; of textiles and clothing exports, from 702 to 11%; and of metallic products, machinery and equipment exports, from 412 to 15%. D. Evolution of Manufactured Exports 110. Manufactured exports expanded from about US$3.5 billion in 1980 to over US$12 billion in 1988. The sharpest increases took place in metal products, machinery and equipment manufacturing, accounting for over two- fifths of manufactured exports in 1988. Transport equipment (which includes the automobile and autoparts industries) has been the most dynamic subsector (i.e. accounting for close to one-third of total manufactured exports). Other large export subsectors include chemical products and processed food, which made up 22% of manufactured exports in 1988. How- ever, exports of oil products and petrochemicals displayed a declining trend, falling from 21.7% of manufactured exports in 1985 to only 6.7% in 1988. E. Evolution of Total Imports 111. Total imports peaked in 1981 at the end of the oil boom at about US$24 billion, but contracted sharply in 1983, 1984 and 1986 when they amounted to US$8.5, US$11.2 and US$11.4 billion, respectively (Graph VIII). Private sector imports accounted for the largest share in imports except for 1983, fluctuating sharply over this period. During Jan-Nov. 1988, they accounted for about four-fifths of the total. Imports of intermediate goods have been the single largest category since 1980, accounting for over three-fifths of total imports. On the other hand, imports of consumption goods, which rose sharply during the implementation of the macro-stabiliza- tion program--drawing undue attention from some critics of the trade open- ing--are still just 102 of the total. Finally, an issue that produced mixed attitudes in official circles is the sharp surge in imports during 1988 (an increase of almost US$6.7 billion compared to 1987). While there may be some concern about the speed of the reduction in the trade surplus, there is also the conviction that imports helped to support the macro- stabilization program, preventing the price freeze from producing supply shortages and stimulating price competition. It is not yet clear if the Graph VI: Manufacturing Exports Graph VII: Evolution of Mfg Exports Billions of US Dollars Billions of US Dollas 1212 -- _____ _-- 1010 8 Food * Oil Derivtive 8 4 £J Chemicals F Iron A Steel 6 2 Metal. Mach. & Equip I e Transport Equip 1979 1980 1981 1982 1988 1984 1985 1986 1987 1988 TotaleI 2 -Total Mfg Exports Public Mfg Exports J CPrivate Mfg Exports0 1984 1985 1988 1987 1988 Figures for 1988 are Jan-Nov. Figures for 1988 are for Jan-Nov. - 49 - increase was temporary due to expectations of reversal of the trade open- ing, uncertainty about the policies of the new Administration, the appreci- ation of the real exchange rate, or permanent because of the changes in the trade regime. However, a detailed review of the evidence is called for in order to make a more definitive judgement on this point. F. Evolution of Manufactured Imports 112. Manufactured imports grew sharply in 1980-88. The largest abso- lute increase was recorded in metal products, machinery and equipment, followed by food products. Imports of machinery and equipment accounted for about half of the total imports every year during the period. 113. In 1988, import growth was highest in livestock activities, fol- lowed by food, beverages and tobacco, textiles and clothing, and railroad machinery and equipment. Increased livestock imports resulted from insuf- ficient domestic supply, which in turn, was caused in part by the implemen- tation of price controls introduced under the macro-stabilization program. Imports of food, textiles and electronics responded largely to expanded consumer expenditures. Railroad equipment imports resulted from the accelerated investment program (Graph IX). G. Trade Ratios 114. Since the introduction of the first phase of the trade reform in 1983, trade ratios have showed a clear tendency to increase. The export- to-GDP ratio, for example, rose from 15.4Z in 1985 to 17.3Z in 1986, and expanded even further to 19.6% in 1987--the highest level ever, falling to 16.52 in 1988. The import-to-GDP ratio also rose from 10.5Z in 1985 to 12.2? in 1987 and to 14.2Z in 1988. H. Private Sector Perception of the Trade Reform 115. This section is based on mission interviews with the private sector both in Mexico City and in Monterrey. The Mission interviewed entrepreneurs and managers in twenty-five enterprises. No prepared format was used and interviews emphasized judgmental conclusions by entrepreneurs and high level managers. Moreover, interviews included a broad spectrum of firms, including large holding companies and small firms, as well as, foreign-owned companies. 116. The above-mentioned limited enterprise survey indicates that the private sector considers the trade reform irreversible. Most firms recog- nize that the traditional way of doing business was not sustainable even before the introduction of the trade reform and some firms in fact began to introduce efficiency-increasing measures before 1985 and to invest in new plants and to reconvert the old ones for exporting. 