SWP768 Export Incentives and Turkish Manufactured Exports, 1980 - 1984 Branko Milanovic WORLD BANK STAFF WORKING PAPERS Number 768 WORLD BANK STAFF WORKING PAPERS Number 768 Export Incentives and Turkish Manufactured Exports, 1980-1984 Branko Milanovic international Monetary Fund Joint Library JUL 3 1 1995 Internabonal Bank for Reconstruction and Development Washington D.C. 20431 The World Bank Washington, D.C., U.S.A. Copyright (© 1986 The Intemational Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing January 1986 This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The World Bank does not accept responsibility for the views expressed herein, which are those of the authors and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank; they do not necessarily represent official policy of the Bank. The designations employed, the presentation of material, and any maps used in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status of any country, territory, city, area, or of its authorities, or concerning the delimitation of its boundaries, or national affiliation. The most recent World Bank publications are described in the annual spring and fall lists; the continuing research program is described in the annual Abstracts of Current Studies. The latest edition of each is available free of charge from the Publications Sales Unit, Department T, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from the European Office of the Bank, 66 avenue d'1ena, 75116 Paris, France. Branko Milanovic is an economist in the Country Programs Department II of the World Bank's Europe, Middle East, and North Africa Regional Office. Library of Congress Cataloging-in-Publication Data Milanovic, Branko. Export incentives and Turkish manufactured exports, 1980-1984. (World Bank staff working papers ; no. 768) 1. Foreign trade promotion--Turkey. 2. Turkey-- Manufactures. I. International Bank for Reconstruction and Development. II. Title. III. Series. HF1583.4.M54 1985 382'.63'09561 85-26614 ISBN 0-8213-0677-4 Abstract Since the introduction of the stabilization program in Turkey in January 1980, export growth has occupied a crucial role in Turkey's development strategy. This represented a major departure from the hitherto followed inward-looking strategy of import substitution. In the changed environment the success on the export front became a sine qua non for the success of the new policy as a whole. The growth of exports and penetration of foreign markets captured popular attention to a much greater degree than in the past. The Government, which centralized the administration of export incentives and increased their importance, gave strong indications that it would follow through with this policy. Turkish exports responded strongly: in the period 1980-84 they increased from $2.9 billion to $7.1 billion; manufactured exports - particularly encouraged by the Government - increased even faster, doubling their share in total Turkish exports from 36 percent (in 1980) to 72 percent (in 1984). This remarkable growth was, moreover, accomplished in conditions of stagnant or declining world trade. Increased export incentives were instrumental in stimulating export growth. These incentives included continuous depreciation of the exchange rate, often in excess of the inflation differential between Turkey and its major trading partners, payment of export tax rebates (to compensate for indirect taxes), access to subsidized export credits, and duty-free imports of necessary inputs for exporters. Explicit policy of export encouragement, maintained by the Government throughout the period under study, produced both a change in the mentality of industrialists, many of whom for the first time seriously looked to the foreign markets as possible outlets for their output, and rendered exports - in conditions of depressed domestic demand - more profitable. The direct export subsidies (i.e., excluding exchange rate depreciation) were so calibrated as to differentiate between industries. They have consistently favored capital goods industries, and were the lowest in consumer goods industries. Often such a differential treatment of industries has been successful in prompting those with a potential to export - and possibly a long-run comparative advantage - to enter foreign markets. But there were also cases where high levels of incentives yielded only meager results. The efficiency of the export incentive policy - measured by its departure from the "optimal" structure - did not show significant improvement in the period 1980-84. Yet although particular aspects of that policy could have been improved, its overall drift both in terms of direct export incentives and, in particular, exchange rate adjustment was correct. This is borne out by the remarkable growth of exports which took place in this period, but even more so by the structural changes it wrought in the economy, such that the return to an earlier inward-looking strategy seems, at present, difficult to imagine. Acknowledgements