Turkey - Fourth and Fifth Structural Adjustment Loan Projects

Document of The World Bank FOR OFFICIAL USE ONLY RIeport No. 7205 EVALUATION OF STRUCTURAL ADJUSTMENT LENDING IN TURKEY PROGRAM PERFORMANCE AUDIT REPORT OF THE FOURTH AND FIFTH STRUCTURAL ADJUSTMENT LOANS (LOANS 2321-TU AND 2441-TU) AND OVERVIEW OF SALS I-V April 13, 1988 Operations Evaluation Department Vis dcment bs a res"trd ditribution and may be used by socplents only I the perfor ACRONYMS AND ABBREVIATIONS APK - Bureau of Planning, Research and Coordination ASAL - Agricultural Sector Adjustment Loan DIF - Defense Industries Fund DSF - Development and Support Fund EBF - Extra-Budgetary Funds EIE - Electricity Survey Administration FPRT - Fiscal Policy and Tax Reform FSAL - Financial Sector Adjustment Loan LRMC - Long-Run Marginal Cost MAFRA - Ministry of Agriculture, Forestry and Rural Affairs MENR - Ministry of Energy and Natural Resouxces MHF - Mass Housing Fund METU - Middle East Technical University PEF - Petroleum Exploration Fund PPF - Public Participation Fund PSBR - Public Sector Borrowing Ratio QRs - Quantitative Restrictions RUSF - Resource Utilization and Support Fund SAL - Structural Adjustment Loan SEEs - State Economic Enterprises SIS - State Institute of Statistics SPO - State Planning Organization SPSF - Support and Price Stabilization Fund TCZB - Agricultural Bank TEK - Turkish Electrical Authority TPAO - Turkish Petroleum Corporation TUSIAD - Turkish Industrialists' and Businessmen's Association T2PK - Agricultural Supply Organization UTFT - Undersecretariat for Treasury and Foreign Trade CURRENCY EQUIVALENT (Annual Average) Name and Abbreviation - Turkish Lira (TL) 1980 US$1 = TL 76.04 1981 US$1 TL 111.22 1982 US$1 = TL 162.55 1983 US$1 = TL 225.46 1984 US$1 = TL 366.68 1985 US$1 = TL 521.98 1986 US$1 TL 674.50 1987 US$1 = TL 860.58 FISCAL YEAR March 1 - February 28 through 1981 March 1 - December 31, 1982 January 1 - December 31 from 1983 M0 OrMAL *I ONLY THE WORLD SANK WAOmon. O.C. ao33 U.S.A. othee of DWCeae Opeseque Evdiuat. April 13, 1988 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECTs Evaluation of Structural Adjustment Ledini in Turkey Attached, for information, is a copy of a report evaluating structural adjustment leading to Turkey. It contains the Program Performance Audit Report on Structural Adjustment Loans IV and V (Loans 2321-TU and 2441-TU) and Overview of Turkey SAL. I-V prepared by the Operations Evaluation Department. This docunat hs a tstittdbuion and may be ud by socipisat only i the pWfoMaace of their official duties. Its contents may not otherwise be disclosd without World Bak authorition. FOR OCUL USE ONLY EVALUATION OF STRUCTURAL ADJUSTMENT LENDING IN TURKEY PROGRAM PERFORMANCE AUDIT REPORT TURKEY FOURTH AND FIFTH STRUCTURAL ADJUSTMENT LOANS AND (LOANS 2321-TU AND 2A1-TU) AND OVERVIEW OF SALS I-V TABLE OF CONTENTS Page No. Preface........ .... . .. ........................ ............. Basic Data Sheets......................................... Evaluation Summary...................... .. # ................. vi Synoptic Overview of Experience................ ......... xx PROGRAM PERFORMANCE AUDIT MEMORANDUM I. BACKGROUND ........................ *............... . II. SAL DESIGN: OBJECTIVES AND POLICIES ................. 3 A. Background: SALs I-III ........................... 3 B. Design of SAL IV ............................... 5 C. Design of SAL V .................................. 9 D. Evaluation of SAL Design ......................... 14 III. IMPLEMNTATION OF SALs IV AND V ..................... 16 A. Introduction . ............. .........0....... 16 B. Export Promotion ......................... ........ 20 C. Import Liberalization ..... o ....................... 22 D. Rationalization of Public Investment .............. 23 E. Energy Sector Policy .......... .............. . .... 25 F. Agricultural Sector Policy ........................ 27 G. Reform of the SEEs ................................ 29 H. Summary of Program Implementation ................. 31 IV. MAJOR ISSUES ................ .................... 32 A. introduction . .o................... o............. 32 B. Exchange-Rate Policy and Real Interest Rates ...... 33 C. Budget Control and Fiscal Dynamics ................ 46 D. Foreign Debt and Current-t."count Stability .......54 E. Investment and Export Capacity ................... 57 F. Financial Market Structure and Stability .......... 66 G. Role of Extra-Budgetary Funds ..................... 69 H. Trade Liberalization .............................. 74 I. Reform of the State Economic Enterprises (SEEs) ... 82 This document has a restricted distribution and may be used by recipients only in the performance of their offBcial duties. Its contents may not otherwise be disclosed without World Bank authorlastion. TABLE Ol CONTENTS (continued) V. IMPACT OF STRUCTURAL ADJUSTMENT PROGRAM .............. 85 A. Economic Growth ................................... 85 B. Balance of Payments ............................... 87 C. Financial Activity and Inflation .................. 89 D. Social Impact ..................................... 90 VI. SUSTAINABILITY OF REFORMS ............................ 97 A. Introduction ...................................... 97 B. Institutional Reforms ............................ 98 C. Economic Policy ................................... 103 VII. ROLE OF THE BANK ............................... 104 A. Program Development ............................... 104 B. Quality of Conditionality......................... 105 C. Complementarity with IMF .......................... 106 D. Social Dynamics of Adjustment ..................... 106 VIII. PRINCIPAL CONCLUSIONS AND RECOMMENDATIONS ............ 108 SAL Design and Strategy .............................. 109 Implementation and Sustainability .................... 110 TABLE OF CONTENTS (continued) LIST OF TABLES Tables 3.1 Consolidated Budget Sumary 3.2 New Bond Equity Permits 3.3 Nominal and Real Deposits 3.4 Export Incentives 3.5 Number of Items Subject to a License 4.1 Turkey - Real Exchange Rate 4.2A Foreign Assets and Liabilities Consolidated Balance Sheet of Commercial Banks Private Sector Foreign Debt 4.2B Foreign Assets and Liabilities - Foreign Exchange Deposits 4.3 Real Interest Rates 4.4 Stock of Consolidated Government Debt Outstanding 4.5A Public Sector Borrowing Requirement 4.53 Growth in Tax Revenue 4.5C Forthcoming 4.6 Primary Government Deficit 4.7 Foreign Debt 4.8 Balance of Payments 4.9 Fixed Investment in Manufacturing 4.10 Estimates of Value of Manufacturing Capital Stock 4.11 Manufacturing Sector Output 4.12 Private Sector Capacity Utilization 4.13 Concentration of Bank Assets 4.14 The Principal Extra-Budgetary Funds, 1985-86 4.15 Sources and Uses of the PPF, 1986 4.16 Sources and Uses of the MHF, 1986 4.17 Exports by Destination 4.18 Turkish Export Subsidies 4.19 Tariff Equivalent of Levies in 1987 4.20 SEEs Financial Position and Employment 4.21 Productivity of the SEEs 4.22 Distribution of Value-Added by 500 Large Firms 5.1 National Accounts, 1980-86 5.2 Exports, Imports and Gross Domestic Product 5.3 Household Income Distribution 5.4 Labor Force, Employment and Unemployment 5.5 Turkish Real Wages LIST OF FIGURES Figures 4.1 Nominal Interest Rates 4.2 Year Inflation 4.3 Real Effective Exchange Rate 4.4 Real Interest Rates (using actual inflation, RA) Real Interest Rates (using lagged inflation, RB) 4.5 Fixed Investment in Manufacturing 4.6 Export Performance, 1975-85 4.7 Effect of Levy on Tariff Revenue 5.1 Manufacturing Real Wage TABLE OF CONTENTS (continued) PROGRAM COMPLETION REPORT I. Introduction ......................................... 113 II. Achievements Under SALs 1-III ............. 119 III. SAL9 IV and Vs Objectives and Policies ............... 122 IV. Impact of the Structural Adjustment Program .......... 128 V. An Evaluation of SAL Designs ......................... 141 VI. Implementation of SAL Programs ...................... 144 VII. The Role of the Bank ................................. 161 EVALUATION OF STRUCTURAL ADJUSTMENT LENDING IN TUREY PROGRAM PERFORMANCE AUDIT REPORT TUREY FOURTH AND FIFTH STRUCTURAL ADJUSTMENT LOANS (LOANS 2321-TU AND 2441-TU) AND OVERVIEW OF SALS I-V PREFACE This is a performance audit of the Bank's fourth and fifth struc- tural adjustment loans (SALs) to Turkey and overview of Turkey SALs I-V. Since Turkey is the only member country which has received a series of five SALs, this audit report in fact is also an evaluation of the series of structural adjustment loans to Turkey. The fourth SAL (Loan 2321-TU, US$300.8 million) was signed on June 27, 1983 and fully disbursed by June 25, 1984. The fifth SAL (Loan 2441-TU, US$376 million) was signed on June 20, 1984, and fully disbursed by June 27, 1985. The audit consists of a Program Performance Audit Memorandum (PPAM) prepared by the Operations Evaluation Department (OED) and a Program Completion Report (PCR), prepared by the Europe, Middle East and North Africa Regional Office. The PPAM is based on a review of the President's Reports (No. P-3543-TU. dated May 26, 1983 and No. P-3783-TU, dated May 24, 1984), the loan agreements, the sunaries of the Board discussions, and the combined PCR and related documents in the Bank's files. Bank staff associ- ated with the SALs have been interviewed and discussions have been Leld with IMF staff. An OED mission visited Turkey in May-June 1987 to review the structural adjustment operations with Government officials and private sector representatives. The valuable assistance provided by representa- tives of the Borrower is gratefully acknowledged. Copies of the draft report were sent to the Borrower in advance of an OED mission that visited Ankara in January 1987. The views expressed by the Borrower in the course of those discussions have been incorporated in the final report. EVALUATION OF STRUCTURAL ADJUSTMENT LENDING IN TURKEY PROGRAM PERFORMANCE AUDIT REPORT TURKEY FOURTH STRUCTURAL ADJUSTMENT LOAN (LOAN 2321-TU) BASIC DATA SHEET (Amounts in US$ million) LOAN STATUS As of 5131187 Original Disbursed Cancelled Reaid Outstanding Loan No. 2321-TU 300.8 300.8 0.0 0.0 300.8 CUMULATIVE LOAN DISBURSEMENT FY85 (i) Planned 300.8 (ii) Actual 300.8 (iii) (ii) as I of (i) 100 OTHER LOAN DATA Original Actual or Loan Dates Re-estimated Initiating Memorandum 09/15/82 09129/82 Appraisal Mission 02/28/83 02/28/83 Negotiations 04/18183 05124183 Board Approval 05/31/83 06123183 Loan Agreement - 06/27/83 Effectiveness 09/27/84 09/19/83 Loan Closing 03/31/85 03/31/85 Actual Completion 03/31/85 03/31/85 STAFF INPUT 1982 1983 1984 1985 1986 1987 Totals Preappraisal - 36.2 - - - - 36.2 Appraisal - 29.2 - - - - 29.2 Negotiation - 7.7 - - - - 7.7 Supervision - 1.7 9.4 0.2 - 1.3 12.7 Other 0.5 - - 0.2 - - 0.7 Totals 0.5 74.9 9.4 0.4 - 1.3 86.5 MISSION DATA No. of No. of Staff Date of Month/Year Weeks Persons Weeks Report Preparaion 10/824 03/22183 Appraisal 03/83 2 9 15 02/22/84 Tranche Review 01184 2 4 8 FOLLOW-ON SAL OPERATIONS Fifth Structural Adjustment Loan 2442-TU approved on June 14, 1984, in the amount of US$376.0 million. - iv EVALUATION OF STRUCTURAL ADJUSTMENT LENPtNG IN TURKEY PROGM PERFORMANCE AUDIT REPORT TURKEY FIFTH STRUCTURAL ADJUSTMENT LOAN (LOAN 2441-TU) BASIC DATA SHEET (Amounts in US$ million) LOAN STATUS As of 5/31187 Original Disbursed Cancelled Repaid Outstanding Loan No. 2441-TU 376.0 376.0 0.0 0.0 376.0 CUMULATIVE LOAN DISBURSEMENT FY85 (i) Planned 376.0 (ii) Actual 376.0 (iii) (ii) as I of (i) 100 OTHER LOAN DATA Original Actual or Loan Dates Re-estimated Initiating Memorandum 03/--/84 03109184 Appraisal Mission 03119/84 03/26/84 Negotiations 05/04/84 05/16/84 Board Approval 06/26/84 06/14/84 Loan Agreement - 06/20/84 Effectiveness 08/20/84 07/03/84 Loan Closing 03/31/86 03/31/86 Actual Completion 03/31/86 03/31/86 STAFF INPUT 1982 1983 1984 1985 1986 1987 Totals Preappraisal - - 15.8 - - - 15.8 Appraisal - - 17.7 - - - 17.7 Negotiation - - 6.2 - - - 6.2 Supervision - - 0.7 15.7 0.5 1.3 18.2 Other - 0.0 0.0 - - 0.1 Totals - - 40.4 15.7 0.5 1.3 58.0 MISSION DATA No. of No. of Staff Date of MonthLYear Weeks Persons Weeks Report Preparation 01/84 2 5 10 02108184 Appraisal 03184 Tranche Review 03185 2 1 2 04104185 FOLLOW-O SAL OPERATIONS NIA - vi - EVALUATION OF STRUCTURAL ADJUSTMENT LENDING IN TURREY PROGRAM PERFORMANCE AUDIT REPORT TURKEY STRUCTURAL ADJUSTMENT LOANS IV AND V (LOANS 2321-TU AND 2441-TU) AND OVERVIEW OF SALS I-V EVALUATION SUMMARY Background The dirigiste economic strategy pursued by Turkey in the 1960s and 1970s yielded annual economic growth rates ranging from 6Z-7%. It was grounded, however, on the development of a domestic market that had to be increasingly stimulated by expansionary monetary policy. The low ratio of exports in GYP and the increasingly large curreant-account deficit that began to emerge in the 1970. led to the accumulation of large foreign debts and a correspondingly heavy debt service burden. By late 1977 it was ap- parent that the strategy of regimented economic growth based on import- substitution was not sustainable. Turkey's problems were compounded by unfavorable developments in the international economy, especially the rapid increase in oil prices which occurred in 1979. After 1977 Turkey made serious efforts to rescue its economic strategy by initiating programs of economic stabilization and structural adjustment. Stabilization programs were worked out in consultation with the IMF in 1977 and in the period 1980-84. They aimed at restoring price stability, creditworthiness, and a sustainable balance in domestic and external finances. The Government that took power in 1980 believed that a market-oriented economy based on export orientation held the promise for rapid economic growth and improved external payments. It negotiated a first Structural Adjustment Loan (SAL) with the Bank in 1980, which was followed by four more SALs, the last of which was concluded in 1984. The underlying philosophy of the new economic strategy was that a new division of labor would come about between the private and public sectors. The former would concentrate on the production sectors, providing the principal impetus to industrial growth and exports, while the latter would reinforce these efforts by focussing on the development of economic infrastructure. As a first step towards the implementation of this *social contract", the Government initiated steps to reduce the dependence of the State Economic Enterprises (SEEs) on the Exchequer and to free the excessive regimentation that characterized the trade regime. A plethora of incentives was provided to the export sector and a process was begun to simplify the complex trade barricades that had previously shielded industrialists from foreign com- petition. - vii - Turkey's commitment to the program of structural adjustment was solidly expressed throughout the life of the SALa. At the policy level, a variety of measures were introduced to improve the efficiency of resource allocation and to strengthen export competitiveness. Very early in the program attention was devoted to reducing the excessive financial losses of SEEs through cost reduction measures as well as changes in pricing poli- cies. Quantitative controls on imports were abolished and the tariff sys- tem was rationalized. The exchange rate was recognized as an important tool for maintaining export competitiveness and improving resource alloca- tion. Steps were taken to improve resource mobilization and to strengthen the tax collection machinery. As a first step towards the development of an integrated debt management policy, the debt recording system was in- proved with assistance from the Bank. Strict controls were applied to the paze of monetary expansion and wage increases were moderated. The economy responded positively to the new economic strategy. GDP grew at an annual average of close to 52 during 1980-85 and the share of industry rose from 25% to about 32% principally at the expense of agri- culture. Reflecting the greater openness of the economy tc international trade, the share of foreign trade in GDP rose from 19% to about 37? over the same period. Merchandise exports, which financed only about 40? of imports in 1980, increased their share to around 70? in 1986. The highest growth has been in industrial exports, principally manufactures, which accounted for 70% of total exports, after having a share of only 36? in 1980. An inflation rate of almost 100?, which existed at the beginning of the program, was reduced to about 35% in 1985. Progress was made in the revitalization of Turkey's capital markets and the supervision of the bank- ing system was improved. In general, Turkey's entrepreneurs and industri- alists demonstrated new orientations as they responded to the vistas opened to them by the liberalization of commodity and capital markets. Despite this encouraging progress, however, only a minor dent was made on the level of unemployment. On the social front some declines in real wages occurred. This was practically inevitable given the imperatives of adjustment, but in return the less privileged also shared in the econom- ic growth that occurred. Much unfinished business still remained at all levels and in all sectors at the time the SAL program was terminated in 1985. The succeeding sector adjustment loans have addressed major sector issues, but a number of macro-economic problems still remain. These are taken up in later sections. SAL Desian: Objectives and Policies of SALs IV and V The design of SALs IV and V followed logically the program laid out in SALs II and III. The program became more focussed, with agricul- ture, energy, and finance coming under sector loans. Fewer conditionali- ties were imposed. Substantial progress was made in numerous areas. The trade regime was opened and liberalized, and the policy of flexible adjust- ment of the exchange rate provided a stimulus to exports. Institutional -viii - improvements were made in the financial seector, such as development of the capital markets, establishment of a bank auditing system, and an inter-bank market. Improvements were made in tax administration, and the value-added tax (VAT) was introduced. The public investment program was streamlined. Most importantly, the outward-looking and market-oriented attitude became entrenched in the economy of Turkey. In both agriculture and energy, poli- cies were designed to increase supply and lessen demand by raising con- trolled prices to international levels. The program also emphasized im- provements in the technical and management abilities of the ministries involved in agriculture and energy. But there were omissions in the design of SALs IV and V. A stronger response was needed to cope with emerging macro-economic disequilibria. Little provision for mid-course correction was made. The Bank may have limited its leverage by limiting the program to five SALs and disbursing the fifth too soon. Each of these problem areas will be discussed in turn. Macro-Economic and Sector Preparation. The Bank's economic and sector work program intensified after the structural adjustment program got under way. Sector work was focussed on agriculture, industry, energy, education and infrastructure development. Much of the macro-economic work was of high quality and was focussed on key areas of the adjustment pro- cess. For example, the report on 'Industrialization and Trade Strategy, (1982) and the follow-up report titled "Trade Policy Issues in the Struc- tural Adjustment Process" (1984) were important milestones in the develop- ment of the policy dialogue on structural adjustment. In SALs II and III, the Bank tended to accept the Fund's analysis of fiscal, monetary, and real exchange-rate policies as the center of analy- sis. Lesser attention was paid to the budgetary process and to the effects of potential fiscal instability deriving from the Government's debt build- up, until the report on 'Fiscal Policy and Tax Reform' was completed in 1986. The subject of high interest rates as a factor that could handicap investment was discussed in the Country Economic Memorandum of September 1983, but the analytical implications of exchange rate and interest rate interaction and the fiscal dynamics of debt expansion would have surfaced much sooner with the early development of a policy-oriented, macro-economic framework. This was an issue from SAL III which the Bank continued to flag, especially in its review of the Fifth Five-Year Plan (1985). Final- ly, the management of the external debt (as opposed to recording it), might have been taken up more vigorously before SAL V, when it became a live issue. Mid-Course Correction. Little provision was built into the SAL design for mid-course corrections. The problems in simply freeing the prices of SEEs from regulation without cost-cutting incentives could have been recognized and corrections made in SALe IV and V. The sluggishness of private investment and factors contributing to it could have been recog- nized. Analysis of the causes for this and remedies proposed could bare been built into the SALs. The growth of the Extra-Budgetary Funds (EBFs), and their unde.ining effects on the integrity of the fiscal system, and the trade regime could have been anticipated. It could have been seen that - ix - the basic data needed to evaluate the social impact of the program was not being collected, and that could have been written into SALs II or III. These were problems that became evident during the SAL period, and remain problematic now. They are discussed in some detail in Chapters IV-VI. SAL Duration and Timina. The SAL program, which commenced in 1980, coincided with an economic crisis resulting from a severe external payments imbalance. The initial thrust was therefore concentrated on im- proving macro-economic management so as to put the economy on a stable path that would promote longer-term economic growth and employment. This was essentially the task of the first two SALs. The momentum of economic sta- bilization was reinforced and complemented by IMF Stand-By Arrangements which coincided with the Bank's own program. The last three SALs began to address some of the hard-core, longer-term policy and institutional con- straints. Towards the end of the disbursement period for SAL IV, however, there was some evidence that hard-core problems of economic stabilization and adjustment persisted. The rate of inflation remained at a troublesome level and unemployment was also high. Economic policy-making was becoming fragmented and there was a loss of budgetary control over a good part of receipts and expenditures as well as over debt management. There was a case, therefore, for giving the Government more time to complete its prepa- ration of a macro-economic framework prior to the negotiation of a fifth and final SAL. Much unfinished business still remained at the macro-eco- nomic and sectoral levels. While succeeding sector loans played a useful role, they were limited in scope and could not address broader macro-eco- nomic policy issues. These needed to be addressed within the context of further SALs. Implementation of SALs IV and V - Maior Issues The policies embodied in the design of SALs IV and V were many and far-reaching. The adequacy of these policies to the task of adjurting the structure of the Turkish economy towards a more outward-looking and market- oriented stance is treated in Chapter II. This portion of the report fo- cusses on several elements that were common to the efforts to actually in- plement the policy reforms. A more complete analysis of the steps taken with regard to each policy is given in Chapter III. Besides elements designed to create a desirable macro-economic framework the loans included policies specifically tailored to the agricul- ture, finance and energy sectors. Under SALs IV and V the Government made important improvements in import liberalization and export promotion. The number of goods subject to quantitative restrictions (QRs) was more than halved in 1984, and the share of imports subject to QRs fell even more. On the export side, reductions were made in 1984 in the amount of subsidies, while the dollar value of exports grew by 25Z in the same year. In agri- culture the number of goods subject to price controls was reduced from 25 to 14, while in the energy sector prices were kept at international levels. In capital markets, 1984 and 1985 saw important advances as government bonds began to be publicly auctioned. More private companies began to sell their paper on the Istanbul Stock Exchange. The efforts at instituting policy reforms in Turkey, while in general successful, were uneven at times. The passage of a new law or of a regulation embodying a reform was frequently. met by the creation of some other mechanism for protecting the interests of those who gained from the earlier status quo. An excellent example of this process was the creation of the BBFs. Originally designed to more expeditiously provide key governmental services--through decentralization--the EBFe also contributed to circumventing efforts at fiscal reform, as well as the moves toward import liberalization. Because EBFs were outside the normal budgetary process, they escaped some of the rationalization pressures that actions to reduce the central government deficit engendered. At the same time, by raising revenues via specific dollar levies on imports, they offered a new channel for establishing protection for domestic firms in a period when traditional QRs and tariffs were being eliminated or reduced. The EBFs then were, at least in part, a possible vehicle through which groups standing to lose from the reforms could safeguard their interests and slow the structural adjustment efforts. Other areas of the reform program show similar patterns of pro- gression followed by some regression. Public investment was reduced and rationalized in the years of the SAL program prior to 1985. In that year, however, public investment skyrocketed. In agriculture, steps to reduce credit and input subsidies that had been steady and successful until 1986 were partially reversed in 1987. A gradual reduction in export subsidies begun in 1984 was reversed in late 1986 because of a decline in export earnings. These revisions, however, occurred after program termination and must be seen in the context of looming elections--more as political expedients than as changes in development strategy. This pattern of progress and retrenchment is understandable if one views the Government as an entity made up of groups with conflicting inter- ests. Efforts at reform will necessarily meet with resistance and the creation of new obstacles. A well implemented adjustment program must be cognizant of this dynamic and seek to offset it. In Turkey this was not always the case. Although the later SAL documents were aware of the exis- tence of EBF9, no efforts at controlling their detrimental effects on the trade regime were attempted. Of course, the benefit of hindsight makes it much easier to identify those institutions or changes that would serve to undermine adjustment. Given that groups that stand to lose will attempt to protect themselves in unforeseeable ways, and that when implementing a program one does not have the benefit of hindsight, it may be reasonable to simplify or limit the scope of SAL programs so as to insure that effective implementation and monitoring can be carried out. For example, the spurt in public investment in 1985 points to another problem, that of relationships between elements of the program. The rapid growth of public investment in 1985-86 was in large part due to EBFs and municipalities, as well as the SEEs, rather than investments by the central government itself. All had been given more decision-making authority, authority that was deemed necessary to improve the operation of each of these entities. In the absence of a well thought-out, medium-tern - xi - strategy (not in the sense of projects or targets, which abounded, but in the form of a policy-oriented decision tool containing interrelated policy measures and their expected results) there was insufficient check on decen- tralized authority and an investment boom ensued. This episode illustrates the need for an integrated, analytical framework for the SAL program that takes into account Interactions among its elements. It also suggests that the scope of the program be sufficiently limited so that such a framework can be produced. While there were major accomplisbments in design, sequencing and implementation of the policies and measures undertaken in conjunction with the five SALs, a number of issues surfaced during the evaluation process. Because of their technical nature, they are only briefly summarized here. They are discussed at length in Chapter IV. High real interest rates turned out to be a central issue. The policy of gradual depreciation of the Turkish lira and financing the fiscal deficit kept real interest rates high relative to world levels. Over the latter portion of the SAL period the real lending rate in Turkey was about 10-20 percentage points above the world rate. High real rates also increased the service on the domestic debt, adding to the fiscal deficit. High real rates contributed to or even determined the continued low level of private investment, ultimately which limited growth in export capacity. With a current-account deficit financed by borrowing abroad, slower export growth means a rising debt service ratio. Sigh real rates also contributed to the fragility of the financial system as non-performing loans accumulated. These conuected problems of instability at the macro level could have been recognized early in the SAL process by the macro analysis that was needed to produce an analytical, medium-term framework for the program. Once they were recognized, policy could have been formulated to deal with them. At the sectoral level, major issues arose in connection with (i) financial market structure, (ii) the EBs, (iii) trade liberalization, and (iv) reform of SEEs. The banking system remained oligopolistic. This characteristic is reflected in the tiny share of business done by foreign banks and in the large interest differential the Government pays on its borrowing from the banks at "competitive, auctions. The EBFs weakened budget control by taking activity off-budget, while they partially dis- torted the trade regime via their variable and specific import levies. The import regime has been made opaque by the E3B levies, which account for one-fourth of all duty revenues. The EBF levies are set in ways that ef- fectively protect import-competing SEES. The.SEE reform policy, however, dictated that they set prices to maximize profits. Rather than increasing efficiency in the SEEs, the combination of the reform policy and the EBY levies created protected monopolies of both the SEEs producing tradeables and those producing non-tradeables. Many of the sectoral problems can be broadly viewed as a reaction of economic actors to a structural reform program that runs contrary to their interests. Their reactions could have been anticipated in SAL design, and early recognition and policy action built into a program capable of mid-course correction. - Zii - Imact of the Structural Adiustment Pogram Even with an elaborate model it would be difficult to analyze the effects of changes in the institutions, outlook and information that were a critical part of the SAL process. Instead. one can examine the performance of the Turkish economy in those areas in which SAL policy reforms were focussed. The immediate aim of the reforms implemented in 1980 was economic stabilization, in the sense of a reduction of the high rate of inflation and accompanying external payments deficit. While strict causality cannot be established between reforms implemented during the SAL program and changes in major economic variables, it would be correct to presume that the policy reforms did have a positive effect in bringing about change. The excess absorption that characterized the economy in the years before adjustment, was significantly reduced through a reduction in the growth of consumption, primarily through a fall in real wages and interest rate increase; consequently, the current-account deficit in the balance of payments was reduced sharply. The major factor in this performance was a rapid increase in the growth of merchandise exports - particularly nanufactured exports - which grew phenomenally - 23.2Z annually in U.S. dollars - between 1980 and 1985. According to the figures published in the World Bank's World Development Report, 1987, this was a higher rate of growth than any other country during this period.1 TURKEY: KEY MACRO-ECONOMIC AGGREGATES, 1980-1987 1980 1981 1982 1983 1984 1985 1986 1987 Estimate Annual Averaze Growth (percent) Real GDP -0.5 3.6 4.5 3.9 6.0 4.2 7.3 6.8 Consumption -3.4 -0.6 3.7 5.1 4.9 3.0 9.6 7.2 Fixed Investment -10.0 -2.0 3.4 2.7 5.5 12.5 13.9 8.0 Inflation (2) (year average) 110.9 36.8 23.1 31.4 48.4 44.0 34.6 48.8 In Percent of GDP Exports 5.1 8.3 10.9 11.3 14.3 15.1 12.8 15.0 Imports 14.2 15.6 17.0 18.4 22.2 21.8 19.3 22.0 External Trade 19.3 23.9 27.9 29.7 26.5 36.9 32.1 37.0 Current-Account Deficit 5.6 3.6 2.2 3.6 2.8 1.9 2.6 1.4 Budget Deficit 3.7 1.8 1.7 2.6 5.1 2.8 3.2 3.8 Public Sector Borrowing Requirement 8.8 6.0 4.0 5.7 6.8 5.2 7.1 8.7 Unemployment Ratio (2) 14.8 15.2 15.6 16.1 16.1 16.3 15.5 15.0 Debt Service Ratio (M) 28 24 25 28 25 29 36 38.0 Total Debt/GNP (Z) 28.0 28.6 32.8 35.6 41.5 47.9 55.9 57.6 Gross External Debt (US$ billion) 19.0 19.2 19.7 20.3 22.0 25.8 31.8 33.1 Source: World Bank. l/ See World Bank, World Development Report, 1987, (New York, Oxford University Press), Table 10, pp. 220-221. - xiii - The high rate of inflation which characterized the onset of the program was also reduced, despite pressures arising from the depreciation of the Turkish lira in terms of foreign currency, rising interest rates and pricing policies of SEEs. Debt management efforts were less successful. The debt accumulation necessary to finance the current deficit, combined with the servicing of debt that was previously rescheduled, has meant that the debt to GNP rat.a\ has risen from 342 in 1980 to about 572 in 1986. and the debt service ratio has correspondingly risen from 25Z to about 361. Sharp currency swings have also meant a higher than expected dollar equivalent debt base. A factor behind the rise in the debt servicing ratio is the shift in the blend of outstanding debt with the share of short-term debt, rising from 131 in 1980 to about 222 in 1986. In sum, therefore, macro-economic management still continues to be a difficult task. Policies of trade and financial liberalization were implemented as integral components of the structural adjustment process in Turkey. Re- flecting the greater openness of the economy, the siare of foreign trade in GDP increased from 192 to 322 between 1980-86. An encouraging feature of structural adjustment was the increase in the share of industry in GDP and a marked rise in the share of manufactures in exports. Turkey also ex- panded its share in all its traditional markets, captured new export mar- kets and increased its share of international trade. However, Turkey's export structure is still focussed on a few traditional manufactures where capacity utilization rates border on 702-80.- A major effort to diversify Turkey's export structure will require substantially new investments now. This is crucial for longer-term development. It requires a reduction of inflation and lowering of real interest rates. The structural adjustment program has made less of a dent on the performance of SEEs. These enterprises have improved their financial per- formance as a result of being given the right to raise prices in an oligop- olistic market. But there is no evidence as yet of a general rise in pro- ductivity. This will need addressing in the future through appropriate policies. At the institutional level, the structural adjustment program contributed toward the strengthening of the Government's capacity to design and evaluate the public investment program and to streamline its debt moni- toring and recording systems. The establishment of the Capital Market Board (1981), the introduction of a new framework for the Stock Exchange (1983), and legislation of a new banking law and regulatory system (1983-85), represent major contributions to the strengthening of the process of efficient financial intermediation. Work in these areas is being continued in the context of subsequent financial sector adjustment loans (FSALs). The financial markets are, however, still oligopolistic and the problem of banking costs as well as the high level of real interest rates remain to be addressed in the future. Assessing the social impact of the program is hampered by a lack of data or by its poor quality or coverage. Figures on income distribution point to a relative worsening of the agricultural sector vis-a-vis the - xiv - urban sector. Income distribution in the urban sector seems to have wor- sened at the expense of the middle class--consisting mainly of wage earn- ers. Using 1980 as a base year. real wages were 402 lower in 1986. At the same time unemployment in 1986 was about the same as in 1980, even though it could have been worse without economic growth engendered by adjustment. The decline in real wages that occurred, nevertheless, performed a constructive role in the adjustment process since it helped improve the competitiveness of the trading sector by compensating for the cost-push effects of high real interest rates. The reduction in farm price supports and the phasing out of input subsidies contributed towards public-sector savings, while the decline in the wage bill of SEEs assisted their finan- cial performance. It seems fair to point out that the deterioration in Income distribution was becoming entrenched before the program began be- cause of a failure to adjust to the deteriorating economic situation. Sustaining the Benefits Sustaining the benefits of structural adjustment requires action on three fronts. First, it is essential that inflation be reduced to man- ageable dimensions as soon as possible. This requires measures to lower interest rates through a reduction in the Government's borrowing require- ments in the bond market as well as a slow-down in SEE price increases through measures to enhance their productivity. Second, a substantial increase in private investment in the production sectors, especially those areas oriented to export markets. This too requires policies and measures to reduce inflationary expectations as well as a lessening of the process of *crowding out* in financial markets--a result of the Governments' need for loanable funds to meet its debt service payments. Third, the mainte- nance of a firm macro-economic policy framework. Efficient resource allo- cation requires continued progress towards making the trade regime less opaque; the recovery of fiscal control and improved debt management; and improved structuring of the Public Sector Investment Program (PIB). Roles of The Bank and IMf The structural adjustment process that was initiated in Turkey in 1980 resulted in severe shocks to key points in the economy and the adminis- tration. The 'rules" governing Turkey's foreign trade, the incentive system that shaped resource allocation and the role of the bureaucracy in determin- ing the nature and extent of economic regimentation, were all drastically changed. Yet the stock of knowledge available at that time to chart a new economic strategy was very slim. Apart from a few economic reports which had begun a 'fermentation processo in government circles towards the end of the 1970s, no serious sector work had been done to address the more glaring defi- ciencies that had contributed to the economic crisis in 1980. With the onset of the program of adjustment, however, the momentum of economic and sector work increased in scale and density. This work was generally well focussed and operationally relevant. It enabled the Bank to engage the Borrower in a meaningful policy dialogue and to help strengthen - ZV - key institutions. The most notable of such assistance was the reform of the protectionist regime and the shaping of new directions in Turkey's export orientation. The public investment programing system was strengthened by the provision of analytical work as well as the availability of training facilities (in conjunction with EDI). Assistance was provided for the estab- lishment of a Capital Market and work towards improvement in the efficiency of the banking system initiated. The debt monitoring system was streamlined. While these efforts at strengthening Turkey's economic management capabili- ties were essentially in the right direction and very fruitful, there were gaps. The absence of a policy-oriented macro-economic analytical framework that embodied exchange rate and monetary policy and fiscal dynamics handi- capped the Bank in providing mid-term corrections to the policy prescriptions and the program. Delays in initiating appropriate work on debt issues pre- vented it from providing advice and assistance to Turkey in managing as well as curbing the emerging debt service problem. In the same vein, the absence of a suitable monitoring device for measuring the social impact of the pro- gram has regressed the ability of the Bank to advise the Borrower on the appropriate countervailing measures required to target assistance to the poor and less privileged. The Bank coordinated its program with the IMF in a systematic fashion and the existence of the Fund's program of economic stabilization provided an anchor to the Bank's prugram of structural adjustment, but the evaluation points to the need for improved coordination of their respective work with respect to macro-economic policy analysis. Developments in the economy and in economic policy since after the two programs were terminated (1985) also raise questions about the timing of the severance. These are discussed in greater detail in Chapter VII. The Bank committed over $2 billion in assistance to 28 projects in Turkey during 1980-85. Many of these contributed directly or indirectly towards attaining the balance of payments objectives of the adjustment pro- gram over the medium term. Project loans were supplemented by sector adjust- ment loans (SECALS) in agriculture, energy, and financial markets. The Bank's net financial resource transfer continued to be positive but showed a declining trend. TURKEY: WORLD BANK ASSISTANCE, 1989-86 (USS milI ton) Comiltmente Repayments Policy Based Loans Bank Total (Principal Project Sector Total Exposure Disburse- and Net Year Loans Loans SALs ComItAents X ments Interest) Transfers 1989 8A1.9 - 275.0 618.0 7.7 812.6 188.5 179.1 1981 280.0 - 89.0 56.9 9.8 464.0 188.6 290.4 1982 848.8 - 804.5 647.8 10.7 56.2 209.8 298.9 1988 878.9 - 80.8 674.7 12.2 486.8 278.9 212.4 1984 418.8 - 878.9 794.8 18.1 628.5 84.1 28.4 19085 588.5 8W9.9 - 888.6 16.5 686.1 411.1 225.8 1986 975.9 891.9 - 1275.9 18.4 684.5 562.7 71.6 Source: IBRD, World Debt Tables, 1986/87. - xvi - To sum up, the Bank's involvement has been timely, appropriate, and helpful. Its economic and sector work undertaken during and after the pro- gram of SALs has, overall, been rigorous and relevant, though there is, how- ever, a question about the exact relationship and division of labor between the Fund, the Bank, and the Borrower on the development of the critical macro-economic policy framework. The Bank's contribution to institution building so far has shown dividends and its continuing assuciation in the context of SECALS is improving the Borrower's capability to implement its economic strategy efficiently. The supervision that was undertaken during the life of the SALs and the ongoing dialogue in the context of economic and sector work strengthened the program of stabilization and growth; the estab- lishment of a Resident Mission in Turkey could help in this process. Conclusions Recommendations arising from an evaluation of the five SALs are summarized belows SAL Design and Strateny (a) Technical Issues o Close attention to the development of a pelicy-oriented, macro- economic, analytical medium-term framework very early in the SAL program was necessary and would have lengthened the program's perspective and strengthened its objectives and its sustainabil- ity. The need for a medium.-term framework was cited in SAL III but one--in the full sense defined above--was not produced. The process of formulating a medium-term framework would have identi- fied the problems of stagnant private investment and macro-econom- ic instability (discussed in Chapter IV). o Greater focus on debt management, in addition to debt monitoring --both domestic and foreign--should have been a visible component of the program for macro-economic adjustment.. This would have brought the problems of potential fiscal and balance of payments instability to the surface earlier. o The mechanics of macro-economic adjustment within the framework of a liberalized trading system and freedom of capital movements should be investigated closely. The problems in the financial system and the connection between exchange rate policy and real interest rates would have been revealed early enough by such an investigation. o Close study was seeded during the design of SAL V of the impact of EBFs on the budgetary process, foreign debt, and the trade regime, though in all fairness, their negative aspects were less evident at that time. The existence of economic organizations such as the Extra-Budgetary Funds to provide the appropriate public services is not in question here. The problem lies in the technique for their financing, especially through import levies. The levies also reduce the transparency of the tax system. - xvii - (b) StrateRy Issues o A further SAL should have followed upon SAL V, as proposed by the Country Department. Lingering problems remained on such issues as debt management, tariff liberalization, budgetary processes, pub- lic investment management, overall demand management and the con- trol of inflation. The timing of SAL V could have been linked to a medium-term macro-framework; establishment of the medisAm-term framework would have reaffirmed the Governments's commIcment to the combined programs of stabilization and structural adjustment. Implementation and Sustainability o Close attention is necessary to monitor problems and evaluate results during and after a program. A suitable mechanism should be devised for this purpose at the commencement of the exercise, as it could serve as a tool for any necessary mid-stream correc- tions. o Statistical and qualitative data relevant to the design of a program need to be readily accessible to the Bank, so that appro- priate countervailing or other measures may be built into the design and, where appropriate, the conditionalities. o Conditionalities imposed on the borrower should be within its capacity to fulfill in terms of the prescribed time horizon, as well as the technical, administrative and financial resources available to it. Also the conditionalities should not be so opaque or ambivalent as to be meaningless to monitor. SALs II and III contained conditionalities that were perceived to be excessively rigid; these were later corrected. The conditionality clauses relating to performance in SALs IV and V were too opaque and too loose, allowing different interpretations by the Bank and the borrower, respectively. "Satisfactory progress, left too wide a scope for discretion or arbitrary interpretation. o Sustainability of a structural adjustment program of the kind executed by Turkey depends, inter alia, on at least six crucial elements: - First, the existence of a political environment that is con- ducive to the implementation of the spirit and letter of the agreed conditionalities, during and after disbursement of the loan. This couitment was solidly expressed and adhered to for most of the period covered by the SALs, with some excep- tions in 1983. - Xviii - - fSecond, the existence of a suitable analytical framework that permits both the Bank and the borrower to identify and re- dress potential problems as they appear. This framework should integrate the micro-economic aspects of structural adjustment and the macro-economics of stabilization as com- pletely as possible. - Third, a close and continaous dialogue between the Bank and borrower, based on the agread analytical framework, needs to be maintained at all times and followed through after the loan is fully disbursed. A mutually rewarding interaction has occurred, supported by operationally meaningful economic and sector work and related sector adjustment and project loans. - Fourth, a solid institutional base needs to be created within the key ministries and agencies of the borrower, to keep the program on track and to ensure continuous overview, follow up and needed additional efforts. - Fifth, a favorable international trading environment, fres of discrimination against exports, needs to be maintail.ed. While market access has been generally good for Turkish ex- ports, enabling it to dramatically increase the share of exports in GDP, a resurgence of protection in foreign markets has restricted access for some commodities. - Six, careful attention needs to be paid to ensure that the resource transfer does not turn negative till the fundamental adjustment process has been completed. Overall Assessment The structural adjustment program that was implemented during the five years 1980-85 had as its cardinal objectives: (i) the attainment of a sustainable improvement in the current account of the balance of payments, and (ii) an improvement in the efficiency with which resources were allocated. The first was to be achieved by changes in the macro-economic policy framework and the second by making the trade regime more neutral. Recognizing that structural adjustment had to be predicated on economic stabilization, reduction of inflation became a major goal. Performance on the current account of the balance of payments has been an outstanding one. Exports increased dramatically and the share of exports as well as of foreign trade and industry in GDP, rose markedly. Turkey increased its share of markets in industrial countries and its share of world trade also rose. Turkey also expanded its share of the interna- - xix - tional market for construction contracts. The progress achieved so far represents genuine and long-lasting improvements, offering the promise of rapid economic progress in the future--but this only if export capacity is stimulated to expand. Important institutional developments have also occurred. The establishment of a bond market and the creation of the Istanbul Stock Exchange filled a vacuum that existed in the intermediation of financial savings. The rules and regulations governing business prac- tices were strengthened and the banking system benefitted from a streamlin- ing of the auditing system as well as from improved supervision. Some of the negative aspects of captive linkages between the banking system and industry were addressed. Sector loans in agriculture, energy and finance have improved the policy framework and strengthened institutional develop- ment in areas vital to longer-term adjustment and growth. Improved effi- ciency in these sectors will provide avenues for enhanced employment of the rapidly increasing labor force. In keeping with the philosophy of re- stricting government activity to the provision of an appropriate environ- ment for free enterprise, the process of *privatization, of SEEs has been the object of focus. There is sufficient prima facie evidence, therefore, for the con- clusion that while much has yet to be accomplished, major goals of struc- tural adjustment have been reached. The task of economic stabilization continues to be a difficult one as the economy wrestles with a high, albeit reduced, rate of inflation, high interest rates, high debt service and high unemployment. Consolidating the progress achieved so far and generating future economic growth, therefore, depends crucially on moderating the growth and size of the external debt. Attention hLa also to be paid to improving the framework for domestic financial resource mobilization and recovery of full fiscal control. Demand management is critical, since it has proven to be one of the causes of the high level of inflation. Final- ly, Turkey's continued economic progress and stability depends on the con- tinuation of a free international trading environment and enhanced net resource transfers. Bank support during the adjustment process has been appropriate, timely and well sequenced, overall. The policy dialogue has continued to be strong throughout the program and a spirit of trust has been built into the relationship. TUE SiTUCRARAL AD.DuSnert LOANS I - V SYNDPTlC OVERVIEW OF EMPHttlNCE 1. MIM S 110E Dld Actual. Reasons for Deviation Economic and A gel l-focused economic Modest focus on sector-wide Macro-economic analyeis of exchange rate. monetary, Sector Work and sectop work program seues. prior to 1900. Sconomic fiscal. and debt dynamic. ws deemed to be the In fielda strategic to reports finalled in 1979/80 province of the IMP. Rigorous work in this area the adjustment program. dealt with maro-economic and began in the Bank only in 1906. public sector investent problems. They initiated a Oferentation procea.s Scale and density of ESt accelerated as the SAL pro- gram developed. Work was in general wll focussed and operationally relevant. Never- theless, a range of crucial ale- met& in monetary, fiscal, and exchange rate analyias and, in particular, their interrelation- ships were Insufficiently analyzed. Nature of Implemntation of princi- Borrower demonstrated a high Termination of the SAL program after 1988; insuf- Policy pal recomendations in degree of commitment to adjust- ficlent mcro-*conomic guidelines built Into Dialogue. Et and policy adjustments sent during the life of the SAL Sector Adjustment loans. Political exigencies in SM and supporbing program. While wacro-economic related to Impending general elections (November eector projects. mnagement has suffered somewhat 1987). Eergence of Extra-Budgetary Funds. recently, overall OT commitment to stabillamion and adjustment continues to be strong. Design of Support policies intended Progree was appropriately focussed Insufficient recognition of the relstion- Program to shift froe an inward- and sequenced in an overall sense. ship between the exchange rate mad mgnetary- looking development 0missione: Insufficient focus on fiscal dynamics, handicapped the development etrategy to an outward- overall debt management policies, of adequate debt management policles. Problems oriented market econogp (even though debt monitoring were noted in CE4's and Preeidenta* Reports but emphaszaing export-led systems were wel I developed) and comprehensive studies were Initiated only after growth. Devise strategies development of a policy-oriented. termination of the SAL program. There was a to reduce inflation and analytical wedium-tare macro-eco- Bank asnagement perception that the policy ensure a better balance nomic framework, early in the life framework for structural adjustment could be between resource avails- of the program. Flexibility for anchored within a space of five SALs and could bility and use. Improve mid-term policy corrections was sustainable afterwards. fiscal and financial limited. The horixon of adjust- framework; strengthen went was limited to five SALs performance of State Eco- while it was clear that Turkey nomic Enterprises and de- needed more time to adjust and velop energy resources as corrective measures needed to be well as agricultural and applied for a longer period. 国国国国国国国国国国日日国国日日国国国国国■■■国■■園国■■国園国■国■■■■＿ ・―;，、，・―----―。：…―。―---&&--―・・，・―---，・，…―--,;,,; elme =d Ab MT6 Pø~ . &ok prffemd Froff~lon af målle =~ Invo~s Mom øf P&116 0~ øf ~ mål le In- * %hl 16 9~ hvoolmønt R«;« and r~r« mr48~ e~ bi, 0» ~h~ P~ræ; 6~ Examleo w~ rff~ ;4 ør~ at w~ wd~rø %Bd* W4 olob P4OPIM" wd em- liucc~ ul øff~ " *,D ex~ l*a^m uf MMICIPOINlag. Snoffec41v8 **t I« priarlikr proj~ 3 r~ lowal øf "Ile tnvøøt~ ~Iteri" et ~ 1r4 prøje~ øød 6~ 1- ahlft acchaml G~ 1416» duriv4 a" U end M. Thi* wo *Ømm, J~ Jelæt le"l" onal pluming at lava«~ frøm emm- «~ I»d In ~ z¥ and V. tut ~ Il Itv la 0~ 109. al"I~Iff »d fachuriftil *D *~ Øl In- ~ 49 ud at WL V mm l=41 af "Inawan øf lava~ t, pmo~ øg. frook~ re; lounmo public Invas~ t accolm~ , *mi*;~ øf ~ 10 In- omkvik"Ing bc d~pler. lam cif vo~ . *,* ømc 0 lanco In ^ K~ =Orm detielt oweral l. *~ wo In onorm Enenw omm~ lo" proc~ en Dmølø~ & %k~ Iff"Mod Priolne Fricifte. 80, In 6~ ~ Øklocæml b*al*. Obo~ øf ckilled C~ ftakløft polialeg. ~rø . r- =rvstion. lhø~ effør" to d*- engift"ra md tochalelwmw hW~ Mho ~Ion =d f *mb an øf v*ter ;n"l"Imel ec"billikr. pro~ 4ffølal~nt. mwlarablan. An en- w 4~lov prollrom le b*, nu umporl-1 by 5~ Adju~t Lmm. Aarloultur* oblælsto offiel~ Public lftv~ 6~16~ *;tb ~er~1 im priel"e pollezý and 0~ offøl~ P~ løft hy laprW*d loprwing prø~ vllw mød affl- omkaidl~ akhrib~ bat ØDT fjo ørielng pol 1c14w.and CIM" la~ hoblad te 0 bo. d~ 16m410" In tø~ øf t~ h~ r c~ alm, M4 sloo hf^ ob"dy "røm~ lnø af faromre. og Inceative abriachure la"Itutiong. Affar eerly 9~ ",h pram~ *up~ In ar~ en prlaing pol lel«. 