117. Private sector representatives in their discussions with the Mission noted some areas of concern regarding the implementation of the trade reform. First, some private sector representatives found some of the Graph VIII: Structure of Total Imports Graph IX: Selected Import Groups In Billions of USS. Billions of US Dollars 25 16__ 14- 20 12 1O 101816179 a 40 179n-M-1t84 196 1988 1987 1988 1979 1980 1981 1982 1983 1984 1985 1988 1987 1988 I Food Paper E Chemicala M Consumer Goods [- intermediate Goode =Capital Goods Metal. Mach A EquliE Transport Equip --- Total Figures for 1988 are for Jan-Oct. - 51 - macro policies incongruent (i.e. such as incompatibility between the trade reform and price controls as well as maintenance of subsectoral programs). Second, in the opinion of some private sector groups, the existence of oligopolies in commercial outlets has reduced the realization of reduced tariffs into decreased domestic prices in some products. Third, almost all private sector representatives found the increase in the minimum tariff rate to 1OZ in 1989 as too high in one step. Fourth, virtually all private sector representatives expressed a feeling that Mexico's trade partners have to show reciprocity and reduce their own trade barriers to Mexican exports, given that Mexico has taken important steps to open its trade regime. Perception of the Macro Outlook 118. Most private sector representatives informed the Mission that during the reform process, erratic behavior of the real exchange rate and high real interest rates made general business conditions too uncertain for deciding on new investments. The inability of translating cost increases into higher prices under the macro-stabilization program is also reducing profit margins. One entrepreneur stated that he would be willing to incur losses on his operations in the short-term and still be prepared to make investments if he believes that the pursuance of policies under the govern- ment program would restore lasting macro stability. 119. On balance, virtually all private sector representatives expressed a feeling that there is more to be accomplished to ensure that the trade reform sticks. First, most private sector representatives expressed an opinion that the excessively low cost of labor may not be sustainable over long periods, that the cost structure in the economy is still inflationary and that the overall efficiency of the economy is low. Hence, most private sector representatives noted that production efficiency should be enhanced to benefit from trade reform. Lessons of Experience 120. Mexican entrepreneurs regard the real exchange rate as the single most important variable influencing their decisions on production and investment, and that a more predictable real exchange rate is crucial in establishing the climate and incentives for stimulating investments. The macro-stabilization program created the conditions for domestic price stability, which in turn, is a precondition to ensure a more predictable real exchange rate. However, the policy of a wage-price-exchange rate freeze is not likely to be a viable long-term strategy, and the importance of maintaining a more predictable real exchange rate as an important complement to the trade policy reforms should not be overlooked in future macroeconomic policy actions. 121. Mexican exports have traditionally been very sensitive to real exchange rate variations in the short-run. The Mexican experience clearly shows that maintaining a disequilibrium real exchange rate is not sustain- able over long periods. The lesson of experience is that maintaining the - 52 - exchange rate policy in line with macro fundamentals is crucial for reducing uncertainty and to encourage growth in investment and exports. 122. Another lesson of experience is that exchange rate devaluation and interest rate increases have to be complemented by more fundamental policy actions to prevent the scenario in which initial price increases result in a loss of confidence in the capacity of the Government to control domestic inflation. Moreover, failure to attack inflation at its roots (i.e. fur- ther reductions in the public sector borrowing requirements) can jeopardize structural reform efforts, especially trade liberalization. Impact of the Trade Reform on The Private Sector 123. The introduction of the trade reform and reduced domestic demand led the private sector to begin to introduce efficiency measures and cut costs. Most firms interviewed by the Mission indicated that they reduced personnel, made efforts to penetrate into new export markets or started exporting, and raised productivity on a limited scale. The need to be efficient gained importance. As the profit margins fell as a result of increased import competition and the price freeze under the macro-stabili- zation program, firms had to intensify the introduction of efficiency measures. Some investments for export markets also started on a limited scale. Recession-induced reductions in domestic sales led firms to orient their sales to export markets. Large firms and holding companies were better placed to introduce the adjustment measures. In terms of the impact on prices, most tradeable consumer non-durable prices fell in real terms as a result of the trade opening. Moreover, low tariffs on inputs helped reduce costs. 