I have greatly benefited from encouragement and advice - often at an early stage - by Jayanta Roy and Firouz Vakil. The text was expertly typed by Barbara Mondestin and Zeny Guzman-Scott. I am thankful to all of them. Table of Contents page I. Introduction 1 II. Administrative Organization and Incentives Policy 5 III. Export Tax Rebates 9 Development of the Scheme 9 Distribution of the Tax Rebates 13 Subsidy Component of Tax Rebates 17 IV. Export Credits 27 Export Credit and General Short-term Credit 27 Subsidy Component of Export Credits 37 V. Foreign Exchange Allocation and Duty Free Imports 42 VI. Real Exchange Rate 46 VII. The Combined Export Subsidy: Optimal and Actual 48 Total Export Subsidy 48 Optimal and Actual Subsidies 57 VIII. Subsidies, Real Exchange Rate and Volume of Exports 62 IX. Subjective Perception of the Importance of Incentives 66 X. An Assessment of the Role of Direct Export Incentives 72 I. Introduction The January 1980 Stabilization program represents a major turning point in Turkey's economic history. Throughout the 1970's, the country followed a policy of import substitution, which resulted in 1977-78 in foreign exchange crisis and debt-rescheduling. Although Turkey succeeded to reschedule between 1978 and 1980 $9.2 billion of its debt (representing about 60 percent of the total outstanding and disbursed debt at the end of 1978) the foreign exchange crisis persisted. Many essential imported inputs became unavailable, and constraints on the import side resulted in a decline in Turkey's GNP in 1979 (-0.4 percent). This was for the first time since 1954 that Turkey's GNP decreased; in per capita terms GNP fell by about 2.5 percent, which to a generation used to a steadily rising standard of living, represented a major shock. Political instability compounded-or probably contributed to-the economic slowdown. It is against the backdrop of such developments that the then new minority Justice Party Government, which took office at the end of 1979, with Mr. Suleyman Demirel as the Prime Minister, decided to announce a new economic program. The general philosophy of the program can be characterized as liberal on the domestic side, and outward-oriented in foreign trade. In order to combat inflation (which at the time was running at 80 percent) the program envisaged reduction of the public sector deficit (by cutting current expenditures, and allowing State Economic Enterprises to fix their own prices, thus reducing the need for government subsidies), and tighter control of monetary aggregates. The program also argued in favor of total lifting of price controls in the private sector. It thus undercut spreading shortages and black market dealings. In the foreign sector the program enunciated two principles which were to guide Turkey's export policy in the years to come: setting of a realistic rate of exchange, and the removal of the anti-export bias. These general principles were simultaneously translated in a set of measures. The policy changes, with respect to exports, were the following: - Turkish lira was devalued by almost 50 percent against the dollar. The exchange rate for general exports increased from TL 47.10 to TL 70 for a dollar. The multiple rates system which existed - with five rates, ranging from TL 35 for agricultural export to TL 48 for tourist expenditures - was eliminated except for imports of fertilizer, fertilizer inputs, and agricultural plant protection chemicals (their rate was raised from TL 35.7 to TL 56 for a dollar). - A set of export encouragement measures was introduced. Administratively, all authority for promotion of exports was vested in "Directorate of Incentives and Implementation" (TUD, Tesvik ve Uygulama Dairesi) within the State Planning Organization. Duty-free imports of inputs for export production were to be facilitated, and export credit (at the rate below the market) to be made more easily available. Foreign exchange - 3 - retention scheme, under which exporters could keep 50 percent of their net export earnings, and the export tax rebates, which compensated for indirect taxes, were to be expanded. The export performance of Turkey between 1980 and 1984 can, without exaggeration, be termed spectacular. The value of exports expanded from $2.9 billion in 1980 to $7.1 billion in 1984 (at an average rate of 25 percent per annum). In volume terms growth averaged 34 percent per annum. Concurrently with the overall expansion of exports, the structure of exports underwent a major change, both in terms of its product composition, and country destination. As Table 1.1 illustrates industrial exports, which in 1980 represented 36 percent of total Turkish exports, have by 1984 doubled their share. Their growth averaged almost 50 percent per annum in dollar terms, and was continuous throughout the period. In the same time, exports of agriculture and livestock remained almost stagnant in value, and their share in total exports decreased from 57 to 25 percent. In terms of country-destination of exports, there