9~ Wicalibure. pol lø rwonwin occorred In 111117. ewom øf xWro" ~rb-torffi fimm- l:mfita hav* Ifter~ med mor- P~Itø r~ prinaleml lp due %D price m~ ~ C cial *~ ro. r*di~ mil hør"Mingl clelm m hu~ Inerwoooo and m*lalt"lan af G~ ly/ &hw Fri~ In~lø~ prognumg Gr4 hav* bom ~ cad. tb ovId~ øUlpoffliame FK,ýr. fi - tb= f~ ~ ømt pr~lvltw bo* å~. A b~ ~- c~ m~r plan for priva%lzoblan af røle øf 8~ to om~ lm ~ hao b~ propend; «id~ pri~ omher. øf proor~ In *his 6~la ~r~ øf 4~16 socell~ Frø~ In ~ at Tr"e pr~loft In key ~rk~ m4 ør~ ; P~ lan af & ~ omme. derIvifte fme tb* doellning 6~ ø~ 11h1« In mrkot *~ 1 an In- er*ation øf en Ø~ lvø la~ 9~ zr«. C=tinuing 40~ 10 Inflotion. duero at Ihlv* fe wm~ k. sløv @r~ af pri"te j~**kwolb In 9~ o~ 191 ly -Uat . In d~ aft 996~ . GW end mø ;her~ ~rø af In~ In CM. Eklonce at Pr-whe 8~ ^- 6=11~ 8~ ~Orm~ vith D~ mm~ t pol l aloe ~ *%røn~ F6smeftko 6 =§h alth ;0w0 i mimrp Inar~ le *ni~ at min@ aum -4- ~b Info~ m Ilhaml~ lo" W4 p~ 46 flemof~ ma. sobs~ 191 rloo " i h» ab~ §md. ~ I~ at & ønw øf ~ Ihlvøe at In «» dmøre øf lo~ fin~ dob* Øwe to bo omkir~ . ubleb " mm In 8 flow- byen~ . ~rønt mac~ p"- Ible ozo~ rolmø øøllgW. ø~ ~Ialt kalved. Lmm ~ - Kalil - gel roadg Rseasons for Deviation Overall GbJetives a1tisfactory performance on debt Reduction in current ac- mnagement: increase in debt count deficit and improve- service ratio and debt to ONP ment In debt service ratio. profite. Financial Reduce high rate of infla- Capital mrket and Stock Exchange Inadequate #lecat management has forceded Sector tion and atrengthen bank- established. Auditing and hank out* private demand for lomnable funds. Devalopment ing aWatem and financial supervision improved. Neverthe- whil keeping intereet ratee high and pri- and Inflation market. ees. banks saddled by non-per- vate investment below necessary levels in forming lomee; high Intermedia- priority sectors. Inflation has been the tion costs and high Interest reeult of coat-push as well s demnd-pul rates. manking system also con- factore. atricted by lank-industry link- ages. Marked decline in rate of inflation initially but signs of resurgence lately. Social While recognizing that While W grew strongly. consump- Program did not provide for eatiefac- Impact etructural adjuetmant en- tion grew at a slightly lower tory mnitring mechaniem to examine tall social costs, mir- rate. but *till higher than its social Impact and to target counter- tain as far as poseible, groeth in simiter countries. De- vailing sistance. Program was also the physical quality of terieration in the urban-rural accompanied by declining expenditure. life. term of trade. Higher concen- on health and education. Inadequate tration of income dieparities in growth in private investent In production urban areas and amore even distri- sectors and high rates of laterear Inhib- bution of income in rural areas. ited employent expanloan activities in Wages and salaries as a share of urban and rural areas. total value added fell white prof- its almost doubled, principally due to operation. of State Economic Enterprises. teploymint remained at the sm level as at the com- mancement of the program. Sharp fall in real wage. in manufacturing s*etr. 4. SUSTAINASILITY Policy Implement policies end The high level of real interest Delay in integrated, policy-oriented Development measures to increase the rate. constraina Investment and macro-econoaic analyasi of exchange rate, flexibility and efficiency compounds the debt service bur- Interest rate, debt problems. Eaing of of the economy to adapt den, both domestic and external. stabilization poliaies after termination of to changes in the economic The high level of external debt, DF and SAL program. Inadequate mr-eco- environment. Specifically, the medalitles of financing debt, nomic policy conditionality In sulsequent to open the econoov to and the rise in the debt service sector adjustent plane. Protracted develop- International competition ratio are complicating the eater- mant of analytical mdius-ter macro*conomice reduce direct controle of nal payments altuation and my frameork to complement the atructural adjust- the financial mrkete; re- ultimately create a constraint on program. duce the Government's Investment and economic growth. role in produciln and High level. of domestic Interest distribution of gion-public rate. are also increasing the goods. and reduce the share of non-performing oane in fiecal deficit. bank portfolios and weakening the etructure and viability of some banks. - KaIv - gi RAmna f>r Deviation Ineitumnal Streng~n C tcro-ecenosio Cupilal iteend stnbul Stock Fiac,. budgeary and public investmont Development pollcy analyale capbiliity Eange establiahed. bak om- plannlng proe have been cneined s mal l se er plenning 9ervielan and auditing e~ by frag.snlelion ef polley wshing proce s and public inveteon pro- atrengthoned. Døbt Riecrding with e estabi n oaw ef amtonos graming. Strengthen in- ty~etee 99opulerlmed and elremr- agenci e (Era Et dge"ary Fund.> Whick e1tutional fraoe~erk for lined. Capacity for evalumtion have seigned revenue saurces and powers finanelal resurce oablI- ef public inv~esteen projects to aptnd. DsMå ~naeent polilles need action. ltvitalize iscal etrenglhened. Fical aya~am o*ronge~ling. Stlat ulbilitieo and and budgelary Ørocees. aeangthned by inroduction of sarvice enterpies have 0onopoly pricing lirove det mnagesment Value Added Te* (VAT). Aselo- powere hile Stel* Economic Enterpri~ee capabi lity. Strengthen tance provided in the caslet ef are *hield d by entry barriere. except efficiency ef State Eco- promting privabaionf intat In a fe cases. namic Enterpriesa. Econesic lEntrprieos. Support provided to *ADog~ for effi- iency loprovement in teatile 6ector. 8. ASSES The atructural adjusteno program had to broad Ghjeives: (i) oatinment ef a austainable improvement in the current account ef the balance af paymenta, and (l) an iaprevement In resoaurce allocative etffciency. E~anorl etabi liatiMn mas a nece~ary and complementery gael. Mhm balance ef paymnte performance an current aceun ham been autotanding and Turkey's ahare ef vorld trade has rien. The e ~nay hae becamb møre open and te e ~ructure of produchtan kao akifted sway from an excesaive dependence an agriculture. Develapment a band arke6 and aefsbl imet of he tock Exchang have etrngthened financial interoedlatln and slapo towarde improving the auperviston and auditing procedures for cmmercial bank. and change governing hank-indu~ry l inkages, have reduced the passib I ties at financial criee. Socter loen* In agriculture, energ and finance. have abrengthened the pol icy fraeork and intiutilena in anes critical to long-ter. grmeth and adjuatmon. The pro øf privatization ef state economic enterprises continue. The ab~ence of releble Houaehold Survey data handicapped an asseemnt ef the ascial impact of the program. l e clear, how- ever, that the remeval ef price diatertlone and oubaidlem ~cabined with reduced epIndlures an health and education, and the fal i in real wages thab aceapanied the proga., advormely affected as groupa. On the other hand, all egasnta of ociety benefltted from the increaed neutrality in the trade regim the reduction in inflation. epansion in eoleo activity and the maintenance et a high rate of ecanemic groth over a noven year period. l% weuld b* correc6 to infer that vithou etructural adjus~tanøi alt sag- ente of Gociev ould have been orse off over the long run. To mu up, shille mre hae ye6 to ba accompliahed, major gste ef etructural adjustmant have been reched. Caneotidating pro- grema, however. requiros maderaion of the ete and growth of *x1ernal debt. rmduction ef te high rate af inflation, further is- provemn in finoncial reaurco mobi iation, recovery øf flecal balance and cantret over budeatary proca- and øraeen of investent in general and in the export and eecial eectora in particuler. gank aupport during the edjustmsnt program has been oppropriato, timely and well squenced. overall. The pol lc, dialogue hae been etreng and * epirit ef trua has been built up. The establiahmen ef a Reasident Kiasion hould help to atrengthen the relationohip. EVALUATION OF STRUCTURAL ADJUSTMENT LENDING IN TURKEY PROGRAM PERFORMANCE AUDIT MEMORANDUM TURKEY FOURTH AND FIFTH STRUCTURAL ADJUSTMENT LOANS AND OVERVIEW OF SALs I-V I. BACKGROUND 1.01 Turkey's economic policies from the early 1960s until 1972 contri- buted to the achievement of rapid growth in GNP (6.6Z a year, 1963-72), with substantial growth in industrial capacity. During the Third Plan period, 1973-77, Turkey maintained, even increased, its growth rate (7.7% average, 1973-76) with increasing recourse to external borrowing, accumulating a heavy load of short-term debt. 1.02 After the mid-1970s, however, external constraints to sustained growth became increasingly severe. They included a sharp rise in Turkey's import prices (including oil), and a significant deterioration in its terms of trade. High interest rates resulted in an increased external debt burden, and recession in Western Europe led to reduced export prospects and migration and remittance possibilities. Continued expansionary policies delayed Turkey's adjustment to the changed international environment, so that by 1977 the country was in a severe foreign exchange crisis. After 1977, the govern- ment made a series of efforts, exemplified in stabilization programs worked out in close consultation with the IMP (1977, 1980, 1983, 1984), to restore price stability, creditworthiness, and a sustainable balance in domestic and external finances. 1.03 Toward the end of the 1970s, the financial situation deteriorated again, partly as a result of the 1979 increase in oil prices. In view of the severity of the external financial squeeze, the Government accepted in January 1980 the need for a temporary sacrifice of growth and of social ob- jectives in order to bring down inflation, 1mrove the balamne of paymentt, and restore creditworthiness as quickly as possible. 1.04 In development issues there was a change in sectoral concerns to favor energy and agriculture, at the expense of manufacturing, especially as far as allocation of public sector resources was concerned. There was a new emphasis on the division of labor between the public and private sectors, with the former expected to concentrate on the provision of infrastructure, while the latter was to provide the main impetus to industrial growth and exports. In the external sector much greater weight was given to exports relative to import-substitution. Finally, and most significant, was a shift toward reliance on market mechanisms as against centralized planning. - 2 - 1.05 After the announcement of the January 1980 program, the IMF released the remaining tranches of the Stand-by Arrangement that had been agreed in July 1979. Soon after, in March 1980, the first Structural Adjustment Loan (SAL) (US$200 million) was signed making quick-disbursing funds available to Turkey in support of the reforms already initiated. The major objectives of the SAL program were to reduce inflation, increase foreign exchange earnings. and improve domestic resource mobilization in the public and private sectors. Priority was given to export incentives and institutional arrangements in support of exports; studying the protection and incentive system; improving external debt management; eliminating the overall public deficit in 1980; reviewing measures to encourage private savings and time deposits; harmoniz- ing public investment and adjusting its composition to the new priorities. 1.06 In June 1980, a new three-year Stand-By Arrangement was signed with the Fund involving SDR 1.25 billion (about US$1.63 billion), with SDR 460 billion in the first year, SDR 400 million in the second year, and SDR 390 million in the third year. The key conditions of the new Stand-By were that the exchange rate was to be kept flexible; the financial position of the public sector was to be improved, mainly through restructuring the operation policies of State Economic Enterprises (SEEs) (i.e., raising prices); mone- tary conditions were to be stabilized by observing strict limits to Central Bank lending; and interest rates were to be adjusted to reflect market condi- tions. The conditions set by the Bank and IMF were mutually reinforcing, with the Fund taking the lvad on exchange rates and aggregate demand manage- ment issues while the Bank focused on public investment and the protective system. There was extensive exchange of information between the Bank and IMF during the preparation of all SALa. 1.07 The Bank and Fund program that began in 1980 combined stabilization and structural adjustment. The Fund focussed on the stabilization aspect of the program at the macro level, and the Bank on the adjustment aspects at the sectoral level. The combination in Turkey seems to have been successful. Real output and exports grew rapidly after 1980, the exchange rate was ad- justed effectively, and the inflation rate was reduced. However, the appro- priate sequencing of stabilization and adjustment programs was not addressed formally during the Turkey SAL program; it remains an open issue. 1,08 In the years 1981-85, the Bank provided four more SALs, bringing the total to five. Under SALs I-III US$879.5 million was disbursed in the period 1980-83. Like SAL I, SALs II and III continued to provide support to Turkey's January 1980 reform program. Priorities of this program are men- tioned in paragraphs 2.04-2.05. SAL IV (US$300.8 million, effective September 19, 1983) and SAL V (US$376 million, effective July 3 1984) con- tinued to press for the development of a medium-term strategy that would link the SAL process to a sustainable longer-term growth path. SALs IV and V also emphasized sectoral objectives, especially in the financial, agricultural and energy sectors. 2n addition to auditing SALs IV and V. the present report presents an overview of the entire SAL program. - 3 - II. SAL DESIGNs OBJECTIVES AND POLICIES 2.01 The five SALs to Turkey were made in support of policies intended to shift from an inward-looking development strategy emphasizing import-substi- tution to an outward-oriented strategy emphasizing export-led growth. The shift in strategy was to be market-oriented, with an increase in dependence on market signals to allocate resources. The SALs supported policy programs designed to implement this change in strategy, and disbursement of the loans was conditioned on specific elements of these policy programs. The proceeds of the loans were to finance high-priority imports needed to support the development program. The design of the policy programs evolved from a focus on stabilization and re-orientation of macro-economic policies in the crisis of early 1980 (SAL I) to a balance between macro-economic and sectoral poli- cies by SAL V. The sections of this chapter first review briefly the design of the SAL I-III policy programs, then discuss the design of SALs IV and V, and conclude with an evaluation of that design. A. Background: SALs I-III 2.02 SAL I. In January 1980, the Government of Turkey announced the package of policy reforms that would evolve into the SAL program. In March 1980 the first SAL was signed making quick-disbursing funds available to Turkey in support of the reforms already initiated. The major objectives of the SAL program were to reduce inflation, increase foreign exchange earnings, and improve domestic resource mobilization. Priority was given to increasing export incentives; improving external debt management; studying the system of protection; eliminating the budget deficit; encouraging private saving, and rationalizing and reducing the level of public investment. SAL I and its Supplement were made in response to Turkey's immediate external and internal problems. By the end of 1980, the effects of the stabilization and adjust- ment measures began to appear, with exports growing and the rate of inflation falling. 2.03 SAL II. The second and third SALs further supported Turkey's January 1980 reform program, and the measures taken subsequently. The action program under SAL II marked the transition to a more comprehensive process of structural adjustment, to be pursued by subsequent SALs over a number of years. In addition to new reforms, SAL II continued to support policy re- form that did not reach full implementation under SAL I. The SAL II program emphasized: (a) domestic resource mobilization; (b) export promotion; (c) liberalization of the import regime; (d) a reduced and rationalized public investment program; (e) development of energy resources; (f) increased pro- duction of agricultural exportables; and (g) reform of the SEEs. With varia- tions in form and emphasis, this would be the agenda of the entire SAL pro- gram. 2.04 To make progress on the points emphasized above, the following spe- cific policy actions were to be adopted under SAL II by the Government. (a) Financial Policies. To improve mobilization of domestic re- sources, a major fiscal reform was planned. This included revision of existing taxes an1 introduction of new ones such as a VAT, and strengthening tax administration. Saving was to be stimulated by increasing real interest rates to positive levels. A new Capital Market Commission was to be established to develop the capital markets. (b) Export Promotion. This was to be achieved by adjustment of the exchange rate, export incentives, mainly through credit, and im- provement of the administrative structure for exports. (c) Import Liberalization. The import regime was to be liberalized by abolishing the quota list and moving many items to two liberalized lists for licensing, for commodities with and without domestic com- petitors. (d) Investment. The allocation of public and private investment was to be improved, with priority to export instead of import-substitute projects. (e) Energy. Public investment in energy was to be given priority. Energy prices were to be raised, and public agencies in the sector were to be strengthened. (f) Agriculture. Export policies, financial incentives and institu- tional reforms were to increase production and exports in agricul- ture. (g) SEE Reform. Access to the Central Bank and Treasury for SEEs was to be reduced, and interest rates on their borrowing increased. Measures were also to be taken to reduce over-staffing and to up- grade management. 2.05 SAL III- The third SAL was approved in May 1982. before the com- pletion of SAL II in November of that year. The SAL III program continued implementation of the basic agenda set out in SAL II. The program was downsized into five major program areass (a) Macro-economic Policies. These included further tax reforms and capital market development, continued exchange-rate adjustment and measures for export promotion, and further import liberalization. Cb) Public Investment. The rationalization of the public investment program begun under SAL II was to continue, with a reduction in the number of projects and reorientation away from sectors that were expected to attract private Investment and toward infrastructure, supported by a strengthening of project evaluation capacity. (c) SEE Reform. A major reform package for SEEs was to be designed and enacted, aimed at improving their performance by increasing their autonomy and market orientation; interim reforms were to continue. (d) Agriculture. This sector was to be given priority in the re- oriented public investment program, to promote growth including exports. (e) Eneray. The energy sector was also to be given priority in the public investment program, with the aim of reducing imports. B. Design of SAL IV 2.06 The SAL IV and V programs continued the agenda set in SAL II, with a somewhat more comprehensive set of adjustment policies. The general line of policy remained to shift from an inward-looking to an outward-oriented strat- egy, and to place greater reliance on market forces to guide resource alloca- tion. The points of emphasis were the same in SALs IV and Vs (a) a realistic and flexible exchange rate; (b) incentives to encourage producers to export; (c) tight monetary conditions to restrain domestic demand and control inflation; (d) deregulation of interest rates to encourage private savinges (e) steps to improve institutional efficiency in key sectors (agricul- ture, energy, finance), CM) rationalization of the public investment program, (g) initiation of SEE reforms to reduce the burden on public finances and Improve their (the SEEs) efficiency; and (h) progressive liberalization of imports including the elimination of quantitative restrictions and the rationalization of the tariff structure.1/ ,1 See President's Report, SAL IV, No. P-3543-TU, para. 10, and President's Report, SAL V, No. P-3783-TU, para. 7. With the same points in the program in SALs IV and V, the two i ris-, be considered as one, with some, but not many, adjustments to policy s' 2.07 The policy actions in the SAL IV program agreed to by tt Ja tk ao the Government of Turkey were classified as follows (i) Agaregate Level - Medium-Term Framework - Fiscal and Monetary Policy - Public Investment - Reform of SEEs - !mport Liberalization - Export Incentives (ii) Sectoral Level - Financial Sector - Agriculture - Energy Proposed actions in each of these areas under SAL IV is summarized below. The classification of policy actions under SAL V, outlined in para. 2.18 below, is essentially the same as in SAL IV, consistent with the characteri- zation of SAL V as an extension of SAL IV. The continuity of the programs since SAL II should be evident from a comparison of this list with the SAL II program summarized in para. 2.04 above. Emphasis changes but the essence remains the same. 2.08 Medium-Term Framework. The Government was to develop a medium-term framework as part of its annual program. The Bank would examine whether the program showed continued stabilization, balance between resources and expen- ditures, and realistic projections of macro-aggregates. The program was to be the first year of a Five Year Plan that also exhibited these characteris- tics. Satisfactory progress in development of the annual 1984 program was a condition of tranche release. 2.09 Fiscal and Monetary Policy. On the fiscal side, the budget deficit would be reduced to less than one percent of GNP, and several measures to improve tax administration were proposed. Monetary policy would adhere to the IMF Stand-By Arrangement. Liberalization of the financial markets and flexibility of the exchange rate would continue. 2.10 Public Investment. The rationalization and reduction of the scale of public investment were to continue under SAL IV. The share of public investment in total investment was to be reduced from 60% to 502 over the medium term. Rationalization included redirection of the program toward infrastructure, identification of priority projects, and concentration of resources on these projects. Improvement of resources for project prepara- tion, evaluation, and selection through training was to continue. Satisfac- tory progress towards rationalizing public investment and concentrating re- sources on high priority projects was a condition of tranche release. 2.11 Reform of SEEs. The freeing of SEE prices under SALs I-III had resulted in the SEEs earning aggregate positive profits by 1982, but at the time of SAL IV no improvement in SEE efficiency was discernable.2/ An SEE reform decree approved in Hay 1983 would alter SEE operation in several ways. Output prices were free to be set by the SEEs, personnel were to be governed by a system separaze from that applied to government employees, and the SEs were to have greater autonomy in formulating investment plans, within the framework of the national plan. Budget transfers to the SEEs were to be reduced further. The SAL IV program contained no additional action recommen- dations. 2.12 Import Liberalization. Liberalization involved shifting gradually from a quota system in which imports were prohibited unless otherwise speci- fied, to a license system, and eventually to a tariff system in which imports were free unless specifically controlled. The quota list had been abolished in 1981, and items were being moved from a license list to a tariff list and a special levy list that had been established in 1982.31 Import guarantee deposits were being reduced, and license administration impro.ed and eased. Based on a forthcoming Bank-sponsored Protection Study, the Government would adopt a plan to rationalize the tariff structure to increase the competitive- ness and export-orientation of industry.A1 Implementation of the plan would begin in 1984. Satisfactory progress in import liberalization was a condi- tion of tranche release. 2.13 Export Incentives. The export system had been partially liberal- ized during SAL9 II-111, with the removal of export licensing, except for tobacco and opium. Export incentives were provided by preferential credit, duty rebates on the import of inputs, and a tax rebate in lieu of VAT rebate. A major additional incentive was the policy of a flexible exchange rate, which resulted in effective real devaluation. The resulting growth of ex- ports during SALs I-III was impressive. Concern was expressed in SAL IV that recovery of domestic demand and reduction of interest rates for export-com- peting borrowers might slow export growth. The Government agreed to review 2/ See President's Report, SAL TV, No. P-3543-TU, para. 56. 31 See President's Report, SAL IV, No. P-3543-TU, para. 62. The levy list was to grow in importance as a source of finance for the Extra-Budgetary Funds. This is discussed at length in Chapter IV, Section F. 41 IBRD, Industrialization and Trade Strategy, Report No. 3641, February 1982. the system of export incentives under SAL IV, with the purpose of sustaining export growth. Vhile external debt management was not specifically listed among the areas for policy action in the President*e Report (see para. 2.07 above for the list), the issue was discussed following export incentives. With assistance from the Bank and the Fund, the Government was to establish a computerized reporting system for external debt. A timetable was to be agreed with the Bank. 2.14 Financial Sector. Here SAL IV clearly continued the policy of the previous SALs. The high level of real interest rates was a source of concern in SAL IV, as it had been earlier. The financial crisis in 1982. when a major brokerage house failed, was attributed partially to high real rates. The Government had taken steps to reduce the cost of intermediation and re- duced deposit rates during SAL III, and would continue to pursue these poli- cies under SAL IV. The financial transactions tax was to be reduced, and preferential lending rates set at zero in real terms. The banking law was to be revised, including introduction of a standardized accounting system. The work of the Capital Market Board would continue, and the Government agreed to study measures to further the development of the equity market, and the pos- sibility of introducing medium- and long-term bonds with returns indexed to time deposit rates. The Government would begin to issue bills and bonds to finance its deficit. Satisfactory progress towards reduction of the finan- cial transactions tax was a condition for tranche release. 2.15 Agriculture. During SAL9 11-111, the Government took steps to rationalize the agricultural sector and to increase its efficiency and ex- port-orientation. Incentives for production and export were introduced. The process of replacing support prices by commodity floor prices was begun. The Ministry of Agriculture was re-organized to consolidate service provision. During SAL III, the sector provided a net trade surplus. These policy lines were to continue under SAL IV. Commodity floor prices were to reflect world market prices. Subsidies to fertilizer prices were to be eliminated over three years, and subsidies to water charges and irrigation development would be reduced. To allow the reallocation of agricultural resources towards exports, the Government planned to review its policies of self-sufficiency. Tne Government agreed to take measures to improve the efficiency of the sys- tem of agricultural credit supply, and of the sales cooperatives. An inter- agency planning group would be established. Rationalization of public in- vestment in the sector would continue within a constant share of the overall level of public investment. 2.16 Energy. In 1981, petroleum imports were 40Z of Turkey's total Imports, and equalled 80Z of export earnings. With domestic lignite re- sources and hydropower potential, dependence on imported energy sources could be reduced. The Government intended, under SAL IV, to increase the efficien- cy of institutions in the sector, including planning and inter-agency coordi- nation, to increase conservation, to focus investment on high priority proj- ects, and to increase the participation of the private sector in energy development. During the SAL IV period the Government would develop an energy action plan, in consultation with the Bank, aimed at these objectives. A new energy conservation law was to be passed in 1983, petroleum prices were to be held above world market prices, and distortions between the prices of differ- ent energy sources were to be identified and removed. The program aimed at increasing energy sources, increasing efficiency, and utilizing the private sector. 2.17 Conditionalities. Tranche conditionalities in SAL IV were satis- factory progress int (a) development of a 1984 annual program, (b) rational- izing public investment and concentrating resources on high-priority proj- ects, (c) import liberalization, and (d) reduction in the level of the finan- cial transactions tax. C. Design of SAL V 2.18 SAL V was a continuation of the SAL IV program in most respects. The points of emphasis in the program remained exactly as in SAL IV, as set out in para. 7 above. Several steps that had not been implemented under SAL IV were carried over to SAL V, most notably development of a medium-term framework. The policy actions in the SAL V program agreed to by the Bank and the Government of Turkey were classified into essentially the same areas as in SAL IV. They were as follows: (i) Macro-economic - Medium-Term Framework - The Revised 1984 Program - Public Investment - Reform of SEEs - Import Liberalization - Export Incentives - External Debt Management (11) Sectoral Level - Financial Sector - Agriculture - Energy This classification of policy actions can be usefully compared to that of SAL IV, set out in para. 2.07 above. The revised 1984 program replaces monetary and fiscal policy, and external debt management is added. 2.19 Medium-Term Framework. The Government was to complete its draft five-year development plan for 1985-89 during SAL IV; this was to serve as the medium-term framework that had been an issue since SAL III; the nature of - 10 * the planning process from a centralized framework toward a more indicative one. The plan was to be based on the themes that have run through the SAL programs outward-orientation and market forces. Provision of infrastructure by public investment, and the Importance of growth in private investment were to be emphasized. The targets of the plan would includes (i) export growth in the range of 82-112 per year, with a consistent exchange rate and incentive policy: (U1) reduction of the PSBR from its expected 1984 level of 3.5% of GNP; (iii) decrease of the share of public investment in total investment from 60Z to 50Z; (iv) allocation of public investment to high-priority sectors, and away from manufacturing (v) real GNP growth of 6 per year; and (vi) maintenance of the debt-service ratio near 252. Since the plan did not yet exist, the Bank's projections, consistent with the above, were presented in the President's Report. There was no discussion in the President's Report on SAL V of how the policies to be followed would achieve the targets and what other circumstances would have to materialize to make it possible. Completion of the plan, compatible with the aims of the structural adjustment program, was a condition for tranche release. 2.20 Revised 1984 Program. The revised program contained a familiar list of measures that were consistent with the SAL themes. These included (i) Government reorganization, with concentration of economic policy in the Prime Minister's office; (ii) Import liberalization; (iii) rationalization of export incentives; (iv) liberalization of the foreign exchange regime; (v) positive real rates of return on saving and free lending rates; (vi) reduction of the financial transactions tax and the withholding tax in interest payments; and (vii) reaffirmation that SEEs could set their own prices. - 11 - The Public Sector Borrowing Requirement (PSBR) was to be reduced from 4.82 of GNP in 1983 to 3.5% in 1984. Monetary policy would continue to adhere to the IMF Stand-By. The President's Report expressed concern about the level of interest rates and its effect on investment and stated that the Government's objectives were tot (i) increase the role of the market in the determination of interest rates; (ii) reduce the differences between preferential and non-preferential lending rates; and (iii) encourage the availability of medium- and long-term credits for the productive sectors of the economy.1] There was no discussion of how point (iii) was to be done, or how points (i) and (ii) would reduce interest rates. 2.21 Public Investment. SAL V continued the SAL IV program on public investment, described in para. 2.10. The Government was to undertake a de- tailed review of the public investment program, with the objectives of speed- ing up high-priority projects and eliminating allocations to low-priority ones, and improving planning capability and institutional performance. The review would also confirm that the program was compatible with resource availability.i/ Satisfactory progress in carrying out and implementing the results of the review was a condition for tranche release. 2.22 SEE Reform. SAL V continued the line of policy on SEE reform from SAL IV, discussed in para. 2.11. The Government was to explore selective privatization of th* SEEs, and had enacted a law providing for the sale of income-sharing bonds in the SEEs to the public. 2.23 Import Liberalization. In its 1984 Import Regime, the Government shifted from a positive list system that specified permissible imports, leav- ing all other goods prohibited, to a negative list system in which all goods could be imported, unless specifically controlled. The number of items re- quiring a license was reduced by over 50%, and tariff rates were reduced substantially. The effect of the tariff reductions on effective protection was not clear, however. In addition, the Levy list of goods subject to vari- able levies to provide revenue for the Extra-budgetary Funds (EBFs) was 51 President's Report, SAL V, No. P-3783-TU, para. 40. 61 President's Report, SAL V, No. P-3783-TU, para. 49. - 12 - broadened.?/ Administration of the Import regime was centralized in the Undersecretariat for Treasury and Foreign Trade under the Deputy Prime Minister. The Undersecretariat issues licenses, decides the status of goods between the free and licensed lists, and sets tariff and levy rates. Satisfactory progress in the elimination of quantitative restrictions and rationalization of the tariff structure was made a condition for tranche release. 2.24 Export Incentives. The SAL V program continued policy on exports from SAL IV. The exchange control system was further liberalized, with banks allowed to set the exchange rate in a 6Z band around the parity rate that is adjusted daily by the Central Bank. The surrender requirement on earnings from merchandise exports was reduced, and banks were permitted to trade in foreign exchange on their own account. The Government was to review the system of export incentives under SAL V. 2.25 External Debt. Design and installation of a computerized system for recording external debt had begun under SAL I. The design phase of this project was expected to be completed, and installation and testing to begin, with Bank and Fund support, during SAL V. Satisfactory progress in the in- stallation of the computerized debt management system was a condition for tranche release. 2.26 Financial Sector. Substantial progress had been made on liberali- zation and development of the financial sector under SAL IV. The transac- tions and interest withholding taxes were reduced, a new banking law was passed, the Treasury began to sell bonds and bills to the public, and several foreign banks were allowed to enter. Real interest rates remained high, after an inflationary burst in late 1983. SAL V continued the program under SAL IV. To maintain momentum and enable the Bank to monitor progress, satis- factory progress toward the following goals was made a condition for tranche release: (i) elimination of the tax on inter-bank transactions, (ii) preparation of a review of a standardized accounting system for banks, (III) issue and diversification of the means for marketing treasury bonds, (iv) preparation of a program for introducing a standardized accounting system for companies making public issues of stocks or bonds, 1 The President's Report, SAL V, No. P-3783-TU, mentions this at para. 60, but provides no estimate of the magnitude of the fund levies. - 13 - (v) enactment of a law on external auditors. (vi) opening the Stock exchange, and (vii) study of fiscal incentives to induce banks to reduce operating costs. Many of these items were discussed end recommended in earlier SALu, but this was the first time they were written into the conditionalities. Most of them were already underway at the start of SAL V. 2.27 Agriculture. Progress in the agricultural sector under the SAL program had been substantial, and the same policy lines were to continue under SAL V. The recent Bank report on the sector recommended (i) greater use of market forces, (ii) support for institutional reform, (III) continued rationalization of public investment in the sector, and (iv) improvements in credit supply.11 The Government was to work with the Bank in formulating an action program for the sector, which would provide the basis for an agricultural sector policy loan. 2.28 EerAV. Progress in the energy sector had been substantial under the SAL program. The Bank had recently forwarded to the Government a paper recamending a crash program to continue energy development. The following elements were cited in the SAL V President's Reports (i) completion of ongo- ing, high priority power and lignite projects; (ii) improvement in the effi- ciency of the existing thermal power plants, distribution system, and mines; (iii) review of the feasibility of constructing new generating plants; (iv) expansion of existing lignite mines; and (v) increasing efficiency of energy use. The Government was to prepare, with the Bank's assistance, an energy action plan by the end of 1984. Satisfactory progress in the preparation of the plan was a condition of tranche release. 2.29 Conditionalities. Tranche conditionalities in SAL V were estab- lishment of a Special Account at the Central Bank for a proposed revolving fund for disbursement of SAL funds, completion of a Fifth Five-Year plan, and satisfactory progress in: (a) carrying out and implementing the results of the review of the public investment program; (b) elimination of quantitative restrictions on imports and rationalization of the tariff structure; (c) installation of the computerized debt management system; 6/ IBRD, Special Agricultural Sector Review, Report No. 4204, June, 1988. - 14 - (d) financial sector reforms; and (e) preparation of the energy program.9/ D. Evaluation of SAL Design 2.30 The design of SALs IV and V followed the program that had been laid out in SAL II. The program became more focussed, with agriculture, energy, and finance coming under sector loans. Fewer conditionalities were imposed. Substantial progress was made in numerous areas. The trade regime was opened and liberalized, and the policy of flexible adjustment of the exchange rate provided a stimulus to exports. Institutional improvements were made in the financial sector, such as development of the capital markets, establishment of a bank auditing system, and an inter-bank market. Improvements were made in tax administration, and the VAT was passed. The public investment program was reduced and re-oriented toward ir,:rastructure. Most importantly, the outward-looking and market-oriented attitude became entrenched in the economy of Turkey. But there were flaws in the design of SALs IV and V. Analysis in the Bank at the macro level could have been more searching, focussing on interrelations between economic variables. Little provision for mid-course correction was made. The Bank may have limited its leverage by limiting the program to five SALs and disbursing the fifth too soon. Each of these prob- lem areas will be discussed in turn. 2.31 Macro-Economic Preparation. Several problems of design of SALs IV and V were rooted in weak preparation of a policy-oriented analytical macro- economic framework. Here preparation included identification and analysis of emerging problems and prescription of policies to remove them. Little atten- tion was paid to the budget process and to the effects of fiscal instability until the report on Fiscal Policy and Tax Reform (FPTR), 6734-TU, in 1986. Little convincing analysis of the causes and consequences of high real inter- est rates was done. The role of exchange rate policy in influencing real interest rates was not recognized. Management of the external debt, as dis- tinct from recording it, came up only in SAL V. Many of these analytical points and interconnections would have had to be faced if a serious effort was made, early on, to develop the analytical medium-term framework. This was an issue from SAL III but it was never resolved. The Bank's review of the fifth Five Year Plan in June 1985 stated: While this (reducing the Plan target GNP growth rate) would lay to rest some of the fears that the Plan quan- tification implies a reinflation of the economy regard- less of the availability of resources, it does not answer the concern with the development of a medium-term framework to be used as a working document for guiding the future course of the structural adjustment process.10/ 9/ President's Report, SAL V, No. P-3783-TU, Annex III, Section III. 10/ Turkey - The Fifth Five Year Plan in the Context of Structural Adjustment: A Review, Report No. 5418-TU, July 20, 1985, p. 13, para. 2.7. - 15 - 2.32 Throughout the SAL programa, the Bank produced medium-term projec- tions and targets. These were calculated for the product side of the na- tional income accounts. They did not integrate the financial sector into the projections, or take account of the links between the budget deficit, interest rates, and exchange rate policy. A medium-term framework would at least be a scenario that integrated these elements into a consistent set of projections that showed the form of the behavioral equations and response elasticities adopted and how, over time, the policy change would relate to the targets which were reasonably expected to be met. In SALs II and III, the Bank tended to accept the Fund's analysis at the macro level.11/ perhaps it should also have considered developing itc own capacity, eapecially so because it was dealing often with other macrcvariables than the Fund and, most importantly, adopting a longer time perspec%tve. 2.33 Mid-Course Correction. Zittle provision was built into the SAL design for mid-course corrections. The problems in simply freeing the prices of SEEs without boundaries which would have been imposed by competi- tion in a non-monopolistic environment, as well as without any clear system of management incentives, could have been recognized and corrections made in SALs IV and V. These could have been aimed at providing the incentive to become profitable through efficiency gains rather than through monopoly pricing. The sluggishness of private investment could have been either anticipated--given the general level of real interest rates--or at least recognised. Analysis of the causes for this and proposed solutions could have been built into the SALs. The growth of the EBFa, and their destabilizing effects on the fiscal system and the trade regime, could have been anticipated. It could have been seen that the Government was not collecting the basic data needed to evaluate the social impact of its own program, and that obligation could have been written into SALs II or III. These were problems that became evident during the SAL period, and remain problematic now. They are discussed in some detail in Chapters IV-VI. 2.34 SAL Duration and Timinst. The lacuna in macro preparation and the problems mentioned in para. 2.32 would have been more likely to be detected and acted upon if the Bank had remained flexible on the issue of the number of SALs, and held back on SAL V until the medium-term framework 111 The division of labor was stated well in a 1986 memo: *Thus, the Fund's attention to the country's monetary and fiscal balance, the maintenance of external competitiveness, and State Economic Enterprise (SEE) financial performance complemented the Bank's efforts in the areas of SEE reform, import liberalization, export promotion, rationalization of private investment, debt management, financial sector reforms, and the development of a medium-term framework." Country Department Memorandum to Director, OED, July 25, 1986. , - 16 - was produced and analyzed. The issues raised in the FPTR report might have been recognized earlier. The Bank might have maintained more leverage over policy if it had not been so clear on termination. Firmer conditionalities could have been acceptable in SALs IV and V. The issues of the purpose, duration, and actual conditionality of the SAL program, while they are problems raised here specifically by the program in Turkey, go beyond this case. III. IMPLEMENTATION OF SALs IV AND V A. Introduction 3.01 The policies embodied in the design of SALs IV and V were many and far-reaching. The adequacy of these policies to the task of adjusting the structure of the Turkish economy towards a more outward-looking and market- oriented stance has been treated in Chapter II. This Chapter focuses on the process of actually implementing the policy reforms. It begins by discussing the reforms that were implemented in the major areas addressed by the SAL programs and to some extent their impact (Chapter V contains a more extensive look at the impact of the SAL program). The Chapter ends with an assessment of the problems faced in implementing the reforms. 3.02 The main areas in which new policies were instituted or former initiatives continued weres * domestic resource mobilization * export promotion * import liberalization * rationalization of public investment * energy sector policy agricultural sector policy reform of the SEEs Domestic Resource Mobilization 3.03 The program agreed to under SAL9 IV and V included policies to improve domestic resource mobilization by new tax bills, further develop- ment of capital markets and the promotion of savings by maintaining posi- tive real interest rates on deposits and deregulation of interest rates on loans, Most of these measures were initiated under SALs II and III. SALs IV and IV sought to improve the implementation of those measures. Tax Reform 3.04 During the period of SALs IV and V efforts to improve resource mobilization produced mixed results. The main objectives of the tax reform were to increase the elasticity of the tax system with respect to growth in - 17 - income and to restore supply-side incentives. Vhile the tax reform imple- mented during the period of SAL II brought a number of improvements, espe- cially the implementation of the VAT, tax revenue as a percentage of GNP did not increase significantly (see Table 4.5A). In fact, it even declined from 20.62 in 1983 to 16.32 in 1984, but this declining trend was reversed in 1985 with the introduction of VAT. Tax revenues reached 20.32 of GNP in 1985 and 18.61 in 1986. Early estimates for 1987 revenues collected by the central government gave a figure of 20.5% of GNP. 3.05 The decline in tax effort in 1984 was due to the personal income tax revenue. It has declined substantially as a fraction of GNP since 1982, with a slight recovery in 1986. The decline was due to the series of reforms beginning in 1980 that included measures that directly reduced personal income tax revenues.12/ Duriig the SAL IV and V periods there were a number of changes in the tax law. Marginal tax rates on income were further reduced combined with an upward adjustment of brackets. Since 1984, wage earners whose incomes are subject to withholding tax have been granted a tax refund in proportion to expenditures in certain goods and services up to a monthly limit determined by taxable income level. 3.06 Tax rates on financial transactions and withholding taxes on in- terest income were also reduced. The financial transactions tax was re- duced from 151 in 1983 to 3Z in 1984 and to 1% in 1986. The withholding tax on interest income was reduced from 30% on CDs and 20% on other bank deposits to a unified 102 rate in 1984; the 25% withholding tax on divi- dends was abolished. 3.07 To offset some of the reforms in personal and financial income taxes certain tax-exempt SEEs lost their exemption in 1984 and have been subject to full taxation since then. The corporate tax rate was raised from 40% to 46Z in 1986. 3.08 There were also other developments with respect to domestic re- source mobilization. A substantial portion of central government finance moved off budget during SALs IV and V. Revenues of the EBFs represented about 201 (3.51 of GNP) of the central government budget in 1986. If one ignores the EBFs, it appears that the size of the central government as a percentage of GNP has fallen between 1981 and 1986. However, when the EBFs and local governments are included, government expenditures have actually increased to 25.5Z of GNP in 1986. The growing role of the EBFs is further discussed in Chapter IV, Section F. 3.09 Due to the lack of elasticity of tax revenues and rising govern- ment spending, the budget deficit as a fraction of GNP has not shown a 12/ The discussion of tax measures implemented since 1980 can be found in PPAR for SALs II and III. - 18 - downward trend since 1983. The data are summarized in Table 3.1. In 1986, the deficit was 3.1 of GNP, up from 2.82 in 1985. A rapidly increasing component of the deficit is interest payments, which have risen steadily from 1.62 of GNP in 1983 to 3.7Z in 1987. This suggests a problem of fis- cal instability that is discussed at length in Chapter IV. The tax reforms undertaken during the SAL period have IVroved the efficiency of the tax system. Nonetheless, continued effort is needed to increase the revenue yield of the tax system. This is essential for the restoration of stabil- ity to the fiscal system in Turkey. Capital Markets 3.10 An important effort to raise funds for investment via the capital market is growing, with signs that more companies are trying to raise funds through the issuance of bonds and equity rather than borrowing from commer- cial banks. The corporate market is expanding rapidly and the commercial paper market is just getting off the ground. In 1986, the Capital Market Board, which was established after the enactment of the Capital Market Bill under SAL II, approved TL 111 billion of bond issues (see Table 3.2). 3.11 Turkey's capital market still remains a saall one, and in 1986 it accounted for less than 31 of the financial system. The size of the market is also extremely small when compared to the stock markets in other devel- oping countries such as Korea and Malaysia. 3.12 The Istanbul Stock Exchange started its operations at the begin- ning of 1986, adding a new dimension to the capital market. The volume of transactions exceeded expectations. Volume in 1986 was TL 1,118.7 billion of which the bulk (TL 965.8 billion or 861 of all transactions) was in government securities. Corporate equities are still only thinly traded. On the other hand, new share issues are increasing; the Capital Market Board permits for share issues increased by 181 in 1985 over 1984 and 352 in 1986. Interest Rates and Deposits 3.13 The Government's policy of maintaining positive real interest rates on deposits and deregulating interest rates on loans continued under SALs IV and V. Deposit rates have been generally positive, in real terms, since 1981, often exceeding 201 (see Table 4.3). Deposits have been in- creasing in real terms but the rate of increase has fallen. While the positive real rates have spurred deposit growth, their exceedingly high level has caused other problems. One of the effects of high real rates has been to jeopardize the financial system. The issue is further analyzed in - 19 - Table 8.1: CONSOLIDATED BUDGET SUMMARY (in billions of TL) 1988 1984 1965 198Lb 1967Ah REVENUES 2,814 2,640 4,476 6,851 10,126 Tax Revenue 1,984 2,872 8,29 5,991 8,69s Direct taxes 1,149 1,882 1,$26 2,961 Indtrect taxes 5 9e6 2,064 8,646 Non"ex revenue Le 86 468 647 o8 1,28W EXENDITURE 2,618 8,764 5,265 6,166 11,648 Personnel 671 925 1,275 1,628 2,40g Other current 891 559 611 1,266 1,65 Investment 476 68 969 1,614 1,006 Transfer* to SEEs 582 275 181 188 480 Other Transfers 596 962 1,412 2,246 8,641 Primary Budget Balance Le -118 -569 -192 -178 283 Interest Payments t1 875 695 1,081 1,666 Foreign Borrowing 181 264 427 682 1,666 Domestic Borrowing 6o 111 166 899 6OW Budget Belance -299 -944 -767 -1,257 -1,628 Increase in Accounts Payable Net 181 -56 -74 -170 a Finance Balance -166 -1,00 -61 -1,427 -1,528 FINANCING 168 1,0 61 1,427 1,628 Repayment -180 -294 -592 -1,557 -2,696 Foreign -160 -286 -426 -764 -1146 Domestic -86 -56 -172 -798 -956 Borrowing 846 1,294 1,458 2,934 8,618 Foreign 59 569 197 824 672 Domestic 289 725 1,2568 2,88 8,641 Government Bonds 199 195 670 1,517 2,686 Treasury Bille (Net) -98 284 24 668 686 Central Bank 72 19 266 257 276 Other 116 56 76 -82 61 FINANCING SURPLUS 299 945 786 684 926 Percent of GNP Revenues 26.6 15.8 16.2 17.4 Expenditures 22.6 26.6 19.0 20.5 of which: Transfers to SEEs 2.6 1.5 0.7 6.4 Budget Balance -2.6 -5.8 -2.8 -3.1 Finance Balance -2.6 -5.8 -8.1 -2.7 Le Includes special revenues A funds and annexed budget revenues. Lb Estimates. g Defined as Total Revenues loss all Expenditures except Interest Payments. Source: Treasury. - 20 - Chapter IV, Sections A, B, D and 9. An awareness of the inordinately high real interest rates is expressed in the President's Reports for SALs IV and V.13/ The source of the problem, however, was thought to be operationally inefficient banks and unnecessary taxes. This analysis was embodied in the condition that the second tranche of each loan be conditional on weatisfac- tory progress, in lowering taxes on financial transactions. Table 3.2: NEW BOND AND EQUITY PERMITS (TL billion) 1983 1984 1985 1986 Corporate Equities 35 64 76 102 Corporate Bonds 16 11 32 111 Sources Capital Market Board. Table 8.8: NOMINAL AND REAL DEPOSITS (TL bi II1one) 1962 1988 1984 1985 1e A. Total Deposits (Nominal) 2.857 8.988 4.,98 7.998 10.989 B. Price Levl (WPI of SIS 1981006) 127.9 165.7 249.1 856.8 402.8 C. Real Deposits 1.856 1.88 2.gee 2.246 2.248 0. Growth Rate (percentage over previous year) 28.12 0.64 7.86 12.8 0.88 Sources Central Bank. 3.14 Despite the new changes in banking laws, and related regulations put in place under the Financial Sector Adjustment Loan I (PSAL), there are still doubts about the strength of the banking sector. It is hoped that FSAL II will improve the situation in part by dealing directly with the level of real interest rates, the fiscal deficit, the structure of the banking system and corporate distress. B. Export Promotion 3.15 The Government's OStatement of Development Policies, of 1983 and 1984 continued to emphasize export promotion; during the SAL IV and SAL V periods, export incentives would be rationalized. The main vehicle for ex- port promotion would be the maintenance of an active real exchange policy 131/ See the President's Report on SAL IV, Report No. P-3543-TU, para. 77, and the same report for SAL V, No. P-3783-TU, para. 80. - 21 - rather than other direct incentives. There are four major direct export incentives available to exporters in Turkeys export tax rebates, subsidized export credit, duty free Imports of inputs to export production and subsi- dies from EBFa. The export tax rebates are direct subsidies since the rebates are not tied to taxes exporters pay. From November 1983 until late 1986 direct incentives were reduced, as can be seen in Table 3.4. Table 8.4: EXPORT INCENTIVES (percent) I II III IV 1088 10e4 1985 1988 196 198U 196 1986 Export Tax Rebate 11.5 11.1 9.7 1.* 7.2 6.9 8.1 7.6 Duty-Free Imports 5.4 2.9 4.9 4.1 5.8 4.0 11.6 6.2 Export Credit 6.5 1.1 6.6 6.6 e.6 6.6 6.6 0.e RJSF e.0 e.8 4.0 4.0 2.o 2.6 9.7 2.6 TOTAL 28.4 16.1 16.0 10.1 156., 12.9 17.0 16.0 forces Mi lanovic, Export Incentives and Turkish Manufactured Exports, 1986-1984. World Bank Staff Working Papers, Report No. 768, January 1986, p. 49, and Forouten, Trade Policy in Turkey in 1986 and 1987: An Evaluation, draft, Report No. 27151, March 1987, p. 12. 3.16 After three years of rapid export expansion, in 1983 the Govern- ment felt that the export orientation of the economy was well established and adequate enough to reduce the direct incentives and implied subsidies. It was also felt that the momentum of exports could be maintained by a policy of gradual depreciation of tLe Turkish lira. Therefore, in 1984 the overall subsidy contained in the direct incentives schemes was reduced to 15.22 from its maximum of 23.4Z in terms of export value in 1983. With the establishment of the Resource Utilization and Support Fund (RUSP), which granted a 4Z cash incentive to all exports, the total subsidy increased to 18.6% of exports value in 1985. During the first three quarters of 1986 total direct subsidy was reduced. At the end of the third quarter it was 12.92. However, in December of 1986, due to disappointing export perfor- mance, the Government increased the foreign exchange allocations for duty- free imports for exporters. At the end of the third quarter of 1986 the foreign exchange allocated as a percentage of exports was 19.6. According to preliminary estimates, this figure jumped in December 1986 to 61.4% of exports, implying that the subsidy element contained in this incentive increased from 4Z to 11Z of exports at the end of the fourth quarter. Although the other components of the total subsidy, the export tax rebate and RUSF subsidy, were reduced, the increase in the foreign exchange allo- cation subsidy more than offset those reductions and the overall subsidy - 22 - increased to 17.8Z of exports at the end of 1986. The Government also re- introduced the subsidized export credit scheme which was discontinued in January 1985. Therefore, the policy of reducing direct export incentives and implicit subsidies was reversed. The prospects for the current year (1987), based on the information available, point to an increase in export subsidies. 3.17 The other key element of export promotion has been the continued devaluation of the Turkish lira. Although there was slight real apprecia- tion of the Turkish lira in 1985 this situation was corrected in 1986, when there was a 10Z real depreciation of the Turkish 1ira relative to its 1984 level. The pace of devaluation has continued in 1987; the path of the real exchange rate is given in Figure 4.3. C. Import Liberalization 3.18 One of the objectives of SALs IV and V was to support the Govern- ment's policies of further import liberalization. In the *Statement of Development Policies* of 1983 and 1984 import liberalization was viewed as one of the medium-term objectives of the Turkish Government. Considerable progress has been achieved under the SAL program in abolishing quantitative restrictions (QRs) and lowering the tariff structure. However, while the importance of the average tariff level and QRs have been declining, the increasing use of levies by the WBF since 1984 has partially reversed the liberalization process. Total import protection as a percentage of the CIF value of imports vas higher in 1986 than in 1985. 