124. The lack of sufficient financial resources presented a major difficulty in introducing the adjustment measures. Large public sector financial needs crowded out the private sector. However, large companies used retained earnings for investment while some companies bought finance companies (casas de bolsa) to fund their financial needs. Multinational corporations used finances from their parent companies. 125. Until mid-1988, the private sector had not yet made fundamental changes. First, the margin of protection provided by the undervalued exchange rate reduced import competition at least until mid-1988. Second, mixed signals on the permanency of the trade opening until the new Govern- ment took office in late 1988 may have discouraged undertaking large investments. Third, the existence of a large degree of excess capacity and high real interest rates discouraged investment. 126. As a result of the decline in the real exchange rate and reduc- tions in QRs and tariffs, domestic producers began to face stiff foreign competition in late 1988. The private sector firms now feel the need to intensify efforts at adjustment through undertaking reconversions and intensifying efforts at increasing efficiency. - 53 - 127. Conclusions. The Mission's interviews with the private sector indicated that the private sector is still concerned with the general busi- ness conditions and short-term macro outlook and specifically with the length of time the price controls need to be maintained due to macro con- siderations. Additionally, private sector representatives believe that the existence of high real interest rates and excessive domestic regulations reduce their response to the new incentives. Furthermore, most entrepre- neurs interviewed by the Mission felt that short-term variations in trade policies would have a destabilizing effect on enterprise operations and would impose serious constraint3 on the medium-term restructuring of the economy. Enterprises indicated that if tariff levels were reduced below previously established targets without prior notice, they could also be raised again at short notice. Moreover, private sector representatives stressed the need for better and more frequent consultation as well as being better informed of what the new policies would be. Private sector representatives also underscored the need for eliminating ad-hoc measures to strengthen the credibility of the reform process. While the private sector believes that it has to accept a decline in profit margins, it also needs assurances that it would be allowed to compete with the public sector on a level playing field. To assure this, the private sector representa- tives interviewed by the Mission expressed their desire to see that new government actions continue to reduce the size of the public sector- -through cutbacks in public spending and an acceleration in the privatiza- tion program--to ensure that financial resources are available for private sector investment needs as the authorities have already agreed to undertake under the new adjustment measures, being supported by the three Bank adjustment loans mentioned earlier. I. Impact of Trade Reform on Public Sector Enterprises 128. On balance, the tariff reform has led only to limited efficiency improvements in public sector enterprises partly because of the imposition of price controls on tradeable outputs produced by large public sector enterprises such as PEMEX, SIDERMEX and FERTIMEX and partly because most of the large enterprises are in utilities or in non-tradeable sectors that have remained untouched by the trade reform. The trade opening, however, had a positive impact on a large number of small and medium-sized public sector enterprises, which began to face foreign competition. Most of the small and medium scale public sector enterprises have been or are in the process of being privatized. Although under the macro-stabilization program, public sector enterprises took a number of corrective measures to survive with significantly scaled-down government subsidies by laying off workers, modernizing plants, and cutting overhead costs, there is still scope for improving efficiency. Freezing public sector prices under the macro-stabilization program, however, has led to financial difficulties in some public sector enterprises such as in FERTIMEX and these, in turn, have reduced substantially the investible resources of these entities. As a result, financial difficulties led to delays in needed investments, increased borrowing requirements and the need for continuation of government subsidies. - 54- 129. Price controls on outputs also reduced the impact of the trade reform on public sector enterprises. For steel and fertilizers, for example, low domestic consumer prices--sustained by continued government transfers--made tariffs generally redundant. In fertilizers, the con- trolled domestic price is about 60Z of the import price, while in steel controlled prices of about 100 products are 20Z-30Z lower than interna- tional prices. This situation impedes import competition, while exerting a great financial burden on private steel manufacturers. Hence, the benefits of tariff reform in increasing import competition in those sectors can only materialize after government subsidies are phased out ana domestic prices are decontrolled. 