was a shift toward Middle Eastern countries: exports to that region increased by almost 54 percent per annum on the average, and their share in Turkey's total exports passed from 17 percent in 1980 to around 40 percent in 1981-84 period. The most dramatic increase took place in 1981 when exports to the Middle East almost quadrupled. In the last two years, however, the growth of these markets seems to have somewhat tapered off and the share of exports to Middle Eastern countries in total decreased. The structural shift toward the Middle East occured slightly at the expense of the share of EEC countries (whose share, however, seems to be rising toward its historical level in the last two years), and more significantly, of the East European countries. The share of the latter in - 4 - 'Turkey's exports dropped from 17 percent in 1980 to only 4 percent in 1984. It is interesting to note that in the beginning of the period under study, exports to East European countries were no less important than exports to the Middle East; by 1984, however, exports to the Middle East were almost ten times greater. Table 1.1: Commodity Composition of Turkey's Export (t million) 1980 1981 1982 1983 1984 Agriculture and 1671.7 (57.4) 2219.4 (47.2) 2141.2 (37.3) 1880.6 (32.8) 1749.2 (24.5) Livestock Mining Products 191.0 ( 6.6) 193.4 ( 4.1) 175.3 ( 3.1) 188.9 ( 3.3) 239.7 (3.4) Industrial 1047.4 (36.0) 2290.1 (48.7) 3429.4 (59.7) 3658.3 (63.9) 5144.5 (72.1) Products Total 2910.1 ( 100) 4702.9 ( 100) 5746.0 ( 100) 5727.8 ( 100) 7133.5 (100.0) Growth Rates (% p.a.) 1980-84 Total Exports 28.7 61.6 22.2 -0.3 24.5 25.1 (value) Total Exports 10.9 68.7 24.0 13.9 23.1 34.3 (volume) Industrial 33.4 118.6 49.7 6.7 40.6 48.9 Exports (value) Note: Data in brackets are percentage shares. Table 1.2: Exports by Destinations ($ million) 1980 1981 1982 1983 1984 EEC Countries 1242.1 (42.7) 1502.9 (32.0) 1755.4 (30.5) 2010.2 (35.1) 2731.7 (38.3) Other OECD 437.6 (15.0) 760.8 (16.2) 800.6 (13.9) 750.0 (13.1) 1008.0 (14.1) members* Eastern Europe 490.6 (16.9) 326.9 ( 7.0) 323.2 ( 5.6) 245.3 ( 4.3) 283.7 (4.0) Middle East 494.8 (17.0) 1893.3 (40.3) 2540.1 (44.2) 2430.1 (42.4) 2756.7 (38.6) Other countries 245.0 ( 8.4) 219.0 ( 4.7) 326.7 ( 5.7) 284.3 ( 5.0) 353.3 (5.0) Total 2910.1 ( 100) 4702.9 ( 100) 5746.0 ( 100) 5727.8 ( 100) 7133.5 (100.0) Growth Rates (% p.a.) 1980-84 EEC countries 13.2 21.0 16.8 14.5 35.9 21.8 Middle East 112.2 282.6 34.2 -4.3 13.4 53.6 * Includes non-EEC members of OECD. Note: Data in brackets are percentage shares. II. Administrative Organization and Incentives Policy As mentioned above, the January 1980 measures centralized administration of heretofore dispersed incentives policies into TUD within the State Planning Organization. (In January 1984 the name of TUD was changed into Tesvik ve Uygulama Baskanligi, TUB. In this text all references will be to TUB.) This measure has helped both the transparency, and the implementation and monitoring of the incentives policy. The only area of incentives about which the data are not collected in TUB relates to tax rebates for exporters which are paid by the Central Bank, although the right to tax rebates is conferred by the certificate issued by the TUB. Granting of any incentives is linked with the issuance of the export incentive certificate ('Ihracat Tesvik Belgesi'). Exporters apply to the TUB for the certificate, and on the basis of it receive preferential credit, foreign exchange allocation (for imported inputs) and export tax rebates. For preferential credit the certificate is presented to the commercial bank which, after extending the credit, discounts all or part of it at the Central Bank. Similarly, for foreign exchange allocations, the certificate gives the exporter the right to purchase foreign exchange from the commercial bank, which is later repaid by the Central Bank, as well as the right to import duty-free inputs up to the amount of purchased foreign exchange. The former aspect has gradually, with the liberalization of the foreign exchange transactions and the near absence of the overvaluation of the TL, become of much less consequence.l/ The duty-free aspect, however, remains important. Finally, export tax rebates are paid back to the exporter after 1/ Obviously, firms can also finance imports by their own foreign exchange holdings (obtained through export retention rights, or otherwise). - 6 - the completion of exports. The certificate, in addition to fixing the amount of credit, foreign exchange allocation, and tax rebates, also details the export commitment on whose realization the granting of the incentives depends. The system of incentives as it evolved since 1980 thus contains the following direct incentive schemes for export-oriented production. (1) Export tax rebates which are supposed to reimburse exporters for the indirect taxes. They are almost entirely (95-97 percent of total rebates paid out) geared toward exporters of manufactured products. (2) Export credits at the rate lower than the one