3.19 During the SALs IV and V period the trade regime continued to transfer more items to the free list, reducing the importance of QRs. In 1983 the number of items requiring a license totalled 821 (List II) (Table 3.5). This number was brought down to 369 in 1984. Total import authorization of items on List II amounted to $5.2 billion in 1983; if the 1984 Regine had been in force, import authorization would have been needed for only about $1.2 billion worth of these imports. Liberalization was thus extended to cover approximately three quarters of the previously licensed items. In 1985 the number of licensed goods increased to 523. However, there was a substantial reduction in 1986 and the number of items requiring licenses were cut to 290. Further improvements came in 1987 when more than half of the items requiring licenses were transferred to the free list (meaning complete liberalization), leaving 111 items on the licensed goods list. However, despite the sharp reduction in the number of goods subject to license in 1987, their (1985) import value is still relatively high, over two billion dollars, or 18.62 of total (1985) imports. Moreover, the positive impact of the gradual dismantling of QRs is limited by the fact that all the 'liberalized* goods have been shifted to the fund list and, hence, are subject to a levy. - 23 - Table 3.5s NUMBER OF ITEMS SUBJECT TO A LICENSE Year List II 1963 821 1984 369 1985 523 1986 290 1987 111 Sourcet UTFT. 3.20 Excluding the QRs, the overall Import protection rate is composed of four elementst tariffs, levies, a surcharge, and a stamp duty. The sum of protection granted by each element makes up the overall import protec- tion, which is measured as percentage of CIF value of imports. The tariff rate has declined from 13.4Z in 1985 at the end of SAL V to an estimated 9.41 of imports value in early 1987. This decrease has been offset, how- ever, by an increase in the dollar-denominated levies whose revenues accrue to various EBFs. These levies were introduced in 1984 as a tool to raise revenues for social projects by taxing importation of luxury goods. At that time their tariff equivalent rate was estimated to be 2Z of imports, 18% of the average nominal tariff rate. Presently the tariff equivalent of levies is 6.1Z, or about 60% of the average nominal tariff rate. The role of the EBFs both on the trade regime and on public finance is discussed In greater detail in Chapter IV, Section F. 3.21 In addition, other existing import barriers have been made more restrictive in 1987. The stamp duty on imports was raised to 62 of imports CIF value from 4? in 1986. Also, the 2? import surcharge whose revenues accrue to the Support and Price Stabilization Fund (SPSF) has been in- creased to 42 in 1987 fEm 2? in 1986. As a result of all these develop- ments, total import tax and tariff revenues as a percentage of CIF value of imports increased from 21.2? in 1986 to 25.2Z in 1987. D. Rationalization of Public Investment 3.22 Rationalization of public investment was one of the major objec- tives of the Turkish Government under the SAL program. The Government's policy with respect to public investment changed significantly. Instead of focussing on leading economic growth the policy was designed to (a) re- strain the size and growth of public investment program, (b) establish pri- ority areas for public investments and cancel low priority projects, (c) shift the sectoral composition of investments from manufacturing to essen- tial infrastructure, and (d) improve the efficiency of selection of public - 24 - investment. At the request of the Turkish Government the Bank prepared a Public Sector Investment Review in 1981, an exercise repeated in 1987. The report supported the Government's new approach to public investment policy. 3.23 Successful efforts were made especially under SALs II and III to constrain the level of public investment in line with available resources. The number of on-going projects was reduced from 9,000 in 1980 to 7,400 in 1982 and to about 5,500 in 1983. Priority was given to agricultural proj- ects with export potential, energy, communications and transportation. Projects in those areas received larger allocations while the share of manufacturing was reduced. Under SALs IV and V the Government was to con- tinue with its efforts to rationalize public investment and to progres- sively lower its share in total investment, aiming at equality between the levels of public and private fixed investment. 3.24 During the SALs IV and V periods the Government continued to em- phasize priority areas in public investment. Toward the end of SAL V the overall level of public investment began to show signs of accelerating growth. Total public investment increased in real terms by only 0.8% in 1982, 1.72 in 1983 and 3.32 in 1984. Investments by SEEs, however, rose only 7.2Z overall in real terms between 1983 and 1986. In 1985 there was a substantial increase. Revised estimates by the State Planning Organization (SPO) indicate that, in real terms, the public fixed investment grew by 162 in 1985 and by 13.7Z in 1986. This growth in public investment was at least partially due to the fragmentation of public sector resource manage- ment caused by the growing number of EBFs and the increase in resources made available to municipalities with a backlog of investment needs. In- vestments by EBFs grew from 0 in 1983 to an estimated TL 480 b in 1985. At the same time investments by municipalities more than tripled in real terms to TL 685 b in 1986. 3.25 Since 1984 several measures have been taken to increase the avail- ability of resources to local governments. These included an increase in the revenue share of central government tax collections, the transfer of the collection of the property tax to local governments and the creation of various EBFs which would be used in part to finance municipal investments. As a result, the municipalities' share of public investment has increased. In 1986 it was estimated that local governments financed 15? of the overall investment compared to 7.8% during 1981-83. 3.26 Renewed growth in public investment is of concern not necessarily because Turkey's public investment expenditure is inefficient. Rather it creates pressure on the PSBR for both internal and external funds. Both contributed to the overall deterioration of the macro balance in 1986. Despite cuts in the size of the public investment program there are still some 6,000 projects, which means that resources have to be spread too thin- ly among a large number of projects. - 25 - 3.27 The need for continued rationalization of the public investment program is discussed in the Bank's second public sector investment review done in 1987.14/ The review was carried out when the analysis of the fifth Five Year Plan, covering the period 1985-89, revealed that the Plan's overall public investment targets were too high if viewed against the resources likely to be available. The report noted that the major adjustments needed were institutional planning capability in agencies, ministries and the SPO. Furthermore, there is still no effective monitoring of on-going projects and expenditures. The fragmentation of the public sector resource management by the EBFs and the municipalities, which increases the difficulty of controlling the public investment program, was also noted. E. Energy Sector Policy 3.28 During the SAL program the energy s ctor was one of the most im- portant items on the Turkish Government's agenda for both public investment aad policy formulation. The goal was to reduce the energy deficit and the cost of energy supply. Under SALs IV and V the Government was to maintain rational energy pricing policies that were initiated in 1981, improve con- servation measures, strengthen the planning and project management capabil- ities of the Ministry of Energy and Natural Resources (HENR) and other energy agencies, and encourage oil exploration and private investment in the energy sector. Energy Pricing 3.29 Since 1981 the Turkish Government has been pricing energy sources at levels which broadly reflect the cost of these resources in interna- tional terms. Energy prices have been adjusted frequently to keep domestic and international prices in line. Bulk electricity rates were increased by 13Z in real terms from 1982 to 1984 and then by 50Z in real terms in 1985, and early 1986.15/ By mid-1986 the average bulk price was estimated at about 90Z of the long-run marginal cost (LRMC) of production, compared with only 50Z in 1983. Retail prices have lagged somewhat but nevertheless were estimated to be about 70Z of their respective LRMC. 3.30 However, there was a slippage in early 1987. In February an 8.9% price increase was announced which fell substantially below the expected inflation rate for 1987. Due to election year pressures, the Government indicated that further increases during 1987 would be difficult. The Gov- ernment adjusted electricity prices by about 351 in early 1988. 14/ Turkey: Adjusting Public Sector Investment, March 30, 1987, Report No. 6603. 15/ Bulk rate refers to large purchases of electricity. - 26 - 3.31 There were adjustments of the prices of other energy sources also. After falling in 1982 and 1984, lignite prices increased in real terms in 1984 and 1985. The average operating subsidy, which had been approximately US$3.2 per ton in 1983, fell to less than US$1.0 per ton in 1985. With a further increase in prices in October 1986. all operating subsidies were eliminated. Prices of petroleum products continue to be maintained at international levels. Energy Conservation 3.32 Until 1986 little effort was made by the Turkish Government to launch a sustainable energy conservation program despite the evidence that such program would generate major savings. To design, implement and coor- dinate energy conservation activities Government announced in early 19C6 that the Electricity Survey Administration (EIE) would be given full au- thority to deal with this issue. Legislation has been prepared to transfer the authority to EB. Earlier energy conservation efforts under SALs IV and V were not very successful due to the fact that energy conservation proceeded on a piecemeal basis with a few ad hoc studies conducted by sev- eral departments and agencies. Institutional Changes 3.33 Under SAL9 IV and V the Government sought to strengthen planning and management capabilities of the MENR and energy agencies, especially the Turkish Electrical Authority (TEK). Project execution by TEK has improved and efforts have been made to rationalize and improve investments in the energy sector. However, further improvements might be achievable through better coordination among various energy related agencies. In addition, the shortage of skilled engineers and technicians continues to be acute in the energy sector. Given the high level of public investment in this sec- tor relative to others, the requirements for skilled manpower will grow. In 1985 the Government introduced new measures to attract and retain skilled personnel to the energy sector, but it is too early to assess the impact of those measures. Oil Exploration and Private Investment 3.34 The Turkish Government has been trying to promote oil exploration and private investment in exploration since 1980. In that year several decrees were issued to encourage Turkish and foreign private companies to engage in exploration activities. In 1983 the Petroleum Law was amended to provide liberal incentives for joint ventures and since then the Turkish Petroleum Corporation (TPAO) has been actively seeking such ventures. By 1985, 280 licenses were issued for exploration, of which 158 were held by TPAO, 11 by joint ventures with TPAO and 111 by foreign oil companies. This compares with a total of 16 issues in 1982. TPAO has prepared a long- - 27 - term exploration plan which was found satisfactory by the Bank. The Gov- ernment has been increasingly more interested in the participation of the private sector in energy. Further progress in this area will require a regulatory climate that is perceived to be stable, consistent and well defined. To facilitate the creation of such a climate the MENR has been working on a three-year work agenda that includes the preparation of regu- lations and guidelines for various issues for private firms that might be willing to invest in the energy sector. 3.35 In line with the Bank's Energy Sector Strategy Paper prepared in 1984, which was discussed with the Government and agreed upon, the Turkish Government designed a new energy development program.16 To support the implementation of this program the Bank is extending an Energy Sector Adjustment Loan in the amount of US$325 million to Turkey.17/ The changes envisioned are comprehensive and provide a medium-term framework that addresses the three interdependent sets of issuest (a) the institutional framework; (b) economic investments; and (c) improved efficiency. F. Agricultural Sector Policy 3.36 Support of agricultural policies continued under SALs IV and V, along the lines of SALs II and III. The broad strategy was one of stimu- . lating more efficient production by reducing input subsidies, maintaining production floor prices, and promoting a crop mix which would better re- flect Turkey's comparative advantage. Investment incentives were to be maintained by realistic exchange rates and export incentives at competitive levels. The policy also emphasized a supportive public sector role. The * export-oriented approach to agriculture implied changing government policy from food self-sufficiency to increased net agricultural contribution to the balance of trade. 3.37 SALs IV and V emphasized the implementation of policies to improve productivity and efficiency since the arable land frontier was reached in the early 1970s, meaning that agricultural growth in the 1980s would have to come from increased productivity. 3.38 During the SALs IV and V periods, agricultural exports increased substantially and there was a shift from unprocessed to processed foods. In line with liberalization of trade tariffs, duties on agricultural im- ports were reduced. In June 1986 a decree was issued which opened up fer- tilizer distribution to private fertilizer manufacturers, ending the 16/ Turkeys Energy Sector Strate&y Paper, Report No. 4973-TU, September 1984. 171 President's Report, Turkey: Eneray Sector Adjustment Loan, Report No. P-4513-TU, June 1, 1987. - 28 - monopoly on fertilizer distribution of the Agricultural Supply Organization (T2PK), an SEE. The implementation of monitorable actions under SAL9 IV and V are discussed below. Public Investment in Agriculture 3.39 Under SALs IV and V the Turkish Government was to allocate at least 112 of public investment to agriculture. This target was not attained; 92 of total public investment expenditures went to agriculture in 198& and 7Z in 1985. After the end of the SAL V period in 1986, 6Z of the public investment went to agriculture compared to the 1986 program of 102. These results are not consistent with Government's intentions of raising yields and efficiency. To improve yields more ample allocations must be made to irrigation projects and technical services. This point is made in the agriculture section of the most recent public investment review.181 Institutional Reform 3.40 Institutional reform was initiated in SAL II and continued through SAL V. There were achievements but they have been slow coming. The Ministry of Agriculture, Forestry and Rural Affairs (MAFRA) has been reor- ganized in order to decentralize decision making, while merging the staff and facilities of overlapping field services. Under SALs IV and V broad responsibilities were assigned to MAPRA's Bureau of Planning, Research and Coordination (APK) to monitor performance and expenditures. APK was reor- ganized and a few of its senior personnel were sent abroad for training. However, APK is still not able to effectively function as a planning and monitoring unit.191 The Government generally recognizes the weaknesses of APK but measures to improve its effectiveness have been implemented slowly. Improvement of APK's effectiveness is one of the important - objectives of the current Agricultural Sector Adjustment Loan (ASAL).01 3.41 Recently the Government has implemented a new 'Farmer Incentive Premium* scheme under which farmers obtain receipts for the sale of their farm produce, to be used for income tax purposes. Farmers with receipts totalling to a predetermined level are able to obtain credit at concession- al interest rates. Moreover, under this scheme the Agricultural Bank (TCZB) is able to issue loans to poorer farmers. TCZB will only require co-guarantees, rather than its usual collateral requirements. This should help to met the needs of small and medium farmers. 181 Turkey: Adjusting Public Investment, Report No. 6603-TU. March 30, 1987. 191 Ibid. 20/ Turkeys Agricultural Sector Adjustment Loan (ASAL), Loan 2585-TV. -29 - Support Prices and Input Subsidies 3.42 The number of supported commodities was reduced from 25 to 14 in 1984. It was the intention of the Government in 1981 to eliminate all fertilizer subsidies within five years. The average subsidy rate as a percentage of purchase value was 46% in 1980. This was reduced to 401 in 1984 and to 262 in 1986. Total public outlays for fertilizer subsidies decreased in real terms by 35% from 1984 to 1986. However, subsequent to the SAL program, there was a policy reversal in 1987. The average subsidy rate for fertilizer has been increased to 331. Agricultural interest rate subsidies were also increased. Despite rising inflation, the rate on se- lected agricultural loans was reduced from 341 to 221, in nominal terms, effective October 1, 1986.211 In addition, new subsidies for pesticides, livestock medicines and milk production were introduced and there was an increase in animal feed subsidies. 3.43 The concern over the recent slippages in the policy reform was discussed with the Government officials in MAFRA and their response was that those new measures were designed to reduce the effects of unfavorable developments (declining real farm incomes, dumped beef imports from the EEC and radiation problems which hindered agricultural exports). They stated that the new measures are not open-ended subsidies like higher support prices and their costs are predictable. Whether the Government has se- lected the best instruments to reduce unfavorable developments with respect to the agricultural sector is open to question, but the new support mea- sures are policy reversals and undermine the progress made earlier in the SAL program. G. Reform of the SEEs 3.44 Since the 1920s, the SEEs have been prime actors in the industri- alization and growth of Turkey based on import-substitution. However, since the late 1970s, the operating losses and growing investment demands of the SEEs have placed substantial burdens on the budget and added to inflationary pressures. Vhile some SEEs were well run and profitable, in aggregate they showed increasing losses through 1980. Therefore, reform of the SEEs has been a major goal of the structural adjustment program. There have been some major achievements during the SAL program period, but more remains to be done. 211 Agricultural interest rates were reduced to 22? only for about 401 of TCZB's loans to farmers (loans for fertilizer, livestock, and loans below US$600 based upon sales receipts). The remaining loans to farmers are still at rates between 30% and 352. This compares with the average annual price inflation of about 40?. - 30 - 3.45 The Turkish Government has since 1980 attached particular emphasis to the reform of SEEs. Policy actions were taken and new laws were passed. During 1980-82, the Government implemented a number of measures to improve the short-term financial structure of the SEEs, redirect their investment programs and finance them increasingly from non-budget sources, and under- take legal reforms to institutionalize these changes. There were efforts to change the role of the public sector from leading economic growth to supporting private sector development from both domestic and foreign sources. Following the reforms in 1980 the SEEs were encouraged to raise their prices; since then they have adjusted their prices periodically. 3.46 SAL IV called for some boundaries on SEEs' growth to consolidate these improvements. In 1983, SEE investment was to be held to 6.5% of GNP, operating profits were to equal 0.6? of GNP and, therefore, overall outside financing was not to exceed 5.92 of GNP. The actual figures for 1983 did not meet these targets. According to the SAL V President's Report the numbers were 7.2? for investment, 0.42 for profits and 6.4Z for overall financing. These shortfalls are not mentioned in the SAL V documentation. Instead, the SAL V President's Report cites the need for further improve- ment firm by firm in the managerial expertise and personnel of the SEEs. It also called for a study of the possibility for privatizing some SEEs. A technical assistance loan designed to improve operational efficiency at certain SEEs that began in 1985 was later aborted, however, when the plans for possible privatization signalled uncertainty as to the future of each SEE. 3.47 The SEEs contributed significantly to the public investment boom in 1985 and 1986. According to new estimates, public fixed investment grew by 16? in real term in 1985, and by 13.7? in 1986. Over half of the in- crease in public investment in 1985 and 1986 was accounted for by the SEEs, which raised their fixed investments by almost 10? in real terms each year. It appears that most SEE investments were on infrastructure projects--com- munications, energy and transportation--rather than manufacturing. How- ever, the financing of those investments is of concern since in many cases funds were borrowed externally and internally, further contributing to the growth of debt. Net foreign borrowing by the SEEs increased by 48? in 1985 and by a further 70? in 1986. 3.48 The 1987 program called for a reduction of nearly 50? in foreign borrowing but it is not clear if this goal could be achieved since the investments started in 1986 will probably require further financing. Net domestic borrowing by the SEEs was also up by 38? in 1985 over 1984 and by 65? in 1986. Both of those figures are higher than the inflation rate, meaning in real terms the indebtedness of the SEEs increased. 3.49 While profits have been generated by SEEs it is not clear whether this is due mainly to price increases and the exploitation of monopoly power or also to improvements in efficiency. Evidence provided in Chapter IV, Section H, suggests that the former is largely the source of higher profits. - 31 - 3.50 Another element of the SEE strategy has been to reduce their num- ber through sale to the private sector. A master plan for privatization of the SEEs has already been completed and in May 1986 the Government obtained legislative authorization to sell SEEs to the private sector. H. Summary of Prosram Implementation 3.51 The efforts at instituting policy reforms in Turkey, while in general successful, were characterized by periods of progress followed by periods of regression. The passage of a new law or reform of a regulation embodying a reform was frequently met by the creation of some other mechan- ism for protecting the interest of those who gained from the earlier status quo. An excellent example of this process was the creation of the EBFs. Originally designed to more expeditiously provide key governmental ser- vices, the EBFs nonetheless undermined efforts at fiscal reform, as well as the moves towards import liberalization. Because EBFes were outside the normal budgetary process, they escaped some of the pressures for control that actions to reduce the central government deficit engendered. At the same time, by raising revenues via specific dollar levies on imports, they created a new instrument for protecting domestic firms in a period when traditional quantitative restrictions and tariffs were being eliminated or reduced. The EBFs then were, at least in part, a vehicle through which groups standing to lose from the reforms could still safeguard their inter- ests and retard the structural adjustment efforts. 3.52 Other areas of the reform program show similar patterns of *two steps forward, one step backe. Public investment was reduced and rational- ized in the years of the SAL program prior to 1985. In that year, however, public investment skyrocketed. In agriculture, steps to reduce credit and input subsidies that had been steady and successful until 1986 were par- tially reversed in 1987. A gradual reduction in export subsidies begun in 1984 was reversed in late 1986 because of a decline in export revenues. 3.53 This pattern of progress and retrenchment is understandable if one views the Government as an entity made up of groups with conflicting inter- ests. Efforts at reform will necessarily meet with resistance and the creation of new obstacles. A well implemented adjustment program must be cognizant of this dynamics and seek to offset it or to reduce its impact so that the general trend remains largely unaffected. In the case of Turkey this was not always done. Although the later SAL documents were aware of the existence of EBFs no efforts at controlling their detrimental effects on the trade regime were attempted. Of course, the benefit of hindsight makes it much easier to identify those institutions or changes that will serve to undermine adjustment. Given that groups that stand to lose will attempt to protect themselves in unforeseen ways, and that when implement- ing a program one does not have the benefit of hindsight, it may be reason- able to simplify or limit the scope of SAL programs so as to ensure fully effective implementation and monitoring. - 32 - 3.54 The spurt in public investment points to another problem, that of relationships between elements of the program. The rapid growth of public investment in 1985-86 was in large part due to EBF9 and municipalities, and the SEEs as well, rather tLan investments by the central government itself. All had been given more decision-making authority, authority that was deemed necessary to improve operation of each of these entities. In the absence of a well articulated medium-term strategy, however, there was no check on this authority--or, for that matter, on the composition of invest- ment made and its relation to the country's overall development strategy- -and a boom ensued. This episode points to the need for an integrated framework that includes interactions among the elements of the program- -both in terms of policies and expected results--and the need to keep the scope of the program limited. IV. MAJOR ISSUES A. Introduction 4.01 The major problems that emerged during the evaluation of SALs IV and V were in essentially the same areas as those discussed in the SALs 11- III evaluation. The issues for the SALs IV-V evaluation are (i) exchange- rate policy and real interest rates, (ii) budget control and fiscal dy- namics, (iii) foreign debt and current-account stability, (Lv) investment and export capacity, (v) financial market structure and stability, (vi) the role of EBFs, (vii) trade liberalization, and (viii) reform of SEEs. The problems in these eight areas form an economic seamless web, with each influencing or exacerbating several of the others. So any initial entry point into the issues will be somewhat arbitrarily chosen. Sub-sections A through H will outline the analysis of each main problem area, summarizing the relevant data and other evidence, and tie the issue areas together. 4.02 The central issues in the following analysis involve the causes and consequences of the high level of real interest rates. The discussion will begin with analysis of the effects of exchange-rate policy and deficit financing on the level of real interest rates. It will continue by outlin- ing the effects of high interest rates on the budget, the balance of pay- ments, investment, and other sectors. In each case, a dynamic loop is closed as the effects of high rates feed back to exacerbate the situation that contributed to them in the first place.22/ An example here will introduce and illustrate the type of dynamic problem that will recur in the analyses of potential instabilities in this section. Bond-finance of the fiscal deficit since 1984 has tended to push up interest rates, which were 221 These dynamic problems were first discussed in Bank documents on Turkey in the SALs II-III PPAR, and fully analyzed in the Report on Fiscal Policy and Tax Reform, Report No. 6374-TU. - 33 - already quite high. The Government has been selling debt of average matur- ity1 of approximately one year to the public, mainly the commercial banking system, at nominal interest rates of 502-60%. The interest is tax-exempt, so for the marginal bondholder with positive taxes, the equivalent pre-tax rate is around 80% (see Figure 4.1 for the data). The inflation rate has been in the range of 25Z-40Z since 1984 (see Figure 4.2). If the expected rate of inflation is near the middle of that range, the Government is bor- rowing at a real rate of 202 before tax adjustment and 502 after tax ad- justment, at the margin. This means that an interest expense of 202 to 50% of the outstanding debt is added to the fiscal deficit each year in real terms. If the real interest rate on outstanding debt is higher than the growth rate of the tax base available to finance the interest expense, the debt will grow exponentially relative to the tax base, even if the budget exclusive of interest expense is in balance. The real growth rate of tax revenue has been highly variable in Turkey since 1980. It has averaged 6.3? per year, but fluctuations have ranged from -20.7Z in 1982 to 32.2Z in 1985. Thus it may be that the condition for exponential debt growth is met (see Table 4.6B for revenue growth). The continuing rise in the real defi- cit to be financed adds to the upward pressure on interest rates, compound- ing the problem of fiscal instability. Here, once more, the dynamic loop is closed, with potentially rapid growth in debt and interest expense and rising real interest rates. 4.03 The problem of fiscal instability will be discussed fully in sub- section (B), where it will be placed in context and evaluated. First the report will turn to the question of why real interest rates are so high -- higher than the growth rate -- in the first place. B. Exchange-Rate Policy and Real Interest Rates 4.04 A persistent problem in the Turkish economy since the start of the structural adjustment program in 1980 has been the high level of real in- terest rates. Real interest rates influence strongly investme-t decisions in the private sector. Here their effects on capacity creation in manufac- turing and resulting increasing export supply are an issue. Furthermore, the level of real rates is an important determinant of the cost of financ- ing government debt service; too high a level can destabilize the govern- ment budget. High real rates also contribute to potential instability in the financial sector. Each of these problem areas is discussed in a sepa- rate sub-section later in this report. One source of high loan rates has been the cost of intermediation in the banking system. This was a problem area identified early in the SAL process. SAL II supported various mea- sures aimed at reducing inefficiencies and transactions taxes in the bank- ing system. SALs III-V, the two FSALs, and the PPAR on SALs II-III - 34 - Figure 4.1 Nominal Interest Rates 100- 90 70- 60- 84:6 84:12 85:6 86u12 8& 812 - dOsitI + gv bond -- lan - 35 - Figur* 4.2 YEARLY INFLATION (t/(a+12» 140,- 110- 90 70 40 40 20 79:67&*12 060312 816 81t12 2682:12868312 84:684&1286686128&6 812 -- CP m - 36 - maintained this focus on efficiency in the banking sector, as well as broadening the financial markets. The structure and potential instability of the financial system are discussed in sub-section E later. Here the focus is on the international determinants of the general level of real interest rates, rather than the spread between borrowing and lending rates in the banking system. 4.05 Given the openness of Turkey's financial markets, one can view the real interest rate as being determined by the world real rate plus the expected rate of depreciation of the Turkish lira in real terms plus a risk premium that increases as the stock of national debt denominated in Turkish lira grows. To attract a net inflow of capital that finances the current account deficit, monetary policy must hold the interest rate above this equilibrium level. Each of these factors influencing the level of real interest rates is discussed in the following paragraphs. 4.06 As part of its policy of export promotion, the Government has followed, since the maxi-devaluation of January 1980, a policy of nominal devaluations designed to progressively lower the real exchange rate. The real exchange rate, as calculated using trad, weights and indexed to December 1982 - 100 is given in Table 4.1. Its path since 1979 is shown as Figure 4.3. From March 1980, just after the devaluation, to the end of 1986 the real exchange rate in Figure 4.3 shows an average annual deprecia- tion rate of 4.32. The depreciation has not been steady, however. The rate rose at the beginning of 1982 and again in 1983. There was also an appreciation of nearly 131 from February 1984 to February 1985. But each of these periods of appreciation was reversed, producing an overall down- ward trend. This has been generally accepted policy, at least implicitly approved by the Bank.231 Since February 1985, the annual rate of depreciation has been about 112 per year. The persistence of the real depreciation since 1981, and the support of the depreciation policy from the international community, suggest that over the medium term continued real depreciation is expected by the financial markets. The expected rate of depreciation may be as low as the average of 4.32, as high as the more recent 111, or even higher if trade performance falters. 231 See, for example, the World Bank President's Report on SAL III, Report No. P-3543-TU, May 1983, or for a more recent discussion, World Bank, Trade Policy in Turkey in 1986 and 1987: An Evaluation, Draft Report No. 27151, March 19, 1987. - 37 - Tabe 4.1: T1IKEY - RM. EXCHANG RATE (<ienthly IZddox) 1979 198 1981 19m2 198 1964 1985 196 JAN 1N.4 189.6 111.9 116.1 17.A 938 191.9 96.9 PEm 111.4 191.9 109.6 116.1 17.9 92.6 lu.7 94.7 mAR 1f.4 111.1 1N.4 M1A. 197.8 9.2 118.7 8.9 APR 171.6 117.6 lo.1 118.4 1N.1 96.7 lo.1 66.7 mAY 177.8 197.6 16.2 111.6 114.7 97.6 M.5 96.6 JUN 161.4 162.8 168.9 18.4 108.9 99.1 94.9 86.6 JU. 140.8 1f1.6 104.2 116.9 102.3 96.5 98.2 86.7 AUG 148.7 100.6 111.7 11N.4 11N.8 97.6 98.2 86.2 se 148.9 104.7 101.9 1M. 96.6 90.2 98.8 85.6 OCT 165.7 111.8 98.9 115.3 96.7 98.6 96.8 6.@ NOV 158.s 111.1 96.7 168.7 99.6 11.8 95.6 68.2 DEC 164.2 111.4 111.9 10.1 99.2 99.8 94.0 82.8 Loft: The real rata is construted uing trade weights for f ive countrie: U.S., Germany, France, Italy and Seltzerland. The trade welight are averaged for 198-e. (Dec. 1962 = 1a) This effecbive reta le then correc~ed for in- f loion using the respecIve coneumer pris indlcas. A deoline in the Index menns a rest depreletion of the Turkih lira. Seeo: Wsorld Bank e6hmötes. -38 - Figure 4.3 Real Effective Exc hange Rote (1982:12-100) L1o 170- 160- 180 - 140- 130- 120 - 110, 100 - 90 - 71:17186 79:12 U0:6 60:12 B1:6 81:12 82:6 B812 8&,6 S&12 B4:6 84:12 866 8612 8186 86:12 - 39 - 4.07 Since the beginning of the structural adjustment program in 1980, opening the financial markets internationally and achieving convertibility of the Turkish lira have been policy objectives.241 The results have been impressive. By 1986, residents in Turkey and Turkish workers abroad were able to hold a wide range of foreign exchange assets as well as assets denominated in lira. Turkish firms, as well as local governments, banks, and even the EBFs were able to borrow abroad, and encouraged to do so. Table 4.2A presents data that suggest the range of potential substitution between foreign exchange and lira on both the asset and liability sides. For example, at the end of 1986 the foreign-exchange deposit component of total deposits, including foreign-exchange deposits, was 23.11, as shown in Table 4.23. Given this relatively free flow of capital between Turkey and the rest of the world the policy of real depreciation, when expected, has the effect of raising the internal real interest rate above that of the rest of the world. 4.08 Given the openness of the Turkish financial system, uncovered interest arbitrage requires that the nominal return to a Turkish asset equal the nominal return on a foreign asset plus the expected rate of nomi- nal depreciation plus the risk premium on Turkish lira.25/ This will leave both lenders and borrowers marginally indifferent between foreign- exchange and lira denomination of their contracts. With open financial markets, this arbitrage condition can be expected to hold in Turkey. The real exchange rate is defined as the nominal rate times the ratio of home to foreign price levels. So the rate of nominal depreciation is by definition equal to the rate of real depreciation plus the home versus foreign inflation differential. When the rate of nominal depreciation in the nominal arbitrage condition is replaced by the rate of real depreciation plus the inflation differential, we obtain the real version of the arbitrage condition. The real return on the domestic (Turkish) asset must be equal to the world real rate plus the expected rate of real depreciation plus the risk premium for borrowers and lenders to be indifferent to the currency denomination of their contracts. It is useful to notice that this real arbitrage relation does not assume that purchasing power parity holds; it only uses the definition of the real exchange rate. To generate a capital inflow, monetary policy must hold the real rate above the level given by the real arbitrage condition. To paraphrase one Central Bank officialt monetary policy keeps interest rates high enough so that the capital inflow finances the current account deficit. 24/ See the President's Report, SAL I, No. P-2725-TU, February, 1980, para. 26 for a discussion of the need to foster international capital flows. Paragraph 30 of the same report cites the importance of *rational exchange rate policies" as well as the need for capital flows. The same themes are echoed in the President's Reports for SALs III-V, No. P-3273-TU, No. P-355543-TU, and No. P-3783-TU, respectively. 25/ William Branson and Henderson, 'The Specification and Influence of Asset Markets*, in Handbook of International Economics by Peter Kennen and R. Jones (editors) (North Holland, 1985). - 40 - Table 4.2k FOREIGN ASSETS AND LIABILITIES Consolidated Balance Sheet of Comercial Banks 1986 Turkish Foreign Lira Exchange Total FX/Total (millions of TL) (M) LIABILITIES Deposits 12,788,986 2,805,564 15,594,550 18.0 Total Liabilities 20,043,584 4,397,212 24,440.796 18.0 ASSETS Loans 10,938,099 3,645,257 14,583,356 25.0 Interbank 1,112,872 2,203,259 3,316,131 66.4 Total Assets 19,888,641 4,552,155 24,440,796 18.6 Sources Central Bank. Private Sector Foreign Debt (medium and long term) Stock Share in Outstanding Total Credit (mil of US$) (2) 1980 535 61.1 1981 440 44.6 1982 394 40.8 1983 399 46.7 1984 425 60.0 1985 359 37.2 Sources World Bank, World Debt Tabless Central Bank of Turkey. - 41 - Table 4.2B: FOREIGN ASSETS AND LIABILITIES Foreisn Exchanye Deposits Share in US$ TL* Total Deposits (mil) (bil) (2) 1980 (a) 543 49.0 6.6 1981 (a) 473 63.2 4.2 1982 (a) 585 109.2 4.6 1983 (b) 792 224.0 7.3 1984 1180 524.8 10.5 1985 2061 1188.9 14.9 1986 3166 2399.2 23.1 1987 (c) 3699 2877.6 (a) Figures are for a foreign exchange deposit scheme now defunct. (b) Includes TL$83 billion under the new deposit scheme. (c) March, 1987 * The year end stock is converted using the year-end exchange rate. Sources Central Bank. - 42 - 4.09 The real arbitrage condition can be used to analyze the level of real interest rates in Turkey during the period of structural adjustment. The real rate should be higher than the world real rate by at least the ex- pected rate of real depreciation plus the risk premium. The data of Figure 4.3 suggest an expected rate of real depreciation of perhaps 5-102. Cal- culations of real interest rates in Turkey are sumarized in Table 4.3. The difficulty is in the estimation of the expected rate of inflation. A rational expectations approach would use the realized inflation over the period of maturity of the loan or deposit. This version of the real rate is shown as RA in the table. A static expectations view would use the most recent observation as the expected rate. This version is shown as RB in the table. Both RA and RB are shown in Figure 4.4. Econometric study of the time-series behavior of the inflation rate in Turkey since the mid- 1970s suggests that the best predictor is the most recent observation, so it is not surprising that RA and RB are fairly similar. An estimate of the US dollar real interest rate in world markets is also presented in Table 4.3 and shown in Figure 4.4. The differential between the Turkish and the world rates is clear from the figure. Over the latter portion of the period the real lending rate in Turkey was about 10-20 percentage points above the world rate. 4.10 The calculation of the risk premium on lira-denominated government debt is harder to estimate than the real interest rate or even the expected rate of real depreciation. The risk premium is, in principle, an increas- ing function of the stock of outstanding lira debt as a fraction of total financial wealth of the population holding the debt. This is a basic con- clusion from the economic literature on international capital-asset pric- ing.26/ Now rapidly the risk premium can be expected to increase as the lira debt rises is unclear. In addition, we have no aggregate data on fi- nancial wealth with which to compare the stock of lira debt. Nevertheless, the latter has risen rapidly since 1984, and is undoubtedly contributing to upward pressure on real rates in Turkey. The level and growth rate of lira debt are shown in Table 4.4, in current and constant prices. The growth rates in constant prices are substantially greater than real GDP growth, which may be a reasonable proxy for financial wealth. Thus it seems safe to conclude that growth in lira debt has raised real interest rates rela- tive to world levels.271 26J The rising risk premium since the end of 1984 can be approximately measured by the growing difference between the real lending rate and the world rates, shown in the lower panel in Figure 5.4, less the expected rate of real depreciation and the margin needed to attract a private capital inflow. 271 This conclusion was also reached by the Bank's report, "Fiscal Policy and Tax Reform', No. 6374-TU, May, 1987. - 43 - Table 4.8: REAL INTEREMT RATES (yerly %) US Dollar Deposit Ratcs(1) Landing Ratos (2) L.R (8) minimum MaximuM Datc RA RB RA RB RA Re RA aB 86: -27.6 -119.6 -17.8 -169.1 7.8 6.2 86:6 -26.6 -97.5 -15.1 -7.1 -4.9 6.2 8:9 -21.9 -N.8 -1.8 -6.1 -8.1 2.2 86:12 -9.4 -79.1 9.8 -69.9 9.4 -1.8 81:8 16.6 6.4 11.8 1.2 12.9 2.1 81:6 22.6 8.6 17.2 8.8 18.8 10.7 81:9 32.8 19.7 17.7 4.6 18.1 16.0 81:12 86.8 81.8 16.7 16.2 18.7 11.5 82:8 82.2 26.4 17.1 11.8 20.6 10.6 82:6 28.8 82.8 11.7 17.2 18.1 14.6 82:9 28.8 82.8 8.7 17.7 10.1 18.1 82:12 16.4 10.1 9.8 15.7 7.6 12.7 88:8 8.2 28.6 -8.2 17.1 9.8 6.2 88:6 -1.7 28.1 -18.1 11.7 6.1 6.8 88:9 -8.7 20.1 -20.2 8.6 6.8 16.2 88:12 4.2 11.7 -12.8 -4.8 7.5 6.7 84:8 2.6 8.2 7.4 18.6 26.6 U.6 11.8 6.8 84:6 16.6 -6.6 28.6 6.6 49.7 28.7 16.7 9.4 84:9 28.2 2.5 24.8 4.6 47.1 26.4 11.8 18.6 84:12 26.9 16.4 22.8 11.8 59.8 48.9 8.6 14.1 85:8 26.6 8.8 28.6 10.2 68.6 47.8 12.6 8.8 8M:6 N6.4 21.6 81.8 28.9 68.8 60.6 12.8 6.2 865:9 10.4 28.2 81.8 24.6 66.7 58.4 7.6 16.8 86:12 mJ.8 26.9 31.7 22.8 66.5 61.1 6.8 11.9 86:8 21.6 28.6 60.4 6.1 7.2 68:6 21.7 81.8 54.2 16.7 4.7 86:9 21.7 81.8 64.8 8.6 4.6 86:12 16.7 81.7 54.1 9.7 87:8 8.7 (1) RA Is the nominal rate loss the actual rate of inflation in thO fo losing 12 months as omasurod by tm UPX of the UTFT. (2) RB la the nominal rat less eth rate of inflatlon in tm provlous 12 months as ~sasured by the YI of tm UTFT. (8) Six month LIBR loss tho US II for the prior or fot lowing six monthe convertd to an annuaiod ratu. Sourcm: Central Dank, SP0 and MUF. - 44 - Fgure 4.4 Real Interest Rates (using actual innlatio% RA) 40- 30- 20- 10 -10 -20- 80:3 80:12 81:12 82:12 8312 84:12 8&12 wfd + op~t Rt Min. Landng R Real Interest Rates 30 20- 10 0- -10 -20 -30- -40- -80- -70 -80 -100 -110- 80:3 80:12 81:12 2u12 M12 84:12 8:12 S&12 W~d eb + Dplt Re 9 Min. lnding Re - 45 - Table 4.4s STOC OF CONSOLIDATED GOVERMIENT DEST OUTSTAleING (December at each year) 1low 1061 196 1968 1964 1965 19ow (Milione of TL) Total Debt 190,847 247,996 8,414 416,40 070,701 1,522,157 2,888,470 dovernment Bonds, ) 1 yr. 141,188 19,965 135,84 3,teU7 80,687 1,081,652 1,511,825 Goverament Sl, < 1 yr. 46,714 9,18 152,491 56,210 389,954 490,305 822,151 (MIillionsof 198 TL) Total Debt 189,647 190,011 191,851 171,619 289,098 298,981 841,812 Goveramet Bonds, > I yr. 141,138 128,144 195,188 148,019 146,297 197,218 221,M68 Govenoment SI Is, < 1 yr. 46,714 07,767 86,218 28,191 98,691 98,712 120,264 Annual Percent Change Tota I Debb -29 1 a -11 26 16 15 Governmeat Bonds, > 1 yr. 8 -18 -15 41 -2 85 12 Government BI Il, < 1 yr. -68 80 27 -78 8 0 28 Memo Item: Consumer Price Index (SIS) 18I 1 3 177 242 868 528 884 Sources SPO and Treasury. 4.11 The basic conclusions, at this point, are the following. The general level of real interest rates in Turkey has been high relative to world rates through most of the period since the first SAL in 1980. This was noted in the PPAR on SALs II-III, and remains the case. Costs of in- termediation and the burden of non-performing loans (discussed in sub-sec- tion E below) account for a large spread between lending and borrowing rates, but not for the high overall level of rates, especially relative to international conditions. The high level of overall rates relative to world levels has two sources in policy. The first is the evident policy of gradual real depreciation of the Turkish lira. If this is anticipated by market participants, as it seems to be, it puts one wedge between domestic and world rates. This may account for 5-10 percentage points of the real differential, as noted in para. 4.06. This policy seems to have been sup- ported by the SAL program. The second is the persistent large fiscal defi- cit that cumulates into a debt stock that raises the risk premium. Reduc- tion of the fiscal deficit has been a principal policy objective of the SAL program, but little real progress is evident. - 46 - C. Budset Control and Fiscal Dynamics 4.12 A principal objective for macro-economic policy in Turkey during the SAL program was reduction of the fiscal deficit to a sustainable level. This would be a level of the deficit that could be financed in a non-infla- tionary manner, given the growth in the capacity of the Turkish and world economies to absorb the debt of the Turkish government. Success in meeting this objective has been partial, at best. The fiscal situation has been described in detail in the recent FPTR Bank report. Here the major points will be reviewed as they relate to the SAL program, emphasizing the links between the fiscal situation and other salient policy problems. 4.13 The fiscal deficit was to be reduced by a combination of expendi- ture reduction and revenue raising. Major items in the SAL program were reduction of public investment and concentrating it on priority projects with feasible completion times, reform of tax administration, and reform of the tax structure, including simplification and introduction of a VAT. Each of these areas will be discussed later. First the results for overall fiscal balance will be discussed. The data for the consolidated government budget and the PSBR are summarized in Tables 4.5A-B. These are updated from the FPTR report. The PSBR includes SEEs, local governments, and re- volving funds and EBFs. In some years the PSBR is twice the consolidated deficit, so the two fiscal measures can yield very different readings.28I 4.14 During the period of the SALs the fiscal deficit of the consoli- dated government, SEEs, and local governments, the sum total of which rep- resents the PSBR, was not reduced in line with the capacity of the Turkish economy and the rest of the world to accommodate non-inflationary increases in the money supply or increases in the issuance of government debt. That is, the Government of Turkey, through its fiscal deficit, supplied more growth in its debt--monetary base plus bonds--than the private sector, on aggregate, wished to absorb into their portfolios. The excess of growth of supply relative to demand required a rise in the real rate of return, as is shown in Figure 4.1 for the period since 1983. 281 The PPAR on SAL I notes that accounts that would allow the regular calculation of the consolidated government budget deficit or the PSBR were not prepared in 1980, making any attempt at fiscal control more difficult to monitor. Recent revisions in the budget figures show that such problems still exist. - 47 - 7eål I4.1A: nu=2C M M (hi ie. at IL. md pr.entge ef o S#Q 14GB1984198l1om 19=7 CTL) ( (TL) (B) (TL) (5) (TL) (5) (TL) c C~nsolidated Budge6 -106 -2.6 -907 -4.9 -767 -2.8 -128 -4.2 -12 -2.9 NM..Pn elea t S 478 -4.2 -mR -9.2 44 -8.1 -1461 -3.7 -1~47 -2.4 LA»# %oernmente -18 -0.1 37 0.2 46 0.2 -149 -0.4 -10 -0.8 EFtr.-and.uter, Pund. 0 0.0 71 0.4 161 0.6 977 1.0 -9 -0.2 n .vln Pund. 0 0.0 76 0.4 149 0.8 19e 0.4 44 0.0 -m4 -4.0 -190 -7.1 -127 -4.6 -280 -4.0 -4 -.6 Finenaing Freien Sorrowingf(te l) 21 0.8 20 2.4 8m0 0.7 Brinaft 948 8.4 21M8 8.5 270 4.4 Rsv nte 727 2.6 122m 8.1 1~0 8.7 Oeslc eurroeig 106 .6 1421 8.6 25 4.9 Canrl bank (ne) 48 1.2 215 0.7 2 0.5 Tranury lande 40 1.8 724 1.6 110 2.2 Treery lii1 (nel) 244 0.9 8 1.7 80 1.2 other (a~) -2 -0.1 -21 -0.6 468 0.9 M: bage .!e dofferences ln the orefasnl et capital ging. the figuro. for 15-87 arg n* coferable with the., of eriler yer. The figurs for 1987 are forecate. §our~ World bak lmotco. - 48 - Tkto 4.=: C M Z TAX I (<blMamn oL nd enofeptae et OM 1WD 9 ia 6 im Q imo a 1984 i 19M a 198 i Ca~ de m ida d tade6 914 29.6 1,= M.2 1,»4 18.8 2,190 19.0 2,4 14.5 4.~8 14.6 4.227 18.9 7venofers ta Local Omormfine 78 1.2 97 1.1 124 1.1 184 1.0 440 1. 84 .6 Lacel Oovernaane 49 0.7 aB 0.7 w7 0.6 1o 0.6 18 0.7 04 1.8 Eltra-budgetary Fundo go8 8.8 187 4.8 TOTAL 914 20.6 1,447 22.1 1,814 17.3 2,101 20.8 2,991 16.8 8.21 20.8 9,801 29.9 Raal Orat ui of Tas Revomnu a 12.0 -20.7 20.0 -17.8 12.2 25.8 Noma Itoms: Current 4,48C.M 0,88.8 8,788.0 11,851.9 8874.8 27.715.2 9.117.8 CPI 100.0 18.8 178.6 W7.ö 8m.4 510.8 5.9 L Data are dlflated cSlng Ce P with 19~<0. Only Central Onarncool mnd O rovenus are Included In the calaomtion. jorm:: Figur=s for the Central ove t 1980- ar fras Anuan I 0ue SDatas . Table 4.1. 198a-8 fram the placej ftlley tser . Chaptr UI. tub 2.10. Leal Oevornmun6 date fram the uas ompma , hap.ter tt. Table 2.4. E date fram the regln' luarterly lånitoring gul*shit. - 49 - 4.15 According to the data presented in the FPTR report, the PSBR has been about 62 of GNP since 1983, with some changes in its components--the consolidated budget, SEEs, and local governments. In 1987 the PSBR is estimated to have been 5.62 of GNP. The data of Table 4.5C show that con- solidated budget expenditures were reduced from around 232-24% of GNP in the early 1980s to around 20Z since 1984. But tax revenues fell from around 202-212 of GNP in 1980-81 to 142-16% in 1984-86 (Table 4.5B). Total revenue fell from its peak of 222 of GNP in 1981 to 162 in 1984. and then increased to above the 1981 level in 1986. The EBFs contributed nearly 52 of GNP to total revenue by 1986. The decline in tax effort to 1984-85 levels explains the weak overall performance of aggregate fiscal policy during SAL9 IV-V. The failure to bring the fiscal deficit into line with the targets for inflation, foreign debt accumulation, and domestic interest rates has caused deterioration or potential deterioration on all of these fronts. 4.16 The Government has not resorted to money finance as the primary means of meeting its borrowing requirements since 1984, thus avoiding an immediate inflationary impact. Rather it has made use of relatively short- term government bonds and bills. The average maturity of these is somewhat less than one year. While the issuance of these bonds has provided a new financial instrument and generated a new financial market, the real inter- est rate attached to these bonds has been excessive; it has been far above the growth of tax revenues since 1983. The real interest rate in Figure 4.4 has been around 202, and revenues have grown at an annual rate of 62 in real terms. 4.17 The budget data are disaggregated in Table 4.6 to show the Ipri- marym deficit and the interest expense. The primary deficit is defined as the deficit less interest expense. The data of Table 4.6 show that inter- est payments have risen from 0.92 to 2.82 of GNP from 1982 to 1986. They have risen from 4.82 of total government spending in 1982 to 13.32 in 1986. The primary deficit has been close to zero since 1981, so interest payments account for nearly the entire deficit. This means that the Government is borrowing to finance its interest payments. In 1986 interest payments in the consolidated budget were 1.1 trillion lire, and bond and bill issue net of amortization was 1.4 trillion (see Table 4.5C). The high level of real interest rates contributes to the level of interest expense, raising the growth rate of the debt. As the fiscal policy report FPTR emphasized, financing a fiscal deficit at a real rate of interest that exceeds the growth rate of tax revenues given the prevailing tax structure draws the country's finances into an inherently unstable position. With interest expense growing faster than tax revenue, for any given ratio of non-inter- est government spending to GNP the ratio of debt to GNP will grow continu- ously. 4.18 This fiscal instability is composed of several components that can, at least in principle, be eliminated by policy. First, the Government is borrowing at marginal effective interest rates that are higher than those paid by private borrowers. The effective rate to the Government, after adjustment for tax exemption, is over 80, while commercial paper - 50 - Tai 4.$¢ Cu~T 01T *IIA (le bi llee øf 7L 1986 1961 111 18 1984 18 1908f 1967f 914 18» 1481 314 2848 4478 61 16126 Ten R1venu1 70 111P 111 14 373 32 ome1 6816 DI~re ta*es 471 711 011 1149 132 1836 2661 Indir~o6 temo *7 479 7m8 811114 1110 "Ntn r~venmes 164 W. I16 m 416 647 10 1260 e mIuI7 1171 1mo8 1 11 8784 - 20a a10ø 116^^ Per*enl 8m1 df 411 671 IM 3275 12 24 Our curront 11 198 16 81 89 811 12 14 Invismtmont m1 m m 479 o68 mI 1614 16 Transfer* to 8EEe 171 191 34 mg 7 181 1 4 Oh~r Troasfere 199 207 32m m1 su3 1412 2240 M641 Prlmery Budge6 alene* f. -184 -41 -74 -116 -569 -192 -176 2m8 Interes P*ymet 31 75 7? 161 878 866 1*1 10 FtoeIn sernming 9 34 a8 11 264 427 2 1~ Da s6le noremlng 22 41 24 6 111 16 Iff w$ oudg.6 alane -116 -117 -151 -299 -844 -767 -1287 -1623 Incrose In Accounta Payable Nb I I -l 181 -SS -74 -170 6 Finanoe Belmme -16 -117 -161 -169 -10I -61 -1427 -1623 16 117 181 168 1116 01 1427 128 Rpeyme -87 -40 -16 -1m -24 -M02 -1667 -20 Forelgn -7 -22 -54 -156 -236 -420 -764 -1149 Deesele -10 -27 -84 - -SS -172 -7M8 -9M0 sorroing løt 1i6 169 848 1296 1488 2164 8618 Forelgn 2 47 42 6 e 69 197 624 72 Dømestle 177 119 127 200 726 1266 2811 8641 ofverment Bedm 17 28 68 191 le 67 1617 29m0 Troneury 81il (Nei) 11m 6 74 -96 284 244 66 68m Centralenk 82 72 1 l66 m57 276 Oer 6 -4 6 116 56 76 -62 61 FNANCING SUPLUM 16 117 181 219 6 7m8 584 e2n Peroenf et 4P Røvenmf 20.8 21.8 16.6 20.6 16.3 16.2 17.4 Epndlture 24.8 23.1 18.8 22.6 29.6 19.9 29.5 of whichz Transfers to SE.9 11., 2.6 2.6 1.6 4.7 6.4 Bude Baane. -8.7 -1.8 -1.7 -2.6 -6.8 -2.6 -8.1 Fineneo selenoe -8.7 -1.8 -1.7 -2.6 -. -$.1 -2.7 Le Ineludeg spelial revue and fun and annexed budgo6 revenuem. LI Etstme. e DefIned am To l Revenmes los m11 ExpedItures e*ep Zateres6 Pa~tene. h t: Tre*oury - 51 - Table 4.6s PRIMARY GOVERlT DEFICIT Estimate im 1901 1902 io 1984 198 1986 1917 (Billitone of T) Primary Budgeb Deficit Le 184 42 74 118 go9 192 176 -288 Intereet Paymensa a1 75 77 181 875 595 1'81 1,3s8 Foreign Borrowing 9 84 58 181 264 427 82 i,m Demsetic Borrowing 22 41 24 SA 111 168 899 am Overall Budget Deficit 18 117 151 299 044 787 1,257 1,528 Overall Cash Budget Defit / 165 117 131 168 1,@" 8ol 1,427 1528 Percent of 0W Primry Budget Deficit A 8.6 6.4 0.8 1.0 8.1 6.7 6.4 Interest Paymente 6.7 1.1 - 0.9 1.0 2.6 2.1 2.3 Foreign Borrowing 0.2 9.6 0.6 1.1 1.4 1.5 1.7 Demeetic Borrowlag 0.5 0.8 0.8 6.4 6.8 0.6 1., Overall Budget Defleit 8.7 1.8 1.7 2.0 5.1 2.8 8.2 Overall Cash Budget Deficit 8.7 1.0 2.1 1.5 5.4 8.1 8.6 Memo Items Interest/Total Expenditures 2.9 4.9 4.6 6.9 10.0 11.8 18.8 15.6 / Overall Budget less Intereet Paymente. L Overall Budget of the Consolidated Government plus changes In net accounts payable. Source: SPO and Treasury. rates are around 55%. The possible causes of this discrepancy are dis- cussed in sub-section E below. One is that the high borrowing rate for the Government results frem an oligopolistic equilibrium in the bond market that perpetuates the original high rates of 1934. Another is that the Government is using high rates as a subsidy to offset the losses banks make on non-performing loans. In either case, policy could be devised that - 52 - would bring the Government's effective rate below that of private borrow- ers. A second source of high rates, as discussed in sub-section A earlier, is the expectation of gradual real depreciation of the lira. Elimination of that expectation might reduce real interest rates by 5 percentage points, 10 at most. If the government borrowing rate were reduced to the commercial rate and overall rates were reduced by 5 points by a change in exchange-rate policy, the nominal government rate could fall from 801 to 501. This could reduce the level of interest expense by 37Z, from 2.82 to 1.81 of GNP, and from 13.32 to 8.31 of government spending. Policies that remove these two sources of high borrowing rates would leave Turkey still facing the core problem of any country with a real borrowing rate larger than the growth rate of revenue: debt growth that is out of control. While the SAL documents make allusion to the excessively high level of real in- terest rates and stress the need for reductions in the fiscal deficit, no explicit connection has been made between the two. The problem can be expressed in terms of two choices for the future. It may be that the ulti- mate success of the structural adjustment program will depend on these macro-economic policy choices. Either there muet be a future fiscal reform of sufficient magnitude to reduce the real value of the government debt stock, or the Government will have to -esort to printing money to meet its debt obl:.gations. In the absence of a substantial fiscal reform that takes the risk premium out of Turkish debt and reduces real interest rates, fi- nancing the deficit at such high real rates trades off low inflation today for higher inflation tomorrow. 4.19 Given the short-term nature of the debt obligations, the effect of bond finance has been to raise nominal interest rates. The above argument also implies that it has raised inflationary expectations. At a constant nominal rate of interest, the increase in expectations of inflation would lower the ex ante real rate of interest. It would also lead to capital outflows as holders of Turkish assets sought higher returns abroad unless the rate of expected devaluation also increased. In fact, nominal interest rates may have fallen during 1986 in Turkey, and there were no significant capital outflows. This implies either that the Government's borrowing strategy has not yet affected the public's expectations of inflation, or that the problem is seen as being so far in the future as to have a negli- gible effect on near-term rates of inflation. If inflationary expectations have not yet been significantly altered, nominal rates of interest well in excess of the ex post rate of inflation also are indicative of a high ex ante real interest rate. This argument reinforces the view that real rates, as well as nominal rates, are abnormally high in Turkey. 