130. Another factor which prevented public sector enterprises from fully benefiting from trade liberalization are the public procurement practices mentioned earlier. Because of discretionary public procurement practices, public sector enterprises purchase most of their inputs domestically--generally at prices higher than international prices--while the private sector is free to import inputs at international prices. Although tariff reductions on outputs created competitive pressures for both public and private sector firms, tariff reductions on inputs helped private firms using imported inputs, whereas public sector enterprises did not benefit as much from tariff reductions on inputs. This differential impact of tariff reductions on inputs for public and private sector firms underscores the need to phase out discretionary public procurement policies, as supported under the PERL. 131. The PERL approved by the World Bank Board in June 1989 forms a part of a program of Bank support to the Mexican Government's objectives of stabilizing its economy and resuming growth. It is designed to assist in the improvement of the efficiency of public enterprises (PEs). It aims at reducing the heavy burden that PEs impose on the economy and, particularly, on the public sector budget. The program inrludes: (i) a "disengagement" component to continue with the sale, liquidation %nd merger of PEs; and (ii) a program of reforms in the policy and institutional environment for PEs to improve the efficiency of the enterprises to be retained by the Government. VII. SUSTAINABILITY OF TRADE REFORM PROGRAM 132. The loan documents for TPL I and II identified three major risks concerning these operations: (i) possible inconsistency between the macroeconomic policy framework and continued structural changes; (ii) pressures to reverse the liberalization process as misallocations in the import dompeting sectors become more severe; and (iii) adverse external developments. 133. The first major risk identified referred to a possible inconsis- tency between the macroeconomic policy framework and trade liberalization. For example, expansionary fiscal policies would trigger balance of payments - 55 - problems and an appreciation of the real exchange rate could potentially derail the trade opening and the export drive. Were this to occur, a reversal of the trade reform might take place. A second risk identified concerned possible misallocations to inefficient import-competing sectors as the trade adjustment process deepened. In particular, failure to transfer resources to expanding and efficient sectors could lead to undue pressures against trade liberalization. A third risk concerns adverse external developments. 134. TYe Loan documents for both operations commented that the domestic risks woulc be reduced if the Government aims att (i) maintaining a real exchange rite consistent with its balance of payments objectives; (ii) ensuring budgetary discipline and the control of domestic inflation; and (iii) continuing the trade policy reform program. Regarding the external risk, sufficient international reserves and the finalization of Mexico's external financing package were expected to assure adequate financing for implementation of the economic recovery program. 135. Very large or sudden changes in the trade regime could be damaging because they could lead to unnecessary economic disruptions and could also undermine the sustainability of the trade reform. Sustainable trade reform requires that transitional difficulties and social adjustment costs be explicitly addressed in the design of the policy reform, and that reforms b6 carefully phased and complemented with ameliorating policies to minimize social adjustment costs and ensure that private costs do not reach levels that would endanger the liberalization process. On the other hand, the history of liberalization episodes demonstrates that without substantial initial action, the liberalization effort will not be credible. Economic agents will not respond in the manner desired, and those adversely affected will exert undue pressure on the government to reverse the policy reform. Moreover, the costs of trade liberalization will tend to be immediate and concentrated on established interest groups, while the benefits will be delayed and diffused and often primarily benefit potential investors. 136. Since the beginning of the stabilization process in 1982, the Mexican economy has only seen years of negative GDP growth, and entered 1989 with a lower standard of living compared to the beginring of the decade. Wages have fallen in real terms by around 402, and open unemployment in 1987 was around 4.5?, compared to 3.8Z in 1981. 137. If the economy does not show .igns in the short to medium-term that the economic adjustment program is bea=ing fruit--meaning enhanced growth and opportunities for resources to flow from contracting to expanding sectors--opposition to the reform could motnt. Furthermore, failure to begin to grow could jeopardize the macro-stabilization program by making it increasingly difficult to reduce the public sector borrowing requirements. 138. GDP is still largely stagnant, but export growth and recovery of private investment would offer opportunities for renewed growth. Private investment would likely recover if prejent high real interest rates start - 56 - to fall, as they already have in the second part of 1989. Moreover, expanding financial resources for private investments is vital for the economic recovery, underacoring the need for fiscal discipline and a successful solution to the external debt problem. 139. The Government maintained a fixed exchange rate throughout 1988 and considered the fixed rate to be the fundamental anchor of the stabili- zation program. The new administration shifted to a slow crawl. Maintain- ing consistency of the exchange rate policy with the macro fundamentals is a key to the success of the trade reform. 140. Other major concerns include the increase in the current account deficit in the balance of payments and the need to secure its financing. If the deficits in the current account of the balance of payments are not reduced to sustainable levels and adequate external financing is not forth- coming, authorities may be tempted to raise tariffs and place controls on imports to reverse the external deterioration. Moreover, a deterioration in the external environment (i.e. oil price drops, increased international interest rates, a recession and/or rising protection in the USA) could complicate the management of the economy. 141. An objective of trade-policy reform is to induce a shift of resources from less efficient sectors to more efficient sectors. As is well known, there are costs and delays in the movement of factors which prevent the productive sectors from responding immediately to a policy change. The shift of resources will require an increase in the share of GDP devoted to investment. As a share of GDP, gross domestic investment fell from over 27% of GDP at the beginning of the 1980s to 18? in 1986 and in 1987, recovering gradually to 23% in 1988. As a result, a backlog of needed investments has accumulated. A growth in investment would require increasing domestic savings, as well as, attracting higher levels of external savings, in part through an increase in direct foreign investment. Hence, growth in domestic and external savings are needed to finance an expansion in investment--vital for the economic recovery. In this context, obtaining relief on obligations on comercial bank debt is crucial for the sustainability of tL macro balances. 142. Fiscal adjustment during 1982-88 was extraordinary. The Govern- ment undertook major fiscal reforms beginning in 1982, with an improvement in the primary account of the consolidated public sector of about 14 percentage points of GDP, reversing it from a deficit of 7? of GDP in 1982 to a surplus of 7Z in 1987. Public sector investment fell from 10Z of GDP to 6Z between 1982 and 1986, declining even further in 1987, while current spending--excluding debt service--also decreased significantly from 26% to 24Z. Unfortunately, these cuts were offset by the effects of higher interest rates, and, consequently, on the cost of servicing domestic debt, leaving total public spending in 1986 at around 472 of GDP. 143. A possible deterioration in public sector finances could also dera'i the reform process. If the authorities expand public expenditures beyond prudential limits, this will have a devastating impact on the - 57 - stabilization achieved thus far. Conversely, if the Government receives a windfall--e.g. from an oil bonanza--its will to continue its role in the economy may be eroded. Hence, continued implementation of prudent fiscal policies is needed for the sustainability of the trade reform. VIII. WORLD BANK'S PERFORMANCE 144. The Bank has supported Mexico's adjustment process since 1982, through a growing dialogue on trade policy reform, and expanding the dialogue gradually to other areas, including public sector management and finances, domestic regulations, and financial sector reform. The dialogue on trade covered a number of other issues indirectly related to trade reform, and eventually led to loans in the trade, agriculture, steel, and fertilizer sectors. There was a clear government commitment to the program. In fact, the Bank played only a supportive role in the adjustment process. However, TPL I and II did strengthen the position of the reform- minded elements in the Government. 