4.20 A possible alternative--not recommended here--to resorting to the inflation tax or to continuing to issue domestic debt might be to finance an increasing portion of the deficit with external debt denominated in foreign exchange. The advantage to the Government would seem to be that such loans are currently available at much lower international real rates of interest. This would lessen the explosive growth of the overall deficit inclusive of interest payments. The Turkish government, however, does not have unlimited access to the international financial marketes eventually it would face a rising supply curve of funds as the fragility of the fiscal - 53 - situation becomes evident. This means that such an option would only be available for as long as its borrowing did not reach those limits. More- over, following such a strategy would put pressure on the Government to raise foreign exchange so as to service its accumulated stock of foreign debt. The problem of servicing government foreign debt is further compli- cated by the policy of devaluation of the Turkish lira that implies a steady rise in the domestic currency cost of servicing external obliga- tions. If the additional expenditures are not met by rising tax revenues generated by the devaluation (say because of increased tradeable sector activity) then this would increase the deficit. 4.21 To summarize the points on the fiscal situation at the macro level, the government debt is on a path that is unstable in the sense that without major changes in policy, the ratio of debt to GNP would grow indef- initely. In other words, the ratio of interest expense to the ability to meet it is increasing. An essentia' element in this process is the high level of real interest rates. Growt the debt adds to its risk premium, raising real rates. The high leve.. f real rates adds to the deficit, increasing the growth rate of the dt.-. A real interest rate on the debt well in excess of the growth rate tf tax revenue makes the growth of debt accelerate. Thus, the whole process becomes self-inductive. The need to reduce the fiscal deficit and the general problem of high real interest rates have been well-recognized by the SAL program. However, the effect of the policy of gradual adjustment of the real exchange rate on real interest rates was not recognized until the PPAR on SALs 11-111. Furthermore, the interconnections between the fiscal problem, real interest rates, and exchange-rate policy and their implications for fiscal stability were not recognized until after the program was completed. In fairness, from the beginning of the SAL program much emphasis was placed on closing the fiscal deficit via increasing tax revenues and cutting expenditures.29/ Had such measures been successful, the vicious cycle of ever growing interest payments widening the overall deficit would have been avoided. With no deficit to finance, the Government would have had no interest expense. Still, the efforts at raising revenues showed only mixed results, and have not significantly reduced the PSBR. Given this fiscal situation and the high real interest rates, a more timely analysis of the fiscal imbalance may have permitted or encouraged the Government to deal with it earlier. 291 All of the SAL President*s Reports make reference to Turkish efforts to close the fiscal deficit by raising taxes. The original SAL loan included a statement by the Government that it would close the fiscal deficit in 1980181. Efforts to reduce the PSBR centered both on generating profits at the SEEs that would cover a portion of their investment plans and reductions or rationalization of other public investment projects. - 54 - 4.22 Several micro issues in public finance are also important for evaluation of the SAL process, and in particular SALs IV-V. These have been analyzed in detail in the FPTR report, so the treatment in this report can be brief, with referral back to that report. The particular issues are tax effort and tax reform, and public investment and SEE reform, which were focal points in the SAL documents, and the emergence of the EBPs, which were not. Adequate control of the overall fiscal deficit has been compli- cated by the creation of the autonomous EBFs, which revenues are specially earmarked taxes and which have the power to borrow domestically and inter- nationally and to spend the proceeds. These funds began to grow in 1984 during SAL V. As of 1987 they have generated a surplus so they have not yet contributed directly to the fiscal deficit. But this situation may not persist. The autonomous nature of these funds, furthermore, complicates fiscal policy by reducing control, increasing the chances for cross subsi- dization and generally complicating the ability of the Government to formu late a tax regime that affects domestic resource allocation in a consistent manner. The ZBFs are discussed separately in sub-section F below, and the SEEs are discussed in sub-section H. 4.23 The issues of tax reform and tax effort were taken up in SAL II and have been central to the program since then. Tax collection procedures have been substantially improved, and low-revenue nuisance taxes elimi- nated. A VAT was introduced as part of SAL III. The tax went into effect in 1985 and became a major source of revenue. The revenue side of the consolidated and PSBR budgets are shown in Table 4.58. The increase in tax efforts, as demonstrated by the ratio of revenue to GNP in 1983 (SAL III), is apparent. Improvements in tax administration during this period greatly outweighed the effects of elimination of nuisance taxes. This was noted in a positive assessment in the SALs II-Ill PPAR. However, after 1983 tax effort fell off significantly, with the tax ratio falling from 19.02 In 1983 to 14.5Z in 1984. Subsequently it increased to 15.9Z, especially with the introduction of the VAT in 1985. D. Foreign Debt and Current-Account Stability 4.24 The phenomenal growth in Turkey's exportri during the SAL program re-established Turkey's access to the private capitil markets. Partly as a consequence, since 1982 the composition of Turkey's external debt has shifted markedly toward private short-term debt and away from public long- term debt. Short-term debt has increased from 112 of total debt in 1981-82 to an estimated 22% in 1986. This change in the composition of the exter- nal debt is part of another potentially disturbing dynamic process at work in Turkey which developed during SALs IV-V. The current-account deficit was essentially financed by short-term borrowing in 1985-6. As was noted in sub-section B above, monetary policy kept real interest rates high enough to generate the financing capital inflow. The capital inflow, in turn, has cumulated into a significant component of external debt, increasing the debt-service ratio. The ratio of service of public and publicly-guaranteed long-term debt to total exports of goods and services is the standard Bank measure of the debt-service ratio increased from 28X in 1980 to 36.01 in 1986 (Table 4.7). Interest expense is absorbing a rapidly-increasing proportion of export earnings, leading to the accumulation of further growth in external debt. - 55 - Table 4.7: FOhlt4i OUT (bit ioe Il of U; end of period) 196 1961 192 1963 1984 1905 19 1967 /a TOTAL 19.0 10.2 19.7 20.3 22.0 25.8 81.* 3.1 Medium and Long Tere 10.4 18.9 18. 18.4 18.8 21.1 24.9 25.5 Publ is 14.9 15.2 16.1 6.10 17.0 19.4 23.9 23.7 Private 0.5 0.4 0.4 0.4 0.4 0.4 0.5 0.9 IMF 1. 1.8 1.5 1.0 1.4 1.3 1.1 0.9 Short Tere 2.0 2.3 1.8 2.1 3.2 4.0 0.9 7.8 Medium and Long Term Percent Of Total N 98 91 89 85 82 73 77 Short tere Percent of Total 18 12 9 11 1 18 22 28 Deb Service Ratio (3) 2 24 25 20 25 29 so 88 Debt/NP () 20 29 as s 42 40 so 6 Debt/Experte (X) g18 10 240 25 228 227 294 n.s. /a As of June 36, 197. Sources World Bank Debt Information System and Under Secretariat of Treasury and Foreign Trade. 4.25 The balance of payments data for 1980-86 are shown in Table 4.8. There che growth in export revenues in dollars over the SAL period 1980-85 is evident. Merchandise export revenue, in dollars, grew by 2842 over that period. Other export revenue, mainly from services such as shipping and construction, grew by 4132. This is the growth that re-opened private -capital markets to Turkey. Export revenue flattened in 1986, but growth seems to have resumed in 1987. Growth in export potential is discussed in some detail in sub-section D below. The more interesting aspect of the data in Table 4.8 is seen in the comparison of the current-account deficit and short-term capital. In 1983 and 1985-6, the inflow on short-term capi- tal account essentially financed the current deficit. If errors and omis- sions are aggregated with short-term capital, the fit is even better. The short-term inflow came through various channels, including foreign borrow- ing by Turkish banks and EBFs and deposits by Turkish workers abroad. It does not include the inflow of migrants' remittances, however. The net result was an accumulation of short-term debt, which is evident in the debt data. - 56 - Table 4.8s BALANCE OF PAYMENT (0illione of UM8) 19o0 1901 192 198 1984 19e" 196 Trade Balance -4,808 -8,84 -2,620 -2,90W -2,942 -2,078 -8,081 Merchandlee Exporte 2,910 4,708 S,0e0 6,906 7,809 4,2U5 7,588 Merchandise, Importe -7,518 -8,667 -8,518 -0,896 -10,881 -11,280 -10,664 Other Goode end Services, Net -974 -680 -601 -698 -679 -86 -896 Credite 762 1,816 2,038 2,041 2,86 8,148 8,20 DebIts 1,78 1,94 2,689 2,784 2,945 8,184 8,4 orker'e Reittenc» 2,071 2,490 2,140 1,618 1,007 1,714 1,684 CURRENT ACCOMT BAUANCE -8,408 -1,919 -98 -1,8 -1,407 -1,018 -1,628 Direct Foreigon Investent 1 96 66 40 118 99 125 Modium nd Long Toem Capits s 68 127 -820 44 -609 626 Dra»ing 2,284 1,972 1,780 1,624 1,161 1,169 2,670 Amortlato» -1,628 -1,289 -1,608 -1,944 -1,107 -1,8 -2,146 Short Tom Capitai -2 104 81 1,088 8 1,650 1,478 Other Capital L 1,892 88 916 784 e81 899 251 Errore and Omestone, Net 1,484 649 -76 07 817 -818 -6 CHANGE RN REEVES & -512 -283 -801 -264 207 -20 -645 Memorandum Item Current Accoent to ONP (1) 5.6 8.8 2.2 8.6 2.8 1.9 2.6 La Includes ExceptIonel Financing and Counterpart Items. A OffIcIal Reservee includIng IMF borrouingei minus sign lmplies an incresse. Source: Central Bank. - 57 - 4.26 In attempting to overcome the balance of payments problems of the late 1970's, the initial SALs provided for improving the collection of data and computerizing the statistics on the Turkish external debt. This effort fell short of debt management, however. The burst of foreign borrowing by SEEs and local municipalities as well as EBFs, and the inflow of short-term borrowing, has all been well recorded by the debt information system. But the maturity structure and modalities of borrowing have not been as well managed. To some ctent, a conflict has arisen between the Government's attempts to decentralize authority and the Central Bank's responsibility to finance the current-account deficit on the one hand, and the Government's efforts to maintain a reasonable debt servicing profile. The borrowing by the SEEs, EBFs, and municipalities creates a future claim on the Government at a time when the debt-service ratio is rising rapidly. The focus of the early SALs on information management was correct. However, SALs IV-V might have followed through more effectively in shifting the focus to debt management. E. Investment and Export Capacity 4.27 Real investment in the manufacturing sector in Turkey has been in a protracted slump since the late 1970s. The downturn in private investment began in 1978. Even after seven years of growth, which began in 1981, private investment in manufacturing in 1987 is at a level that is just 652 of its 1978 level. The downturn in public sector investment in manufactur- ing began as part of the structural adjustment program in 1981. It has steadily shrunk since then, reaching 30Z of its 1980 level in 1987. The combined effect has been to leave total investment in the manufacturing sector in 1987 at just 54Z of its peak level of 1977. The shrinkage in public sector investment has not been offset by an increase in private investment. The SALs II-III PPAR noted the need for a revival of private investment and suggested that the low level of investment might lead to a capacity constraint on exports of manufactures. This constraint would appear when the manufacturing sector reaches full capacity utilization and further increases in exports can be attained only by reducing home demand. The PPAR also argued that a revival of investment is an essential element of the transition from the stabilization phase to sustainable growth. This transition is the focal point of the SAL process. These points remain valid for the SALs IV-V evaluation. Investment in manufacturing remains in a slump and may threaten the future sustainability of export-led growth in Turkey. 4.28 The data on investment in the manufacturing sector in real terms (1975 prices) are presented in Table 4.9 and summarized in Figure 4.5. There the paths over time of public, private, and total investment are displayed. Public-sector investment in manufacturing rose from TL 15.0 billion in 1975 to TL 18.0 billion in 1980. Then, under the reductions in public-sector investment that were an essential part of the structural adjustment program, it began to shrink. By 1987, it was down to TL 5.5 billion, in 1975 prices. Private investment peaked at TL 31.4 billion in - 58 - Table 4.9: FIXED INVESTMENT IN MANUACTURING (in 1975 prices. TL billions) Year Public Sector Private Sector Total 1975 15.0 27.0 42.0 1976 15.1 31.2 45.9 1977 16.2 31.4 47.5 1978 11.2 28.3 41.3 1979 16.5 19.3 36.8 1980 18.0 16.6 36.1 1981 16.5 16.3 34.1 1982 13.8 16.3 31.0 1983 8.8 16.5 25.3 1984 8.9 17.6 26.4 1985 8.5 18.4 26.6 1986 6.8 19.6 25.7 1987 5.4 20.5 25.0 sources SPO. - 59 - Figure 4.5 Fixed Investment in Manufacturing bilions of 1975 IL 45- 40- 36- ,30- 25- 20- 16 Q 10- 1976 1976 1977 1978 1979 1980 1981 1982 1983 1984 1965 1966 1987 l Pablic + Private ToalW I4 - 60 - 1977. It then fell to a low of TL 16.3 billion in 1981. Growth since then has brought it to TL 20.6 billion in 1987. Bank reports generally discuss investment in terms of growth rates. In those terms the positive growth since 1981 in private investment can disguise its persistent low level. Tot&l investment in manufacturing peaked at TL 47.6 billion, in 1975 prices, in 1977, and then fell to a plateau around TL 25 billion in 1983-87. This is a very substantial reduction in manufacturing investment, especially for a program aimed at growth led by exports of manufactures. 4.29 The levels of the capital stock in the manufacturing sector in- plied by the investment data can be computed, given an assumed rate of depreciation and a consistent assumption on the initial level of the capi- tal stock. The growth rate of the capital stock is given by the ratio of investment to the existing stock less the rate of depreciation. Given the initial investment data in Table 4.9, an assumed growth rate of the capital stock in 1975 and the depreciation rate yield an implicit initial level of the capital stock.301 With this initial estimate of the capital stock, the assumed depreciation rate, and the actual investment data, we can iterate forward to produce a consistent estimate of the time series of the capital stock. 4.30 These calculations are shown for the private and total capital stocks in the manufacturing sector in Table 4.10 under alternative assumed depreciation rates of 52 and 102, with an initial growth rate of 5Z. Given the private investment level of TL 27 billion in 1975 and a depreciation rate of 52, an initial growth rate of 52 implies an initial stock of TL 270 billion, as shown at the top of the first column of data for the private stock in Table 4.10. Under these assumptions, the stock grows to TL 340 billion by 1987, with a plateau at TL 331-332 billion in 1980-84. The aver- age annual growth rate from 1975 to 1987 was less than 22. If the assumed depreciation rate is 101, the initial estimate of the private capital stock consistent with 5Z growth is TL 180 billion, as shown in the second column of private capital stock data. With the higher depreciation rate, in this case the stock peaks at TL 220 billion in 1979 and shrinks thereafter to TL 196 billion in 1986-7. With a 101 rate of depr6ciation the growth of private investment since 1981 would have stopped the decline in the stock, that is to say capacity, only in 1986. 301 For example, if the rate of depreciation is 5Z (.05) the ratio of investment to existing capital that would produce an initial growth rate of 5? is .10. With a private investment level of 27 billion in 1975, this implies a capital stock of TL 270 billion. - 61 - Table 4.10t ESTIMATES OF VALUX OF MANUFACTURING CAPITAL STOCK (1975 Prices T.L. billion) Private Sector Total 5 percent 10 percent 5 percent 10 percent Depreciation Depreciation Depreciation Depreciation 1975 270.6 180.4 420.9 280.6 1976 284.1 189.2 441.9 291.8 1977 301.1 201.5 465.3 311.0 1978 317.5 212.8 490.0 327.7 1979 330.0 219.9 506.8 336.0 1980 332.9 217.4 518.3 339.2 1981 332.9 212.1 529.8 341.4 1982 332.5 207.3 536.2 341.4 1983 332.3 202.9 538.9 338.3 1984 332.3 199.1 538.8 329.8 1985 334.1 197.0 538.3 323.2 1986 337.0 195.7 538.1 317.6 1987 340.7 195.7 536.9 311.6 Mehodolonys The growth rate of the cgpital stock is given by g-I/K-6, where 6 is the depreciation rate. The investment data are provided in Table 5.9. The initial growth rate g in 1975 is assumed to be 5Z. The initial capital stock in 1975 is then calculated as K(75) - I(75)/(g+5). The subsequent years' capital stocks are calculated from K(t+1) - K(t) - 1(t) - k(t). Source: Calculated from real investment data. - 62 - 4.31 The same calculations for the total capital stock in manufacturing are shown in the last two columns of Table 4.10. With an initial growth rate of 5Z and a 52 depreciation rate, the implied initial estimate of the total stock is TL 421 billion. This grew to a peak of TL 539 billion (1975 prices) in 1983, and then fell slightly to TL 537 billion in 1987. The average annual growth rate in this case was 21. With a 1OZ depreciation rate, the initial estimate for the total stock was TL 281 billion. This rose to a peak of TL 341 billion in 1981-2, and then shrank to 312 billion by 1987. In this case the average annual growth rate from 1975 to 1987 was .092. To summarize the capital stock calculations of Table 4.10, with a 5Z depreciation rate the private stock grows at an annual rate of 1.92, with a plateau in 1980-84. Since the rate of public sector investment in Table 4.9 falls through the 1980s, the total stock has an average annual growth rate of 2Z, but it is falling after 1983. With a 102 depreciation rate, the private stock shrinks after 1979 and the total stock after 1983. Their average annual growth rates in the 102 case are .07 and .09 percent, respectively. 4.32 The estimates of the capital stock in manufacturing in Table 4.10 provide maximum estimates for the economically effective capital stock. They assume gradual rates of depreciation. But many of the economic shocks to the Turkish economy since the 1970s have rendered uneconomic some of the capacity in place in 1980 or installed since then. Movements of world energy prices, first up and then down, and increases of prices of inputs produced by the SEEs, all since 1979, have made some of the pre-existing capital stock obsolete. The very redirection of the economy toward an outward-orientation would have made some stock obsolete. Thus it seems likely that old vintages of capital in the manufacturing sector could not now cover variable cost and should not be counted as capacity. This point was made in the SALs II-III PPAR. The estimates of Table 4.10 overstate the growth in the economically effective capital stock in manufacturing by an unknown amount due to the changes in the structure of the economy that have been the objective of the SAL program. 4.33 The causes of the low level and weak recovery of investment in manufacturing are several and it would be quite difficult t9 quantify their contributions to the problem. Two of these are readily identified, though. The reduction of public sector investment to less than one-third of its level in the late 1970s was an explicit component of the SAL program. Under the heading of "Rationalization of Public Investment, the program success- fully cut back the level of public investment and redirected it toward investment in infrastructure. This implied cutting investment in manufac- turing activities that, at least in theory, could be performed efficiently by the private sector. Unfortunately, high real interest rates, tight cred- it, and uncertainty seem to have prevented the private sector from picking up the slack. This can be seen in the data of Table 4.9. Even while public investment in manufacturing was being reduced by the program, private in- vestment was being crowded out by high real interest rates. Tight credit and high real rates influence private investment in Turkey via several channels. The level of real rates is shown in Figure 4.4 above. With nomi- nal loan rates around 552-602, the real rate since 1983 has been around - 53 - 20Z. This is a prohibitive cost of capital. Even if finance were available at lower rates for fixed investment, the realization that this would be the rate on subsequent working capital would deter investment. In addition to a high cost of capital in real tems, the investor in Turkey faces a system of credit rationing. As discussed in sub-section E below. due to the con- centration of the financial system, particularly the commercial banks, loan rates in Turkey are set at non-competitive rates, high for the Government and lower for the private sector. Turkey seems to be a good example of *equilibrium" credit rationing, as described by Stiglits and Weiss (1981). As the banks raise loan rates to borrowers of unknown riskiness, the risk in their loan portfolios rises. This happens because to finance a project at a higher rate, the borrower must expect a higher return. But to obtain a higher return, the borrower must assume more risk. Since the banks are aware of this--loan portfolios in Turkey contain a large and growing pro- portion of non-performing loans--they set a loan rate lower than the mar- ket-clearing rate and ration credit. So the borrower faces a high capital cost of fixed investment, a future high real rate on working capital, and the costs of ensuring that he or she is not rationed out of the market. The high level of real interest rates is undoubtedly an Important factor caus- ing the low level of private investment in Turkey throughout the SAL pro- gram. 4.34 The last set of factors suppressing private investment are likely to be uncertainty about the economic environment and about policy. Uncer- tainty about the economy for manufacturers may be characterized by varia- tions in the growth of aggregate demand and unpredictable fluctuations in the real exchange rate. The growth rate of real final demand since 1980 is shown in Table 4.11. The same table also includes figures for manufactur- ing output. After two years of negative growth in 1979-80, the growth rate of manufacturing output has fluctuated between 52 and 102 during the SAL period. Some of this fluctuation can be explained by shifts in demand poli- cy. The path of the real exchange rate over time is shown in Figure 4.3 above. While the trend is clearly negative, during the SAL program there were several periods of real appreciation. For manufacturers or.ented towards exporting, this variability of the real exchange rate creates un- certainty about profits. For example, how certain should an exporter con- sidering an investment in 1986 be about his expectation for the path of the real exchange rate in 1987? Perhaps it would rise toward the trend line, or perhaps continue the recent path of depreciation. In the face of this uncertainty, the prudent investor may postpone investing until some of it is resolved. - 64 - Table 4.11s MANUFACTURING SECTOR OUTPUT (real growth rate, Z) 1979 1980 1981 1982 1983 1984 1985 1986 Production /a Total na na na 8.8 9.8 11.3 4.4 11 Private na na na 4.2 10.3 12.9 8.9 9.3 Public na na na 15.4 9.3 9.2 -1.7 13.6 Production jb -5.2 -6.4 9.5 5.4 8.7 10.2 5.5 9.8 Exports -10.3 33.4 118.6 49.7 6.7 40.6 18.7 -0.5 GDP at factor cost -0.6 -0.5 3.6 4.5 3.9 6.0 4.2 7.3 Domestic Demand 1.6 2.8 4.7 5.5 5.5 4.5 /a Manufacturing Production Index, SIS. /b Manufacturing Output from the National Accounts, SPO. Sources SIS, SPO. 4.35 Another source of uncertainty is the future course of other ele- ments of the trade liberalization policy besides the exchange rate. Import protection has been reduced in a somewhat unpredictable way, with B levies replacing to some extent QR and tariffs.31/ On balance, it appears that the level of protection was reduced during the SAL period, but may have been increased since. The uncertainty of the trend in protection is compounded by the reduction in the transparency of the import regime as variable EBY levies replace more formal protection. On the export side, there were several changes in the subsidy and incentive schemes during the SAL period which reduced the overall rate of subsidization. However, several of the EBFes provide subsidies, and the overall subsidy level was raised at the end of 1986 in response to the stagnation of exports that year. These variations in the liberalization policy tend to undermine its credibility, and make the future course of policy more uncertain. Any pub- lic tension between the international institutions and the Government con- cerning policy adds to the uncertainty. Again, faced with the perception of these policy vagaries, the prudent investor may well decide to wait. 31l This topic is discussed more fully in sub-section H. - 65 - 4.36 The consequence of the poor investment performance in manufactur- ing may be the appearance of a capacity constraint on exports of manufac- tures. The data do not give a clear signal that such a constraint is begin- ning to bind, though. The average annual growth rate of exports of indus- trial goods from 1980 to 1984 was 402. In 1985 these exports grew by 16.51, and in 1986 they fell by 11.21. However, the real exchange rate appreciated from February 1984 to March 1985, growth slowed in the OECD, and exports to Iran and Iraq fell sharply. As is shown in Table 4.12, capacity utiliza- tion in private manufacturing has been rising since 1982. The rate is par- ticularly high in textiles and its increase particularly large in metal and metal products, Turkey's largest industrial export sectors. These are re- cent signs that a capacity constraint is being approached. Table 4.12: PRIVATE SECTOR CAPACITY UTILIZATION (1982-1984) (weighted) 1985 1986 Code Sectors 1982 198 1984 1965 I II III IV I II III 81 Food-Bevrage-Tobacco 74.6 75.8 71.9 71.9 72.4 68.7 75.6 69.7 74.5 78.4 74.5 82 Textiles-ClothIng-Leather 70.2 75.8 75.2 75.8 74.6 74.9 75.4 76.9 78.6 77.9 79.1 88 Forestry Products 57.7 69.7 54.7 69.6 52.9 57.9 61.5 64.6 59.4 6.0 71.5 84 Paper-Printing 71.8 78.9 75.7 75.2 78.2 74.6 74.8 75.9 74.1 78.8 79.9 85 Chemicals 69.6 72.9 74.1 74.5 75.1 74.8 77.9 71.7 69.9 75.4 69.8 86 Se I Products 69.8 72.0 79.5 69.9 79.4 69.0 9.1 62.8 76.9 76.7 80.9 87 Metal 68.6 65.8 67.1 72.6 71.6 74.6 67.7 76.9 72.2 72.9 75.6 88 Metal Products-achinery 59.7 62.2 69.8 66.6 69.9 69.4 64.8 79.2 69.9 70.8 69.9 89 Others 71.6 71.9 72.5 78.9 72.9 72.8 77.1 78.6 6.5 66.6 64.4 TOTAL 66.8 69.6 72.6 72.7 72.6 72.9 72.5 72.7 72.9 78.7 73.4 La Le According to the latest avallablo figures for 1987, capacity utilization of the largest industrial plants increased to 6OX. Sources Istanbul Chamber of Industry. 4.37 The poor performance of manufacturing investment during the SAL program and since, and its implication for future export performance pro- vide an example of the instability associated with high real interest rates in Turkey. The high level of interest rates has contributed to the low level of private investment since the late 1970s. This, in turn, is likely to slow export growth in the future, if it has not already. Slower export growth means a larger deficit in the current-account to be financed by a surplus on the capital account. This will, in turn, require higher real interest rates in Turkey, worsening the problem. 4.38 The role of policy as a contributor to the low level of private investment raises an issue that was discussed in the SAL II-III PPAR under the heading 'Medium-Term Strategy and Reviving Private Investment'. One objective of the SAL process is presumably to formulate policy to move the economy through the interval between a short-term stabilization program and - 66 - longer-term sustainable growth. The high level of real interest rates and uncertainty about the f4ture course of policy work uAnast this objective. The process of developing a medium-term strategy with a solid analytical basis might have brought the problems of instability associated with high real interest rates into focus earlier and the publication of a credible strategy document might have resolved some uncertainty by committing the Government to a particular program of liberalization and industry policy. The formulation of a medium-term strategy of this sort was not achieved during the SAL program. F. Financial Market Structure and Stability 4.39 One of the most troubling aspects of the Turkish economy at pres- ent and throughout the SAL program since 1980 is the fragile situation of many of Turkey's financial institutions. Real interest rates have been very high for most of the period. With exceptions in 1983 and 1986, the management of domestic demand has been restrictive. This combination of high real interest rates and depressed home demand has made it difficult for firms oriented toward the domestic market to service their debts to the banking system. In this sense, the business sector in Turkey is in a con- dition similar to that of the major international debtors, and the banking system is in a condition similar to the large international creditors. The difficulties involved in recovery of assets from a bankrupt firm and the still prevalent difficulties in the auditing system provide incentives for the banks in Turkey to capitalize interest payments on loans that are es- sentially in default, expanding the base of bad debt. These points were noted in the PPAR on SAL. II-III; they remain valid in 1987. 4.40 It is conon knowledge in government and financial circles in Turkey that commercial banks are routinely rolling over loans to insolvent debtors rather than declaring the loans non-performing and taking their losses. The size of the stock of these non-performing loans is apparently such that recognizing them might technically bankrupt the individual banks holding the loans. The potential loan losses exceed the book value of the banks' capital and reserves. The published balance sheets of the banks show non-performing loans as only 5.72 of total loans as of the end of 1986, versus 8.8? in 1985. These figures mask the true extent of the prob- lem, however, because they do not include any loans that would have been classified as non-performing had the bank not issued a new loan to cover the principal and interest of the old loan as it came due. This is what is meant by capitalizing the debt service on the non-performing loan. In discussions with the audit mission, estimates of non-performing loans have varied from a low of 2? to a high of 25-30Z for a few banks. 4.41 The presence of these financial problems has not gone unnoticed in the Bank. The efforts in the SAL and FSAL programs have concentrated on identifying mismanagement. An institute has been set up to improve the skills of bank managers. A sectoral adjustment loan, FSAL I, was specifi- cally designed to strengthen the regulatory apparatus of the Central Bank and the banking division of the Undersecretariat for Treasury and Foreign Trade which currently share supervisory authority for the banking system. - 67 - The staff of auditors has been increased, regulations concerning accounting standards have been standardized and external auditing of commerzial banks is about to be implemented. These efforts are to continue with a second sectoral adjustment loan. They should improve management and oversight of the system, but they will not deal with the underlying problem. 4.42 The problems of the system were discussed in the SALs II-III PPAR. There the systemic nature of the disequilibrium was emphasized. An occa- sional collapse, such as the Kastelli crisis in 1982 when a number of money-market brokers failed, is not an isolated event. It is the unstable system's reaction to a normal disturbance. The existing analyses of the financial system, including the SALs IV and V donuments, have generally not attempted to determine the extent to which the problems of non-performing loans stem from the structural adjustment process itself, the macro-econom- ic situation, mismanagement, or bad luck. The interactions between the structure of the Turkish financial system, the shocks to which it has been exposed and the problems of non-performing loans and potential instability were not explored under SAL9 IV and V. Shoring up the regulatory author- ity, improving the quality of management, and improving the quality and timeliness of information will only partially solve the problems of the financial system. 4.43 Few of the measures applied in the SAL program have seriously addressed issues of market structure in the financial system. The high cost of intermediation has been a source of concern since SAL II, but the possibility that it is due to the oligopolistic nature of the system has not been effectively addressed. In Turkey, three banks control 75Z of the private commercial banking system's assets and liabilities and one-third of the total banking system's assets, as can be seen in Table 4.13. These banks also are the market for government bond and bill issues. The oligop- olistic nature of the system may be seen in the spread of the government borrowing rate over the commercial rate. Since the first issue in 1984, the government rate has been 452-502, tax-free. Since 1984, the Government has developed a sealed-bid method for its bond auctions. Nevertheless, competition among the banks has nt resulted in a reduction of the rate. This does not necessarily imply collusion among the four banks. They may be in an expectational equilibrium where each expects the others to con- tinue to bid at the pre-existing rate. This is the high rate that the Government set at the initial sales in 1984 to ensure their success. Since there are only three major banks, each can monitor the others' bids. The other small banks can be content to accept the major banks' bid rates. So the equilibrium can continue, with profits for all the banks. An indus- trial-organization analysis of the banking system, using a model of an oligopolistic center with a competitive fringe might yield uiqful policy insights. The policy of opening the banking system to foreign competition has not worked well so far. With 3% of the total banking business, the foreign banks have simply joined the fringe. - 68 - Table 4.13 1 CONCENTRATION OF BANK ASSETS (billions of TL) 1980 1985 Number Assets Share Number Assets Share Held (2) Held (M) Public Banks 13 892.1 51.5 13 8,243.2 51.6 Development Banks 2 53.0 3.1 2 357.8 2.2 Foreign Commercial Banks 4 47.8 2.8 16 563.0 3.5 Domestic Commercial Banks 24 737.5 42.6 20 6,826.4 42.7 3 Largest Commercial Banks /a 3 515.4 29.8 3 4,979.8 31.1 Largest Four Banks lb 4 899.8 52.0 4 8,442.9 52.8 TOTAL 43 1,730.4 100.0 51 1,5990.4 100.0 ]a Is Bank, Akbank, Yapi ve Kredit Bankasi. Lb Is Bank, Akbank, Yapi Kredit and the Agriculture Bank, a public entity. Sources Banker's Association of Turkey. 4.44 A second issue of market structure is the interlocking nature of commercial banks and their borrowers. Turkish banks both lend to their owners and hold securities in the conglomerates of which they fom a part. The most recent revisions to the banking law nominally limit credit to a individual to 102 of a bank's net worth. But when indirect lending is accounted for, this limit could rise to much higher levels. The failure of one group of companies can then effectively bankrupt the associated bank. The interlocking ownership of banks and industrial companies contributes to the build-up of non-performing loans. The lack of an arm's length associa- tion between borrower and lender can lead to transfer pricing and other devices designed to shelter income. It facilitates the transfer of the earnings of profitable industrial members of the group to those incurring losses, thereby disguising the actual state of the latter to the benefit of the stockholders. In the event that the entire group does approach bank- ruptcy, it is easier for the deposits of the bank to be used to attempt to bail the group out. This is a further argument for more stringent com- petition policy in the financial sector, with greater separation of finan- cial and industrial companies. 4.45 The existence of a stock of non-performing assets complicates the conduct of macro-economic policy, especially monetary policy. New loans to refinance interest on the non-performing loans crowd out investors with potentially viable projects. With a large share of total credit going to meet the needs of SEEs and other public agencies, adding the requirements of the insolvent debtors substantially reduces the credit available for borrowers with viable projects. - 69 - 4.46 In addition, the pool of viable projects is decreased as the addi- tional demand for credit from insolvent borrowers tends to raise the domes- tic interest rate. If the insolvent borrowers have first crack at new moneys because denying them would bankrupt the bank itself, one response of the endangered banks is to raise the interest rate they charge to other borrowers, hoping thereby to increase profits. However, with this strat- egy, as the interest rate is increased, the pool of borrowers becomes more risky and the probability of default rises. This means that at some point the expected return to the banks begins to fall as the interest rate rises. The ability of monetary policy to control aggregate demand or inflation is thus limited by the sensitivity of the vulnerable banking system to move- ments in interest rates. 4.47 Since the financial problems are generally known to exist, though their magnitude is debated, how do the banks in question continue to attract the deposit funds to continue to finance the non-performing loans in their portfolio? Some deposits are covered by insurance which shifts the losses in the event of default from the depositors to the Central Bank which runs the insurance fund. The interest rate on these deposits, how- ever, is set by the Central Bank, which limits the ability of the banks to raise deposit rates of interest to attract funds. The fastest growing source of deposits has been foreign exchange deposits which do not carry a guarantee, but which are not subject to interest rate controls. These deposits are also more attractive to the banks as the reserves they are required to hold against them carry market interest rates and are therefore less costly than domestic currency deposits. 4.48 Aside from increasing its deposit base the bank can meet its cash flow needs by generating profits in other areas. One of these other areas is the banks' government bond portfolio. In December 1986, the holdings of the securities by the banking system, principally the TL 2.3 trillion in government debt, was 111 of the banks' total assets. These bonds were earning 491 tax-free in June 1987, equivalent to a pre-tax rate of over 801. At the same time private corporations were issuing commercial paper at 522 before taxes. This situation may reflect an attempt on the part of the Government to pass an implicit subsidy to the banking system by paying above market interest rates. The margin of government interest payments in excess of commercial borrowing rates may finance the banking system's con- tinued roll over of non-performing loans. Implicitly, the Government may be acceding to the non-competitive behavior of the banks in the bond market. G. Role of Extra-Budgetary Funds 4.49 The growth of the Extra-Budgetary Funds (EBFa) is an institutional development during the period of SALs IV-V that has already had large im- plications for the fiscal system and trade regime. They have reduced budget control and the transparency of the import regime. The EBFs are also begin- ning to play an increasingly important role as international borrowers. The EBFs are off-budget funds with assigned revenue sources such as levies on particular transactions and a mandate to make expenditures in specific areas, generally of a social nature. An example is the Development and - 70 - Support Fund (DSF), established in 1984. The DSF provides support for agri- culture and livestock projects and school dormitory construction, and cred- it for municipalities for land acquisition. Its revenues come from variable import levies, transfers from other funds, and domestic borrowing. Its activity level was TL 62.2 billion in 1985 and is estimated at TL 113.1 billion in 1986. Under the terms of a recent decree selected EBFs may be required to surrender up to 312 of their revenue to the exchequer. The decree, however, does not seem to impose any ceiling on expenditures. 4.50 Although a few of the smaller funds have been long established, for example the Petroleum Exploration Fund (PEP) was established in 1979, their growth has been most rapid since 19e4. By 1987 some 80 funds existed, controlling revenues equal to about 202 of the central government's reve- nue. The growth of the funds has moved a significant fraction of government social spending and revenues out of the budget. The EBFs have earmarked tax sources and borrowing power. They represent a loss of fisual control at the level of the central government. The basic data on the largest funds are summarized in Table 4.14, taken from the FPTR report, Table 7.1. By 1987 the largest fund was the Public Participation Fund (PPF), which is the agent of privatization in Turkey. The PPF builds and operates dams, high- ways, bridges, waterworks, and other infrastructure, and sells revenue participations in them. It receives revenue from a petroleum consumption tax and it has authority to borrow domestically and abroad. The 1986 activ- ities of the PPF are summarized in Table 4.15. The PPF acts much as a high- way or port authority in the U.S., except for the assignment of the petro- leum consumption tax. The second largest fund, the Mass Housing Fund (MHF), provides a different example. Its 1986 activity is summarized in Table 4.16. The MHF's revenue sources are almost exclusively earmarked taxes--half of the variable import levy revenue, supplementary VAT, petro- leum product taxes, and a $100 tax per exit on Turkish citizens. It pro- vides credits for construction of small housing units and supports land acquisition. It also makes transfers to the DSF for complementary activi- ties. The other major funds' revenue comes principally from earmarked taxes. The DSF gets the other half of the variable import levies; the Resource Utilization Support Fund (RUSF) taxes bank loans; the Support and Price Stabilization Fund (SPSF) taxes trade. So the PPF is the only major fund that does not receive most of its revenue from earmarked taxes. Aside from their control over earmarked tax sources (about which, more below), the EBFs resemble their counterparts in the U.S. and presumably other OECD countries. They provide credit, subsidies, and other supporting services to sectors or activities that are perceived to meet social needs. The Federal Home Loan Bank Board, the Tennessee Valley Authority, and the New York Port Authority are, to some extent, analogs in the U.S. The principal structural - 71 - XÉL*M li MieC3a. 013m- urTM PUs. 190-0 ost. 1985 nevenu. Punds Wn Etstblished nevenae Prjoch maje r ets e et d en e majer Ares u me (bl ten EL) %fen. Ifdt~hry uer60 Pund - IP) i9 - 1u4.1 om feass te~ en ae åuee r ta IfgestvmI in *he dires6 tean eerpergie end dgstl defense Induefrt . perenet 1,~ nn-aa revene. DoMemfås., end Supper P.nd 0> 19%4 8.8 118.1 -uT en luxury laperte and t~ens- Price mespprt e InIel fe- tara frme other fund.. steek and social pr~gr~es. Ml.. Houasin Pund n$) 194 212.4 818.7 Tax en luxury Idprte ehere e Sueldla.d hueing eredi . suppminentery VA? and trnofer. frnu ther funds. P*lteum CGnuestlen innd (PW|P 1984 8.0 0. Tas en fucst ce ean. Fin~fing nvs.t.ente of local eferunte od hl~a7 md srt dlreoteree. Pemrel au. EmplersIlen Pund (P) 197m 78.8 f. 90.0 Ten en deaeble peroeum. Tr~nser te other fund. (einly SPF). Pubila Prleiosaon Fund (PPPM 1m84 22=.7 431.7 Uer eh~rge. on pu lt infra- Pul* Invet~nl ln stru~tue3 leuence of revenus- onfrstrtre. ~haring rilueatmee sft.a of 0etroleun Ceusslen fs. s~crae Uhl Baalten supper UUP) 19m 112.8 172.2 Tax en bank ..n.. Exper6 end inafvecten subidy. Supprl und Pril* StbIlllasion Pund 190 207.7 go0.0 Ts en Imports ad egrIeul~al Fert§lste subeldy. i Prewlolana. & Refer to 19j4 and t no6 Incluosod n the ote. * 72 - Table 4.15: SOURCES AND USES OF THE PPF, 1986 (TL billion) Sources Uses - Sale of revenue-sharing - Maintenace expenditure 15.7 certificates 60.0 - Investment expenditure 314.0 - Operating income 153.7 Construction of dams 140.0 Bosphorus bridge (18.0) Karakays dam (64.0).- leban dam (113.5) Ataturk dam (38.0) Oymapinar dam (22.2) Altinkays dam (15.0) Construction of - Interest income 12.0 highways 38.0 Potable water works 16.0 - Sales of Treasury bonds 25.0 Second Bosphorus bridge 77.0 - Foreign borrowing /a 71.0 Free-trade zones 4.5 Priority regions 36.5 - Petroleum consumption tax 50.0 - Repayment of revenue- - Balance at Ziraat Bankasi 30.0 sharing certificates 92.0 TOTAL 421.7 TOTAL 421.7 /a For the second Bosphorus bridge. Sources SPO and UTFT. Table 4.16: SOURCES AND USES OF THE MHF, 1986 (TL billion) Sources Uses - Share of supplementary VAT - Housing credit 284.0 on tobacco, alcoholic, to cooperatives (21.7) non-alcoholic beverages 88.0 to individuals (52.3) - Petroleum product taxes - Transfers Io DSF 10.0 (from PPSF) 111.0 - 50Z of levies on luxury - Tourism infrastructure 5.6 goods imports 32.0 - Land acquis4tion 11.0 - Tax on travel 18.0 - Others 1.4 - Premium payment from Imports without waiver 21.8 - Others 20.2 - Balance from 1984 24.7 - Balance carried forward 3.7 TOTAL 315.7 TOTAL 315.7 Sources SPO. I - 73 - difference between these and the EBFes in Turkey is the pre-emption by the latter of some of the central government tax sources, and the manner in which the tax rates are set. The effect of the latter process on the trade regime was discussed briefty in the SALs 11-III PPAR; the growth of the EBFs makes it more salient now. 4.51 The variable taxes, mostly in specific terms, are set by the appropriate Deputy Undersecretary for Treasury and Foreign Trade, in con- sultation with the director of the relevant EBF. Many of these taxes are variable specific levies on imports. As items are removed from quota lists, they are moved to the 'free list', which makes them potentially subject to fund levies which in select cases are set at levels that provide 'efficient protection'. Levies will be gradually reduced as domestic firms modern- ize to become competitive internationally. Several heads of firms that produce import-competing products confirmed this view of the process that sets levies. Again to paraphrase, the levy is set to keep the import price slightly above the price of the import-competing good. Since the levies are variable, they can be changed administratively in response to market conditions. This makes them more effective than fixed tariffs as protective devices, and makes information about the tariff-setting process more ob- scure to agents who are not intimately familiar with particular market conditions. The fact that the levies are specific requires that anyone desiring to compare them to tariffs or to compare one levy rate with others must make detailed calculations. This adds to the obscurity of the import regime. The gradual shift in emphasis from QRs to tarI ffs to variable levies has thus decreased the transparency of the import regime in Turkey, increasing the returns to those with specialized knowledge of the protected sectors and the levy-setting procedure. The growth of the EBFs has been an essential part of this process. They are the vehizle by which the protec- tiv4 effect of the import regime has gradually been removed from easy scrutiny. 4.52 The earmarking of taxes for the EBFs effectively removes them from fiscal control and increases the fiscal deficit. The levies are set with protective effects as the principal objective. The revenue flows to the EBFs, which are free to spend it subject to limitations described in para. 4.50. In some important cases they also are able to borrow. As is shown in detail in sub-section G on the trade regime, the imposition of earmarked levies increases the fiscal deficit because the levy revenues are spent while the fall in the quantity of transactions in response to the levy reduces normal tax or tariff revenue to the budget. A good case can be made for the existence of self-financing EBFs to provide specific social ser- vices. But the use of earmarked variable levies, as in Turkey, simultane- ously undermines fiscal control and the liberalization of the trade regime. Careful scrutiny of this aspect of the growing EBFs was possible during the SALs IV-V periods. Reform on this dimension will be important for the sus- tainability of the process that began under the SALs. 4.53 The relevance of the EBFs to the design of the SAL program is that within that program there was pressure to control governent spending and raise tax revenues while rationalizing public sector investment. The EBF - 74 - was a means of meeting these pressures In the short-run. These new enti- ties initially avoided the scrutiny afforded the traditional entities in- volved in public investment: the line ministries and the SEEs. The streamlined, *can-do* structure of many of the EBFs and their close link with the highest levels of the government made it possible for services to be provided at less cost than would have been the case had traditioinal government agencies been used. 4.54 The quasi-independence of the EBFs, however, also poses a threat. Unlike the borrowing authorities that are a usual part of government struc- ture in any country, such as the Federal Home Loan Bank Board in the U.S., for example, the EBFs have allocated to them the revenue from a dedicated tax. Furthermore, they exercise a substantial measure of control over the taxes' variable rate. Since the tax is in many cases an import levy, it can easily be used for protection, even at a micro level that might be suscep- tible to corruption. Since the taxes are generally specific instead of ad valorem, it is difficult to estimate their actual levels. Thus while the EBFs may increase the effectiveness and flexibility of the provision of government services in the face of cumbersome budget procedures, they also tend to increase the PSBR and reduce both contrl and transparency of the fiscal process &ad the trade regime. i. Trade Liberalization 4.55 Liberalization of the import regime and rational *ation of the system of export subsidies with the intention of eventually eliminating it have been policy objectives of the Turkish government since 1980. Tariffe have largely replaced quotas in the Import regime, and the system has been changed from implicit prohibition ef imports to implicit freedom. However, the proliferation of EBFe sincP 198A, with their levies on imports, is con- straining progress on the import side. The export success of Turkey, at least until 1986, has been phenomenal. But less progress has been made in eliminating export subsidies. These have been increased in 1987 in view of the stagnation of exports in 1986. On both fronts, export subsidies and import duties, a conflict between the fiscal balance and trade liberaliza- tion has developed. 4.56 The data on exports by destination are shown in Table 4.17. From 1980 to 1985 total export value in U.S. dollars increased by 273Z. The increase in exports to the Middle Eastern countries was over 6502. But the Middle East was not the sole source of export growth. Exports to the OECD area grew by 245Z from 1980 to 1985, an annual growth rate of 18%. In 1986, however, total exports fell by 6.3Z in current dollar terms. Exports to the OECD increased by $186 million, or 4.52, while exports to the Middle East fell by $910 million, nearly 30%. All of the decrease to the Middle East was accounted for by Iran and Iraq. The reduction in export growth to the OECD area may be more worrisome than that to the Middle East, however. While exports to Iran and Iraq fell by $910 million, imports from those countries fell by $1410 million. The symmetry probably reflects the fall in 1 75 - IThle, 4.1, EXPORTS BY DESTINATN (i a111n of U.S. dollars) 19 191 1962 198 1964 1986 1986 OECD Countries S,679.7 2.96.? 2,56.0 1,764.1 39.? 4.1$6.2 4292. EEC Countries 1,242.1 1.W. I.7.4 2916.0 t.731.7 8188.3 392.1 West sermaey e4.0 648.2 797.4 87.6 1,279.7 181. 1,446.6 France 168.9 215.7 194.6 190.5 200.6 218.8 298.7 Italy 216.4 246.1 827.5 422.8 61.2 562.2 57t.6 England 194.5 146.4 169.6 247.0 260.6 518.7 884.2 Othr 151.8 249.9 88.7 822.? 469.4 496.1 606.4 Other OECD Countries 487.6 76.6 66.6 76.0 1690.0 972.6 1629.1 USA 127.4 267.9 251.6 281.7 868.2 S6.6 649.8 Switzerland 125.4 268.7 828.9 266.5 853.2 126.4 162.8 Oters 164.6 229.2 225.1 281.6 261.6 888.4 817.6 Eastern Europe 494.8 382.2 245.8 288.7 884.5 810.5 Middle East Countries 494.6 1,098.§ 2.640.1 2,05.1 2.759.7 3,28.2 2,865.9 Iran 64.6 288.7 791.1 1,067.7 760.9 1,78.9 564.4 Iraq 184.8 569.6 610.4 820.6 984.4 961.4 558.8 Libya 61.8 441.5 284.6 164.8 142.6 56.6 185.6 Saudi Arabia 48.6 187.4 857.0 864.7 876.6 480.0 857.4 Egypt 26.8 72.1 145.6 76.2 146.6 141.6 145.2 Other Countries 245.6 .0 826.7 264.8 U 279.8 48.2 TOTALS 2,919.1 4.M.9 6,746.9 5,724.9 7,133.5 7,959.2 7.46.S Le Include Sudan, Syria, Jordan, Israel, Labanon, Abu Dahbi, Babrals, qatar, Oman, Dubat, Kuwat. Source: 8S, SPO. the price of oil. This reduces Turkey's oil import bill, but its income effect on the oil exporters reduces their imports which are Turkey's ex- ports. On balance, and in purely foreign exchange terms (excluding the employment effect), Turkey gained. The slowdown in growth to the OECD area, however, may reflect more fundamental problems. The growth rate of total demand in the OECD in 1986 was 2.52, not particularly slow given recent history. So the problem may be in supply elasticity and structure in Turkey. High real interest rates and slack domestic demand have contributed to a low level of investment in manufacturing in Turkey since the 1970s. This could mean that exports are approaching a capacity constraint. - 76 - 4.57 The most covious area of conflict between trade policy and fiscal balance is the use of government subsidies to exporters to increase the incentive for private sector producers to export. This conflict has been recognized and was Inctrporated into the SALs by reference to the various export promotion subsidies as temporary. These subsidies include tax re- bates in excess of taxes paid, access to duty-free imports, preferential finance, and subsidies from EBFs. Estimates of the subsidy rates, measured by the ratio of payments to total export values are shown in Table 4.18. The term "tax rebate" is a misnomer since the subsidy rate is not related to the amount of taxes paid by the exporter. It was originally justified as a substitute for a VAT rebate in the context of the GATT, which prohibits export subsidies but allows VAT rebates. Since Turkey instituted a VAT in 1985, an actual VAT rebate was added, making the prior rebate scheme purely a subsidy. In 1985 the total subsidy rate from the tax rebate and duty- free import schemes varied across commodities between 1OZ and 202 of export value. The subsidy rate was highest on capital goods and lowest on consumer goods. The combined subsidy rate from tax rebates and duty-free imports continued essentially unchanged in 1986.32/ Table 4.18: TURKISH EXPORT SUBSIDIES (in Z of exports) 1980 1981 1982 1983 1984 1985 1986 Export Tax Rebate 0.0 3.6 10.1 11.5 11.1 9.7 7.6 Duty-Free Imports 5.5 4.4 4.2 5.4 2.9 4.9 6.2 Export Credit 16.6 12.5 6.4 6.5 1.1 0.0 0.0 RUSF 0.0 0.0 0.0 0.0 0.0 4.0 2.0 TOTAL 22.1 20.5 20.6 23.4 15.1 18.6 16.0 Sources Milanovic, Export Incentives and Turkish Manufactured Exports, 1980-1984. World Bauk Staff Working Papers, No. 768, January 1986, pg. 49, and Foroutan, Trade Policy in Turkey in 1986 and 1987t An Evaluation, draft No. 27151, March 1987, pg. 12. 4.58 Steps were taken from 1984 to 1986 to phase out preferential fi- nance and EBF subsidies for exports. Preferential credit was ended in 1985. The RUSF provided a non-selective cash incentive of 4Z for exports in 1985. This subsidy was eliminated in two steps during 1986. The estimated total subsidy on export value in 1985, shown in Table 4.18, was 18.6Z, half from tax rebates and a quarter each from duty-free imports and RUSF. The esti- mated rate in 1986 was 162. In response to a fall-off in export growth in 1986, however, preferential finance and a selective cash subsidy for ex- ports by the SPSF were re-introduced at the end of 1986. The SPSF subsidy varies across 110 commodities with rates reaching a maximum of 16.7% for 321 Foroutan, F., 'Trade Policy in Turkey in 1986 and 1987: An Evaluation', Draft No. 2715-1, March 1987. - 77 - chemicals. The level of the credit subsidy is not yet clear. But the over- all subsidy rate in 1987 is likely to be above that in 1985. Prefer- ential finance and the SPSF subsidy were introduced in late 1986 in re- sponse to the stagnation in export growth in 1986. There is evidence, however, that one effect of export subsidies has been to promote fictitious exports. According to Rodrik (1987) the value of Turkish imports regis- tered as being received by OZCD countries averages 16? below the value of exports to OECD countries recorded in Turkey.331 The over-invoicing is con- centrated in a few countries, particularly Germany where in 1984 the dis- crepancy was 451 below the value of the exports registered in Germany. Figure 4.6, taken from Rodrik, shows the time series of Turkish exports to West Germany as registered by Germany, compared to the values registered in Turkey. The difference is the level of "fictitious exports.' The arrows in the figure show the points where export incentives were increased (arrow points up) or decreased (arrow points down). There is a clear positive relation between changes in incentives and the fictitious export gap. Rodrik's regression analysis shows that export subsidies were insignificant in explaining the growth of exports as valued at the destination. It also showed that movements in the real exchange rate variables that represent the Iran-Iraq war significantly affect actual exports. If this analysis is correct, the increase in export subsidies may only have a ficticious effect on exports while worsening the fiscal imbalance. 