145. The Bank's help especially proved to be useful in exposing the excessive costs of protection. The Government also recognized the bene- ficial effect of announcing the reform program within a multi-year frame- work--as suggested by the World Bank. Apart from the intellectual impact of the Bank, the financial assistance, including the cofinancing package, helped to finance the Government's program, enabling the introduction of the reforms. 146. With the Trade Policy Loans, the Bank's policy dialogue with the Mexican Government deepened at the right time and focused on a politically feasible program. The Bank team working on Mexico was able to convince Mexican officials, who were initially against the pace of the reform, of the need for and the urgency of the reform. 147. The Country Programs Division prepared TPL I with the support of the Industrial Development and Finance (IDF) Division and the latter was responsible for TPL II. After the reorganization of the World 3ank in 1987, the supervision of the Technical Assistance component of the TPL I and the TPL II operation beca"s the responsibility of the Trade, Finance and industry (TFI) Division, with the Country Operations Division (COD) in charge of supervising the macro conditionality. The COD and TFI Divisions coordinated their activities closely. The lack of a clear macro condition- ality may have inhibited close supervision of the macro framework while domestic inflation rose to 160Z p.a. and international reserves built up at a high rate. While the Mexican Government did neither seek the advice or consent of the World Bank when introducing the macro-stsbilization program, which reduced tariffs to 202, based on its analysis of the economic circum- stances, the World Bank came out supporting it. While supervision of TPL I was very close and staff intensive, formal supervision of TPL II concen- trated on key issues, but was less thoroughly documented and lacked detailed analysis, which was carried out in the context of the green cover Trade Impact Study Report mentioned earlier. - 58 - IX. PROJECT RELATIONSHIP 148. The success of the dialogue with the Government can be attributed to at least three factors. First, the Bank essentially supported the Government's program of reform. The Bank did, however, press for more far- reaching reforms, especially during TPL I negotiations. The dialogue, from the Bank's viewpoint, took into account that the Government included reform-minded as well as conservative elements. It is likely that the Bank's policy dialogue strengthened the former elements, resulting in faster adoption of the reforms. Had the Government (GOM) been single- mindedly opposed to the types of reforms advocated by the Bank, it is doubtful that the reforms could have been introduced. 149. Second, the Bank became involved in the dialogue at a propitious time. In 1982, when the new Administration took office, the Bank was able to make a forceful case in supporting its position while policies were still being debated and formed. It was also helpful that this new Adminis- tration was seeking financial assistance, and included a sufficient number of reform-minded technocratic elements less steeped in Mexico's traditional politics than had been the case under previous administrations. 150. Finally, it was also helpful that the GOM was petitioning to accede to the GATT, and was therefore required to fulfill conditions that paralleled those that the Bank judged necessary for a successful introduc- tion of the trade reform. X. OPERATIONAL ISSUES A. World Bank-IMF Relations 151. Since the Bank has not made a Structural Adjustment Loan (SAL) to Mexico, it has relied on the IMF to monitor macroeconomic performance, as necessary. Most of the World Bank's sectoral adjustment lending, there- fore, was carried out concurrently with IMF programs. The one exception was TPL I, which was approved by the World Bank Board in 1986 after dis- bursements under the IMF Extended Arrangement had been suspended. Nonethe- less, as condition of effectiveness, the Loan required an agreement on a financial package to cover Mexico's external financing needs for 1986, and since commercial lending was contingent upon Mexico's reaching agreement with the IMF on an adjustment program, this became a de facto condition of effectiveness as well. Furthermore, at the time the TPL I was presented to the World Bank Board, the IMF publicly announced its support for Mexico's macroeconomic policies and its intention to approve a new agreement with Mexico. TPL II also included in its second tranche release a condition for an appropriate macroeconomic policy framework. World Bank and IMF staff dealing with Mexico coordinated efforts adequately, and when different economic forecasts were prepared, the issues were discussed and resolved. - 59 - B. Disbursement Performance under the Loan Operations 152. Although TPL I disbursed fairly quickly, TPL II encountered diffi- culties in disbursements. A number of factors identified and discussed below are responsible for the slow disbursement performance under TPL II. First, eligible import categories were reduced de facto by the exclusion of imports which are subject to NTBs. Moreover, the authorities could not produce quickly enough the required documentation for imports, including those: (i) financed at the parallel exchange rate; (ii) financed by short-term loans; and (iii) private sector imports subject to international bidding. Second, disbursements were slowed down somewhat by the simultane- ous existence of four pclicy-based lending operations, with one operation (Steel Sector Restructuring Loan) having a positive list, two operations (TPL II and AGSAL) having negative lists, and one operation (Fertilizer Sector Adjustment Loan) having both positive and negative lists. Third, inadequate documentation and processing arrangements at the commercial banks, Central Bank and at BANCOMEXT, including use of inappropriate codes for goods, also contributed to the slowdown in disbursements. 153. Suggestions for improvement. For future policy-based loan opera- tions, it is suggested not to exclude financing of imports subject to NTBs as this exclusion in effect defeats its purpose by raising protection pro- vided to those import categories. In this way, the excessive administra- tive burdet of monitoring large numbers of import items would also be avoided. 154. The problems encountered in the disbursement performance under the TPL II operation focused the Government's attention on taking actions to begin to resolve the issues which slowed down disbursements. As a result, disbursements under the new generation of policy-based loans (i.e., ISPL, PERL, and FSL) have been rapid. C. Technical Assistance under TPL I 155. TPL I contained US$11.0 million in technical assistance to improve customs administration and the collection of trade statistics. Although the component to improve customs involved mainly a relatively straightfor- ward purchase of computers and other equipment to systematize data collec- tion, it ran into problems and very little was disbursed (as of end of October 1989, less than US$0.5 million was disbursed under the technical assistance component). Several factors played a role in this unfavorable outcome. First, the administration of this component involvea five govern- ment entities and hence coordination became a formidable issue. Second, the technical assistance covered very difficult areas such as customs and anti-dumping (a completely new area in which large firms tried to recapture part of their loss in protection). Third, executing agencies did not have budgetary approval and counterpart funds to allow them to utilize the funds under the technical assistance component. Consideration should be given to providing a free-standing technical assistance project in the future. - 60 - D. Cofinancing 156. In addition to the net resource transfer achieved by the TPL operations, the World Bank also played a critical role in the preparation of a large loan operation by the foreign money center commercial banks to finance Mexico's financial gap. The financing package--incluiing new money and the rescheduling of existing debt--had the highest amount to that date and carried partial World Bank guarantee for some of the new money facili- ties to induce the commercial lenders to provide new financing. At that time, the prevailing view of the international financial community was that the new money package for Mexico would not have been forthcoming without the World Bank's direct involvement. Clearly, TPL II provided a catalytic effect on the new money package. The difficulties which Mexico faced in obtaining the commercial new money even after having secured the World Bank's partial guarantees underscore the critical role of the World Bank in the preparation of the new money package. 157. The Bank involvement in the preparation of the new money package for Mexico prompted the World Bank Board to review the Bank's role and policy in high-debt countries, particularly regarding the use of the Bank's guarantees. This review concluded that further use of the Bank's guarantee power for commercial new money packages would require exceptional circum- stances such as those encountered in Mexico at that particular time. XI. CONCLUSIONS 158. Progress under the trade reform in Mexico has been impressive. Import licensing coverage is now one-quarter of what it was in June 1985, official reference prices for imports have been completely abolished, and the maximum tariff is now one-fifth and the average tariff one-half of what it was in June 1985. Progress, as of mid-1989, went beyond the Government's original schedule, GATT commitments, and the targets in TPL II. The progress achieved in trade reform was comp'emented by other reforms, particularly in the areas of macroeconomic stabi ization, initia- tion of a program of economic deregulation, and the introduction of finan- cial liberalization measures. While the full impact of the trade reform will take some time to materialize, the initial results are promising- -incentives have been redirected