4.59 The other link between fiscal policy and the process of trade liberalization has been the growing importance of EBFes since 1984, during the tenure of SAL V. The conflict with trade liberalization is that the source of funding for these organizations is often earmarked import levies. The authority to set variable levies provides a substantial incentive to succumb to domestic pressures for protection from foreign competition. This problem was discussed in the SALs 11-111 PPAR and in the previous section. The SALs IV-V audit mission found that the fund levies continue to be set so as to provide systematic protection to domestic import-competing firms, many of them SEEs. Thus to a substantial degree reductions in formal tariffs have been offset by the variable fund levies. The fund levies on imports were intro"uced in 1984 as a means of financing social expendi- tures. They were t3 apply to luxury imports. The coverage of the levies rapidly increased, so that by 1987 fund levies applied to 572 composite commodities, covering approximately 302 of total imports. The levies are set as specific dollar amounts. The tariff equivalent values of the levies in 1987 are shown in Table 4.19. The highest levies in percentage terms are on capital goods and agricultural imports. For example, non-electrical machinery and transport equipment have levy rates of 25.9? and 36.4%, re- spectively, and live animals and plants have a rate of 64.1?. The overall levy rate, as measured by the ratio of revenue to import value, was 5.2% in 331 See Rodrik, Dani, "External Debt and Economic Performance in Turkey, forthcoming in T. Nas and M. Odekon, (eds.), Liberalization and the Turkish Economy, Greenwood Press, (1987). - 78 - Figure 4.6 Export Performance, 1975-85 000 6000 total = 4000- o 2000 0 1i75 1976 1977 1978 1919 1i80 1981 1982 1983 1984 198" - 79 - TIabl J.' TMPP JIVALUO 8' LM IU IN 1 07 (1) (8) <() (4) (83 (8) Valuo of Tariff th6ber NumbIer EeIvalent Zapr Overall Affected .t et Ak~rage Töte i with TorIf 160pr. Ies Lvioo Levy I.porte Levi EkuIvolont 3rN Gras (60M)(4n 2) (cmialuf.) (Mtml (m:*M()+6 01-0 Live mim4a. plant. 39.74 18 4.1 0.7» 78,484 84.4 34.9 07-14 Frulte end vectables 197,180 a 18.6 0.91 221,9~ 9. -14.0 1W-1O Mant produote 218,001 28 18.9 1.00 218.622 99.7 .15.9 17-19 Sugar, .cfeeinors 9.8~7 9 1. 0.04 11.80 0.8 U.0 20 Fr.l proceosin 3,~ 7 18.6 0.02 8.821 100.0 .18.6 21-24 everage, em~are6te. 80,101 19 74.8 1.91 CG,97 6. 618.7 25-2 pktal 0~, ~cmne 10 8 32.9 0.00 1M,84 49.8 68.7 27-3m Chemlmle 8,711 0 18.8 2.08 3.298,378 10.2 1.8 40 ubber and slaceih 8,222 21 6.0 0.01 97,724 24.0 1.6 41-48 Loather and fur 13,388 8 13.0 0.06 47.98 82.0 4.2 44-49 Vood and paper 62.619 87 4.7 0.18 148,297 48.7 3.8 50-89 Tent; lco 211.79 25 3.3 0.30 294.8m8 71.9 2.4 60-6m Clothing 9,25 nt 41.6 0.12 20,18 47.7 +19.8 89-71 Ola s and cam 7.27 18 11.2 0.1 57077 12.8 1.4 73 Irn and ~teel C8,289 29 9.7 1.95 1.059.08 64.8 6.3 74-79 Non-förrou mam 6,= 10 54.8 0.10 2.870 3.0 1.6 g0-~8 ~et1l produ~ts 17,979 u 20.1 0.1 4,.70 84.2 6.9 84 Netelebrle chInery 780.08 44 25.9 8.44 1.882,40 47.8 3.0 85 Elocrl omchinery 95,8 48 20.4 0.87 88.498 14.8 8.0 af-9 Transporuatlon ouipment 44,888 16 38.4 4.5 782.601 85.6 20.2 90-91 Mauring Mqulpanl6 34,416 14 28.8 0.29 185.88 20.8 5.9 92 ElmbIrenIe euipum% 8.878 7 108.2 0.10 78,888 4.8 4.6 94-99 NiualImaneoua 8.775 47 19.8 0.04 22,646 29.9 8.9 TTALS 3.40.05 572 20.1 11.240.471 80.3 8.1 Tariff .~uivalen Agro: 1-24 82.349 112 26.4 888.817 81.5 25.4 Tontil lo 80-68 221.604 67 4.9 315.043 70.3 8.8 Con~umer good. 748.952 179 20.0 808,880 82.2 16.8 Intorldiate good. 1,3 784 211 1.8 7.045.486 19.0 2.1 Capitel good4 1,82,08~ 1m2 29.1 8.M.678 40.3 11.7 TOTALS 3.406.86 872 o0.1 11.240.471 30.3 6.1 soua: world &ak emuimat*s uninG data of UPU and 8PO. - 80 - 1985 and 6.12 in 1986. The administrative, variable, and selective nature of the fund levies makes them an effective instrument for protection. The EBF levy system is an important step backward from the import liberaliza- tion program that began with SAL I. The beginning of this trend was noted in the SALs II-III PPAR, and it grew rapidly during SAL V. 4.60 The fund levy system increases the budget deficit to an extent that has not been estimated. The levy raises revenue that will be spent by the fund. Thus if the funds maintain balanced budgets on current expendi- ture, a levy increase will not change the PSBR directly. The levy increase, however, will reduce import quantity in two ways. First by raising the domestic price the quantity of imports will fall by the price elasticity of demand times the levy increase. This is shown as the decline from B to F in Figure 4.7. At the same time the higher protection will allow more inefficient domestic suppliers to meet overall demand, further reducing imports by the supply elasticity of demand times the levy increase. This is the change from A to E in Figure 4.7. The overall lose to the Govern- ment, which raises the PSBR, is simply the tariff revenue lost because of the lower quantity of imports after the imposition of levies. This is the sum of the two shaded boxes in the figure. 41 4.61 The other, more general problem, with the EBFs is that they frag- ment the tariff structure and its control. If the EBFe continue to ex- pand, a tariff structure will continue to develop in which effective rates of protection once again show a high level and a wide variance across dif- ferent sectors of the economy. The EBF levies in 1987 contribute an esti- mated 6.12 of an overall rate of nominal protection of 25.5%. They are highly selective and can be varied by administrative discretion. The evi- dence is that they are varied to provide systematic protection to domestic producers. In combination with the numerous other variable taxes that are earmarked for EBFs, this has created a tax regime with distortional effects on resource allocation that are unmeasurable. All that is knowable at this time about the system is that it is protective to a high degree. 341 One can present the loss in somewhat more mathematical terms. Denote the level of the formal tariff rate by t, the (negative) price elasticity of demand by d, the price elasticity of domestic supply by s, and the increase in the levy as a percentage of the pre-levy, tariff-inclusive price by 1. Let Q be the original level of imports. If one could separate the fall in imports into the portion attributable to raising domestic supply and the portion attributable to falling domestic demand, then the formula for the loss in tarif revenue would be (a + d)tfQ. This would be the increase in the PSi. due to the increase in a fund levy. - 81 - Figure 4.7 Effect of Levy on Toriff Revenue 34- 32- 29- D 26 24 22 20 14- 18 ulo- 14 2L sd~ 12a ta - 82 - I. Reform of the State Economic Enterprises (SEEs) 4.62 An important aspect of the SAL process has been reform of the State Economic Enterprises (SEEs). The objectives have been to increase their efficiency, reduce their scope, and reduce their contribution to the fiscal deficit. The Government has freed the SEEs to set their prices at market levels, reduced the size of their investment programs, and substan- tially reduced their borrowing from the Central Bank. These measures were taken beginning with SAL I in 1980. More recently, during SAL IV in 1984, the Government, supported by the Bank, began a program aimed at improving the internal efficiency of the SEEs. The PPF began the process of privati- zation of some of the SEEs beginning with the Bosphorus Bridge. The PPAR on SALs II-III included a critical discussion of several aspects of the SEE reform program; most of the points made there are still relevant. 4.63 The SEEs are a major factor in the Turkish economy; they were the main instrument of the policy of import-substitution. In 1979, before the beginning of the SAL program, they were producing 40% of Turkey's manufac- turing output, and their gross revenue was 19.5Z of GNP. Their employment was 30? of total industrial employment. Their operating profits were negative, and they generated three-quarters of the PSBR of 8.61 of GNP. As the SALs II-III PPAR noted, the SEE system needed reform. 4.64 The mixed results of the reform are summarized in Table 4.20. The main effects that are perceptible there come from freeing the SEEs to raise prices to meet market conditions and cutting their investment programs. Revenue from sales, equivalent to 25Z of GNP in 1980, peaked at 341 in 1984. Profits have become significantly positive. The SEE share in the PSBR in 1986 was down to 3.12 of GNP and is projected to fall to 2.41 in 1987. In 1980 SEE fixed investment was 24.52 of SEE revenue and 32.5Z of total fixed investment in the economy; by 1985, these ratios were 17.2Z and 28.8%, respectively. SEE employment fell from 30% of total industrial employment in 1980 to 28Z in 1981. However, by 1983 it was again above 30?. The effects of the pricing reform are evident in SEE revenues and profits, and the reduction in their investment level is clear. The results for effi- ciency are not obvious, however. To obtain any estimates of movements in SEE productivity, the increases in sales revenue must be decomposed into changes in price vs. quantity, and the value-added component of the quanti- ty change identified. A recent Bank document, "Turkey: Private Manufactur- ing Assessment of the Impact of Past Policies and Future Adjustment Needs, 1987,0 shows that in 1980-81 the SEEs gained in labor productivity as mea- sured by the ratio of value-added at constant prices to the number of em- ployees. The same data show, however, that some of the gains were reversed in 1983. An approximate measure of SEE productivity is shown in Table 4.21. The State Planning Organization (SPO) provides data on SEE sales and inven- tory accumulation, and the SIS provides an index of public sector prices. The sum of sales and change in inventory is total production. Deflating by - 83 - Table 4.20s 8EE FINANCL. POSITION AO ELOYMEN7 (Sill lione of TL) 99, 1961 1962 im nu 1985 1986 Sal.. 1148 1767 2658 8596 681 9819 12617 Carr GP 4,48.2 6,~8.8 8,785.9 11,~51.9 18,874.8 27,715.2 89,117.8 S.ls.. of GNP 26 26 80 81 34 88 81 Pref16s (L~oss) -88 -88 16 -10 848 551 564 Pr Ut (.Less) > s of G -4.8 -6.5 6.11 -1.8 1.8 1.9 1.2 Flud Ifvitmet by the 8EE. 281 489 588 s85 968 1684 1988 Tetal Flxed Invs ~ 868.6 1,241.4 1,648.9 2,188.4 8,867.6 6,661.8 9,245.8 (Prvte & Public) = Pimad Iwvestmn6 as % of T.6M1 82.58 82.9k 82.86 27.46 28.68 28.84 21.68 Plaed Ifv*sm~6 TAtWl Finanelna Reaulred: e dh. o E. 474 624 628 786 920 1284 1244 PM -560 -554 -02 -494 -1805 -1277 -2860 ase 1 et PIM 84.6 112.8 108.9 186.9 78 94 62.9 a of Cm 11 9.5 7.1 6.8 6.0 4.8 8.1 FlamIna o s 8~dsbary Tranfers 158 241 208 292 289 181 140 Tran~fo fr= PrIco ftebIlIzlg Fund 80 74 76 18 178 248 177 &erslag from C~1ral Dank 58 82 80 -5 -56 80 8 96ete Iovesm ank 16 16 59 28 49 19 74 Foreiga Borreing 67 122 104 284 289 480 784 Sh~r6-Ter. Oorr~Ing 158 189 162 78 226 248 111 SEE@ EMPLOYMENT Total / 647,871 521,967 128,617 586,872 e8,f96 858,86 694,490 Tat~ 1 547,871 621,967 628,617 EN0,872 67,986 568,084 68,961 Per 6 mShr* In 8 28 28 88 88 81 82 IadeeteIl Eeploymen Peeh Shmra In Induahrial 80 28 28 80 28 27 27 Promu6 Shar. In Toal 8.69 8.89 8.41 8.72 8.17 4.09 4.26 Civillmn Employ~men Per~a Shr. l Total 8.69 8.89 8.41 8.72 8.6 8.52 8.7 Civilam Eapoyment Ms 0 and Treasury. - 84 - Table 4.21: PRODUCTIVITY OF TI SE1. Number of RPI ear Real Product (RP) EmploMees E) TL millions (TL billions, 1981 prices) (thousands) workers 1982 2,174.6 528,617 4.11 1983 2,282.4 580,372 3.93 1984 2,831.4 658,096 4.30 1985 2,646.5 653,066 4.05 1986 2,693.8 694,490 3.87 Sources SPO. the public sector price index yields the data on real product in the table. The ratio of real product to number of employees is real product per worker in the last column of the table. As the figures do not exhibit a positive trend since 1982, it is not possible to conclude that productivity has been increasing. This suggests that improvements in SIR profitability are due to price increases, not to efficiency gains. A data set compiled by the Istanbul Chamber of Industry shows that the composition of nominal value- added is consistent with the surpluses generated by the SEEs, being the result of raising prices rather than increasing productivity. The figures are presented in Table 4.22. The share of profits in value-added among the SEEs increased from -2.6Z in 1983 to 48.92 in 1985. The corresponding fractions in SPO data are -52.6Z and 36.1t. At the same time the share of profits in private sector firms fluctuated between 24Z-282. The relative increase in SEE profits is consistent with increases in SEE revenue being generated by the SEE price increases rather than productivity growth during the period. Table 4.22: DISTRIBUTION OF VALUE ADDED BY SM LARGE FIRMS (a s percent of total Value added) All Fim Private Firms Public Pias 1982 1988 1984 198 19 1988 104 1985 1982 1988 194 1985 Wages and Salaries 82.8 SS.G 40.4 48.4 424 42.7 8.2 89.4 88.4 74.9 57.4 41.8 Interest Payments 27.0 28.9 22.2 24.8 81.0 20.9 88.1 80.2 28.0 27.5 7.4 14.0 Rents 0.4 0.4 0.4 0.5 e.0 0.5 0.7 0.7 0.2 0.2 e.1 0.2 Profits 19.4 15.2 81.0 84.5 26.1 28.9 28 28.7 10.4 -2.0 85.1 48.9 Net Value Added UW S lee 10 in too 100 10 100 10 10 1lf SoaMls Journal of the Istanbul Chamber of Industry, September, 198W, Ne. 21, pp 154-108. - 85 - 4.65 In addition, the Istanbul Chamber of Industry data show a signifi- cant fall in the share of value-added going to interest expense among pub- lic enterprises relative to the private sector in 1984 and 1985. This came at a time when concessional credits were being phased out. It is possible that the reduction in public enterprise interest expense may be the result of different amounts of leverage between public and private firms in the Istanbul data. But it may also be that the elimination of concessional credit was less complete than expected. It is certainly worth investigating the source of the divergence. 4.66 A more fundamental concern than the SEEs generating a profit is that they efficiently utilize resources. Among the SEEs are public utili- ties and other natural monopolies. In addition, it is argued in sub-section F on the EBFs that their levies are set to protect systematically the in- port-competing SU9. Under these circumstances, a profit-maximizing policy for the SEEs will result in monopoly pricing. This point was stressed in the SALs I1-Ill PPAR. Ideally, the set of SEEs that produce public goods should be identified and subjected to regulation, while the others should have their pricing policies shaped by international competition. In the presence of continued protection, the latter group not producing public goods should also be subjected to modified regulation, rather than left to set monopoly prices. While the two major decrees reforming the administra- tion of SEEs made a distinction between those enterprises supplying public goods and those providing private goods, no distinction has yet been made in defining pricing policies. To avoid the continued abuse of monopoly or oligopoly positions by the SEEs, the market structure faced by each SEE should be taken into account in a systematic regulatory policy. 4.67 Efforts have been made during the SAL process to address cost minimization as well as to boost output prices as a means of covering defi- cits. The principal initiative was the provision of bonuses for managers who showed exceptional skill in efficiently running their businesses. It is unclear how successful these measures were. The laws themselves contain no criteria upon which bonuses should be granted. A master plan for priva- tization of the SEEs was completed under SAL V, and in May 1986 the govern- ment obtained legislative authorization to begin the privatization process. V. IMPACT OF STRUCTURAL ADJUSTMENT PROGRAM A. Economic Growth 5.01 Economic growth in the context of the structural adjustment pro- gram in Turkey was to be achieved through opening up the economy. By pro- viding Turkish producers with a world market it was hoped that growth would be higher and more sustainable. To this end the policy of maintaining a realistic exchange rate, supplemented by export subsidies, was put in - 86 - place. In addition, by lowering and rationalizing tariffs the Government hoped to foster a more efficient use of domestic resources, an allocation that would put less pressure on the balance of payments. 5.02 A second main thrust of the adjustment program aimed at fostering growth was the rationalization of public investment. The goal was to In- crease the return to public investment projects while reducing their number so as to rely more heavily on the private sector for future growth. How. ever, as was indicated in Chapter IV, Section D, growth in private in- vestment, at least in the manufacturing sector, did not replace the reduc- tion in public investment. Total investment in manufacturing fell from a level of TL 45 billion (1975 prices) in 1976-77 to TL 35 billion in 1980 and TL 25 billion in 1983-87. 5.03 The overall growth performance of Turkey during the SAL period of 1980-1985 has been admirable, particularly in comparison with other heavily indebted countries. The growth of GNP averaged 4.62 during the period (see Table 5.1). In 1986 growth was even higher, reaching 7.91. The most dy- namic growth has been in the industrial sector, particularly manufacturing, spurred by a surge in industrial exports that has averaged 41.81 per year in dollars between 1980 and 1985. At the same tine, the share of industry in GDP rose from 251 to 321, chiefly at the expense of agriculture, a dra- matic shift for such a short period of time. Table5.1: mATOmL ACn1Ms. 190-198 lose ees I= nm = Am nu Im a Current Percent Current Percent Real Change from Previous Year Prices of #P Price. of P (Percent) (TL bi.) CL bil.) Consumption 8,781.4 84.1 22,518.0 81.0 -.4 -0.6 8.7 8.1 4.9 8.0 g.e PubIic 544.1 12.8 2,881.7 8.4 8.4 -0.8 0.9 8.1 8.0 8.9 6.8 Private 8.187.8 71.9 20.181.8 72.6 -4.9 -0.7 4.0 8.8 5.1 8.0 0.7 Fixed Investment 88.0 19.5 5,811.8 20.0 -10.0 2.0 8.4 2.7 8.5 12.5 18.8 Public 464.9 10.9 8,227.6 11.6 -.7 10.2 2.2 -4.9 8.8 16.0 18.7 Private 878.7 8.5 2.888.7 8.4 17.8 -9.8 5.8 15.2 8.7 7.9 18.8 Change in Stocks 84.8 1.9 240.2 0.9 89.6 49.6 -2.9 38.6 4.8 51.5 TOTAL DOMESTIC 0EMD 4,679.8 105.5 28,814.8 101.9 1.2 0.6 2.6 4.8 8.8 4.2 10.0 Foreign Balance 244.1 5.5 525.1 1.9 -40.9 *89.1 22.6 -12.7 -0B.6 18.0 GP 4,485.2 100.0 27,789.4 100.0 -1.1 4.1 4.6 8.7 8.9 5.1 0.0 Memorandum Items Agriculture 925.0 28.9 4.700.8 17.2 1.7 0.1 6.4 -0.1 8.8 2.4 7.7 Industry 1,024.2 28.1 8,060.5 20.0 -4.0 7.4 4.9 8.0 ?9.1 6.8 9.0 Construction 211.0 4.6 951.2 8.4 0.8 0.4 0.8 0.8 1.9 2.9 6.8 Services 1,985.7 48.6 11,714.1 42.2 0.9 4.1 8.9 4.4 8.7 4.1 6.0 QDP (factor cost) 4,828.0 26,526.1 -0.5 8.6 4.5 8.9 6.0 4.2 7.8 Surst: SPO. - 87 - B. Balance of Payments 5.04 Remarkable growth of exports has served to greatly improve the trade and current-account of the balance of payments during the period of the structural adjustment program. This was discussed in Chapter IV, Sec- tion D, and is summarized in the Balance of Payments accounts given in Table 4.7. Note that as a share of merchandise imports, merchandise ex- ports have grown from 39? in 1980 to 71% in 1986, narrowing the trade defi- cit in percentage terms. Within exports, the highest growth has been seen in industrial exports, principally manufactures, which accounted for over 70% of total exports after holding a share of only 36? in 1980. Receipts from other goods and services, principally from contracts of Turkish engi- neering and construction companies, have also shown dramatic growth, jump- ing more than fourfold between 1980 and 1986. 5.05 The growth in exports was accompanied initially by a marked shift in the destination of exports. Goods shipped to Middle Eastern countries doubled their share in total exports from 1980 to 1981 (see Table 4.17). The latest figures for the first half of 1987, however, show a decline from a peak of 42.8Z in 1984 to 26.5?. Much of this is due to the fall in the price of oil, a fall that has lately been reversed. 5.06 One note of caution should be sounded regarding export perfor- mance. Empirical work on the growth in exports shows that a portion of the growth was in fictitious exports which grew in response to the various export subsidy programs that were instituted in 1980 (see Chapter IV, Sec- tion F). The responsiveness of actual exports to the real exchange rate was confirmed in the same work. Such findings argue for a move once again to lower export subsidies rather than to delay the exchange rate adjust- ments and compensate it with higher subsidies, a trend that was abandoned in 1986 because of worsening export performance. 5.07 The PPAR for SALs II and III cited reservations about the extent to which the growth in exports was due to the tight control over domestic demand, control that would lessen as the program succeeded. In 1986 domes- tic demand surged and exports fell. This "vent of surplus' argument, how- ever, does not seem overly convincing since domestic demand is currently still buoyant whereas exports have rebounded. 5.08 Imports into Turkey have also grown more rapidly than the economy at large, a result of the outward-oriented policies of the adjustment pro- gram. The share of imports in GNP rose from 142 to 232 between 1980 and 1985. Substantial progress has been made in reducing the dispersion of tariff rates and streamlining the regulations governing imports. Unfortu- nately, the recent policy of providing EBFs with a voice in setting up specific tariffs has increased the dispersion among effective rates of protection and obfuscated the system. - 88 - 5.09 The dramatic growth in foreign trade has produced a such more open economy, openness that can be quantified using the ratio of total trade, exports plus imports, to GDP. As can be seen in Table 5.2, this ratio was only 19.32 in 1980, but by 1985 had reached 36.92 before falling somewhat to 32.12 in 1986. Efforts to maintain this openness are reflected in the on-going work of establishing two free sones within the country, one at Antalya and another at Hersin. According to the Turkish Industrialists' and Businessmen's Association (TUSIAD), the demand for space within these two areas from foreign and domestic firms has exceeded the amount of land available by a factor of three.151 Table 5.2: EXPORTS, IMPORTS AND GROSS DOMESTIC PRODUCT (TL billions) Year GDP Exports Imports X/GDP NGDP (X+4M)/GDP (M (M) (2) (2) (M (FOB) (CIP) 1975 519.1 20.1 69.0 3.9 13.3 17.2 1976 663.9 30.8 82.9 4.6 12.5 17.1 1977 862.9 31.3 104.9 3.6 12.2 15.8 1978 1,274.8 55.4 113.3 4.3 8.9 13.2 1979 2,155.9 75.7 178.5 3.5 8.3 11.8 1980 4,328.0 221.5 613.3 5.1 14.2 19.3 1981 6,411.2 530.7 1,002.4 8.3 15.6 23.9 1982 8,607.1 937.3 1,461.4 10.9 17.0 27.9 1983 11,531.8 1,299.0 2,127.1 11.3 18.4 29.7 1984 18,212.1 2,608.3 4,034.9 14.3 22.2 36.5 1985 27,513.6 4,153.0 5,994.9 15.1 21.8 36.9 1986 39,168.7 5,012.3 7,561.2 12.8 19.3 32.1 Sources SIS. 5.10 Foreign investment has also grown respectably during the years of the SAL program. As Table 4.7 shows, direct foreign investment has in- creased sevenfold to US$125 million in 1986. Inflows are likely to con- tinue to grow as the amount of foreign investment licenses granted by the SPO in the last three years totals US$870 million, roughly two and one-half times the inflows recorded in the balance of payments. 35/ See TUSIAD, The Turkish Economy, 1987, (Istanbul: TUSIAD) p. 113). - 89 - 5.11 The growth and diversification of trade, the dismantling of quan- titative restrictions, the elimination of bureaucratic barriers, and more liberal foreign investment laws have all served to support an outward ori- entation in Turkey and to buttress groups whose livelihood depends on the continuation of such outwardly-oriented policies in the future. Offsetting this optimistic appraisal is the aforementioned burgeoning of EBY levies and the protectionism that they entail. 5.12 While the trade balance improved during the SAL period, high lev- els of interest payments, partially a legacy of previously extensive bor- rowing and the reschedulings of the late 1970s, have added to the current- account deficit. There may be some overstatement of the deficit, however, due to the introduction of Turkish foreign exchange deposits. Inflows formerly counted as workers' remittances may now be counted as capital inflows. Overall, the current-account deficit has fallen from 5.6Z of GNP in 1980 to 2.6% in 1986. 5.13 To insure that the capital inflows necessary to finance the cur- rent-account deficit are forthcoming the Government has resorted to mone- tary policy. In the skittish financial markets efforts to use domestic monetary policy to foster capital inflows have required very high internal real interest rates, thereby retarding the growth of private investment and conflicting with economic growth. C. Financial Activity and Inflation 5.14 During the period of the structural adjustment program important advances were made in improving the instruments of monetary control, strengthening financial institutions and nurturing a capital market. More effective monetary control has been important in reducing the rate of do- mestic inflation. Improvements in banking supervision, including a new banking law, have added to the soundness of financial intermediaries. The auction of government bonds, and more recently the sale of commercial paper, have strengthened capital markets. 5.15 Nonetheless, the financial system shows serlous problems. Chief among them is the level of the real interest rate. At current levels many firms cannot generate the cash flow necessary to service their loans, thereby increasing the de facto non-performing loans of the banking system. Afraid of wiping out their capital, banks have been reluctant to declare loans non-performing, preferring instead to bid up the interest rate on foreign exchange deposits in their search for the funds to roll over their bad debts. At the same time the rates charged customers who are meeting their debt payments are increased to cover the bad debts, thereby creating more delinquent borrowers in a vicious circle. 5.16 Fiscal policy has also had an important impact on the financial system. While bond finance was the cornerstone of the policy to eliminate - 90 - the need for money finance, and thereby bring down the rate of inflation, the rates at which such debt was sold have created a situation in which the need to pay the interest on existing government debt accounts for the ma- jority of new debt issues. In 1986, for the consolidated government ac- counts, the budget, exclusive of interest payments, was in surplus. 5.17 The overall monetary effect of the government deficit, including SEEs and EBFs, as measured by the PSER, on monetary growth has not shown significant improvement since the early years of the adjustment program. Unless a fiscal reform is initiated and real interest rates are brought down the snowballing interest payments could induce the Government to use money finance, with adverse inflationary consequences. D. Social Impact 5.18 Assessing the social costs of the policies that accompanied the five SALs is extremely difficult due to the lack of statistical information that would normally be collected by the Government on basic social indica- tors. No official household income and expenditure survey has been done by the SIS since 1973, although one is planned for the near future. No physi- cal quality of life index is maintained. The methodology used in collect- ing figures on unemployment is currently being overhauled as it has been judged to be unsatisfactory. This statistical vacuum makes any analysis necessarily preliminary. 5.19 As a very imperfect and approximate measure of the social impact of adjustmet one can look at the performance of real GNP between 1980 and the present. While such a measure of social welfare has well known short- comings it is frequently highly positively correlated with other more ap- propriate measures of welfare. Between 1980 and 1986 real GNP growth in Turkey averaged 5.2Z, well above that of other middle-income oil importing countries whose GNP actually shrunk on average between 1981 and 1984. On a per capita basis GNP growth averaged 2.5% between 1980-85, also better than comparable highly indebted middle-income countries. Aggregate consumption during the same period grew slower than GNP at 4.42, still above the growth rate of consumption in similar countries. 5.20 As for the distribution of income either across the population or among various factors of production, information is very scanty. The na- tional accounts do not present data on the functional distribution of in- come among wages, rents, interest and profits. There have been some at- tempts to look at changes in the size distribution of income by academics and institutions outside of the Government. The most recent is a 1986 income and expenditure survey conducted by TUSIAD. The survey covered 1,444 households. It was not primarily designed to assess income distribu- tion. Questions on household monetary and in-kind income were asked, how- ever. The results of the survey are shown in Table 5.3. The figures - 91 - show some slight improvement in the distribution of income when compared with the results of the 1973 government survey even though it could not be ascertained whether these two surveys are fully comparable. Also, given the length of the period, shifts in income distribution may be of the 'normall type, usually associated with overall development of maturing economies. Moreover, only one-half of this period falls under structural adjustment policies. The poorest two quintiles accounted for IS.32 of total income in the 1986 TUSIAD survey as compared to 11.41 in the 1973 study. The majority of this gain came at the expense of the fourth quin- tile (second richest) whose share in total income fell from 20.52 to 19.22. The richest quintile, meanwhile, showed a slight rise in its share from 55.32 to 55.92. The overall Gini coefficient fell from .50 in 1973 to .46 in 1986. Table 5.3: HOUSEHOLD INCOME DISTRIBUTION (Z of Total Income) Quintile OVERALL METROPOLITAN /a RURAL lb 1973 1986 1973 1986 1973 1986 1 3.5 3.9 5.0 5.3 2.5 3.3 2 7.9 8.4 9.3 8.5 6.2 8.0 3 12.8 12.6 13.7 12.1 11.1 12.7 4 20.5 19.2 20.5 19.0 19.9 19.6 5 55.3 55.9 51.6 55.1 60.3 56.4 Gini Coefficient 0.50 0.46 0.45 0.46 0.56 0.47 Ja The classification in the 1973 survey was non-agricultural. JL The classification in the 1973 survey was agricultural. Source: TUSIAD, *The Socio-Economic Situation and Outlook of the Turkish Household, 1986, pg.23, and Celasun, "Income Distribution and Domestic Terms of Trade in Turkey, 1978-1983, Middle East Technical University (METU) Studies in Development, Vol. 13:1,2, 1986, pp. 193-216. 5.21 There is a marked difference in the survey results for the rural as opposed to the metropolitan areas. The structural adjustment process seems to be associated with growing concentration of income in the cities and with a more even distribution of income in rural areas. The TUSIAD figures show a loss of 4Z in the share of income held by the richest rural dwellers. In the cities, however, the richest group gained almost 4% in - 92 - its share at the expense of all but the poorest quintiles whose share of income also rose. The growing concentration of income in the metropolitan areas and the opposite process in the rural areas meant that the TUSIAD figures show an almost equal profile for income distribution across the two regions of the country. 5.22 While income within the agricultural sector is now more evenly distributed, the share of national income going to agriculture has fallen sharply since the start of the structural adjustment process. Table 5.1 shows that agriculture's share in GNP has fallen from 20.92 to 17.2Z be- tween 1980 and 1985. In 1986 the share fell further to 16.52. Figures presented in Celasun (1986) point to adverse movements in the urbanirural terms of trade as responsible for much of the decline in agriculture's share in national income until 1983. Recent reports by the Bank cite con- tinued lagging of agricultural prices behind the general price level.361 5.23 Some preliminary judgments as to the effect of the stabilization on the functional distribution of income in the industrial sector can be drawn from a survey of 500 large industrial firms carried out each year by the Istanbul Chamber of Industry. This survey covers firms whose output represents about one-third of Turkish industrial value-added. Part of their survey presents data on value-added disaggregated into wages, rents, interest and profits and can be seen in Table 4.22. Wages and salaries as a share of total value-added have fallen from 52.6% in 1982 to 40.4Z in 1985. At the same time profits have almost doubled from a share of 19.22 in 1982 to 34.5? in 1985. The reduction in wages and salaries and the rise in profits is due almost entirely to the public sector firms included in the sample. For these firms wage costs have shrunk from 682 of value-added to 41?, a share close to that held in the private sector. At the same time, the share of profits in public sector firms' value-added has increased fourfold. In the 1985 data the share of value-added in public firms going to profits actually exceeded the share going to wages. Data on overall employment in SEEs do not show declines, implying that the reduc- tion of the wage share in value-added has come through one of three chan- nels: 1) the reduction in the wage rate, 2) an increase in the price of output, or 3) an increase in labor productivity not matched by an increase in the wage rate. The earlier discussion of SEE pricing and productivity supports the second channel. 361 See Celasun (1986), 'Income Distribution and Domestic Terms of Trade in Turkey, 1978-19830, METU Studies in Development, Vol. 13:1-2, pp. 193-99, for data on the agricultural terms of trade. The World Bank data for more recent years is discussed in the supervision report of the current agricultural sector adjustment loan, Report No. 2585-TU, dated June, 1987. - 93 - 5.24 The conclusions for the functional distribution of income that one can draw from the Chamber of Industry study are limited. Services and agriculture, which together account for about 70Z of GDP, are not treated at all. Within its limited scope, the survey presents a picture of a fall- ing share of wages and a rising share of profits in national income between 1982 and 1985. The bulk of this movement seems due to the changes in the price and wage policy followed by SEEs. 5.25 While information on the overall functional distribution of income is not available from the Government, there are data on unemployment and the real wages. The two series are presented in Tables 5.4 and 5.5. While unemployment fell from 16.3Z in 1985 to 15.5% in 1986, it is still almost two percentage points over the unemployment rate that prevailed in 1979. The number of unemployed in 1986 is estimated to be somewhat less than its 1984 figure--2.3 million persons. These figures include an estimate of disguised unemployment in agriculture of about 72 or 650,000 people. None of these esimates are based on a direct survey but rather on census data, national accounts and productivity data collected by the SPO. 5.26 In part the stubbornness of the unemployment rate is the result of a closing of other countries' labor markets to Turkish workers, markets that in the past had served to absorb excess labor and ease the unemploy- ment figures within Turkey. According to the figures of the SPO, since 1982 the number of Turkish workers abroad has stabilized at about 1 mil- lion. As the labor force has been growing since 1982 at 1.37Z, the con- stancy of the level of Turkish workers abroad has served to raise unemploy- ment by this rate. Over the period 1982-1986, then, the constant level of Turkish workers abroad has added almost 6? to the cumulative unemployment rate. 5.27 If one concentrates on the creation of jobs, rather than the rate of unemployment, the figures given in Table 5.4 show that the fastest grow- ing sector in terms of job creation has been services. Between 1977 and 1985 the average annual growth rate of service sector jobs has been 3.02%. The next two major sources of job creation, industry and construction, grew at almost half that rate, 1.64%. This rate barely exceeds the overall growth of the labor force which was 1.57? over the same period. Industry, however, has grown strongly since 1980 under the new outward-oriented poli- cies, averaging 8 between 1980 and 1985. As is to be expected in the development process, agricultural sector employment has fallen, by .212 per annum since 1977 and at a slightly higher rate of .272 since 1980. 5.28 While the unemployment rate has remained at very high levels, the real wage has shown a great deal of flexibility. The figures given in Table 5.5 show that in manufacturing real wages have fallen by 37? since 1980, a point already 20? lower than their 1978 average. Figure 5.1 plots the same data. An alternative series based on daily wages reported to the Social Insurance Institute (SII) in many industries shows a much smaller decline in the real wage after 1980, but a tremendous decline in 1978-80 of 62? just prior to the stabilization efforts. (a c åi g g 1 .4 g 0 El 2 4. -; . - 1 1 -: - 1 Es 8: 1 1 l1l 1 i ~. j s .a 3 I3 - s 1 f. e -l 1:1 - -- .z1: eII... . * - 95 - Table 5.5: TURKISH REAL WAGES Nominal Index CPI Real Wage Daily Wage (1980-100) (1980-100) (1980-100) (TL) SSI* MI** SSI NI SIS SSI H 1978 207.9 311.8 48.7 35.7 30.0 162.4 119.2 1979 294.3 483.8 68.9 55.4 47.6 144.9 116.5 1980 427 872.7 100.0 100.0 100.0 100.0 100.0 1981 543.8 1,274.1 127.4 146.0 136.6 93.2 106.9 1982 723 1,617.2 169.3 185.3 180.4 93.9 102.7 1983 944.4 1,940.6 221.2 222.4 237.5 93.1 93.6 1984 1,307.9 2,412.7 306.3 276.5 352.4 86.9 78.4 1985 -- 3,178.4 364.2 510.9 -- 71.3 1986 -- 3,789.5 434.2 687.7 -- 63.1 * SII Social Insurance Institute. ** MI Wage derived from the Annual Survey of the Manufacuring Industry carried out by the SII. Figure is the result of dividing total wage payments (inclusive of bonuses, but exclusive of social secur- ity oand retirement fund payments) by the number of workers em- ployed. Source: SPO based on data collected by the SII and the SIS. - 96 - Figure 5.1 Manufacturing Real Wage (SØ dato deflaed by the 515 CPI) 120 - 110- 100- 90- 8b- 70- 80- 1978 1979 1980 1981 1982 1983 1984 1985 1986 - 97 - 5.29 Some decline in real wages appears to have been necessary to es- tablish the international competitiveness of industrial exports in the newly open economy of Turkey. This implies that at the beginning of the structural adjustment program in 1980, real wages were sufficiently high and the return to capital was sufficiently low so that Turkish firms were not competitive. This does not tell us, however, whether the current level of real wages is sufficiently low enough to allow Turkish manufacturing firms to effectively compete internationally given the existing exchange rate. The growth of industrial output and industrial exports would seem to be causal evidence that Turkish firms have been able to compete given the existing levels of the real wage and the exchange rate. The element of their cost structure which has not shown improvement, at least in the pri- vate sector, is interest expense. As has been noted earlier in this re- port, the high interest expense is linked to the policy of expected devalu- ation and the financing requirements of the Government. Were these two items to be corrected the share of value-added going to wages or to profit could be increased without affecting the competitiveness of Turkish firms. VI. SUSTAINABILITY OF REFORMS A. Introduction 6.01 The main objective of the structural adjustment loan program was tc put the economy of Turkey on a sustainable and stable export-led growth path. The adjustment programs from 1980 to 1985 were thought by the Gov- ernment of Turkey and the Bank to be necessary to establish the initial conditions for such growth. The basic approach of the Bank and the Govern- ment was to increase competition in the economy. This was to be achieved by opening the economy to international competition, by reducing direct controls of the financial markets, by reducing the Government's role in the production and distribution of non-public goods, and by reducing the fiscal deficit. In general, the programs aimed at a more market-oriented growth path for Turkey. 6.02 The process of structural adjustment can only be started, but not completed, in a five-year period. This was the objective of the SAL pro- grams. But in reality the process is never completed; all economies face the need for continual structural adjustment as their economic environments change, even though of a less drastic and more evolutionary type than the basic adjustment necessary to return to macro-economic equilibrium. In this sense, one purpose of the SAL programs was to increase the flexibility and efficiency of the economy of Turkey in order to make it more able to respond and adjust to challenges of economic change. To achieve this in- crease in flexibility and efficiency, the Government undertook institution- al changes that would be fully effective only if they were sustained beyond the SAL period. In Chapter III on implementation it was pointed out that - 98 - even the implementation of some institutional changes, such as development of stock and bond markets or introduction of a VAT, would extend beyond the SAL period. Attaining a stable and sustainable growth path also requires policies that stabilize. In Chapter II it was pointed out that some as- pects of policies under the SAL program and since might constrain longer- term growth and even contribute to potential instability. This chapter of the report addresses the issue of sustainability of the structural adjust- ment process by pulling together the relevant points from Chapters III and IV, and suggesting policy corrections that might improve the probability of longer-term success. 6.03 The question of sustainability can be broken down into discussions of the degree to which genuine and productive institutional change has been achieved, and whether economic policies have contributed to the sustaina- bility and stability of growth. These discussions extend those in the SALs II-III PPAR under the same headings. Under the first heading of institu- tional reforms questions arise about the growing role of the EBFs, espe- cially as they influence the budget and the trade regime, the continued removal of export subsidies, the structure and stability of the financial system, and the lack of a medium-term strategy that would tie the SAL pro- gram to the longer-term growth path. Under the second heading of economic policy the main questions involve the causes and consequences of high real interest rates. Here fiscal and monetary policy and their contribution to potential instability at the macro-economic level are discussed. B. Institutional Reforms 6.04 EBFs. The first major issue of institutional reform for SALs IV- V is the growth of the EBFs. This began with the MHF during SAL V, and the growth has been rapid since. The fiscal structure of the EBFs, with assigned revenue sources, limits budget control. The administration of their import levies has effected a partial retreat from liberalization of the import regime. The protection that the levies provide to the SEEs strengthens their monopoly power domestically. The existence of quasi- governmental economic organizations that provide housing credit (the MHF) or operate infrastructure with marketable output such as bridges or power plants (the PPF) makes sense and is not unique to Turkey. An organization that provides public support and control for the defense industry Defense Industries Fund (DIF) might also be appropriate in Turkey. The problems arise from the proliferation of inappropriate EBFs, the particular way the EBFs are financed, and the extent of their operating freedom in Turkey. 6.05 The growth of *inappropriate EBFs since 1984 has been extremely rapid. By 1986 there were approximately 80 EBFs, gathering the equivalent of some 20Z of the central government's tax revenue. The activities of the largest 8 EBFs are summarized in Table 4.14. The MHF, PPF, and DIF togeth- er accounted for half of total EBF revenue in 1986. So elimination of the -99 - other EBPs would remove half of the problem. This might be a good first policy goal. But the effects of the financing technique of the largest three would remain. 6.06 In Chapter IV it was noted that the prerogatives of the EBFs re- duce budget control and increase the fiscal deficit. The import levies are set to provide protection for domestic industry. Their revenues are spent by the NBF. The level of spending is therefore determined, at least in part, by the quest of domestic industry for protection, not necessarily by a decision on the desirable level of activity for the EBP. The EBF levies also increase the overall fiscal deficit. While the EBFs spend their reve- nue, their levies reduce the quantity of imports, and therefore revenue from the normal tariff. This point applies to any EBF tax that is addi- tional to an existing tax. 6.07 Imposition of the EBP import levies is a retreat from liberaliza- tion of the import regime in several ways, as discussed in Chapter IV. Originally applied to luxury imports, they now apply to a wide range of other imports, as shown in Table 4.19. The average levy rate in 1987, as measured by the ratio of levy revenue to total import value, is 6.12. The levies reduce the transparency of the import regime. Since they are spe- cific instead of ad valorem, comparison of levy rates is difficult. The levies are changed frequently with administrative discretion. This makes it difficult for economic agents not already intimately familiar with mar- ket conditions to enter the market, creating rents for the agents that have acquired that familiarity. The availability of this administratively flex- ible instrument may also provide an incentive for misuse. 6.08 The EBF levies provide systematic protection for the industrial SEEs. This is clear from the discussions of the audit missions for both SAL9 II-III and SALs IV-V with the senior personnel that deal with EBF levies. They all perceive the protective effect of the levies. This pro- tection provides the import-competing SEEs an effective monopoly position domestically. The policy of setting profit-maximizing prices ensures that they exercise their monopoly power. This has undoutedly contributed to the shift in income distribution toward profits in the SEEs (discussed in Chap- ters IV and V). The form of EBP financing thus undermines trade liberali- zation, provides monopoly power to the SEEs, and shifts the income distri- bution against wages. 6.09 The existence of economic organizations such as the EBFs to pro- vide the appropriate public services is not in question here. The MYF and the PPF have their equivalents in many countries, and the DIF may well be appropriate for Turkey. Careful analysis may suggest that others of the BPs also make sense. The problem is the technique for their financing, especially through the import levies. Dedication of revenue sources to the EBFs gives them implicit, but automatic, priority over activities that - 100 - compete for general revenue resources. The levies reduce the transparency of the tax system, and the import levies provide systematic protection. This erodes the sustainability of the reforms supported by the SAL process. 6.10 Export Regime. Substantial progress was made during the SAL period on liberalizing and rationalizing the export regime. The real de- valuation in 1980 (see Figure 4.3) and the gradual real depreciation since have provided price incentives to exporters, and have allowed for the grad- ual elimination of export subsidies. The subsidies were preferential cred- it rates, a tax rebate for exporters in lieu of a rebate of the VAT, duty- free imports of inputs, and a direct subsidy from an EBF, the RUSF. By the end of 1986, the preferential credit rates and the RUSF subsidy had been eliminated. The tax rebate scheme was still in effect, even though a VAT with an explicit rebate for exports was enacted in 1985. Rebates of duties on imported inputs continued. Thus two of the four elements of the subsi- dies had been eliminated under the SAL program. 6.11 Export growth slowed in 1986; total exports were $7.5 billion in 1986, compared to $ 8.0 billion in 1985 both in current prices. In re- sponse, at the end of 1986, after the completion of SAL V, the Government re-introduced preferential credit rates and a direct subsidy from another BP, the SPSF. This was discussed in Chapter IV, Section G. It somewhat reversed the liberalizatiL-n policies under the SAL program by reinstating the two subsidy elements taat had been eliminated. In a sense, the rever- sal may go beyond the starting point because the tax rebate scheme was maintained when the VAT, with an explicit rebate, was introduced. The subsidy rates may be lower in 1987 than in 1980, but all of the institu- tional forms for subsidization are intact, and one, the EBF, has been added. 6.12 The direction for renewed policy reform of the export regime seems clear. It may also be palatable due to the revival of export growth in 1987. There is evidence, cited in Chapter V, that export subsidies mainly result in the overstatement of export earnings. Elimination of these subsidy elements would restore the situation that was obtained at the end of the SAL period. The next policy target might be the ending of the tax rebate scheme, which might be the first step in further liberalization of the export regime. Additional real depreciation of the Turkish lira would compensate for any detrimental effect of further liberalization on exports. - 101 - 6.13 Financial Structure. Three aspects of the structure of the fi- nancial sector that were discussed in Chapter IV could weaken the sus- tainability of the SAL reforms. These are the continuing problem of non- performing loans, the oligopolistic market structure of the banking sector, and the interlocking between banks and industry. All of these problems have been recognized in the SAL loan documents, were discussed in the SALs II.III PPAR, and are being addressed under FSAL II. 6.14 The oligopolistic nature of the financial sector seems fairly clear, as noted in Chapter IV, Section F. Three large banks do 75Z of the private commercial banking business and form the market for government debt. Although the Central Bank has instituted a system of sealed bids for debt issues, the bid rate among these banks has not fallen from the high rate that was set administratively at the first issue in 1984. While the system has been opened to entry by foreign banks, they do only three per- cent of the sector's business. This oligopolistic structure generates a high spread between lending and borrowing rates. Lending rates to the Government and to solvent borrowers are pushed up even higher to compensate for non-performing loans. This shuts out private investment. Under the SAL program constraints on the financial system were removed, lending rates were deregulated, and several structural reforms were put in place; but the fundamental oligopolistic nature of the sector still remains to be addressed. 6.15 A closely related structural problem is the inter-penetration of the financial and industrial sectors. Banks are attached to industrial groups, with cross-ownership of the bank and the firms in the group. Through this arrangement, the bank can be effectively used as the instru- ment by which profitable units in the group subsidize the units incurring losses, shielding the latter from explicit bankruptcy. Exporters have a heavier weight among the profit-makers so the system uses export profits effectively to subsidize the losing firms oriented toward the domestic market.371 The losses are hidden within the group, and instead of being re-invested in the export sector, export earnings are used to prop up firms that would otherwise be bankrupt. The threat to sustainability comes from two aspects of this interlocking system. First, profits of solvent firms subsidize losers, reducing sources of investment financing in the profit- able sectors. Second, if or when the total group profits become negative, the bankruptcy in the losing industrial firms would unavoidably spread through the captive bank. So the interlocking banking and industrial system increases the vulnerability of the entire financial sector. Policy should aim at increasing competition in the banking system and reducing the equity interconnections between banks and industrial firms. A careful 37/ It is noteworthy that the situation seems to be reversed in countries considered as long-term, successful exporters, in particular those on the Pacific rim. - 102 - industrial organization study of the financial sector would add greatly to the information base available for the formulation of policy in this sector. 6.16 Medium-Term Strateny. Development of a medium-term strategy document that would lay out at least a feasible path from the short-run stabilization program to a longer-run growth path was emphasized in SALs I- III and their PPARs. This medium-term strategy--in the full meaning of the term, including a policy-oriented Integrated analytical framework--was not developed during the SAL period and it did not seem to exist in 1987. The document would have laid out what the SAL program was expected to achieve and how it would place the economy on a sustainable growth path. The point is not development of a detailed plan--or, worse, of projections which would have been rapidly invalidated--but rather of one or two internally consistent scenarios that showed how the major elements of the program fit together and interrelate. Many of the problems of sustainability involve unanticipated interactions between separate parts of the program. Examples are the effects of exchange-rate policy on real interest rates or the protective effects of EBP financing. The effort to develop a medium-term strategy and analytical framework might have revealed these potential negative interactions and enabled policy to be re-designed to eliminate them. Therefore, such strategy would have been continuously adjusted and updated. 6.17 One area in which the development of a medium-term strategy would have improved the prospects of sustainability is private investment, par- ticularly in manufacturing. This is the sector that will be crucial for continued export growth. The basic idea of the reduction of public invest- ment in the industrial SEEs under the SAL program was that it would be replaced by more efficient and market-oriented private investment. As was argued in Chapter IV, Section I, this has not happened. Private investment in manufacturing in real terms was below its 1975-79 level throughout the SAL period. In 1987 it was projected to rise above its 1979 level for the first time in the 1980s (see Table 4.9). The consequence of the stagnation of private investment has been the substantial reduction in total invest- ment in manufacturing (shown in Figure 4.4). This has come at a time when the changing structure of the economy required an increase in investment. It suggests that capacity may be now generally overestimated, and that supply constraints on exports could soon begin to bind. 6.18 The work of developing a medium-term strategy document might have brought this problem of private investment into focus earlier. Analysis of the path of real interest rates and study of credit availability and profit prospects under the stabilization program could have signalled difficulty with investment and stimulated earlier policy analysis of the interest rate and non-performing loan problems. Thus the medium-term strategy is not an end by itself; its preparation is a way to identify potential problem - 103 - areas. Moreover, if broken down into sequential periods, it could have provided a stronger conceptual basis for a logical and operationally mean- ingful staging of policy change. Finally, if anchored within such medium- term strategy, the time horizon of SALs would have notably lengthened, and consecutive SALs could have been merged into larger, farther-reaching loans. C. Economic Policy 6.19 Real Interest Rates. The central problem of macro-economic poli- cy that was identified in Chapter IV, and affects the sustainability of the reform program, is high real interest rates. These are shown in Figure 4.4. The consequences of high real rates include potential unsustainabil- ity of the paths of the domestic and foreign debt, low investment with shortened payoff periods, potential export supply constraints, and insta- bility of the financial system. The causes of high real interest rates have several sources in policy. A policy of anticipated gradual real de- preciation with open financial markets raises equilibrium real rates above external (dollar) rates by the rate of depreciation. The unsustainable growth of lira-denominated debt due to the fiscal deficit adds a growing risk premium to real rates. The need for an inflow of capital to finance the current-account deficit requires the Central Bank to hold real rates above the equilibrium that includes the expected depreciation and the risk premium. The causes and consequences of high real rates were discussed in detail in Chapter IV. 6.20 Deficits and Debts. The high level of real interest rates, com- bined with the fiscal deficit, may have placed the Turkish economy on an unsustainable path of growth of budget deficits and government debt. As discussed in Chapter IV, Section B, for any given ratio of non-interest government expenditure to GNP, if the real interest rate on government debt exceeds the growth rate of real tax revenue, the ratios of debt and of debt service to GNP will grow continuously. This fiscal situation is unsustain- able; eventually a crisis of confidence in the Government's ability to service the debt will develop. During the period of SALs IV-V and since, real interest rates on government debt in Turkey have been higher than the average growth rate of real tax revenue. Policy action to increase the growth of tax revenue and to reduce the real borrowing rate is needed to stabilize the fiscal situation. 6.21 A similar dynamic instability may have developed in the current- account and foreign debt position in the SALs IV-V period and especially since 1984. This was discussed in Chapter IV, Sections C and D. Foreign borrowing in various forms, especially short-term since 1984, finances the current-account deficit. This was shown in Table 4.7. Growing service on - 104 - the foreign debt adds to the current deficit. As a resvlt, ci:,, vice ratio inclusive of short-term debt rises. In Table 4.8 t .eob-9'.- vice ratio is seen to rise from 251 in 1984 to 361 in 1986. I additie, the low level of investment in manufactu:rin mar presage expert su!ply constraints in the future. 6.22 Policy Directions. The two main macro-economic policy steps t!c would increase the prospects for sustaining the SAL reforms are redction of the fiscal deficit and real interest rates. If these are done at about the same time, the depressive effect of a tighter fiscal stance could be offset by the expansionary effect of lower real interest rates. To reduce the fiscal deficit effectively, the budget process must be brought under control. This means that both public investment and EBF expenditure should be subject to budgetary constraint. Reduction of the fiscal deficit would not only reduce current crowding out of private investment, it would also remove some of the risk premium from real interest rates by reducing the eventual level of the Government's debt. It would, therefore, generate two powerful incentives for private sector investment. 6.23 The structural adjustment program in Turkey had at its core four elements: liberalization of the trade regime, liberalization of the finan- cial markets, shrinkage and reform of the SEEs, and growth of private in- vestmeut. Under the program, substantial progress was made in the first two, less in the third and less still in the fourth. Beginning during SAL V, the growth of the BFs began to undermine the objectives on the trade regime and SEEs. The re-introduction of preferential export credits and an EBF subsidy for exports in late 1986 was a further retreat on the trade regime. Progress has been made on financial liberalization but this is offset somewhat by the problem of non-performing loans and the oligopoly structure of the financial sector. Policy steps needed to sustain the structural adjustment program are a reform of the EBF system, restoration of momentum on trade liberalization, consolidation of the bad debt of the banking system, and introduction of competition into the banking system. To restore sustainability on the macro level, it would be prudent to shift from an often inadvertent policy mix of large fiscal deficits and high real interest rates to a (deliberate and decisive) mix of fiscal tightening and lower real interest rates. VII. ROLE OF THE BANK A. Program Development 7.01 The approach to structural adjustment was handicapped to some extent in the initial stages by a relatively limited stock of previously executed economic and sector work. the major focus of earlier sector work had been agriculture, where two reports issued in 1978 enunciated the first small steps towards the development of an agricultural strategy for Turkey. No major work had been conducted in the fields of industry and trade pol- icy. Two economic reports, finalized in 1979/80, dealt with macro-economic - 105 - and public sector Investment issues, as they were perceived at that time. These reports did initiate a "fermentation process', which was captured by the Government. 7.02 As the SAL program gained momentum there was no let-up in the scale and density of economic and sector work, which facilitated a dialogue in the key areas of adjustment. This work continued after the completion of disbursements on the fifth and final SAL in 1985. The most notable of such recent work comprises studies on fiscal policy and tax reform, public investment programuing, and medium-term economic planning. During the life of SAL III economic and sector work was supplemented by a technical assis- tance loan, oriented towards improving efficiency and policy development in SEEs. 7.03 The economic and sector work that was executed during the SAL and post-SAL periods fostered a close and continuous policy dialogue with the Central Bank and especially the SPO and the Undersecretariat for Treasury and Foreign Trade (UTFT)--two organizations which had major responsibility for steering the implementation of the SAL program. The training programs initiated by EDI In conjunction with the rationalization of the public investment program and the assistance towards computerization of public debt strengthened policy analysis and implementation capabilities. 7.04 The structural adjustment program was also reinforced by support- ing project and sector loans in the areas of agriculture, energy, industry and finance. These coincided with the adjustment program in a few cases and followed upon the cessation of the SAL program in some others. While they helped in a number of ways to strengthen the objectives of the program at the production levels, they did not always contain the elements required to keep. the macro-economic policy framework intact. They nevertheless helped the Government of Turkey reach the core objectives of promoting improved resource allocation and increased productivity in the production sectors (with the exception of SEEs) and strengthening the physical and financial infrastructure. B. Quality of Conditionality 7.05 The conditionality that was applied to the tranching of the five SALs was, in general, appropriate and well focussed. Key performance areas were well monitored. There were some exceptions, however, to the generally flexible manner in which conditionality was expressed. Earlier SALs, which put time horizons on the passage of legislation governing the rules of operation of SEEs, and introduced a VAT. were found to be unrealistic and were modified. This facilitated smooth disbursement of the tranches. The conditionality clauses related to the fourth and fifth SALs, however, went to the other extreme. In most cases, conditionality criteria were deemed - 106 - to have been satisfied if actions to ensure adequate progress' had oc- curred. This formulation unfortunately left much to individual discretion and interpretation and thereby anted some of the drive and discipline that must inevitably attend any concept of conditionality. We recognize that an appropriate balance has to be struck between precision and flexibility, but in our view, the formulation that was chosen loosened the control that the Bank was entitled to in the implementation of SALs IV and V and blurred expectations. C. Complementarity with IMF 7.06 A critical component of the adjustment program was the maintenance of fiscal and monetary stability. The implementation of this aspect of the program depended to a large extent on the existence of Stand-By Arrange- ments with the INf. Five such Stand-Bys coincided with the life of the SALs and they helped to reinforce the Bank's objectives. The Fund's Stand- By programs ceased in 1985 and this, to some extent, removed the discipline that had previously existed. As pointed out in Chapter VI, the sustaina- bility of the program of adjustment depends very much on the vigor with which the fiscal imbalance and the problem of high interest rates are addressed. This requires determined efforts to improve financial savings and greater disciplire in public expenditures. 7.07 Experience with conditionality, in the case of the five Turkey SALs, raises fundamental questions about the timing of the efforts of the Fund and Bank. Specificallys (i) Should the two programs have continued side-by-side for a longer period? Should they have been phased differently? (ii) Recognizing that the Bank moved--by necessity rather than by choice--away into sector-based loans, should these loans have had appropriate macro-economic policy frameworks built into their conditionalities? And would that have been at the expense of adjustments of sectoral policies? It it not possible, at this point, to provide precise answers to these questions. Some thought should nevertheless be given to them, recognizing that in the final analysis, it is the Borrower's commitment to reform and its demonstration thereof that ultimately matters. Where this commitment falters or is lacking nconditionalityl can only play a limited role. D. Social Dynamics of Adiustment 7.08 The ability of democratically elected governments to sustain ad- justment programs that reduce levels of consumption and real income for reasonable lengths of time greatly depends on the public's perception of - 107 - the manner in which the array of adjustments measures impact on them as individuals or groups. Experience with adjustment programs of the kind implemented in Turkey have shown that the more dynamic segments of society having good access to factors of production and distribution networks us- ally reap the better rewards and this is how it should probably be, if higher productivity is an objective. Structural adjustment also involves some retrenchment of labor and accentuation of unemployment (which may continue for some time if the economy is bereft of the infrastructure needed to put redundant labor to work). In the case of Turkey, some redundancies were necessitated as the SEEs began to shed labor in the initial stages of the SAL program. The anti-protectionist measures also added to these pressures and the slower than warranted growth in private investment compounded difficulties still further. To what extent this was as it should have been, having productivity as an objective--and this without any trade-offs--is open to questioning. Neither the present state of economic knowledge nor accumulated experience on structural adjustment provide any clear answer to this question. 7.09 Hence, the sustainability of the policies and measures, as well as the program's success, depended to a large extent on the Government's con- tinuing commitment to reforms. Such a commitment had essentially to be predicated on a full knowledge, inter alia, of the economic and social im- pact of the actions it was taking. While mechanisms are in place in Turkey to measure economic impact, there is no formal apparatus for monitoring social impact. While sympathetic to the view that public resistance to the reforms are possible if the 'social impact' is widely disseminated and discussed during the life of the program, especially if it militates against the welfare of the poor, equally, a failure to understand the im- plications of the policies and measures for social welfare could lead to a concertina of problems in the future. This could result in the program being ultimately discredited and with the measures backfiring. 7.10 The data available on 'social impact' of adjustment has been pre- sented in Chapter V, which deals with the impact of the overall program. While no official statistics are available, private survey data reveal a mixed picture; it essentially confirms the feeling that income polarities have been accentuated between the rural and urban sectors and that some income deterioration has occurred, especially for fixed-income and salaried workers. The Bank's study on public investment points to a deterioration in rural welfare deriving from neglected rural infrastructure, which inhib- ited employment expanding activity. It also points to the sharp fall in expenditures on health and education during the span of the SALs and the need to revive investment in these areas as early as possible. 7.11 When conditionalities are designed provisions should be made for monitoring the social impact of adjustment. The methodology for this work needs to be developed, bearing in mind the need to be cost effective and - 108 - also taking into account the concerns of the Borrower. If, in the case of Turkey, an adequate monitoring mechanism was in place, the Bank would have had the information necessary to advise on the design of countervailing measures to minimize adverse social impacts, inter alia, through invest- ments in health and education and the upgrading of rural infrastructure. VIII. PRINCIPAL CONCLUSIONS AND RECOMMENDATIONS 8.01 The program of structural adjustment that was initiated in 1980 in Turkey was a response to a deteriorating economic situation which gathered speed towards the end of the 1970s. A program of structural adjustment was also needed to lay the foundation for sustainable, longer-term economic growth. The Government that took office in 1980 believed that this should be achieved within the framework of a market-oriented economy. It nego- tiated five structural adjustment loans with the Bank. These were rein- forced by an equal number of Stand-By Arrangements with the Iff and ran parallel with the Bank's program. 8.02 The policies and measures undertaken in conjunction with the pro- gram of adjustment have been analyzed in depth in Chapters IV-VII. The corrective measures that were implemented between 1980-86 were comprehen- sive and well focussed. They helped to change the structure of production and trade, reduce inflation, strengthen export competitiveness, and restore economic growth to a respectable level. The progress achieved so far, though significant, still represents only the first steps on the road to fundamental structural change. Major efforts are needed to restore the still disappointingly low level of private investment to a state where it could meet the challenges of the future. The domestic capital and stock markets need revitalization and the fiscal and budgetary processes need consolidation, streamlining and stronger discipline. The banking system carries an unknown amount of non-performing loans. The burden of the foreign and domestic debt can constrain investment and growth. The fight against inflation and the need to cope with the problem of unemployment remain uphill tasks. The SAL process has inevitably left its crop of casualties in the exacerbation of income polarities as well as an increased dichotomy between the wealth of cities and rural areas. Many fixed-income earners have paid penalties, *as price distortions and subsidies on which they depended were abolished during structural adjustment. 8.03 Attention to these facets of adjustment will require patience and dedication in the future. An aggregate approach to macro-economic planning needs to be orchestrated. Considerable fragmentation has occurred in the economic coordinating process in recent years, to wit -he advent of new - 109 - institutions such as EBPs, as well as circumventions in the approval pro- cess for the execution of public investment projects. Care is also needed to ensure that the gains achieved so far in the context of trade liberali- sation are not reversed through ad hoc administrative fiat. 8.04 The task of keeping the structural adjustment process on track is not an easy one. A close and continuous policy dialogue with the Bank may be necesiary. This needs to be supported by an array of well focussed and closely monitored project- and policy-based loans. Initiatives that would enable the Bank to overcome portfolio constraints in providing such assis- tance need to be given the strongest possible support. 8.05 Recommendations arising from an evaluation of the five SALs are summarized belows SAL Design and Strategy (a) Technical Issues o Close attention to the development of a policy-oriented, macro- economic, analytical medium-term framework very early in the SAL program was necessary and would have lengthened the program's perspective and strengthened its objectives and its sustainabil- ity. The need for a medium-term framework was cited in SAL III but one--in the full sense defined above--was not produced. The process of formulating a medium-term framework would have identi- fied the problems of stagnant private investment and macro-econom- ic instability (discussed in Chapter IV). o Greater focus on debt management, in addition to debt monitoring --both domestic and foreign--should have been a visible component of the program for macro-economic adjustment. This would have brought the problems of potential fiscal and balance of payments instability to the surface earlier. o The mechanics of macro-economic adjustment within the framework of a liboralized trading system and freedom of capital movements should be investigated closely. The problems in the financial system and the connection between exchange rate policy and real interest rates would have been revealed early enough by such an investigation. o Close study was needed during the design of SAL V of the impact of EBFe on the budgetary process, foreign debt, and the trade regime, though in all fairness, their negative aspects were less evident at that time. - 110 - (b) Strategy Issues o A further SAL should have fallowed upon SAL V, as proposed by the Country Department. Lingering problems remained on such issues as debt management, tariff liberalization, budgetary processes, pub- lic investment management, overall demand management and the con- trol of inflation. The timing of SAL V could have been linked to a medium-term macro-frameworki establishment of the medium-term framework would have reaffirmed the Governments's comitment to the combined programs of stabilization and structural adjustment. Implementation and Sustainability o Close attention is necessary to monitor problems and evaluate results during and after a program A suitable mechanism should be devised for this purpose at the commencement of the exercise, as it could serve as a tool for any necessary mid-stream correc- tions. o Statistical and qualitative data relevant to the design of a program need to be readily accessible to the Bank, so that appro- priate countervailing or other measures may be built into the design and, where appropriate, the conditionalities. o Conditionalities imposed on the borrower should be within its capacity to fulfill in terms of the prescribed time horizon, as well as the technical, administrative and financial resources available to it. Also the conditionalities should not be so opaque or ambivalent as to be meaningless to monitor. SALs II and III contained conditionalities that were perceived to be excessively rigidl these were later corrected. The conditiorality clauses relating to performance in SALs IV and V were too opaque and too loose, allowing different interpretations by the Bank and the borrower, respectively. OSatisfactory progress' left too wide a scope for discretion or arbitrary interpretation. o Sustainability of a structural adjustment program of the kind executed by Turkey depends, inter alia, on at least six crucial elements: - First, the existence of a political environment that is con- ducive to the implementation of the spirit and letter of the agreed conditionalities, during and after disbursement of the loan. This commitment was solidly expressed and adhered to for most of the period covered by the SALs, with some excep- tions in 1983. - 111 - Second, the existence of a suitable analytical framework that permits both the Bank and the borrower to identify and re- dress potential problems as they appear. This framework should integrate the micro-economic aspects of structural adjustment and the macro-economics of stabilization as com- pletely as possible. Third, a close and continuous dialogue between the Bank and borrower, based on the agreed analytical framework, needs to be maintained at all times and followed through after the loan is fully disbursed. A mutually rewarding interaction has occurred, supported by operationally meaningful economic and sector work and related sector adjustment and project loans. Fourth, a solid institutional base needs to be created within the key ministries and agencies of the borrower, to keep the program on track and to ensure continuous overview, follow up and needed additional efforts. Fifth, a farorable international trading environment, free of discrimination against exports, needs to be maintained. While market access has been generally good for Turkish ex- ports, enabling it to dramatically increase the share of exports in GDP, a resurgence of protection in foreign markets has restricted access for some commodities. Six, careful attention needs to be paid to ensure that the resource transfer does not turn negative till the fundamental adjustment process has been completed. - 113 - PROGRAM COMPLETION REPORT TURKEY: FOURTH AND FIFTH STRUCTURAL ADJUSTMENT LOANS I. INTRODUCTION 1. In January 1980 the Turkish Government launched a comprehensive package of new economic policies in response to an economic crisis of unprecedented proportions. The program attracted considerable financial support from abroad. The World Bank participated with five Structural Adjustment Loans (SALs), disbursing a total of US$1.556.3 million in the 1980-85 period. Under SALs I-III an amount of US$879.5 million was disbursed in the period 1980-83; for these loans a Program Performance Audit Report was published in June 1985. The present report is the Program Completion Report (PCR) for SAL IV (US$300.8 million and effective on September 19, 1983) and SAL V (US$376 million and effective on July 3, 1984). The report is based on a mission to Turkey in November 1986. 2. The SALs were designed to support the Government in its efforts to restructure the Turkish economy in a major departure from policies pursued for over two decades. The principal aims of the structural reform were to shift from an inward-looking to an outward-oriented strategy, and to place greater reliance on market forces as compared to direct and indirect Government control omnipresent before. To appreciate the depth and magnitude of the changes in policy and to enable evaluation of design, implementation and results of the SALs, some background on the structure of the Turkish economy and on how the crisis came about is essential. Socio-Economic Environment 3. Turkey's economic program adopted in the 1980's was a far-reaching attempt to change previous policies, some of which were deeply rooted in the past. Perhaps more than in most countries, the roots of Turkey's economic problems and future lie in the difficult adjustment process of a country which, under Kemal Ataturk, cut its traditional links and developed a keen nationalism which has remained largely insular in its outlook, despite considerable modernization. In 1923 Turkey emerged from the War of Independence devastated, with a poor infrastructure and a population which was largely illiterate, imbued with a fatalistic outlook, and engaged primarily in rural subsistence. Trade was paralyzed, and the country lacked artisans, bankers and businessmen, for it was largely Greeks and other foreigners who had fulfilled these roles, and they had almost all departed. Turkey needed capital to develop her natural resources, but feared to borrow it abroad lest it fell once more into the foreigners' grip from which it had just freed itself at such great cost. Turkey thus found itself in an economic vacuum, at a time when it was necessary to raise the standard of living and replace a traditional religious outlook with a more modern, secular one. - 114 - 4. As part of an effort to eliminate the extensive poverty and to modernize society Kemalism, an interrelated system with republicanism, secularism, reformism, populism, nationalism and etatism as its elements, was introduced. Unlike other countries where economic nationalism was a reaction to world depression, in Turkey it was a manifestation of a much deeper feeling of mistrust toward foreigners, rooted in the capitulations and other humiliations undergone in the final years of the Ottoman Empire. As a parallel development of nationalism, statism or etatism was introduced. It was never clearly defined, but was meant to insure through populism against exploitation of the people, combining pervasive state control with a certain element of personal enterprise. For economic, but also for political and psychological reasons, industrialization was chosen as the main vehicle for economic growth. Empoverished Turkey could not now afford to buy foreign manufactured goods, it resisted the image of a peasant-country and had decided to seek development through "modern" European models. Consequently, over a period spanning half a century, from its creation and the advent of the great depression through the end of the 1970s, Turkey followed an inward-looking economic strategy with emphasis on import substituting industrialization and heavy reliance on government intervention. This approach was formalized under the first two five-year plans (1963-72), specifying ambitious targets for industrial growth to be achieved through stepped-up capital formation and import substitution. Non-financial public enterprises (known as State Economic Enterprises or SEEs) were primarily responsible for the realization of the targets, with virtually unlimited access to financial resources and favored with protection from foreign competition. In fact, the trade regime became a major instrument of industrial development and restrictions proliferated. The exchange rate was fixed, while multiple rates were provided for certain basic imported inputs. Balance of payments crises were invariably met with a tightening of quantitative import restrictions, followed by a one-off devaluation of the Turkish lira. The subsequent easing of the foreign exchange shortage led then again to some relaxation of restrictions. Results of Early Economic Policies 5. This model early on provided Turkey with good economic growth, which lasted through the 1960s and into the 1970s. GDP increased at an average annual rate of 6.4 percent during the First Plan (1963-67), 6.7 percent during the Second (1968-72), and 7.2 percent during the Third Plan (1973-77). While a 1973 household income survey showed that income remained unequally distributed, overall there had been a substantial improvement in living conditions and the quality of life in general. Significant gains had furthermore been made in nutrition, housing, educational standards, social and health care. It was achieved, moreover, without the availability of large amounts of oil or other valuable natural resources, and within a basically democratic political framework, even though political stability had its ups and downs. Mixed Results of the Seventies: The Need for a New Approach 6. Turkey entered the 1970s achieving rapid economic growth, while maintaining a low rate of inflation and a surplus in the current account of the balance of payments. Partly as a consequence of this favorable external picture, which itself was mainly due to a rapid rise in workers' remittances, - 115 - the public sector stepped up its investments very considerably, and encouraged the private sector to do the same. Therefore, even when, partly due to the first oil crisis, the current account surplus turned into a deficit in 1974, investment as a share of GNP vent up from 18.1 percent in 1973 to 25 percent in 1977. During 1973-76 real GNP growth rate still averaged 7.7 percent, but it was based on a rapid increase in investment and consumption expenditures, spearheaded essentially by the public sector. The growth rate of public investment in 1974-77 was four and half times higher than in 1970-73; and the growth rate of public consumption nearly doubled. Moreover, the deterioration in the terms of trade was not passed on to domestic prices of energy and other key SEE products and as domestic savings lagged, the budget deficit, as a percentage of GDP, increased, causing a severe internal imbalance. In 1977 the public sector borrowing requirement had reached the equivalent of 11.3 percent of GNP, compared with 2 percent of GNP in 1973; the rate of increase of money supply was accelerated to finance this stepped-up expenditure, and there was a sharp increase in the rate of inflation to around 25 percent in 1977. 7. The impact of all this on volume and composition of Turkey's external debt was dramatic. In the 1960s development had occurred with moderate reliance on foreign assistance. In the 1970s it was supported by large increases in workers' remittances, but after 1974 also by an excessive reliance on foreign borrowing. Of the total increase in debt of $8.9 billion between 1972 and 1977, nearly three-fourths was short term. Short term external debt, which stood at $219 million in 1974 and $1.4 billion in 1975, had increased to $6.6 billion at the end of 1977. An innovative scheme (the Convertible TL Deposits Scheme, CTLD), whereby non-reeidents could open deposit accounts with Turkish commercial banks, while principal and interest payments on these deposits were guaranteed by the Central Bank against all foreign exchange risk, made the fast expansion of short term debt possible. As a result, when a crisis of confidence developed, it hit hard; the inflow of external capital dried up, leaving in its wake a host of domestic and external problems, which could not be addressed by the same set of economic policies and philosophy applied so far. 8. In spite of clear signals that a substantial change in policies was required, the Fourth Plan (1979-83), approved by Parliament in November 1978, was in many essential ways an extension of traditional Turkish development planning and reflected more past successes (and failures) than future needs. It set out to complete the unfinished tasks of the Third Plan, and continued to emphasize a high growth rate of 8.2 percent per year, a large investment allocation for import substitution in basic and intermediate goods industries and reliance on administrative controls rather than on market incentives. Where it addressed relevant issues, the targets set were high - an annual export growth rate of 18 percent and a marginal savings ratio of 34 percent - and it did not articulate the policies or the adjustments needed to accomplish these objectives. In fact, the signals were quite mixed in that while the Plan's stated priority was for export-oriented investment, the allocations favored import substitution. The Plan also failed to fully recognize the private sector as an emerging dynamic element. Direct controls as an instrument of economic policy had been quite effective, when the Government was the only powerful actor on the economic stage, but the situation was now changing. What rendered the plan particularly ineffective was that it was - 116 - based on the premise that the Turkish economy was undergoing a short term balance of payments crisis, but that eventually this crisis would pass and the economy would be able to resume its long term trend of sustained growth. These expectations, however, did not materialize. 9. At the same time there were clear signals in other elements of Turkish society that fundamental changes were taking place. Increased urbanization was upsetting old power structures, leading to a period of succession of weak coalitions and alliances of disparate political beliefs and economic philosophy, which brought the country to the brink of political anarchy and economic bankruptcy. In this environment the timeliness and hence the effectiveness of economic measures taken could not always be optimal and a stabilization program announced in early 1978 had only mixed success. The current account deficit of $1.7 billion in 1978 was less than half the $3.6 billion deficit in 1977. However, the budget deficit of the Central Government remained at about the same level as in 1977 and the losses of SEEs were significantly higher. Inflation worsened; prices increased at an average annual rate of 50 percent in 1978. 10. In March 1979 the Government announced a new stabilization program, which provided the basis for a Standby aggreement with the IMF, concluded in July 1979 for a period of one year. Principal economic policy measures included a substantial devaluation (about 44 percent), SEE price increases, continuation of an export earnings retention scheme and the usual controls on Central Bank credit and public sector borrowing. Towards the end of the year, it was evident that the anticipated improvements had not occurred, in that inflation continued unabated, export growth was unimpressive and the balance of payments position remained precarious. In addition the cost of petroleum imports increased significantly by over $2 billion, thus causing a further squeeze on other imported inputs and hence on capacity utilization. As a result of these developments, economic activity remained depressed. GDP decreased (by 0.4 percent) and inflation went up to some 80 percent. In this bleak economic situation a newly appointed minority Government decided that bold policy actions were needed to initiate fundamental, structural and institutional adjustments in the economy. The Government announced its extensive economic reform program in January 1980. The January 1980 Reform Program 11. The policy announcements accompanying the program indicated a major shift in Turkish economic policies. The Government immediately began the process of structural and institutional changes in key areas with a view to fostering recovery and renewed economic growth over the medium term. The program, which still represents the Government's policy framework in 1987 includes the following main features: (i) a realistic and flexible exchange rate; (ii) incentives to encourage producers to export; (iii) tight monetary policies to restrain domestic demand and control inflation, if necessary at the cost of some growth; (iv) improved external debt management; (v) deregulation of interest rates to encourage private savings and in general to improve domestic resource mobilization; - 117 - (vi) steps to improve institutional efficiency in key sectors by changing from direct controls to greater reliance on market mechanisms and prices by both the public and private sectors; (Vi) rationalization of the public investment program; (viii) initiation of SEE reforms to reduce the burden on public finances and improve their efficiency; and (ix) progressive liberalization of imports including the elimination of quantitative restrictions and the rationalization of the tariff structure. It was clear from the outset that the program was ambitious and far-reaching and would need time to be implemented, as it implied a fundamental change in attitudes, behavior institutions and policies. Impact of Domestic Pragmatism and Assistance from Abroad 12. The prospects for successful implementation of the program were certainly enhanced when the geopolitical position of Turkey was emphasized again by the fall of the Shah of Iran in 1979 and the beginning of hostilities between Iraq and Iran in 1980. The military take-over in September 1980 provided Turkey much-needed political stability. The new team, which included a number of prominent civilians like Mr. Turgut Ozal, the principal architect of the January 1980 package, lent their support and started to implement the program vigorously and systematically. Subsequently a Consultative Assembly was set up to draft a new constitution, which was overwhelmingly endorsed in a nationwide referendum in November 1982. New election and party laws were then promulgated, and parliamentary elections completed in November 1983. A new majority Government formed by the Motherland Party assumed office in December 1983 and reaffirmed its commitment to the January 1980 program. 13. Before discussing the SAL program it may be useful to make some general remarks. First, turning an economy around from inward to outward orientation is easier said then done. The inward oriented policy had been applied successfully; Turkey, as a lower middle income country, exported (goods and NFS) only 6 percent of its GDP in 1965, while the average for that group was about 17 percent. As comparative advantage studies indicate, there is no reason why, with the appropriate set of supporting policies, Turkey should not export some 20 percent of its GDP. It illustrates the potential Turkey still has, even though the problems of penetrating key markets are formidable. 14. Second, the financial help received from abroad for the restructuring process has been very substantial. To meet massive immediate payment obligations negotiations, aimed at debt rescheduling and emergency balance of payments support, started as early as mid-1978. A breakthrough came when in July 1979 a Standby agreement with the IMF was concluded. Between 1978 and 1981 Turkey successfully rescheduled about $9.2 billion of both, official source debt ($4.9 billion) and commercial obligations ($4.3 billion), unprecedented amounts at the time. Thus, by end 1981 an estimated 60 percent of Turkey's external debt had been subject to rescheduling and over 30 percent had been rescheduled twice or more, most with repayment obligations pushed back till 1985. - 118 - 15. The IMF has played a crucial role in Turkey's efforts to obtain large-scale external financing and rescheduling and provided substantial assistance to Turkey since 1978. A two-year standby arrangement of SDR 300 million was approved in April 1978 and replaced with a new one-year arrangement of SDR 250 million in 1979. The January 1980 program was supported by a standby arrangement concluded in June 1980 for SDR 1,250 million ($1,630 million) at that time the largest ever extended to a developing country. In June 1983, an additional one-year standby arrangement was approved for SDR 225 million. In April 1984, a final one-year arrangement was approved for SDR 225 million to replace the 1983 arrangement. Bank program lending started in 1980. When the January 1980 restructuring program was announced, the Bank was quick to react and already in March 1980 the Board approved a program loan, later re-labeled SAL, the first such loan of the Bank, which became the first of a series of five, providing about $1.6 billion of quick disbursing balance of payments funds in a 5-year period. 16. Third, general experience with SAL programs shows that SALs cannot be successful without the full backing of the Government. This experience is confirmed in Turkey. A major reason for the failure of reform programs in the late 1970s in Turkey was exactly the lack of political stability and support. It is clear that from 1980 onwards that the Turkish Government has been fully committed; uncertainty in 1983 during elections and the transition from military to civilian rule translated immediately into slippages. 17. The fourth remark refers to the time period required to really restructure an economy, which is what is at stake in Turkey. Restructuring an economy, which has developed basically for 60 years according to a certain pattern, while the economic philosophy itself had been developed as an integral part of an approach touching upon many aspects of society is nothing less but an economic revolution. Policies have to change as have policy instruments; institutions have to be adjusted and attitudes and behavior altered. Structural changes of such depth and magnitude take many years to implement, even if strong backing from the Government and financial support from abroad is available right from the start. The SAL documents often mention that the full restructuring can only be implemented in the medium term, generally meaning some 4-5 years. Indeed, reality in a democracy generally does not allow politicians a time horizon beyond 3 to 5 years and admittedly also for institutions like the IMF and the World Bank, for many practical and valid reasons, it is hard to go beyond that. The program Turkey embarked on in 1980, however, involves such a fundamental reorientation of its economic policies that, realistically, it will take much longer than five years before it will be substantially completed. Much of the so-called "slippages" in the SALs is more a reflection of the realism that the time required to achieve structural adjustment is a medium to long-term proposition, well in excess of five years. 18. Finally, the Government's obsession with growth sometimes seems to be the Achilles' heel of the adjustment and restructuring program. Twice already, in 1983 and in 1986, the Government, eager to achieve high economic growth, stimulated the economy to the point of overheating. It may in this context be useful, first to remember that the country is used to growth as illustrated before and the electorate may be supportive of change, but certainly not at any price. And second, coming out of the seventies Turkey - 119 - faced the problems of sero growth, a worsened income distribution and a large debt service. By trying to solve these problems, partly through continuation of the same expansionary fiscal spending policies and large external borrowings, the situation only got worse, culminating in a debt crisis, and an inflation rate in 1980 in excess of 100 percent. The situation was turned around after 1980 and therein lies the most imediate benefit of the SAL program. However, the country had also mortgaged its future heavily through foreign debt. Servicing a large debt and restoring its creditworthiness once more, received necessarily the highest priority from the Government. Pursuit of distributional objectives too soon might place undue pressures on the internal balances, it was felt. On the other hand the Government frequently stated its intention to achieve strong growth consistent with stabilization and adjustment and thereby to improve the economic position of all elements of the society. However, to ignore distributional aspects for too long was understood to be a risky and sub-optimal policy. II. ACHIEVEMENTS UNDER SALS I-III The Record 19. Between 1980 and 1985, the Bank has supported the Government's program through sizeable policy-based lending with five consecutive one-year SALs, totalling about US$1.6 billion, to promote a wide range of structural measures at the macroeconomic, as well as sectoral, levels. Date of Amount Totally Loan Approval in $M Disbursed SAL I March 1980 275.0 FY81 and Supplement November 1980 SAL II May 1981 300.0 FY81-82 SAL III May 1982 304.5 FY82-83 SAL IV June 1983 300.8 FY84 SAL V June 1984 376.0 FY85 TOTAL 1,556.3 SAL I dealt primarily with the initiation of studies on the rationalization of industrial protection, debt management and accounting, and financial sector restructuring, laying the groundwork for future policies in these areas. SAL II and III focused mainly on public finance issues: adjustment of income tax rates; introduction of a value-added tax; reform of SEE operations, investment and employment on the basis of market principles; and rationalization of the public investment program. SAL IV and V concentrated on: financial sector reform, including elimination of the transactions tax; improved accounting standards and development of financial markets; trade liberalization, notably elimination of quantitative import restrictions and rationalization of the tariff structure; improvement in resource allocation in the agricultural and - 120 - energy sectors through the phaseout of subsidies and initiation of medium-term action programs; and adoption of a medium-term macroeconomic policy framework. 20. The Performance Audit Report (PPAR) of July 1985 for SAL II and III concluded that the loan designs were appropriate for their goals, while the goals in turn were fully consistent with and supportive of the Government's economic restructuring program. A weakness in the process, however, was felt to be the absence of a medium-term strategy document that systematically linked the short-run stabilization program to a longer run development plan. Such a strategy document would have integrated the public investment program and the structural reforms in at least an indicative framework showing a feasible transition to sustainable growth. The strategy document could then have facilitated private sector investment planning, and mobilized aid for private investment. 21. Judging by the progress made between 1980 and 1983, the Turkish economy has shown an impressive response to the structural adjustment program to which the SALs contributed. Until 1982, actual performance met or exceeded the Government's own targets. The reform program was strictly adhered to, and most fiscal and monetary targets in the stabilization program were met. Major structural reforms in the tax system, new import and export regimes, moves towards economic pricing by SEEs, improvement in public investment, the liberalization of non-preferential interest rates, and the adoption of a daily fluctuating exchange rate, all contributed to substantial short-term progress with favorable longer-term implications in the direction indicated in the Government's program. After two consecutive years of decline, real GNP grew in 1981 and 1982 by 4.1 percent and 4.6 percent respectively. The growth was due primarily to significant improvements in the foreign balance since the stabilization program allowed for only slight increases in domestic demand. There was a slow-down in 1983 when thG growth rate reached 3.3 percent and total domestic demand increased by 4.3 percent with a negative contribution coming from the foreign balance. Nevertheless, the three year average growth was significantly better than that achieved in the period before the launching of the SAL process. And most importantly, the economy, after a chaotic period at the end of the seventies, had been given a sense of direction again and the response from various actors in the process indicated that it was perceived as such. 22. A number of specific accomplishments during this period included the following: - major reforms in the tax system were prepared and executed; considerable progress was made in liberalizing non-preferential interest rates and the real interest rate became positive again; and daily fluctuating exchange rates were adopted; - the rate of inflation, as measured by the wholesale price index, dropped from an annual rate of about 107 percent in 1980 to 31 percent in 1983; - merchandise exports rose from $2.9 billion in 1980 to $5.7 billion in 1983, with export growth being led by the manufacturing sector and involving a significant diversification of markets. The current account deficit fell from $3.2 billion (5.6 percent of GNP) in 1980 to about $2.0 billion (3.4 percent of GNP) in 1983; - 121 - - there was a marked retrenchment of the public sector. Government Consolidated Budget expenditures, which stood at 24.0 percent of GNP in 1980, fell to 19.4 percent in 1983. The budget deficit to GNP ratio, which stood at 4.6 percent in 1980, decreased steadily to 0.5 percent in 1983, while SEE transfers as a percentage of GNP showed a decline from 4.8 percent in 1980 to 2.5 percent in 1983. The financial performance of SEEs also registered substantial improvement, as a situation of chronic deficits changed to one of growing profits. In other areas of the public sector, like the extra budgetary funds, expenditure increased and overall progress in cutting down the public sector deficit was le3s than the progress in some specific departments suggests; - successful efforts were made to constrain the level of public investment in line with available resources, to limit the number of projects to a manageable level, and to ensure that priority projects received larger allocations in order to speed up their completion. Allocations for the energy, agriculture and transport sectors have been increased and the share of manufacturing reduced. Public sector investment in real terms increased by 9.4 percent in 1981, by 2.2 percent in 1982 end by 1.7 percent in 1983, while private investment after having declined by 8.7 percent in 1981 rose by 5.4 percent and 4.8 percent during 1982 and 1983 respectively. As a result, the share of private fixed investment in total investment improved; - substantial progress was made in import liberalization, through the abolition of quotas, the freeing of a large nuber of items from licensing, rationalization of tariffs and a simplification of administrative procedures; - in the energy and agriculture sectors, actions were taken to increase institutional efficiency, adjust prices closer to economic levels and reduce input subsidies in a phased manner; and - measures had been !nitiated to simplify the regulations surrounding banking operations and improve the performance of money and capital markets. 23. This brief sunmary demonstrates clearly the existence of a strong and sustained government commitment without which no structural adjustment program can succeed. One of the major conclusions of the SAL 11-III Program PPAR was that overall, the programs connected with SALs II and III were deemed to have been a success. Exports grew rapidly in response to the Government's policy emphasis, supported by a plethora of incentives and markets were diversified. The flexible use of the exchange rate and import liberalization policies improved resource allocation. Delinking of SEEs from the Treasury improved their financial self-reliance. Fiscal and credit incentives widened the opportunities for many producers, many of whom had previously relied excessively on domestic markets. Financial deregulation and policies for strengthening the functioning of capital markets improved financial intermediation. Finally, institutional reforms in agriculture helped to improve productivity over the longer-term. Major reforms were also instituted to conserve, as well as exploit, Turkey's energy resources. - 122 - Areas for Special Attention under Later SALs 24. As the adjustment program is a long term program there remained a number of things still to be done at the end of SAL III. The tranche review for SAL III (February 1983) concluded that the adjustment program was progressing well and the second tranche was released, but some areas of concern were identified as well. Exports continued their excellent performance, but capacity utilization remained generally low and private investment did not respond to the program as hoped. Inflation and unemployment remained persistently high, as did real interest rates. Not unrelated, the 1982 budget deficit was slightly worse than expected, while substantial excess liquidity was generated in the system. The financial sector remained fairly undeveloped and attention was focussed on its deficiencies when in 1982 the country's largest private brokerage house collapsed. An immediate banking crisis was averted, but it focused Government attention on additional actions required to reform and strengthen the financial sector. Showing flexibility, the financial sector was subsequently made one of the focal points in the design of SAL IV. III. SALs IV and V: OBJECTIVES AND POLICIES General 25. The action programs for SALs IV and V form the fourth and fifth phase of the comprehensive structural adjustment process that Turkey began in January 1980. As with previous SALs in Turkey, both SALs embraced a number of decisions already taken, measures to redress some slippages from earlier SALs, as well as measures still to be taken in late 1983, 1984 and early 1985. 26. SAL IV was preappraised in October 1982, appraised in March 1983, negotiated in May of that year and became effective in September 1983. The loan was disbursed in two tranches, one for $200 million and one for $100.8 million following a satisfactory tranche review. The loan was fully disbursed by March 1984. SAL V was preappraised in January 1984, appraised in March, negotiated in May and became effective in July 1984. The loan was disbursed in two tranches, one for $250 million and one for $126 million, following a satisfactory tranche review. The loan was fully disbursed by June 1985. 27. Although not explicitly stated, it was clear almost from the beginning that the program would comprise at least five SALs. However, it was understood that by that time the adjustment process would not be complete and alternative vehicles had to be found to carry on with Bank assistance for Turkey's structural adjustment process, a process which itself changed as it progressed. Unfortunately, it is not enough for turning the structural reform into a success to take a few major decisions at the top; massive dismantling of old institutions and procedures and building of new instruments and institutions downstream was required to allow forces working on the improvement of resource mobilization and allocation to make their impact. Most of the sector reforms anticipated, therefore, required substantial improvement of institutions before they could be properly implemented: this was especially true for the financial sector and for agriculture, energy, and the SEEs, priority sectors in the adjustment process. This was recognized at an early stage of the adjustment process and the Bank's economic and sector work program (ESW) for Turkey was planned accordingly. Already in SAL II a beginning was made to prepare for the post-SAL era, when the progress made at - 123 - the macro level would need follow-up at the sector level, for which the institutional improvements had to be made early. This strategy became fully visible in SALs IV and V where the groundwork was laid for a period of post-SAL sector adjustment loans (SECALs) for the agriculture, financial and energy sectors. 28. The backbone of the SALs is the "Statement of Development Policies" (SDP), drafted in unison by the Government and the Bank, in which the'former reflects on achievements and looks ahead basically one year, formulating the steps and decisions it intends to take in the continuing reform process. The SDPs for SAL IV and V reiterated that the previous three years witnessed considerable progress in the fundamental restructuring of the Turkish economy. The main thrust of the program remained to redirect the economy towards a development path placing more reliance on market forces and adopting a more outward-oriented strategy. The SDPs for both SALs reconfirmed this commitment strongly. The strategy continued to have four themes: monetary stability, export promotion, an enhanced role for the private sector, especially in manufacturing, and an increased reliance on market forces as opposed to regulation or direct control. Along these lines, the outward-orientation would intensify and lessen the dependence on import substitution and production for the domestic market which, until 1980, had characterized economic management. The SDPs emphasized again that the complete harmonization of the various policy regimes underpinning the new orientation was, of course, a medium-term effort which could only be partially .ndertaken in any given year and therefore the Government had to think of 4-cremental achievements within the broader context of a medium-term framework. 29. SAL IV and V were different to earlier SALs in Turkey in that they continued to pursue macro-economic targets, while adding a much more increased emphasis on sectoral objectives. To support the program of structural change, two distinct types of measures were presented in the SDPs: (i) macro-economic, which included the development of a medium-term framework, SEE reforms, import liberalization, export growth and the continual rationalization of the public investment program; and (ii) sectoral, which included the implementation of specific measures to deal with structural issues in the agricultural, energy and financial sectors. Macro-economic Objectives, Policies and Measures 30. SAL IV and V were similar in scope; SAL V has even been characterized as a "maintenance operation", which it was not. Both have macro-economic objectives, with SAL IV being much more specific in this area. Also, in SAL IV the financial sector is addressed in detail. SAL V is designed around the firm confirmation of policies by the newly elected Ozal Government and the many measures taken by that Government in early 1984. SAL V also is trying to come to grips with the Government's long awaited comprehensive medium-term outlook, but the Fifth Plan (1985-89) was still only available in draft at the time SAL V was designed. - 124 - 31. On the macro-economic front SAL IV reflected the Government's view that inflation had to be brought down to 20 percent in 1983, while fiscal balance had to be achieved before Turkey could regain its target growth path. They were considered short term goals and must be achieved as rapidly as possible in order to enable adequate attention to be devoted to medium-term issues including growth and employment. In real terms, GNP was targeted to grow at 4.8 percent in 1983, representing a moderate improvement over the 1982 level; this was due primarily to a projected increase in domestic demand of 4.3 percent (versus 2.5 percent in 1982) and a 0.5 percent contribution from the foreign balance (versus 1.9 percent in 1982). The impetus to this growth was expected to come from a revival of private sector investment which was projected to grow by 7.0 percent in real terms. SAL IV also quantified growth targets by sector and for the balance of paymLits. 32. After establishing that the positive trend in the adjustment process had been reversed in 1983, SAL V stated that bringing inflation under control was the main short-term objective, as a stable economic environment would make resolving social issues easier. The SDP confirmed that free market orientation would continue to be pursued, while direct state activities should be confined mainly to the development of an infrastructure serving the nation as a whole. Finally administrative and legal procedures would have to be speeded up and be made less wasteful. These statements and the following steps taken by the Government that took office in December 1983 confirmed and strengthened the policies stated and implemented since the beginning of the 1980s: (i) a reorganization of the admi.nistrative machinery which, inter alia, concentrated authority over economic affairs in the Depty Prime Minister's office; (ii) a substantial import liberalization; (iii) a rationalization of export incentives; (iv) a liberalization of the foreign exchange regime; (v) an interest rate policy designed to provide a real rate of return on savings, while largely freeing lending rates to be set by the commercial banks; (vi) a reduction in the financial transactions tax and the withholding tax on interest payments; and (vii) a reaffirmation that, with some exceptions, SEEs would be free to set their own prices. 33. SAL V furthermore concentrated on the Fifth Five Year Plan (1985-89). So far the adjustment process had been conducted without the benefit of a consistent medium-term framework, in the context of the Fourth Plan, as mentioned before, was more a continuation of previous policies rather than a reflection of the new orientation. It was the Government's intention to have the Fifth Plan prepared by mid-1983, and the 1983 SDP stated so, but as the year unfolded political developments led to an election in November and it was decided not to preempt the new Government's economic policies, a sensible decision. 34. At the time of appraising SAL V the Fifth Plan had not yet been approved by the National Assembly, but the 1984 SDP gave a sort of preview. The annual growth rate of GNP for the five-year period was forecast at 6 percent; exports would increase by 8-11 percent; the balance of public/private investment, which stood at about 60/40 in 1984 would become about 50/50 at the end of the plan period; the share of total public investment going to the priority sectors energy, agriculture, transport and communications would not be less than the share achieved in 1983; and the debt service ratio would average about 25 percent. It also was the intention to make the Fifth Plan - 125 - more indicative in nature than its predecessors, showing the Government's desire to design and implement the public investment program in a manner which would be compatible with pursuing the free market system. The basic theme of the Plan would confirm the practised outward-oriented development with an increased reliance on market forces. The Plan would define the respective roles of the public and private sectors, with increasing reliance placed on an efficient private sector as the engine of growth. Private initiative and innovation along the lines of Turkey's comparative advantage would be encouraged. The public sector would provide the infrastructure required to allow the private sector to achieve its potential. This shift in emphasis would also mean that the Government would make greater use of private sector participation in the planning committees. 35. The SDP of 1984 (SAL V) was also very specific about the role of the private sector as envisaged by the Government. At the heart of the structural adjustment program was a development strategy based on sustained export growth over the medium term. Such a strategy required that the private sector be allowed to play a leading role, since export growth was critically dependent on private manufacturers developing their comparative advantage. Not only would the size of the Plan have to be in line with the availability of domestic and foreign resources, but in addition, the public investment program would have to be so designed that it would not lead to preempting of resources for the private sector. 