and restructuring toward more efficient activities has already begun. - 61 - ANNEX I Page 1 of 2 COMPLIANCE UNDER TPL I & II OPERATIONS 1. The following summarizes the conditions under the TPL I and TPL II operations and compliance for both effectiveness and Second Tranche conditions. It also presents the tariff reduction calendar and actual performance as well as reductions in license coverage and in official reference prices. TRADE POLICY LOAN 1 (2745-ME) a. EffectIveness: November 18, 1988 Conditions: Compl Iance (1) macropollcy framework consistency with satisfactorily met the Trade Policy Program (11) meet external financing needs for 1986 satisfactorily met (111) progress with the Reform Program satisfactorily met before effectiveness b. Second Trenches December 1, 1988 Conditiones Compitance (1) satisfactory progress with the Reform satisfactorily met (see Program below) (11) macropolicy framework consistency with satisfactorily met the Reform Program (li1) reduction Sn NTS coverage by at least reduction from 76.55 In SX of domestic production from March March to 62.8X in December 31, 1986 to December 31, 1988 (Iv) publication of anti-dumping regulations published in regulations prior to the disbursement Dario Official' on 25 of the Second Tranche November, 1986 TRADE POLICY LOAN II (2882-ME) Effectiveness: January 20, 1988 Con0itiones: Compliance: (1) macropelley framework consistency with satisfactorily met objectives of the Trade Liberalization Program (ii) continued implemeotation of the Tariff satisfactorily met (see Reform calendar below) ANNEX I Page 2 of 2 Second Tranches August 1, 1988 Conditions: Compilnce (1) progress with the implementation of satisfactorily mat (see the Reform Program below) (Hl) reduction )n NTB coverage by December reduction from 82.8% in 31, 1987 by at least 20% as compared November 1986 to 37.6% in to November 30, 1988 December 1987 (ill) macropolicy framework consistency with satisfactorily met the Reform Program Tariffs The Reform Program aimed at an across-the-board tariff adjustment according to a preannounced schedu.e to decrease the maximum rate from 100Z in early 1986 to 30Z by end 1988. Tariff Reduction Calendar 1986 4/86/86 2/28/87 12/81/87 16/31/88 Maximum 1001 465 40 35% 30% Minimum e1 61 61 O5 01 Dispersion 18.90 16.5 18.81 12.01 8.4% Tariff Levels 16 8 7 7 5 Tariff Reduction Performance: 1985 04/30/86 08/06/87 12/15/87 Maximum 1601 46X 401 201 Minimum 6 O 01 61 Dispersion 18.91 16.O 12.3X 6.91 Tariff Levels 10 8 7 6 General Import Tax An additional policy measure in the Reform Program supported by the TPL II operation was the phasing out of the 5Z general import tax by the end of 1987. The general surcharge was abolished on December 15, 1987. - 63 - ANNEX II TRADE POLICY LOAN I, LOAN/CREDIT DATA (As of 12/31/88) Amounts (US$ Millions) original Disbursed Cancelled Repaid Outstanding Loan 2745 $500 $489.34 0 0 $489.34 Original Loan Dates Actual or Re-estimated Initiating Memorandum 04-30-85 08-16-85 Letter of Development Policy 07/22/86 07/22/86 Negotiations 12/85 03/06/86 Board Approval 02/86 07/29/86 Loan/Credit Agreement 08/08/86 08/08/86 Effectiveness 11/06/86 11/18/86 Loan/Credit Closing 11/30/88 11/30/89 Actual Completion 08/89 08/89 Cumulative Loan/Credit Disbursement FY87 FY88 FY89 (i) Planned $494 5 1 kii) Actual 489.01 0 .32 (iii) (ii) as of Z of (i) 98.9% - 32% MISSION DATA Month, Year No. of No. of Staff Weeks Date of Weeks Persons Report Preparation 11&12/84 3.6 4 14.4 02/01/85 Preparation 4&5/85 2 5 10 05/06/85 Appraisal 09/85 3 8 24 10/85 Appraisal 11/85 3.6 8 28.8 04/86 Negotiations 04/86 1 4 4 12/18/85 Supervision 11/86 2 6 12 12/86 Supervision 09/87 1 2 2 09/87 Completion 1-8/89 2 1 2 08/89 FOLLOW-ON ADJUSTMENT OPERATIONS Mexico Trade Policy Loan II No. 2882 was appyoved on November 11, 1987 in the amount of US$500 million. Industrial Sector Policy Loan of US$500 million was approved on June 13. 1989. - 64 - ANNEX III TRADE POLICY LOAN II. LOAN/CREDIT DATA (As of 12/31/88) Amounts (US$ Millions) Original Disbursed Cancelled Repaid Outstanding Loan 2882 $500 $494.58 0 0 $494.58 Original Loan Dates Actual or Re-estimated Initiating Memorandum 03-09-87 03-09-87 Letter of Development Policy 05/87 05/87 Negotiations 06/10/87 06/10/87 Board Approval 09/87 11/12/87 Loan/Credit Agreement 12/07/87 12/07/81 Effectiveness 12/87 01/20/87 Loan/Credit Closing 06/30/88 12/31/88 Actual Completion 08/89 08/89 Cumulative Loan/Credit Disbursement FY88 FY89 (i) Planned $500 - (ii) Actual 299.42 265.16 (iii) (ii) as of I of (i) 45.9Z - MISSION DATA Month, Year No. of No. of Staff Weeks Date of Weeks Persons Report Preparation 11/86 2 6 12 12/86 Appraisal 03/23 2 8 16 05/87 Post Appraisal 09/87 1 3 3 09/28/87 Supervision l/ 02/88 3.6 11 39.6 02/26/87 Supervision 04/05/88 2 2 4 06/03/88 Supervision 06/88 0.8 3 2.4 07/28/88 Completion 01/08/89 2 1 2 08/89 FOLLOW-ON ADJUSTMENT OPERATIONS Mexico Industrial Sector Policy Loan of US$500 million was approved on June 13, 1989. 1/ The figures include the preparation of the green cover Trade Impact Study Report.