36. The Government considered that the achievement of a satisfactory rate of growth was essential to provide a stable economic environment. To encourage the necessary level of private investment, the Government believed that priority should be given to bringing inflation under control. This would entail tight monetaryMand fiscal policies designed to keep public expenditures in line with public resources as they materialized. While the adjustment process might involve hardship in the short run, it was designed to provide a sounder economic base from which would emerge improvements in the standards of living of the population over the medium term. 37. The SDP for SAL IV included specifics on fiscal policy and on vublic investment. In terms of fiscal policy the Government stated that in 1983 expenditures, estimated to be about 20.8 percent of GNP, were expected to be matched by revenues amounting to 20.2 percent of GNP; this would leave a budget deficit of 0.6 percent which would be a further improvement on the results of 1981 and 1982. If revenues would fail to match the planned 25 percent increase in 1983, expenditures would be pared accordingly. To improve the equity and efficiency of the tax system the Government announced in its 1983 SDP that it had: (i) reduced personal income tax rate on the first TL1 million of taxable income from 39 to 36 percent; (ii) reduced the corporate income tax from 50 to 40 percent (in respect of income accrued in 1982); (iii) abolished the fiscal balance tax on corporate profits and income; (tv) doubled the assessment rates on real estate taxes; (v) linked personal income tax paid by the self-employed to the taxpayer's wealth; and (vi) enacted a law to facilitate payment of tax arrears in installments over the next two years. 38. In the 1983 program public fixed investment was programmed to grow at 5.2 percent in real terms with the emphasis on the completion of ongoing projects. The Government would also continue to rationalize the ongoing - 126 - program and to restrain the growth of public investment. In the context of the longer term goal of an equal balance in the ratio of private and public fixed investment the Government in 1983 set specific targets for public funds going to priority sectors. 39. For monetary and credit policy; the reform of SEEs; and trade liberalization, SAL IV and SAL V were basically the same. The aims of monetary and credit policy were defined to reduce inflation and improve the balance of payments through a flexible exchange rate policy to maintain external competitiveness. Interest rate policy aimed at providing a real rate of return on savings. However, lending rates remained high in real terms and thus hindered investments. Therefore, the Government aimed at increasing the role of the market to solve the conflict and to reduce the differences between the interest rates on preferential and non-preferential credits. 40. Reiterating a strong principle laid down in the early 1920's, the Government stated that State Economic Enterprises should continue to provide the infrastructure and services required for the society as a whole. But the State should not appropriate funds directly for investment in activities that should be the province of the private sector. In this context the Government intended to explore opportunities by offering share capital of certain SEEs to the public and did not intend to protect the monopoly status of any SEE in the industrial sector. The Government also intended to strengthen management to create an operational environment ir SEEs, to encourage managerial responsibility and to implement performance criteria, including target rates of return on assets employed. 41. The Government placed considerable emphasis on bringing about a structural change in the import substitution policy practised hitherto. Incentives should be afforded largely through the exchange rate. Measures were taken to reduce drastically the quantitative restrictions on imports and to reduce tariffs. The stated objective was to complete the transition to a system relying solely on tariffs and to undertake a rationalization of the tariff structure to achieve significant uniform reduction in effective protection across sectors, provided similar developments would take place in the countries that are Turkey's major trading partners. On the export side the Government expressed its intention to maintain the flexible exchange rate policy, to ensure the availability of a satisfactory level of export credits within the credit constraints, and to encourage commercial banks to increase their role in such financing. The introduction of a system of export credit insurance would be studied. Sectoral Targets, Measures and Policies 42. SALs IV and V emphasized the agriculture, financial and energy sectors in order to lay the groundwork for what later in the adjustment process could develop into full fledged sector adjustment loans addressing a broad array of sectoral policy issues, in this way directly and indirectly continuing to support policy changes at the macro level. 43. In view of the importance of the financial sector in the overall structural adjustment process, the Government's objectives after the severe problems of 1982 were to restore and maintain confidence in the financial - 127 - system and to expand its range, reduce the cost of credit, improve the efficiency of credit allocation, allow more competition in the financial sector, strengthen the institutions which supervise the financial system and revitalize the capital market. Immediate actions would be undertaken to redress the short-term problems of high cost of credit for non-preferential borrowers and of liquidity and solvency. The structural problems which were related to sectoral finance, the efficiency of the financial institutions, the development of capital markets and bank supervision would be addressed over the medium term. 44. Specific measures the Government had in mind were: (i) on interest rates to: reduce the excessively high real interest rates; reduce the disparity of interest rates applied to preferential credits and to make the lowest rate equal to the rate of inflation; introduce a deposit insurance scheme; and undertake a study for introducing a system of floating interest rates for medium-term and long-term credits; (ii) on capital markets to: ensure that the Capital Market Board would begin to function effectively; revitalize the bond market by issuing treasury bills and Government bonds at competitive rates; undertake a study reviewing the introduction of medium- and long-term bonds with interest rates indexed to the announced interest rates on time deposits; and undertake a study to devise strategies for long-term development of the equity and stock market including measures in the areas of accounting and auditing standards; (III) on financial structure to: enable firms to revalue their fixed assets and accumulated depreciation; and introduce legislation regulating the accounting profession; (iv) on bank supervision to: enlarge the role of the Central Bank; introduce a standardized accounting system; and improve the system of monitoring bad debts. 45. In agriculture the Government would continue to promote higher productivity and growth of exports through encouraging a crop mix which vould better reflect Turkey's comparative advantage. The outward orientation would include modifying rigid self-sufficiency as a goal and increasing agricultural exports while allowing for food imports as necessary. Improved price signals, reduced subsidies and improved credit availabilities would increase the efficiency of resource allocation. Specific measures envisaged in agriculture were: (i) on investments to: allocate not less than 11 percent of public investments; (ii) on pricing and marketing to: move from producer support to floor prices; phase out fertilizer subsidies over the next three years; review the self-sufficiency policies, including examining the feasibility of removing restrictions imposed on exportation of wheat by the private sector; and revise the seeds decree to enable efficient and expeditious testing of imported varieties of seeds and to encourage formation of joint ventures with foreign firms. - 128 - (iii) on research and extension services to: develop a medium-term planning capability; set priorities for the use of research; and develop medium-term plans for reorganizing and consolidating the Government's diverse extension services, including decentralization of responsibility to the regional level in addition to improving links with agricultural research; and (v) on credit to: develop and implement proposals for improving the efficiency and accountability of the sales cooperatives and study measures to make them financially self-sufficient with their credit obtained on commercial terms. 46. The Government recognized the importance of energy to future prospects for growth and the need to reduce oil imports by increasing the production of indigenous sources of energy and by encouraging greater conservation. Thirty-four percent of the public investment program would be channelled into this sector. The Government intended to improve performance and efficiency under the new SEE decree and to improve planning and coordination in the sector. Private sector involvement in energy development would be encouraged and accordingly the petroleum and mining laws were under revision to reflect this desire. Plans to establish a Planning and Progranming Department in the MENR with appropriate terms of reference and staffing were at an advanced stage. The work program would include the responsibility for coordinating preparation of revised energy demand forecasts to be updated on an annual basis. The Planning and Programming Department would carry out a proper evaluation of a number of large, high priority projects in the energy investment program and would also review existing pricing policies and practices with a view to identifying appropriate relative prices between petroleum products and lignite so as to encourage the most efficient use of energy resources. IV. IMPACT OF THE STRUCTURAL ADJUSTMENT PROGRAM General 47. The structural adjustment program was and is a program of the Turkish Government to reform and restructure the economy. In this effort it has received substantial assistance from abroad; the Bank has been, and still is, a major contributor. The program had three basic objectives: (i) promoting economic growth; (ii) strengthening the balance of payments; and (iii) promoting financial stability. Compared to the starting point in 1980, with a shrinking GNP, a current account deficit of over 5 percent of GNP, a debt service ratio of over 30 and, as a result, economic uncertainty and political instability, Turkey has come a long way. 48. The turnaround in Turkey's economy came quicker and was of a larger magnitude, especially in exports, than originally expected. Due to indecisiveness, rather normal in an election year, a change from military to civilian Government and bad weather causing agriculture to stagnate, the year 1983 showed considerable slippage. However, strong action of the new Government put the adjustment process back on course and progress was made again in 1984 and 1985 under SALs IV and V, but at a price as the next table shows. - 129 - Table 1: GNP, CURRENT ACCOUNT, DEBT, DZBT-SERVICE RATIO AND INFLATIONs 1979-86 1979 1980 1981 1982 1983 1984 1985 1986 GN1 (Z growthlyear) -0.4 -1.0 4.1 4.6 3.3 5.9 5.1 7.0 Current Account Deficit (C of GNP) -2.0 -5.3 -3.3 -1.7 -3.7 -2.8 -1.9 -2.5 Total Gross External Liabilities ($ bin) 15.8 19.2 19.7 20.2 20.7 22.5 26.1 29.0 Debt-Service Ratio 1/ 32.4 30.6 27.3 28.5 30.3 26.1 33.4 37.8 2/ Inflation (% per year)l/ 64 107 37 25 31 52 43 28 l/ Debt service as a proportion of exports and services. g/ Estimated, including interest on short-term debt. 1/ Change in wholesale price index. 49. GNP growth, although admittedly a very limited and narrow measure of success, increased from -0.4 and -1.0 percent in 1979 and 1980 respectively, to an average of 5.0 percent for the 1981-86 period, with a low of 3.3 percent and a high of 7.0 purcent in.1983 and 1986, respectively. This represented good performance, especially because during this period globally growth rates had a tendency to be much below those attained in the sixties and seventies, which is not much help for a country reorienting its economy towards exporting. However, as the table shows, debt has gone up very substantially and inflation, after an initial strong decline, has stayed uncomfortably high. 50. It is too early to judge fully achievements in 1986, but indications are that, although growth may have been high, other indicators give reason for concern. Regarding the balance of payments, gains from a lower oil import bill were largely offset by sluggish growth in Turkey's export markets, especially in the Middle East and grace periods of debt reachedulings have run out. High inflation is one reason why the response of the private sector remains below expectation, which in turn may be a major reason for the public sector to step up activities, endangering the very development model upon which most of the adjustment process is predicated. 51. At the heart of the Government's program throughout the structural adjustment process has been stimulating growth, especially through the private sector, export expansion and inflation control. The Government's preference for growth and its risks has been discussed earlier in this report. Basically, the Government has opted for high growth, which it considered the quickest way to meet unemployment concerns and improve living conditions. For the period 1980-82 growth was based primarily on significant improvements in the foreign balance since the stabilisation program allowed for only slight increases in domestic demand. In 1983, however, the situation was reversed as total domestic demand increased by 4.7 percent with a negative contribution of 1.3 percent coming from the foreign balance. The table below shows that domestic demand continued to be the main carrier of growth in 1984 and 1985 which, with 5.9 and 5.1 percent respectively in 1984 and 1985, has been considerable, especially taking into account the international economic environment. The National Accounts table shows how especially private - 130 - consumption suffered in 1980 and still in 1981 when, through a stabilization program, the economy really had to be turned around. However, during the 1981-86 period private consumption increased again by 5.2 percent on average annually, or close to 3 percent per capita, which is high. In fact, consumption has been allowed to continue to grow at a rate, which left too little for savings to cover the investments made. IhkJU&A: NATIONAL ACCOUNTS, 1980-86 1983 Level 1980 1981 1982 1983 1984 1985 1986 At current In -------------A c t u a 1 ------------ Provisional Prices Percent Real Change from Estimate ITL billions) of GNP Previous Year f%M Consumption 9585.3 83.5 -3.4 0.6 3.9 4.7 5.3 3.5 7.7 Public 1167.3 10.2 6.5 0.9 2.0 1.7 3.0 3.2 7.4 Private 8418.0 73.3 -4.9 0.6 4.2 5.0 5.5 3.6 7.8 Fixed Capital Investment 2180.8 19.0 -10.2 1.7 3.5 3.0 4.7 10.9 11.5 Public 1315.8 11.5 -3.8 9.4 2.2 1.9 1.8 13.3 10.2 Private 865.0 7.5 -17.3 -8.7 S.5 4.7 8.8 7.8 13.5 Stock changes 194.7 1.7 (3.9) (0.8) (-1.0) (0.9) (0.4) (-0.5) (0.4) Total domestic demand 114865.8 104.1 -1.0 1.6 2.8 4.7 5.5 4.5 8.9 Foreign balance 475.6 -4.1 (0.0) (2.5) (1.7) (-1.3) (0.3) (0.5) (-1.2) GNP 11485.2 100.0 -1.1 4.1 4.6 3.3 5.9 5.1 7.0 Uemrandum Items Agriculture 2058.0 17.9 1.7 0.1 6.4 -0.1 3.5 2.4 7.1 Industry 3096.0 27.0 -5.9 7.6 4.6 8.2 9.3 6.3 9.2 Construction 447.6 3.9 0.8 0.4 0.5 0.6 1.9 2.9 5.6 Services A1L8Z SI A.J. LA LA LA LA f lA GOP at Factor Cost 10784.8 100.0 -1.0 4.7 4.3 4.1 5.8 4.2 7.1 NOTE: Figures in parentheses are the contribution in percentage points to the growth in GNP. Source: State Planning Organization. - 131 - 52. Sectoral output has been rather varied. Agricultural growth has shown the largest swings - from a -0.1 percent decline in 1983 to an estimated 7.1 percent growth in 1986. Except for 1980, industry has grown faster and more steadily than agriculture, made possible by the spectacular growth of manufactured goods exports, whose share has risen from less than one-third to nearly three quarters of Turkey's total exports today. Good weather, increased fertilizer supplies, and a general rise in aggregate demand (from about 4.0 percent in 1985 to about 6.0 percent estimated for 1986) have led to the estimate of a sharp increase in the growth of agricultural output (double that of 1985). Private and Public Investment 53. One of the areas for concern reported in the tranche release mission of SAL III in early 1983 was the slow response of the private sector to the new policies. Therefore, in the design of SAL IV and V the ratio of public over total investment was introduced to enable close monitoring.l/ However, because of rather big jumps from one year to the other, for which statistical weakness may partly be responsible, this benchmark is not very stable and the story told depends much on the base year chosen. The pre-1980 data help to understand that private sector investments as a proportion of GDP remained rather stable at between 8 and 9 percent for a period of over two decades, while public investments increased considerably in the early 1980's, coming down in 1983 to go up again thereafter. This gives cause for concern, because the restructuring exercise is modelled on high growth coming increasingly from a more efficient private sector, with investments in the public sector concentrating on infrastructure that supports productive activities. However, the return on such stepped-up public investments in infrastructure will only be substantial if the private sector follows suit. Realizing that the private sector can be enticed but cannot be forced to invest, highlights the difficulties of forecasting private investment flows. Yet, such investment is critically important to the economic reform program. 1/ National Account Statistics are rather weak and subject to frequent changes. In 1981-82 the ratio of public fixed investments/total fixed investments stood at between 60 and 61 percent (it is not clear whether stock changes were included, as they should). The target was to bring the ratio down to 50 percent by 1990. It should be noted that this level or lower had already been achieved in a number of years prior to 1980. - 132 - Table 3: INVESTMENTS IN THE PRIVATE AND PUBLIC SECTOR (1 of GDP) 1962 1972 1980 1981 1982 1983 1984 1985 1986 Program Pub. sector invest. 7.1 8.9 11.8 13.7 12.3 9.7 10.0 11.4 10.8 Priv. sector invest. 8.6 9.0 10.2 8.5 8.5 9.0 8.9 8.2 8.6 PubliclTotal invest. 45 50 54 62 59 52 53 58 56 54. Investments in the public sector have indeed increasingly avoided the productive sectors for the supporting ones as the following tabulation shows: Table 4: FIXED PUBLIC INVESTMENTS IN PRODUCTION AND SUPPORT SUB-SECTORS (as percentage of total fixed public investments) 1980-82 (average) 1984 1985 Agriculture 9.1 9.1 6.8 Manufacturing 24.8 15.7 13.7 Energy 24.7 24.4 23.3 Transportation 18.8 24.4 28.8 It is, of course, of great concern if the private sector would fail to do its shares but it would be even worse if the public sector, to compensate, would step in early, making it even more difficult for the private tictor to follow suit. And yet, this is what may be happening in Turkey at present, although data for 1986 are still too incomplete to draw firm conclusions. Also, in a structural adjustment process the various actors have to be given sufficient time to react to resource allocating new signals and, generally, this takes longer than policy makers plan for. However, after a period of moderate growth, public investments grew by 13.3 percent in real terms in 1985, while private investments went up by 7.8 percent, but by only 5 percent if housing is excluded. Balance of Payments 55. Improvements in the balance of payments suffered a considerable setback in 1983, but in general progressed impressively from a current account deficit of 5.3 percent of GDP in 1980 to a deficit of 1.9 percent in 1985. The 1983 deterioration happened principally in response to a weakening of the trade balance and a drop in workers' remittances, with the current account deficit climbing to about $2.1 billion (3.5 percent of GNP). This was because merchandise exports, which were originally programmed at $6.8 billion, only reached $5.7 billion and workers' remittances. originally expected to reach $2.2 billion, only managed a $1.6 billion showing. Export volume growth, however, maintained a brisk 13.4 percent growth reflecting the continuation of improvements in the competitiveness of Turkish goods in European and Middle Eastern markets. On balance, the performance of merchandise exports has been most impressive over the 1980-83 period. - 133 - 56. The balance of payments improved again in 1984 and 1985. Merchandise exports grew by about 12 percent in 1985 (10 percent in volume terms), with exports of industrial products growing by 26 percent, despite the scaling down of export tax rebates, the elimination of subsidized Central Bank export credits in January 1985, and the absence of significant real exchange rate depreciation. However, the establishment of the Resource Utilization Support Fund in January 1985, under which investors were eligible for a cash grant on the value of manufactured goods exports is estimated to have contributed to an increased level of overall export subsidy from 15.2 percent of export value in 1984 to 18.6 percent in 1985. Import growth in 1985 was limited to 9 percent (7 percent in volume terms), due to the slight decline of oil prices and slower growth in the volume of non-oil imports (10 percent versus 22 percent in 1984). Consumer goods imports in particular, though still comprising less than 6 percent of total imports, were by far the fastest growing category of imports in 1985, as was the case in 1984. It supports the observation that in these years growth was largely domestic demand driven a worrying development. Table 5: SOME MAJOR BALANCE OF PAVMENTS CATEGORIES 1980-85 ($ millions) 1980 1981 1982 1983 1984 1985 Exports of goods & NFS 3,266 5,366 6,980 6,889 8,590 9,958 Imports of goods & NFS 7,854 8,857 8,886 9,221 10,886 11,839 Workers' Remittances 2,071 2,490 2,140 1,513 1,807 1,714 Net Transfers 18 16 105 236 229 236 Current Account Balance -3,408 -1,919 -935 -1,898 -1,407 -1,013 Direct Private Investment 18 95 55 46 113 99 Public M&LT Capital (net) a/ 1,554 425 230 -136 245 -780 Other Capital b/ 1,832 982 668 2,084 1,174 1,969 Change in Reserves (-=increase) 4 417 -18 -96 -125 -275 Current Account Balance (M of GNP) -5.3 -3.3 -1.7 -3.7 -2.8 -1.9 g/ Includes private guaranteed and non-guaranteed debt. b/ Includes errors and omissions. Source: Ministry of Finance and Central Bank 57. The breakdown of export performance in the next table shows that agricultural and mining products moved very little and that the 1980-85 export drivu was entirely based on industrial products: processed food more than trippled, textiles quadruppled and other manufactured products increased almost ninefold. Although this record is impressive it should be realized that exports went up by 62 percent in 1981, but that it took exports another four years to grow by another 69 percent in dollar value; it may even have taken more than five years as exports in 1986 may not reach the dollar value - 134 - of 1985 (exports through August 1986 where down 7.5 percent compared to the same period a year before). These results were achieved by a combination of indirect (exchange rate policy and import liberalization) and direct (tax rebates, preferential credits) measures to enhance the relative profitability of exports versus the alternative of producing for the domestic market. The flexible exchange rate policy of the Government was the most important factor contributing to the growth of industrial exports. The penetration of Turkish products in the Middle East also contributed significantly to the positive results. Here again it is important to note that the more dramatic changes in the proportions as to destination of exports were reached in 1981 at the very beginning of the adjustment process. Init.S: iUOaI lsdorts, ton-as xorts by sector tvestock and sa prAacts M$g *gg 4 1 1 1 nIS 416 usn s so m sas a m mining nt1 2 1M A in2 I 6 310 sExprts by destlnstion Inustral ofe asts tn TexOcmtles 1 2 470 SIS335 0 320 0 95 1 4402 03 3111 12" 811 122 31913 1 Other Wa36i ton 104 113 2452 ISIS o OTA am U n go li As nn a a n" an nu lo soure stat lannin5 areantastion . miy INS 4417150 1462 i9 Otheor Mctis D 21 A z 2 J& U JA 1A3 H M3 1 ITa 1 20 110 123 10 LW1 1U 213 S2 1W0aa 1 thre r S a is,min OrganIzati4o66 - 135 - 58. Worker's remittances recovered in 1984, but started a decline again in 1985, which has continued in 1986. Several factors have accounted for the slow-down of remittances: (M) the number of Turkish workers abroad peaked in 1984. Both, the recent economic downturn in the Middle East and schemes in host countries, which provide incentives for Turkish workers to return home, can be expected to result in further net reductions of workers abroad; (ii) the decline of remittances has been accompanied by the growth of non-resident foreign exchange deposits at the Central Bank (Dresdner Bank scheme) as well as at the commercial banks; and (iii) the decline of remittances may also reflect a heightened perception among non-residents about the likelihood of larger depreciations of the exchange rate in the future. 59. Preliminary data for 1986 indicate that the current account of the balance of payments has worsened considerably in 1986. The performance of exports in 1986 has been disappointing. The deterioration in the financial position of Turkey's oil exporting trading partners was clearly a factor in the weak export performance and counteracted the healthy impact of an oil bill reduction of an estimated $1.4 billion. With regard to the incentives framework, the trade-weighted real exchange rate, after remaining stable during 1985, has depreciated significantly thus far in 1986 (by about 8 percent). However, balanced against this, direct export incentives in 1986 are estimated to have declined as compared to 1985, while the increase in the level of nominal protection observed since 1985 may also be beginning to have a detrimental impact upon export competitiveness. 60. The widening of the current account deficit of the balance of payments in 1986 appears primarily to have been the result of expansionary government policies, which haye succeeded in raising economic growth well above the original target for 1986. In particular, the.continuation of the rapid growth of public investment observed in early 1986 has contributed to the very rapid rise in capital goods imports during the first half of 1986. 61. The capital account of the balance of payments shows how drastically the net inflow of borrowed capital has come down; from $3.4 billion in 1980 to $1.2 billion in 1985. Concomitant with the reduction in the current account deficit in 1984 and 1985 was a pattern of greater medium and long-term borrowing from commercial banks, signalling the improvement of Turkey's international creditworthiness. Medium and long-term credit from banks amounted to about $1.9 billion in 1985 as compared to $1.0 billion in 1984. Short-term borrowing, particularly by the private sector, also grew rapidly during the year. 62. Debt. Since 1978 when Turkey was confronted with, and overcame, a huge debt crisis, it has made great progress in restructuring its debt and has met all of its obligations. Whereas short-term debt accounted for about 50 percent of total debt in 1978 it accounted for only 12.5 percent in 1981 as a result of major reschedulings. This proportion grew to 20 percent over the next three years and to 29 percent as of June 1986. It argues strongly for caution, especially in relation to other recent signals, which seem to point at the public sector stepping in for an apparently hesitant private sector. - 136 - Table 7: EETERNAL DEBT SERVICE (US dollars, millions) 1984 1985 1986 g/ Total Principal Payments 1,437 2,633 2,664 of which: Amortization 1,221 2,382 2,312 IMF Repurchases 216 251 330 Total Interest Payments 1,586 1,753 1,993 of which: If Charges 178 135 110 Total Debt Service 3,023 4,386 4,657 Debt Service/Exports (5) g/ 26.1 33.4 37.8 Debt Service/GNP (M) 6.0 8.3 8.2 g/ Bank Projection 1/ Exports of goods and services. Source: Treasury and Foreign Trade, External Debt DepartWent 63. Net debt, relative to GNP, increased by 5 percentage points between 1984 and 1985 and is projected to increase by a similar proportion by end-1986, with net debt then being almost equal to-balf of GNP. Turkey remains a high debt country. Net external liabilities, measured as ,gross liabilities less the liquid external assets of the Central Bank and resident commercial banks (external claims over which the Government has the ability to exercise some degree of control), rose even faster, from $19.4 billion at the end of 1984 to $26.3 billion at the end of June 1986. Recent developments in the debt profile are a major cause for concern. Debt recording has improved considerably recently, but debt management could still be improved significantly. Public Finance 64. The total public finance sector consists of the Central Government, Local Governments, Extra Budgetary Funds (EBF) and State Economic Enterprises (SEE): each will be discussed briefly. 65. The Central Government balance (excluding transfers to SEEs) has deteriorated since the early 1980s. Even the record revenue increase in 1985, due to the introduction of the Value Added Tax (VAT), was more than offset by the growth of expenditures, particularly expenditures on investment, export-rebates and tax rebates to wage earners. The tax collecion problem has been a source of continuing concern and there have been systematic shortfalls in revenue forecasts. Fortunately, the well executed introduction of VAT in 1985, which replaced eight different taxes, has improved collection - 137 - Table 8: PUBLIC FINANCES 1980-1986; SUMARY TABLE 1980 1981 1982 1983 1984 1985 1986 CENTRAL GOVERNMENT Revenues (TL bin) 838 1,329 1,424 2,156 2,831 4,691 Expenditures (TL bin) a/ 1,073 1,503 1,575 2,533 3,731 5,313 STATE ECONOMIC ENTERPRISES Sales Revenues (TL b1n) 1,146 1,767 2,650 3,596 6,310 9,319 Current Exp. (TL bin) 1,184 1,000 2,640 3,756 5,962 8,768 Central Government (% of GNP) 0.6 0.9 -0.7 -3.6 -1.6 Local Governments (% of GNP) 0.4 0.4 - - 0.2 BBFS (% of GNP) 1.2 0.8 1.1 1.8 0.8 SEEs (1 of GNP) -9.5 -7.1 -8.0 -6.2 -6.2 TOTAL -7.3 -5.0 -7.6 -8.0 -6.8 a/ Includes transfers to SEEs. b/ Figures differ from Government estimates due to adjustments for losses incurred by SEEs on foreign exchange and to the inclusion of gross borrowings to refinance maturing 4-mestic debt. substantially; it yielded already TL 1,070 billion in its first year and the Government is confident it will increase to TL 1,600 billion in 1984, However, the Central Government deficit may well get worse, unless expenditures are curbed, which, possibly with a long postponed general round of real wage increases in 1987 for the public sector and elections in 1988 will be difficult to achieve. 66. Municipal accounts overall have not been in deficit so far in the eighties, but this may change. Having received a revenue boost in 1985 as a result of the Government's decentralization strategy, the municipalities are now able to engage in stepped-up spending, even through borrowing directly from abroad. Given the constitutional steps taken to implement this strategy, it is difficult for the Central Government to now turn around and try to control specific investment decisions at the municipal level. This need not be done, provided decentralization of investment decision, which is basically in line with the Government's broad policies, is accompanied by the decentralization of budgetary discipline, together with technical assistance if required. 67. A major source of higher investment spending have been the Extra Budgetary Funds (EBFs), which are financed through levies and a variety of earmarked taxes. They seem to have expanded again in 1986 and are expected to account for 33 percent of the increase in investment and constitute about 12 percent of total public fixed investment, up from 5 percent in 1985. About 60 percent of their investment financing is being directed towards the transportation and power sectors, whereas another 19 percent is being directed towards the manufacturing for the defence industry and about 10 percent to "other services". The Bank has on numerous occasions recommended, for a - 138 - variety of valid reasons, to reduce earmarking and to make EBFs part of the regular budget, but the Government has so far preferred to keep the flexibility EBFs provide. However, especially at a time when the Government is tempted to invest in lieu of a hesitant private sector, EBFs may make it so much harder to resist as developments in 1986 show. 68. Net earnings of State Economic Enterprises (SEEs) comprise an additional source of public savings. However, the profitability of SEEs has been a matter of concern and all SALs have addressed the issue. In the late 1970s, the aggregated losses of SEEs were about 4 percent of GNP. Their poor financial performance could partially be attributed to price controls, but on the other hand, they received various subsidies. Periodic conversion of outstanding debt into equity (the most recent such consolidation of debt having taken place in end-1983) as well as other indirect transfers, such as tax privileges and interest subsidies, obscured the actual financial performance of SEEs. 69. Under the January 1980 adjustment program, price controls on SEE commodities and services were lifted (with a few exceptions) and SEEs have been required to cover their operating costs and generate funds for their investment programs through price changes and efficiency improvements. Simultaneously, their access to concessionary funds has been restricted, budgetary transfers reduced with the ultimate aim of providing transfers solely for the few remaining subsidized activities, and, with effect from January 1985, exemptions from various taxes withdrawn. Although these were elements of a comprehensive package it would seem that the option of price increases has been executed with vigour, contributing substantially to inflation; it is less clear how far efficiency has progressed. 70. Substantial price increases resulted in a considerable improvement in the financial performance of SEEs, with a situation of chronic deficits changing into one of marginal but growing profits. By 1984, SEE profits (after taxes) amounted to 1.9 percent of GNP. In 1985, their profits are estimated at 1.8 percent of GNP. Nevertheless, their financing requirements are still large, and they are the major factor behind the public sector deficit, as is shown in the Summary Table above as well as in the next table. Table 9: OPERATIONS OF THE STATE ECONOMIC ENTERPRISES (in Percent of GNP) 1980 1981 1982 1983 1984 1985 Profit on Current Operations -0.9 -0.5 0.1 -1.4 1.9 1.8 Investment (excl. depreciation) 9.8 9.0 7.3 6.6 8.1 8.0 Total Financing Requirement 10.7 9.5 7.2 8.0 6.2 6.2 71. The Summary Table shows that on the overall public sector deficit hardly any progress has been made as it has basically remained at around 7.5 percent of GNP throughout the adjustment period; this level is inconsistent with a reasonable rate of inflation. The Government realizes, and has stated - 139 - so repeatedly, that to get inflation under control is one of the cornerstones of its adjustment program. while the persistent high level of inflation is, no doubt. one of the main reasons for the private sector's lackluster reaction so far. A significant reduction in inflation to levels of 15 percent or less is not likely to be achieved unless the overall public sector deficit is reduced significantly to, say, half its present value. Prices 72. After the first oil crisis in 1974 inflation went up and stayed at about 20 percent on average. However, in the 1979-80 crisis inflation went up to just over a hundred percent per year, was brought down to 31 percent in 1983, shot up to 52 percent in 1984 and is estimated to be somewhere just below 30 percent in 1986. As indicated above it may not come down much further unless internal balances are restored by bringing public sector spending under control, primarily through showing more confidence in the private sector, and by curbing domestic consumption. 73. There are several downstream causes why inflation remains so high. One is the price increases of the SEEs granted them to lower their burden on the Government's budget; another to the various new instruments created in the financial sector which tend to increase money velocity, while most analyses concentrate on the monetary mass alone, which, of course, overlooks part of the picture. A major factor is also the high cost of the financial system in general and of the cost of capital in particular. At the end of the 1970s real interest rates were highly negative in Turkey and were estimated at -23 percent in 1978. However, internal and external balances were restored during 1981 and 1982 by arresting appreciation of the exchange rate and increasing the interest rate. The real interest rate turned positive and was some 12 percent on average in 1981. As a result of large budget deficits real interest rates climbed to 21 percent in 1982 and have remained high ever since, suggesting that too much of the burden of external adjustment was shifted to the private sector, for, to compensate for deeper fiscal deficits higher private surpluses are required. Higher real interest rates is the instrument by which those surpluses are generated. 74. It is in this context of concern that the growth of Treasury borrowings has been very rapid over the past few years. In mid-1986, with an estimated inflation of some 30 percent the Treasury could borrow at an interest rate of nearly 50 percent. Issuing domestic debt at such a high interest cost will allow lower money growth but at the cost of future budget deterioration, as debt service will grow explosively at real interest rates staying so far above the real growth rate of the economy. A debt strategy that so clearly sacrifices future budget balance is "ily to fuel inflationary expectations even if favorable external shocks allow a temporary decline in the rate of increase in prices; this in turn will keep nominal (and hence "ex-post" real) interest rates high, thus fueling a self fulfilling prophesy cycle of high interest rates, high public sector debt service, increasing budget deficits, high inflationary expectations and finally back to high interest rates. There is another drawback as well: the interest income from treasury bonds is tax free and to match the net return for the lender other borrowers, the private sector, has to offer rates approaching 70 percent which is not the best incentive to encourage the private sector to invest. - 140 - Recent Measures 75. The authorities, taking into account the continuing nature of the structural adjustment program, introduced several measures in 1986 to slow the growth of public spending and imports as follovs: (a) adoption of stricter controls by the State Planning Organization (Sla) over the issuance of investment certificates to public entities, such as the SEEs and municipalities; (b) the Treasury will no longer authorize municipalities to borrow abroad; (c) municipal investments exceeding TL 200 million now require SPO's prior approval; (d) the share of central government revenue to be transferred to the municipalities was limited at 13 percent for 1986, delaying a planned rise in such transfers; (e) only 40-50 percent of annual budget appropriations for government departments and SEEs will be made available at the beginning of the year starting in 1987; (f) appropriations for 1986 were frozen at 92 percent of original allocations, i.e., reduced by 8 percent across the board; and (g) the cash deposit requirements for imports was raised from 3 to 9 percent of the value of foreign exchange required to import the item. Assessment 76. Impressive progress has been made during 1980-85 in executing a stabilization program, implementing in parallel a structural adjustment process and concurrently turn a GNP decline of 1 percent in 1980 into a GNP growth of 6 percent a year on average in 1984-86; no small achievement, especially considering the unfavorable and sometimes even rather hostile international economic climate of the eighties. In spite of some backsliding, a measure of monetary and fiscal discipline has undeniably been introduced in the management of the economy. Moreover, introducing a realistic and flexible exchange rate and positive interest rates as well as liberalizing trade were the right thing to do to arrive at a proper allocation of resources in which the country's comparative advantages can be fully exploited. 77. National accounts data show that, after a decline in 1980 and 1981 when the stabilization effort was most severe, consumption has been allowed to grow rather strongly, especially private consumption, at the expense of savings. If at the same time investment is maintained at a high level, the shortfall between investment and savings has to be financed from abroad. Most Bank documents on Turkey strongly emphasize the need to cut public investment in order to obtain fiscal balance, as well as to match better the level of savings. Where the balance between consumption and savings should be struck is much less clearly articulated. The macro-data indicate that consumption standards have been kept fairly high, notwithstanding the stabilization and adjustment program; and it may be that Turkey has been able to achieve a balance of these difficult goals. On the other hand, the consumption patterns may have prolonged the need for stabilization. 78. This observation, of course, does not mean that the discussion on rationalizing investments is not useful. It has rightly been an important topic in most Bank economic reports. The discussion in Turkey, however, tends to put more emphasis on whether an investment is public or private, rather than whether it is efficient or inefficient, which, of course, is the more relevant distinction. Nevertheless, it would help indeed to get clearer statements from the Government at an appropriate time, on which projects/enterprises should not be in the public sector. - 141 - 79. As indicated before, achievements in growth were bought at a price: foreign And domestic debt Lave gone up again substantially in recent years and inflAtion has remained uncomfortably high; even to the extent that it may be endangering the Government's very growth model in which an active and efficient private sector is given such a crucial role. However, for the private sector high inflation in general and very high real interest rates in particular form a deterrent to invest. On the other hand, it is hardly surprising that, seven years into the stabilization and adjustment process, the impatience to achieve high economic growth would increase. However, it is important that policy makers resist such pressures until inflation is more convincingly defeated. There is undeniably backsliding going on in the economy at present and, together with the prospects of elections in 1988 an atmosphere of uncertainty is generated, which is detrimental to the private investment outlook. Under these circumstances it is imperative that the Government give the right impetus to regain momentum. Ad-hoc measures cannot change the private sector's perception from an uncertain feeling to one of direction, purpose and continuity in policies. Also, with changes so fundamental having taken and still taking place, time is required for prices, institutions, markets, regulations, laws, decrees and attitudes to adjust to the new signals. By keeping the main signals pointing in the right direction expected results will come, provided authorities allow the process sufficient time and avoid preempting the role envisaged for the private sector. V. AN EVALUATION OF SAL DESIGNS 80. In the Bank report: "Structural AdjusLment Lending: A First Review of Experience", of September 1986, an assessment of performance under SALa I-III gives Turkey the highest rating (a three on a scale of three), while it comments: "Overall progress was substantial in areas designated for reform. The Government demonstrated a strong commitment to the program and fulfilled all major undertakings." There was a delay in the implementation of only one covenant, requiring the passing of legislation concerning the operations of SEEs. As a result, the second tranche of SAL III was held up but the loan was still disbursed within a year. Passing legislation often takes longer than anticipated and the designers of SAL IV and V have learned from experience and succeeded in avoiding such pitfalls. 81. One of the conclusions of the SAL Review mentioned above is that quantitative targets for economic outcomes should be avoided as part of the conditionality. In Turkey the adjustment program is very much a product of Turkey itself, as it should be, and each year a substantial part of the program presented in the SDP was already under implementation, rather than holding out a promise for action sometime in the future. Therefore, in many instances a broad definition of conditionality could suffice and in fact was the only sensible way to proceed. To quote the President of the World Bank from "Finance and Development" of December 1986: "I think it is fairly accurate to say that, in a real sense, conditionality has to originate in the borrowing country if it is to be successful over a longer period of time....That's one of the realities about conditionality that very few people seem to understand. It is not something that can be imposed from outside for a long period of time. Therefore, conditionality that goes beyond the moment of crisis has to be totally acceptable within the country which is seeking to rec&pture growth through reform." - 142 - 82. In SAL IV and V designers have avoided the temptation to load the operation with a multitude of complex conditions, which could not realistically be expected to be fully implemented. The designs also took into account the fact that the sequencing, the synchronisation, as well as the interdependence and anticipated impact of the various measures contained in reform packages will, if many factors are involved, grow exponentially in complexity and become totally unmanageable and unmonitorible. The approach embodied in the economic reform program since 1980 that the Government should concentrate on the basics and leave the bulk of what has to be done to market forces, should also be the guiding principle when designing SALs. Straight-forward and transparent designs minimize the risk of non-performance for non-relevant reasons. It may be desirable to consciously aim at an "ambitious" goal, which is a little above current judgements on feasibility, but not so high as to be unrealistic. Government officials may at times even find this helpful in internal discussions with the ultimate decision makers, the country's politicians. The correct balance of "risk-taking" in the sense of trying to "stretch" the capacity of governments for policy reform in the design of programs is and will remain a matter of judgement. In the final analysis SAL design, indeed, is very much a matter of judgement. However, the judgement should be informed and the subjectivity objectified. In the case of SALs IV and V for Turkey this was basically done through an intensive economic and sector work program which will be discussed later. Although the outputs of the ESW were made available to the Government, it is felt that more training and assistance could have been given to Turkey in the areas of planning and macro-economic management, for instance. In designing and implementing SAL programs there has to be mutual trust in each other's intentions and integrity, and it requires high skills and understanding to maintain this over a period of time as long as a structural adjustment process may take. To conduct the dialogue at a high level of understanding and insight helps in this respect. For this, knowing the facts and appreciating the interdependencies are a must. 83. On several occasions both parties showed restraint in their actions or reactions when conditions where not met as planned. In each and every case the Government was moving into the right direction, but it had to convince the Bank that more time was needed. For example, the intermediate steps to phase out fertilizer subsidies over a period of three years gave rise to much discussion, but the final goal was reached earlier than planned. Second, partly because inflation stayed at a higher level than expected the Government found it impossible to have all preferential real interest rates positive at the end of 1984, although substantial progress was made as noted in the documentation for the Financial Sector Adjustment Loan. It was mutually agreed to monitor developments carefully and to postpone the target date of positive real interest rates for the productive sectors to the end of 1986, which was achieved. Third, the introduction of the VAT, replacing eight other taxes, was originally announced by the Government in its SDP for 1982. It proved to be a much more complex matter than originally anticipated and the Bank agreed that more time was --eded. When the VAT was introduced in January 1985, several experts feared that the complex operation had been inadequately prepared. In the event the introduction of VAT proceeded smoothly. Finally, in 1983 the Bank accepted the Government's %rocision to postpone release of the Fifth Five Year Plan by one year as a sensible decision. Designers generally -143 - resisted loading follow-up SALs with those items which had caused delays in previous ones, but preferred to monitor the issues causing such delays, recognizing that the reform program was basically on track. 84. SAL IV was put together in late 1982/early 1983, as the fourth in a series of structural adjustment loans to Turkey. The three previous SALs had been approved in each of the previous years and there was good reason to plan on a steady continuation for SAL IV and SAL V. The designers did grapple with the question of whether it might have been better to postpone SAL IV for a while. There were a number of uncertainties to consider. Mr. Turgut Ozal, the prinepal architect of the January 1980 New Economic Policies, had left the Government in July 1982. The 1983 SDP was signed by the new Minister. The Cabinet reshuffle, the electioneering thereafter and the change from a military to an Ozal-led civilian Government in December 1983, did not make for clear direction in the economy. It was also decided, probably by mid-1983, to postpone the release of the Fifth Plan, in order to keep the medium-term options open for the new Government. The SAL III second tranche review identified some areas of concern in early 1983. Also, the Government was struggling to pass SEE legislation on time and, as the year progressed, there were some early indications that the SEE Decree might turn out to be less comprehensive than had been expected. One could argue that a six months postponement of the tranche release and the preparation of SAL IV would have allowed more time for the economy and political environment to settle down. However, the SAL program had developed its own momentum and the decision to postpone the Plan had been made with good reason (to allow the new Government time to formulate its own Plan). Moreover, the IMF reached agreement on a one year Standby in June 1983 which helped to continue the momentum. The rather great detail in description and quantification of SAL IV seems to reflect the uncertainty at the time. The decision to not interrupt the program at that juncture was probably the correct call. 85. There is reason to ask the question about timing again with regard to SAL V. A January 1984 pre-appraisal mission reported a loss of momentum in performance in 1983. Bad weather (growth in agriculture turned out to have been -0.1 percent, compared to 6.4 percent in 1982) and adverse price developments on the external front played a role, but slippages in the monetary program and Government inertia related to elections were identified as the root causes. However, the new program announced by Mr. Ozal and his team in December of 1983 was very encouraging. It reconfirmed the reform strategy of January 1980 and showed the political will and determination to go ahead. The new policies, summarized earlier in the report, meant a leap forward in making the economy more outward oriented and responsive to free market forces. At the request of the Government the current IMF Standby was replaced by another one year agreement in April 1984. SAL V was appraised in March 1984, and a number of new features were included (such as financial sector issues) while greater specification was provided for in areas such as the SEE framework. Since the new Government had only recently taken office, a case could have been made for a postponement by about 9-12 months which would have allowed more time to judge the appropriateness of policies. However, without the continuity of the SAL program, one could argue that 1984-85 performance might not have been as good as was achieved. - 144 - 86. In sunary, SAL IV and V were well designed. They responded well to the needs of the country's program, which in turn was a very worthwhile program to support. Both SALs reflect the division of labor whereby the IMF concentrated on the stabilization of the economy, which is a prerequisite for a return to a sustainable growth path, with the Bank focussing on structural adjustment efforts with growth. The successive SALs show a deepening of the understanding of the issues; and the SALs as well as the tranche reviews demonstrate a maturing of a fruitful relationship. The gradually intensifying preparation for the post-SAL era was appropriate and conveys vision. VI. IMPLEMENTATION OF SAL PROGRAMS General 87. This Chapter will evaluate in some detail progress ma4e under both SALs on issues and areas in the structural adjustment program on which the Government had agreed the Bank would concentrate its participation. Performance will be reviewed in relation to the specific tranche release conditions, which were formulated as quoted below. Specific attention will be given to the financial sector and to the Fifth Plan as these were given dominance in SAL IV and V, respectively. 88. For the tranche release of SAL IV the Borrower and the Bank would review: (i) the adequacy of progress in the development of the Borrower's 1984 Annual Program; (1H) the adequacy of progress towards reaching the Borrower's objectives of: (a) rationalizing its public investment program; and (b) concentrating resources on high priority projects in such program; (iii) the adequacy of progress on import liberalization; and (iv) the adequacy of progress towards reduction in the level of the financial transactions tax. 89. It was agreed that the tranche review for SAL V would look into the following: (i) completion of the preparation of a Fifth Five Year Plan compatible, in terms of its strategy and its targets, with the Program; (ii) actions to ensure adequate progress in undertaking a review of the Borrower's public investment program and in carrying out the recomendations of said review; (iii) actions to ensure adequate progress in the elimination of quantitative restrictions and in the rationalization of the tariff structure; (iv) actions to ensure adequate progress based on the agreed plan of action for the installation of a computerized debt management system; (v) actions to ensure adequate progress towards achieving the goals identified in the Program for the reform of the financial sector; and (vi) actions to ensure adequate progress in the preparation of an energy action plan. 90. The agreements have, on purpose, been drafted in rather general terms, dubbed by some as "loose wording", which is missing the point. The careful formulation has to serve and satisfy a wide spectrum of audiences and purposes. The agreement has to carry not only some top Government technicians/negotiators, but also the politicians and as many working level technicians as possible. Turkey has a democratic system, ruling by majority, -145 - which does not mean that everybody agrees with the ideas of the Government. And even among those who agree there are many shades of agreement with the Government's economic program or parts thereof. Even with the Government's strong support in general, some issues remain controversial and the greatest of care should be observed as to how such issues are being integrated into the design of a SAL; the timing of action on such issues is invariably of the essence and the judgement of the Government should be taken very seriously. Furthermore, the tranche release conditions from part of the Loan Agreement, which is published in its entirety in the official Gazette; hence the delicacy of stating conditionality that soon becomes public material in a country with an open press. On the other hand, while the Bank is adjusting gradually to the role it must play in adjustment programs, a certain amount of "substantive conditionality" is appropriate. A guiding principle in designing the review criteria has been the view that formulation of the conditionality should provide room for "interpretation and judgement". However, this arrangement has to work both ways; it sometimes helps to avoid declaring the Government in violation of a particular condition if the country is generally on the right track, but has advanced less than hoped. On the other hand it did give the Bank the opportunity to insist or to ask for more than could be argued had it strictly been written into the agreement. Table 8: COMPARING THE 1983 SDP TARGETS WITH THE 1983 REALIZATIONS 1983 Realization SDP 1983 Over Under Targets Performing Performing GNP growth (2) 4.8 3.3 Domestic Demand Growth (% of GNP) 4.3 4.7 Foreign Balance Contrib. (% of GNP) 0.5 -1.3 Private Investment (% of GNP) 7.0 4.7 Public Investment in Agr. (2 of GNP) 11.0 9.8 Public Inv. in thanufac. ( of GNP) 16.3 16.0 Growth in Agriculture (M) 3.4 -0.1 Growth in Industry (%) 5.5 8.0 Inflation Increase (M) 20.0 30.5 WPI 31.4 CPI Export Growth ($ billion) 6.8 5.9 Import Growth (M) 12.8 4.4 Growth in Workers' Remitt. ($ billion) slight drop of increase 0.6 Current Account Balance ($ billion) 0.9 1.9 Central Budget: Expend. (TL billion) 2,265 2,533 Expend. (% of GNP) 20.8 22.1 Revenue (Z of GNP) 20.2 18.8 Balance (2 of GNP) -0.6 -3.3 SEE: Profits (2 of GNP) 0.6 0.3 Investments (2 of GNP) 6.5 8.0 Financial Requirements (2 of GNP) 5.9 7.9 - 146 - The "Base Year" for SALs IV and V 91. The 1983 SDP became the base for SAL IV. That it was a year of slippages has already been mentioned. How bad it was is shown by the following table. The strong showing of the Motherland Party in the 1983 elections and the affirmative and decisive action taken by the new Government in December 1983 put the adjustment effort back on track, but 1983 had clearly shown how fragile the economy still was. When these data became available by mid-1984 it was too late to affect either the tranche release for SAL III or the design of SAL IV much. And even if results would have been available earlier it probably would not have made a difference, because first "adequacy of progress" in its Annual Program was not a condition for tranche release under SAL III; and second, the strong actions by the new Government had restored confidence and things were moving again. The Fifth Plan 92. Because the changes in policy in January 1980 were so fundamental, both the Government and the Bank agreed that the structural adjustment process had to be placed within the boundaries of a new medium-term framework, and that the best mechanism for doing so would be the Fifth Plan. Its publication was eagerly awaited and both SALs refer to it. Why its release was postponed has been mentioned already. In an effort to bridge the gap, the Bank developed medium term projections in the 1983 CEM, using both the RMSM model and the Computable General Equilibrium (CGE) model. These were extensively discussed with the Government. The Bank's forecasting model is quite weak in the fiscal and domestic budget departmen - priority areas for scrutiny in Turkey - but strong with regard to the external balance. Assuming export sustainability and "fiscal discipline", the debt service ratio was forecast to stay within the 16-20 percent range; the current account deficit would drop to about $500 million by 1985; and growth would average 5 percent per year for 1982-85 (stabilisation period), and go up to 6.1 percent per year for 1985-90 (growth period). There was probably hope that this contribution from the Bank to the discussion would strengthen the hand of the "moderates" in the Government as far as growth is concerned, and maybe it did, but the Plan which was approved by the National Assembly in July 1984 had a growth target of 6.3 percent per year anyway.l/ 93. SAL V was negotiated three months prior to the publication of the Fifth Plan. In its 1984 SDP dated May 1, 1984, the Government reiterated its intention to prepare the Fifth Plan to be compatible with the aims of the structural adjustment process. The Plan targets do, however, reflect certain differences with the 1984 SDP. These include: (i) the ratio of public fixed investment growth over private fixed investment growth is higher in the Plan 1/ In the Fourth Plan (1979-83), which was published in a time of economic crisis, the Government forecast a growth of 8.2 percent per year, indicating its belief that the crisis was just an unfortunate dip which would pass. The Bank's 1979 Plan evaluation mission found a growth of 6.7 percent feasible, although even then the external situation would remain extremely difficult, if not critical in the medium-term, the Bank's report said. The Bank's forecast was obviously very optimistic under the circumstances, but possibly advanced in an effort to create a climate in which the dialogue could be continued. - 147 - (1/1.6) than in the SDP (1/2); (ii) the ratio of public to private investment is higher in the Plan (56/4 in the last year of the Plan) than in the SDP (53/47); (111) the borrowing pattern between short-term and medium- and long-term debt is tilted towards the short-term in the Plan; and (iv) annual CNP growth is 6.3 percent in the Plan versus 6.0 percent in the SDP. 94. A Bank economic mission which visited Turkey in September 1984 to evaluate the Plan, established that a number of assumptions in the Plan appeared optimistic in view of both past performance and future expectations, while others fell within the range of the SDP but, when considered in the total context of the package, reflected ambitious targets. This is further specified and illustrated in great detail in the Bank's Plan Review Report (Report No. 5418-TU) which was published in July 1985 (in three volumes). 95. The 1985 Plan Review Report concluded that: (i) the targeted average annual rate of growth of GNP at 6.3 percent was likely to be ambitious; (ii) the requirements of stabilization would tend to retard a strong revival of private investment for at least another couple of years; and (iii) the Plan would strain domestic resources and the external balance. It considered it imprudent at this stage to threaten the gains from structural adjustment by giving the wrong signals to Government agencies, especially in terms of higher growth targets, which might engender additional financial commitments. Of equal importance, the Plan's scenario regarding the acceleration of growth which, being predicated on high rates of private investment, did not agree with the fact that private investors had been reluctant to invest, other than at preferential rates, in the face of higher than expected inflationary pressures. Another important conclusion of the report was that the Fifth Plan did not provide the type of medium-term framework one had been waiting for. The aim had been to have a document, dynamic and useful for the policy makers rather than satisfying only legal requirements. So, it was unfortunate that the document came so late and that, when it came, it failed to be the working document for guiding the future course of the structural adjustment process.1/ 96. The Government cleared the rather critical Plan Review Report and it was published in grey cover in July 1985. It is regrettable that the excellent analysis in the Plan Review Report did not play a more prominent role in the second tranche review of SAL V. This, together with the tranche review mission's judgement that "the overall performance is satisfactory but not uniformly so" should in all probability have led to making a case for postponing the tranche release. Moreover, in March 1985 and again in May the IMF failed to reach agreement with the Turkish authorities on a new one-year 1/ In the meantime SPO is considering to revamp the planning system by making it more indicative and by preparing a strategy paper to cover a 6-10 year period, together with a three year rolling quantified framework, to be revised annually in the context of the preparation of the annual program along lines recommended by the Bank. The PCR mission in November 1986, however, did not get the impression that the idea was pushed with much vigour or enthusiasm within SPO. - 148 - Standby, originally intended to start in April 1985 when the current one-year arrangement would expire. Several reasons have been advanced for the decision eventually taken by the Government not to try further to agree on another Standby, but in the event it would simply have been virtually impossible to meet If requirements. Another matter with the Bank - the undertaking to reach positive real interest rates for preferential credit by the end of 1986 - was settled by allowing Turkey till the end of 1986 to reach that point. This cleared the way for approval by the Board in June 1985 of a $300 million Agriculture Sector Loan, a factor which was prominently behind the decision to release the second tranche of SAL V just before that. This sequence, judged against economic performance in late 1985 and early 1986 (particularly from the viewpoint of excessive public spending), seems to strengthen the case for the proposition advanced earlier, i.e., that a slower pace in the series of SAL9, and in particular postponing SAL V, might have been the better decision. Public Investment 97. This report has described already how, in its search for growth and consistent with State philosophy, public investment surged in the seventies. This undisciplined proliferation of public sector activity had largely been brought under control during the early years of the adjustment program. Resources were concentrated on a smaller number of major projects; priority projects and later-on priority sectors were identified in line with other macro economic policies. The public investment program has been successful at the macro level in reducing the size of the programs; a number of large projects, the economic viability of which had not been established, were dropped, deferred for further study or substantially reduced in scope. The implementation time of projects has been shortened and the program has been redirected from manufacturing towards productive activities supporting infrastructure. The Bank's involvement in the priority sectors, especially in agriculture and energy through the emphasis on sectoral issues in SALs IV and V, but also through the rest of the lending program, has greatly contributed to bring the public sector investment program in line with other macro policies in the program. This improvement is expected to continue through a number of sectoral structural adjustment loans planned and under implementation. 98. The Government's imaginative initiaLive to have the private and public sector cooperate in the so-called Build-Operate-and- Turnover (BOAT) schemes, especially in mining and power is strongly endorsed. As part of the structural adjustment program the Government has taken a number of far-reaching measures to liberalize foreign private investment but the response so far has been disappointing and the direct private investment activity from abroad has remained low for a country the size and importance of Turkey. Government guarantees for foreign private involvement in the BOAT schemes continue to be a problem and risk to jeopardize investments in the power sector as the program in this sector is basically predicated on the success of the BOAT schemes. 99. The Bank is in many ways involved in the country's public investment program; through its lending program and through the increased emphasis on priority sectors in the SALs as the program progressed. As part of the ESW program reviews are carried out regularly. They help monitor the adjustment - 149 - process. For instance, late 1986 an Agro-industries Sector Review mission visited Turkey and found that the cost of the Investment Incentive System may have gone up to as high as 20-25 percent of the Central Government budget, with a direct cost to the Treasury in cash grants of an estimated $1 billion in 1985. The benefits of the system are not quantified and its effectiveness in producing the desired result!, uncertain. Alsc, incentives to create private economic activities in baccward areas seems to have led, In a number of instances, to less efficient private firms, which, instead of being foreclosed, are beyond the reach of creditors and are being continued by the Government in the public domain. 100. The Bank carried out a Public Sector Investment Review in 1981, an exercise repeated in 1985; the draft report was discussed with the Government in November 1986 and the final report is expected to be issued in Grey Cover in March. The report forms a basis for a continued dialogue on public investment issues and for further Bank involvement in rationalizing Turkey's public investment program. A major recommendation of the report is to . considerably strengthen planning and monitoring capabilities and capacity at the level of the technical ministries and autonomous agencies. In this model SPO would develop its "upstream" activities and emphasize much more its role : in setting macro targets, coordinating the sectors and generally ensuring that * the investment program is formulated and implemented in a way which is internally consistent as well as consistent with the country's structural adjustment program. Comparatively little progress has been made in these areas during the implementation of the five SALs as apparently strong forces are involved, which can only be reoriented gradually and over a considerable period of time. However, progress can be made in this area given the right conditions, which is the case in Turkey since: (i) there is a substantial measure of agreement with the borrower on a macroeconomic framework, comprising a feasible growth scenario and the measures required to achieve it; (ii) comprehensive policy and/or institutional reforms in a specific sector are needed to improve the efficiency of resource use; and (iii) government commitment to a sector adjustment program is evident. Import Liberalization 101. The import liberalization measures started in 1981 with the abolition of the quota list and the liberalization of imports equivalent to about 12 percent of total merchandise imports in 1980. Over the next four years there was further, though relatively less, liberalization of imports. The 1984 Import Regime advanced import liberalization more than any agreement so far had asked the Government to do. It represents a major structural departure from previous regimes. While in the past the import regime specifically emmerated the goods that could be freely imported and the ones whose import required a license, thus virtually prohibiting imports of all the goods not included on the lists, the new regime lists all prohibited and licensed imports and igso facto frees the imports of all other goods. In 1983, the number of imports requiring a license (Liberalized List II) totalled 821 items; in the 1984 regime these were reduced to 369 items. Total import authorization of items on List II amounted to $5.2 billion in 1983; if the 1984 Regime had been in force, import authorization would have been needed for only about $1.2 billion worth of these imports. Liberalization has thus been extended to cover approximately three quarters of the previously licensed items. - 150 - 102. Simultaneously, the Government also announced a general reduction in tariffs. Out of about 500 products whose tariffs were changed, more than 80 percent had their tariff rates reduced. In designing the new tariff structure, the Government followed the general principle that tariffs on basic materials should be in the 0-15 percent range, tariffs on intermediate and semi-finished goods between 10 and 30 percent, and tariffs on finished products between 30 and 40 percent. Further, in addition to the prohibited 1/ and licensed lists, the Government also defined a new "levy" list. Goods on this list, which may either require a license or be freely importable, are mostly luxury goode that would carry a surchage. All the revenues realized from the levies would be transferred to a fund for the construction of low-cost housing. Finally, the new regime also provided for considerable further simplification of the administration procedures applicable to imports, following the centralization of authority on all matters dealing with imports in the newly-formed Undersecretariat of the Treasury and Foreign Trade under the Deputy Prime Minister. 103. Measured against 1983 imports, the liberalization in 1984 is estimated to have removed quantitative restrictions (QRs) from 63 percent of imports previously subject to licensing; the value of liberalized imports represented about 45 percent of total merchandise imports in 1983. Thereafter, there has not been a significant reduction in licensed imports. By 1985, imports requiring licenses accounted for slightly less than 25 percent of total 1984 imports and the number of import items on the prohibited list was reduced to 127, almost half of the number in 1984. About half of the prohibited imports are consumer goods. The 1985 Import Regime reflected additional movements in the direction of liberalizing foreign trade, although its principal theme may be characterized as one of consolidation and rationalization of past gains rather than the introduction of new measures. Export Incentives 104. The excellent performance of exports early on in the program has been mentioned already. Convinced of the structural nature of the improvement no criteria about export performance were included in the tranche review for SALs IV and V. However, exports have not been able to keep up the earlier pace and, because export earnings are so crucial for the overall result of the program, they will need again the Government's attention. The existing incentive scheme was introduced in 1980; exporters may qualify for as many as three schemes, which are quite cumbersome, flawed in design and difficult to administer. Some incentives are non-neutral and controversial and have invoked counter-measures by some of Turkey's trade partners. Basically most export incentives will be gradually eliminated as Turkey has clearly chosen to encourage exports primarily through its exchange rates which has been adjusted regularly to maintain or improve its competitive position. This is evidenced by the fact that the trade-weighted real exchange rate, after remaining stable during 1985, has again depreciated significantly in the first half of 1986 (by about 8 percent), while direct export incentives declined. 1/ The prohibited list comprises 207 items, including some agricultural goods, whose imports were also prohibited in the past. - 151 - External Debt Management 105. The Bank and the IMF have both assisted Turkey to improve its external debt management. Phase I of the installation of a computerized debt recording system, consisting of the design of a system of information flows and the development of input-output formats, was completed on schedule in 1984 and the final report has been published. Phase II, comprising the installation and testing of the computerized debt system was completed in 1985. A notable feature of the system is that it was designed primarily by Turkish analysts, with guidanace from a consultant financed by the Fund. The introduction of the computerized External Debt Data Base (EDDB) is a major step in debt management and it has improved the monitoring considerably. The development of the system into a tool of management by carrying out the necessary analysis, and for which the soft-ware has been developed, is a stage which still has to be implemented; it needs an organizational consolidation of external debt management, which, at the end of 1986, is still too dispersed. SEE Reform 106. The State Economic Enterprises (SEEs) have been important economic institutions in Turkey since the 1920s. They were a major instrument in the import substitution based industrialization that developed the manufacturing sector, which was able to shift toward exporting in the 1980s. In 1980, however, it was generally agreed that the SEEs were too inefficient. In addition to production of the usual services and infrastructure provided by the public sector, about 40 percent of manufacturing output was provided by SEEs. Their aggregate losses in the late 1970s were 4 percent of GNP. Financing these losses was a major source of the budget deficit. Therefore the Government and the Bank agreed in 1980 that the size of the SEE sector should be reduced, SEE efficiency should be improved, and SEE activity should be redirected toward provision of the usual public goods. The SEEs were to be reoriented toward economic and financial objectives. 107. The reform program was aimed at exposing them to more rigorous external economic conditions and at generating internal reforms. They were largely cut off from direct financing by the Central Bank and their access to the budget was eliminated. The redirection of the public investment program toward infrastructure greatly slowed the growth of manufacturing SEEs. A series of internal reform actions: to freeze hiring, increase management efficiency and productivity, and eliminate credit subsidies, were implemented under SALs II-V. A major reform decree was passed in May 1983. The SEE reform so far has been mainly redirection of public investment, cutting access to public funds, and pricing to market conditions. In Chapter IV it has been shown how an operational loss in 1980 of 0.9 percent of GNP has been turned into a profit of 1.8 percent of GNP in 1985. However, the SEE financing requirements still remain high - about 4 percent of GNP in 1986. The reform has still a long way to go as far as SEEs are concerned. 108. The discussion about the subject has been substantially polluted by lumping SEEs (State Economic Enterprises, operating in competitive markets) and PEEs (Public Economic Enterprises, providing public services) together under the heading SEEs. The public enterprise is a phenomenon deeply entrenched in Turkish society and it is a very sensitive and sometimes even - 152 - emotional subject. Reform on this issue should be introduced step-by-step and with caution: it looks like at this stage efficiency should be pursued for all public enterprises, but with regard to privatization SEEs and PEEs are a completely different matter. It is not logical that privileges, which in a transition period would still be available for utilities (PEEs), would also be available for public cement plants or textile factories (SEEs). In other words, it seems prudent for the time being to only seek improvement of efficiency for PEEs. and leave them in the public domain, as they are in so. It would probably allow to pursue a much more vigorous course, including early privatization, for SEEs. 109. The Government's basic philosophy with regard to SEEs, that is to subject them to market prices, is a -nod one and consistent with its basic economic policies. It was, therefor logical to allow SEEs to adjust their prices to market conditions. Firms . h cannot make it under the new rules, however, should not be bailed out a: more by the Government like before. Another Government decision was to rivatize activities which do not have to be in the public sector. In this context the Public Participation Fund (PPF) was established in 1983. One of its responsibilities is to manage, on behalf of the Government, the privatization of individual SEEs. So far PPF has sold revenue-sharing bonds in the Bosphorus bridge and the Keban dam; has sold one textile mill under construction; and is on behalf of Sumerbank, in the process of selling another six unfinished plants. Further, the decree law of 1984 promoted a more decentralized legal structure for SEEs. whereby individual plants or groups of plants should be organized as legal companies. The same law authorized the Council of Ministers to decide on cases of liquidiation and divestiture and gave PPF the responsibility for implementing these decisions. As a consequence of the reform package both, SEE sector investments in manufacturing and subsidies to the total SEE sector, have been reduced in real terms and relative to GDP during 1984 and 1985. 110. These actions indicate the determination of the Government to make progress on this issue. Because of the impact of the sector on the entire economy, the Bank has considered the possibility of designing a comprehensive Public Enterprise Reform Sector Loan (PERL) as a logical extension of its early involvement in this area. Rowever, the sector adjustment loans (agriculture, financial, energy and transportation) as well as various projects aim for the same end to some extent. The Bank's comparative advantage in the SEE area may, in general, not be pronounced enough to t successful, with all the risks this may entail for GOT-Bank relations in general. The Government's determination to continue to move on the SEEs should not be doubted; it may be preferable to assist the Government from the sideline with ad hoc technical and financial assistance if and when the Government judges it useful. Financial Sector Reform Ill. General. The interest rate, of cc 'se, is of prime importance as an instrument to mobilize and allocate finanC Al and capital resources. Very negative real interest rates in the seventies basically encouraged consumption and discouraged domestic savings, reason why in the end close to a third of investments had to be financed from abroad. Such a pattern could not be sustained and one of the first measures to stabilize the crisi* struck economy - 153 - was to lift deposit interest rates to real positive levels again. It had a dramatic impact on the ability of banks and other institutions to mobilise domestic resources. Due to negative returns on deposits in the 1970s the depth of the financial system as measured by the ratio M2/GDP fell from 26.1 percent in 1970 to 15.8 percent in 1980. True to the statement in the first SDP in 1980 the authorities raised deposit interest rates sharply in July 1980 and as they were maintained at positive levels in real terms thereafter, the ratio of M2 to GDP increased to about 23 percent in 1985. 112. Large differentials in the yields offered on the various types of financial instruments have caused significant changes in the composition of Ms during the last few years. Time and savings deposits, yielding approximately 50 percent annually, have grown considerably at the expense of non-interest or low-interest bearing instruments like currency and demand deposits. This has pushed up the average interest cost of bank deposits from about 30 percent in 1982 to about 40 percent in 1985, which in turn has caused a significant increase in the lending rate of banks. Moreover, a combination of high and volatile inflation since 1980 with a system of fixed deposit interest rates has led to short term deposits and, consequently, the banks' inability to lend Turkish lira funds on a long-term basis. 113. The impact of the reform of the Turkish financial system has been particularly strong on the commercial banks. Entry of new banks, including foreign banks, has increased competition and helped to improve the efficiency of intermediation as well as profitability. At the end of 1985, there were 50 commercial banks operating in Turkey. The 12 public sector banks accounted for 52 percent of total assets as against 44 percent for the 19 private banks and 4 percent for the group of 19 foreign banks. However, interlocking ownership between commercial banks and major industrial conglomerates is continuing to be a factor inhibiting competition among banks. 114. Another consequence of the change in the composition of monetary assets in Turkey has been a change in the manner the Government has financed the budget deficit. As demand declined for currency holdings (which earn no interest), the ability of the Central Bank to expand its advances to the Treasury without fueling inflation declined as well. This consideration led the Government to increase its reliance on bond issues as a means of financing the budget deficit. The rising volumes of Government bond issues purchased by comamercial banks have been partly substituting for credits to the private sector. The share of the private sector in total domestic credit extended by the financial sector reached about 20 percent of GDP in 1984 but declined to some 15 percent thereafter, while the share of the public sector went up from around 15 to 22 percent, trends which have continued in 1986. It has inevitably resulted in crowding out of the private sector from the domestic credit markets. The situation is likely to change only to the extent that the inflationary situation improves, to which a balanced budget would contribute greatly. 115. For securities a large but completely unregulated market had emerged in Turkey in the late 1970s as a result of the rigid deposit interest rate structure which produced highly negative real rates in a period of high inflation. Brokers acting as bankers, borrowed and lent money in their own names, but without being subject to reserve or liquidity requirements or - 154 - controls. Initially this line of business proved highly profitable, but the system collapsed in 1982 as brokers could not collect the loans they had made, and in the absence of a lender of last resort, could not honor their repurchase guarantees. The event helped the Government to focus its attention on the financial sector and SAL IV, designed around the 1983 SDP in late 1982 and early 1983, reflects this appropriately and adequately. The overall objective of the Government's financial sector reform effort was to deepen the financial system and to establish, over the medium term, an efficient and flexible system of resource mobilization which would respond rapidly to market forces and offer a wide variety of instruments to both borrowers and savers. Such a system would make resources available to borrowers at positive but reasonable real interest rates and would continue to grow both in real terms and in relation to GDP. 116. The Areas for Improvement. The issues to be addressed in the financial sector related essentially to improved resource mobilization and allocation, and called for measures covering financial policies, instruments and institutions. Financial policies have strong linkages to fiscal and monetary policy, interest rates and selective credit policies. Development of instruments includes measures to integrate the growth of Treasury bonds, commercial paper, and corporate bonds and equities. Institutional development includes strengthening of the banking system and expanding the role of the Central Bank in bank supervision, strengthening the accounting system, developing the capital market, and developing new institutions and services. It was clear from the beginning that the implementation of such a program would require time and coordinated action on a wide front. Therefore, it was decided for the remaining SAL period to progress on a number of issues, which, at the same time, could serve as preparatory work for a possible Financial Sector Adjustment Loan (FSAL). 117. In both SAL IV and V specific targets have been formulated and both second tranche release reviews contain conditions for the financial sector to be met, and they were. With the Bank's Financial Sector Report, and continued under the Fourth and Fifth Structural Adjustment Loans, emphasis began to be placed on financial sector reform. Some important reforms were launched - mainly in the structure and level of deposit interest rates, reductions in the financial transactions tax and the withholding tax on interest earnings, rationalization of preferential lending rates, and setting up of an auctioning system for Treasury Bills. Meanwhile, the Government time and again confirmed its policies and intentions in this sector and, although there have been minor slippages - never in direction but only in time - achievements have been impressive. 118. Achievements. On policy matters, deposit rates have been adjusted regularly to maintain positive real levels; the latest adjustments were made in March and October 1986 to reflect lower inflation rates, while interest on sight deposits, which had been kept at 5 percent per year, was increased to a level compatible with the other rates. Since the deregulation of lending rates on non-preferential credits in 1983, commercial bank rates on short term credit have exceeded 60 percent per year; including a bank commission, intermediation costs, taxes, surcharges, and a 10 percent contribution to an investment fund, the real interest rate is often over 40 percent per year, which is a reason for concern and has reduced the demand for domestic credit. - 155 - The Government is aware of the cost pressures and is lowering deposit rates whenever justified by lower inflation; it has also reduced the reser%e requirements imposed on commercial banks from 20 to 15 percent in 1986 and reduced the financial transaction tax since 1983 in steps from 15 to 10 and then to 3 and finally to 1 percent in 1986; bringing the cost of the banking system down through further rationalization and support to improve efficiency is a goal of the FSAL. To encourage greater use of foreign borrowings by the private sector, the Government, in July 1984, instituted a Foreign Exchange Risk Insurance Scheme (FERIS) for selective credit lines; the dialogue on this issue is continuing under FSAL. Furthermore, the proportion of preferential credit is estimated to have come down from 41.6 percent to 13.5 percent since 1983. Positive interest rates for all preferential credit to the productive sectors was achieved by the end of 1986. Finally, a system of floating interest rates was studied, as recommended (under SAL IV), but the Government decided against their introduction as it was concerned about a possible adverse impact which a system of indexing might have on its fight against inflation. 119. The Role of the Central Bank. The Government's intention is that the Central Bank should increasingly concentrate on the formulation and implementation of monetary policy and on the supervision of the banking sector. The former task requires clear goals of monetary policy and appropriate instruments of implementation, which will take time to sort out given the dominant role (at present) of Treasury in the economic triangle, Treasury, Plan and Central Bank. In the meantime SAL IV encouraged the Government to issue securities priced at competitive rates. It did so in 1984. Initially these securities were sold "on tap" through the Central Bank. In May 1985, a market pricing mechanism was introduced through weekly auctions of six-month, one-year and two-year securities. The introduction of the regular weekly auctions has significantly increased the supply of market-priced securities. The Government is currently considering actions to improve the functioning of the system and to diversify the maturities offered. The secondary market for Gorernment securities has also grown. Coamercial banks account for 85 percc't of the trading, and ten licensed firms for the rest. The growing secondary vrket should offer opportunities for introducing open market operations by he Central Bank as an effective tool for monetary policy purposes. 120. In the context of the task of the Central Bank's supervision a deposit insurance scheme covering funds held in commercial banks was introduced in 1983. The scheme presently provides coverage of prinicipal amounts up to TL3 million. The Government intends to review the scope of deposit insurance periodically and raise the ceiling as necessary to maintain the level of cnverage in real terms. Also, the standardization of accounts for commerctal banks as stipulated in SAL IV will be complied with in 1986. And a conditionality under SECAL - F is the implementation of the Banking Law requirement to submit annual balance sheets and profit and loss statements certified by external auditors. 121. The Central Bank is also policing the April 1985 Banking Law which has introduced certain restrictions on the operations of banks to deal with this and other problems. The restrictions include a 20:1 debt/equity ratio limit, a limit to the maximum exposure to any single customer of 10 percent of a bank's net worth, and limits on a bank's exposure to officials of the bank - 156 - and their relatives, or to companies in which the bank has an equity investment. The Banking Law also contains more stringent capitalisation requirements aimed at improving the soundness of the banks and their ability to weather abnormal loan losses. Although paid-in capital and reserves increased from 2.5 percent of total assets in 1980 to 6 percent in 1984, the ratio is still very low by international standards. Provisions for bad debts have been raised from 0.3 percent of assets in the late 1970s to an average of 0.6 percent in the period 1981-84. 122. The Capital Market Law enacted in March 1981 provided for a regulatory authority, the Capital Market Board (CMB), which was established in February 1982. It is responsible for the regulation and supervision of the primary and secondary markets, including inter alia the approval of public offerings of corporate securities. In response to the actions taken by the CMB, public offerings of equities have risen in the last three years and amounted to TL76 billion in 1985. The CMB "approves" public offerings, depending not just on the applicant's compliance with financial disclosure and other legal requirements, but also on a judgement of the investment merits of securities. This approach carries some potential risks, because it can be perceived as CMB having responsibility for losses suffered by investors who bought "approved" issues. The fact that the CMB's review is based on unaudited data adds to the concern. Therefore, the Government intends in the next phase, to be worked out under the FSAL, to reduce CMD interventions and to have it concentrate instead on enforcing disclosure and audit requirements for securities issues, with a view to ultimately enabling investors and intermediaries to assess the quality of securities on the basis of reliable information. It has been agreed that independent auditing of companies registering for public offerings or seeking listing in the new Stock Exchange will be introduced from 1987, in accordance with the same time frame as that proposed for the banks. This would be followed in due course by the introduction of compulsory external audit for securities intermediaries, other financial institutions and eventually all major corporations. The Stock Exchange in Istanbul opened in January 1986; its set-up was assisted by IFC, which is continuing to follow developments in an advisory capacity. 123. FSAL: Contents and Comments. In spite of the substantial progress made under SALs IV and V much was left to be done and to help the Government deal with it FSAL was developed. The FSAL loan of $300 million was approved by the Board in June 1986 and, having a disbursement pattern similar to the SALe, is scheduled to be quick disbursing in about a year and in two tranches, the second of $100 million after a performance review in February/March 1987. 124. The actions to be undertaken in FSAL are described in the Government's Statement of Financial Sector Policies - 1986. The expected results of the medium-term reform program can only be delineated in indicative terms, because as the financial system becomes more market responsive, targets set exogenously will become less useful. However, as the process of financial sector reform evolves, Turkey's financial system can be expected over the medium term to exhibit characteristics similar to those of more.developed financial systems of other medium-income countries, namely: (a) financial depth, measured by the percentage of Ma to GDP, increasing from the present 20-25 percent to 30-35 percent; (b) the intermediation margins of commercial banks declining to about 3-5 percent of assets from the present 7-10 percent; - 157 - (c) the share of long-term instruments in financing of investments increasing to about 7-10 percent from the present 3-4 percent; and (d) issues of corporate bonds increasing by about 50 percent, equity issues by about 15 percent and the volume of trading in the inter-bank market by about 30 percent, all in real terms. 125. In this regard it is worth mentioning tl-ee points. First, the banks are the main intermediaries providing investment iinance in Turkey. Their inefficiency causes a spread of lending over borrowing rates that is very large by international standards. This has been noted in several previous World Bank reports. With limited administrative and financial resources in Turkey, it may be more productive in the immediate future to concentrate on improving the banking system, instead of trying to develop more sophisticated capital markets. The stock and bond markets play a marginal role in financing investment in Turkey; diverting scarce resources into their development at this time may not be efficient. 126. Second, the target indicators mentioned above are ambitious and wide-ranging indeed. In fact, if "medium-term" would mean about five years, as it usually does, the targets look over-ambitious, if not unrealistic. It looks like some of the lessons presumably learned from SALs about the time it takes to change attitudes and institutions have not fully been taken into account by FSAL's designers. Furthermore, it is difficult to see how such long term objectives can be successfully pursued by a quick disbursing program loan, unless, of course, a number of follow-up FSALs would have been planned, which, looking at the future lending program, does not seem to be the case. To uee an analogy from the financial world: banks lending long term based on short-term borrowings get into trouble rather sooner than later. Also, loading Turkey with a number of far reaching institutional-improvement- seeking-loans over a relatively short period of time will soon start to backfire as neither the Government nor the Bank will be able to properly channel the forces unleashed or provide the follow-up required. 127. The FSAL is entirely built on the adjustment and iwprovement of forward or downstream linkages. This is extremely useful, but efficient functioning of the financial sector should be very closely linked with the macro-economic policies followed by the Government as part of its strategy for the next phases of its adjustment program. The challenge is essentially to achieve an appropriate balance between stabilization and adjustment with growth. The Government has undertaken to pursue monetary, fiscal and trade policies aimed at curbing the budget deficit, controling inflation and maintaining export growth through the revival of private investment. An efficient financial sector will greatly strengthen the Government's program in these three areas. In particular, a smoothly functioning financial sector is a crLcial requirement for the healthy expansion of the private sector. At the same time, financial sector reform can only be carried forward effectively within a framework of sound fiscal and monetary policies. However, the way FSAL has been set up seems to presume that stabilization at the macro economic level has sufficiently been established and the design, therefore, tends to neglect the continued importance and the need for vigilance vis-a-vis the backward or upstream linkages. As if to illustrate this point the Statement of Financial Sector Policies-1986 is rather silent on macro-economic linkages, demonstrating how difficult it can be to maintain a meaningful and leveraged - 158 - dialogue at the sector level, even though there is no other sector conceivably closer to macro economic issues than the financial sector. The last tranche of SAL V was disbursed in June 1985 and the macro policy dialogue has possibly been more voluntary since. Agriculture Sector Reform 128. General. Recent studies have shown that the potential for Turkey's agricultural sector through expansion of area or livestock numbers were largely exhausted by the mid-1970's and growth must now come primarily from increased productivity and changing the crop mixture to reflect better Turkey's comparative advantage at present. Increased productivity would require expansion in the irrigated area through improved efficiency of the implementing agencies (better planning, increased use of contractors), better extension and research programs, expanded imports of improved seeds and appropriate equipment, and increased availability of institutional credit (particularly for small and medium-scale farmers). 129. The demands of an export-oriented approach have led to a change of emphasis from food self-sufficiency to increased net agricultural contribution to the balance of trade. Thus imports of certain foods in which Turkey does not have a comparative advantage have been allowed. Increasing agricultural exports has involved appropriate pricing signals, maintenance of a competitive exchange rate, reduction of export regulations, and improved marketing. Other important issues in the sector include improved sectoral planning, continued progress in the reduction of subsidies, and reform of agricultural SEEs and marketing agencies. 130. Achievements. SALs II and III already emphasized efficiency improvements and SALs IV and V pressed on along more or less the same lines. During the SAL process significant reforms have taken place in the agriculture sector. The share of agriculture in the public investment program stood at 7 percent in 1980 and the Government's target was to increase it to at least 11 percent in 1983 and an average of at least 10 percent for the period of the Fifth Plan. The realization has been below the average so far with 9.8 percent in 1983 and 9.1 and 6.8 percent in 1984 and 1985 respectively, which is reason for concern. Eroject completion rates have increased significantly. In the irrigation subsector, for example, as against only 18,000 ha of new irrigation put into operation by DSI in 1981, 86,000 ha were put into operation in 1984. Moreover, several large project, the economic viability of which had not been established, have been dropped, deferred, or substantially reduced in scope. The system of commodity support prices has been replaced by one of protective floor prices, and the number of affected commodities reduced from 25 to 13. These floor prices have been brought broadly in line with international prices. Livestock and grain exports have been liberalized. Plant protection subsidies have been eliminated and fertilizer subsidies had been gradually reduced. 131. In spite of a relative decline of agriculture as a sector in the economy of Turkey it is still a priority sector and as such was singled out early in the SAL process for "downstream treatment". The main goal was to resume growth, which required an increase in agricultural productivity. Production growth was to come from more intensive use of cultivated area, - 159 - expanded multiple cropping, crop yield increases, and an increase in the productivity of livestock. The Government was not organized to intensify the use of its land or livestock resources. A reorganization, consolidation and regionalization of crop and animal husbandry research was required as well as an improved system to establish and monitor priorities for technology transfer and generation. Increased emphasis on product quality and the needs of the importing countries was also considered necessary. 132. Through the sixties and seventies, Turkey's agricultural policies were inward-looking, stressing food self-sufficiency through subsidized inputs and producer prices. This led to a relatively rapid growth of production, which, with a deteriorating overall macroeconomic situation, could not be maintained due in part to the strain placed upon the budget by the level of subsidization, and the competing demand for resources from the manufacturing sector. Ineffective public investment policies, weaknesses in technical services, and problems in marketing and credit also contributed to sluggish growth. During this period, sectoral exports remained a small fraction of total production (less than 1 percent of agricultural GDP in 1979) due to the overvalued exchange rate and other disincentives to export. 133. As part of the structural adjustment program adopted by the Government in 1980, many incentives were abruptly dismantled, and more market-oriented policies were introduced. Input subsidies and production price supports were reduced and the overall restraint on monetary policy forced a curtailment of agricultural credit. Exports were encouraged through the introduction of a competitive exchange rate policy and other incentives. This shift in strategy initially resulted in considerable disarray in the sector, but in 1982 and 1983 real growth in agricultural GDP recovered to a decent average annual rate of 3.1 percent. During 1984 it reached 3.7 percent. Due to the adoption of a realistic exchange rate and relaxation of export licensing restrictions, agricultural exports increased substantially for agriculture and agroindustries combined, the average annual rate of export growth attained a healthy 9.5 percent, indicating a significant switch from unprocessed to processed exports. By mid-1986, earlier than originally planned, fertilizer subsidies were abolished and the local distribution freed, a major breakthrough long opposed by a strong parastatal distributor. 134. During the first half of the 1980s the prices of seeds and agricultural machinery have been decontrolled, the restrictions on seed imports have been removed, and the time required for testing and registration of new seeds has been reduced. It is very important for increased productivity to have access to high quality genetic material in seeds and since the liberalization some 20 joint ventures with foreign seed firms have been formed. Irrigation water charges and agricultural interest rates have been increased, although complete compliance of positive interest rates on preferential credits had to be postponed till the end of 1986. The Ministry of Agriculture, Forestry, and Rural Development (MAFRA) has been reorganized in order to substantially decentralize decision making, while merging the staff and facilities of overlapping field services. In addition, MAFRA's policy advisory and analysis unit (APK) has been given broader analysis and planning responsibilities, and serves as the "nerve center" for MAFRA. 135. In order to consolidate the progress, to build on the progress already achieved and to assist in fostering continued progress in the structural reform of the agriculture sector an Agricultural Sector Adjustment - 160 - Loan (ASAL) was designed. It basically carries the issues worked on during the SALs a step further, especially as far as pricing and institution building is concerned. The loan of $300 million was approved by the Board in June 1985. It is a SAL-type loan, relatively quickly disbursing in two tranches with the release of the second tranche of $100 million subject to a satisfactory review. The President's Report for the ASAL states explicitly that: "In parallel, general economic stabilization must continue to provide a foundation for uninterrupted sectoral growth". It is noteworthy, however, that the tranche release criteria are entirely sectoral and devoid of macroeconomic elements. Energy Sector Reform 136. General. In the 1970s Government energy policy was almost totally dominated by supply considerations. Pricing policy (with the exception of petroleum products which broadly reflected economic costs throughout the period), demand management and energy conservation were all relegated to low priority. The exigencies of a policy of energy supply at all costs to feed a rapidly growing and largely unconstrained demand resulted, in turn, in the abandonment of systematic maintenance and operations procedures, as well as a tendency to invest in some projects, especially lignite, which were of doubtful economic viability. 137. Achievements. Perceptible-changes in government policies and attitudes by 1984 towards the energy sector resulted in a set of policy initiatives aimed at adjustment in the sector. Government assigned the highest priority to the energy sector and to investments thattwould reduce the energy deficit and costs of supply. An energy sector strategy paper prepared by the Bank formed the basis for two rounds of discussions with the Government in 1984 on the strategy and actions for the energy sector. In these discussions, agreement was reached on the approach and analysis in the strategy paper as well as on a broad action program for the energy sector. Detailed action programs for individual agencies are under preparation. 138. The broad action program is based on a set of priorities which include: strengthening of the planning and project management capabilities of the Ministry of Energy and energy agencies; early completion of ongoing priority projects, particularly in the electricity and gas subsectors; greater Involvement of the private sector in selected aspects of energy development; upgrading of existing facilities in the power and lignite subsectors; improved demand management including indentification of energy saving investments and early enactment of energy conservation legislation; evaluation of alternative options for energy supply; and development of integrated pricing policies for the various energy sources. In particular the Government's decision to open the energy sector to private investment appears to represent a fundamental change in direction in the sector, a change which, however, is obviously consistent with the liberalization policies at macro eccnomic level. The new policy is concerned with both, privatization through the transfer of public assets to the private sector and liberalization through the removal of barriers to private sector investment. The Government's actions in energy pricing have been impressive: as a result of a number of increases in electricity tariffs implemented since January 1984. bulk tariffs are now equal to 85-90 percent of the estimated long-run marginal costs of production, - 161 - compared with 50 percent in 1983. Prices of petroleum products continue to be maintained at international levels. Basically, the progress made under SAL IV and V has paved the way for an energy sector loan, the initiating memorandum for which was approved by the Loan Committee in November 1986. VII. THE ROLE OF THE BANK General 139. The structural adjustment process that began in 1980 was supported by the Bank in several ways. The resource transfer through five SALs in the 1980-85 period, though small by itself still represents 15 percent of the current account deficit over the same period. It was complemented by comprehensive economic and sector work combined with technical assistance in key areas, and a focussed program of project lending. The Lending Program 140. The structural adjustment program had as its objectives the strengthening of Turkey's balance of payments and the creation of a spectrum of economic incentives that would promote rapid growth of production for home consumption and exports in a non-inflationary atmosphere. To support the SAL effort, agriculture, industry, energy, transportation and the financial sector have been singled out as the core sectors for Bank project and sector lending. In agriculture, emphasis is on irrigation, credit, research and extension services. In industry (including DFCs) Bank supported projects assist in the promotion of exports, employment, and increasing operational efficiency. Energy projects focus on power generation based on domestic hydro and lignite resources, as well as enhanced oil recovery and oil and gas exploration. Lending for the transport sector is helping to develop the infrastructure necessary to facilitate exports, and to improve the efficiency of transport operations. Despite much progress in recent years, the execution of Bank-financed projects in the public sector has been slow, due in part to weak management, limited coordination amongst ministries, staffing problems, and financial stringency. Further attention to these areas will be necessary if the broad program for economic re-structuring and longer-term growth is to be sustained. 141. SALs have provided quick disbursing balance of payments support; they accounted for the major part of the increase in disbursements. So far quick disbursing sector loans have kept gross disbursements high in the post-SAL period. Without a Standby agreement at present, Turkey has started to repay the IMF substantial amounts. The next table shows that also repayments of interest and principle to the Bank are increasing rapidly; based on existing commitments as of November 1986 total repayments to the Bank are estimated at $880 million in FY 1990. Economic and Sector Work (ESW) 142. With the onset of the first structural adjustment loan in 1980 the Bank's program of economic and sector work on Turkey was intensified, with a special focus on foreign trade, agriculture, industry, energy, public enterprises, macro-economic planning, public sector investments, domestic resource mobilizaton, and the financial sector. Special studies in these - 162 - Table 1.0: BANK DISBURSEMENTS FOR PRE-SAL AND POST-SAL PERIODS FOR TURKEY (annual averages, $ million) Pre-SAL During-SAL Post-SAL 1978-80 1981-85 1986 Gross Disbursements 213 527 591 Principal Repayments 37 98 185 Net Disbursements 176 429 406 Interest & Charges 62 151 287 Net Transfers 114 278 119 areas were complemented in the context of the Bank's regular economic reports. The ESW underwent substantial changes in the early 1980s and was clearly redesigned to be as supportive as possible of the SAL program and the policies pursued by the Government. As a result of the intensive policy dialogue during the period of structural adjustment lending the ESW program expanded in those areas where the Bank has a comprative advantage and where policy advice could offer high returns. The annual ESW program now routinely has some 15 items of sector work from sector memos to rather comprehsensive sector strategy papers. The average number of staff weeks spent on ESW per year in the 1982-86 period was 435. The ESW program is reviewed at least twice a year with the Treasury and SPO and, therefore, the program also reflects where the Government favors the Bank's involvement. The Government has repeatedly resisted so far studies in sensitive area's like income distribution and employment, probably not because a study would show necessarily deficiencies in those areas of economic policy, differcnt from those in other countries, but rather because of the political fall-out of the Government having agreed to prepare such studies. 143. Economic and sector work on Turkey was reinforced by programs for technical training. Staff from EDI and DYB have jointly conducted a training course on power projects in January 1983 and another on infrastructure projects in September 1983. Assistance has also been provided in the context of debt management. Generally, more training could probably have been given in the course of the SAL program, although there do not seem to have been firm requests for it by the Government agencies. The ASAL and the FSAL both have substantial training components, however. Disbursement and Procurement 144. Administratively, loan implementation in both SAL IV and SAL V were satisfactory. Because of the experience gained by the Ministry of Finance and the Central Bank in earlier SALs, the administration of these loans went smoothly. The Ministry of Finance and the Central Bank bore the main responsibility for administering the loans, and applications for import licenses received substantive approval from the appropriate executive agency. As mentioned before, both loans were fully disbursed, each in two tranches within the year. Starting with SAL IV an "initial payment" or Revolving Fund of $40 million was established, which greatly facilitated withdrawals. Any commodity could be imported regardless of its type so long as the commodity in - 163 - question was not part of the agreed negative list (e.g. alcoholic beverages, tobacco, nuclear reactors, depleted uranium, jewelry and pearls, etc.)Imports were made directly by the users with imports costing $10 million or more processed through international competitive bidding in accordance with the Bank's guidelines for procurement. All contracts of a lesser value were awarded through normal trade channels on the basis of normal procurement procedures of public sector and private sector firms concerned. The procurement procedures of public sector firms had adequate choice of international supplies to ensure reasonable availability and price. As in previous SALs, counterpart funds equivalent to the loan amounts were deposited in a special account with the Central Bank and used by the Government to help finance expenditures associated with its development program. The Bank did not directly influence the allocation of these funds, although indications are that the funds were used on expenditures consistent with Bank priorities. Resident Mission 145. Given the size of the Bank's operations in Turkey there would probably be many advantages in establishing a Resident Mission, not only in cost, but especially in relation to depth, timeliness and continuity of the policy dialogue, as well as to project preparation, implementation and supervision. However, so far the Government has been opposed to it. It certainly makes it harder for Bank staff to keep abreast of things, although the Government generally does not seem to have problems in letting Bank missions have the necessary data and information. The relationship between the Bank and Turkey is very good and cordial. However, because advantages of having an office are so apparent, it is worthwhile to continue efforts to convince the Government of the advantages of having a Resident Mission.