charged on similar projects whose output is not directed for exports. Export credits are. of four different kinds: general export credits, export credits for packaging and delivery, credits for exports of fresh fruits and vegetables, and credits for Foreign Trade Corporations (i.e. large trading houses).l/ The latter two are paid out of a special Export Encouragement Fund. Export credits to Foreign Trade Corporations are thus for the purpose of monitoring separated from the general export credit. More than 90 percent of total export credits allocated through TUB's encouragement certificates were for manufactured exports. (3) Foreign exchange allocations which confer right to duty-free imports of inputs used in production for exports are of two sorts: general foreign exchange allocation, and allocations for 1/ Up to 1984 Foreign Trade Corporations were called Export capital companies. - 7 - temporary imports (i.e. for inputs which will be directly built into exports, like, for instance packaging materials). Almost all of allocated foreign exchange is for manufacturing industries. l/ The purpose of the paper is to review each of these incentives, quantify the importance of each scheme for manufacturing exports, assess their individual and combined subsidy element, and finally, come to a conclusion about the influence and significance of the incentive schemes for expansion of Turkish manufactured exports in the period 1980-84. This period represents a self-contained whole. This is so because its beginning coincides, as mentioned above, with the launching of the stabilization drive, while its end marks also the ending of strong direct incentive policies. In effect, the system of export incentives was by the early 1985 quite different from the one that existed only one year ago. Export tax rebates were lowered, and plans for their gradual phasing out were being implemented; preferential export credits were abolished, and in February 1985 Turkey signed GATT Subsidy and 1/ Foreign exchange allocation for export-related investments is not considered here. Also, there are direct tax rebates for exporters. Exporters can deduct 20 percent of export value from their taxable income (assessed at the corporate rate of 40 percent). Total taxes paid are thus reduced by 8 percent of value of exports (provided, of course, the taxable income is still positive after the deduction). For non-incorporated exporters, whose taxable income is assessed at the rate of 15 percent, the value of the rebate is equal to 3 percent of exports. In our calculations this incentive has not been included because the data on direct tax rebates are not available. Taxable income of enterprises is assessed at the provincial level, when the allowance for exports is also made. No centralized data on the amount of taxes thus saved, exist for Turkey. Furthermore, it is obvious that the actual amount of taxes saved will vary both in function of the overall profitability of enterprises, and distribution of profitability (i.e. if most profitable enterprises earn the bulk of their income from domestic sales, the value of the incentive will be less), and are not thus as easily controlled by the Government as the other incentives schemes whose impact is more immediate and better known. -8- Cornpensations Tax Code which prohibits direct government export subsidies on goods exported to the countries-signatories of the agreement. The system has thus already moved away to some extent from direct encouragement of exports and is likely to do so even more in future.l/ The five year period during which it existed in an almost unchanged form represents a sufficient period to study how, in conjunction with the devaluation of the Turkish lira, the system has contributed to the unprecedented growth of Turkish exports. The study is organized as follows. Sections III to VI (on tax rebates, export credits, duty-free imports and exchange rate policy) survey the export incentives granted to the manufacturing industries and present an estimate of their relative importance -in terms of the subsidy component they contain- for the period 1980-1984. Section VII brings these results together by estimating the total combined export subsidy, and its structure by particular industries, and groups of industries (consumer, intermediate, and capital goods). It also evaluates whether the level of incentive granted to different industries was consistent with Turkey's comparative advantage. Section VIII presents an estimate of the real export exchange rate (which accounts for all direct export incentives and the real exchange rate) and considers its relationship with the volume of manufactured exports. Section IX studies, on the basis of interviews conducted with exporters, their perception of the importance of different incentives. Finally, Section X presents an overall assessment of the role played by direct export incentives in stimulating growth of Turkish exports. 1/ In January 1985 a different type of direct incentive was introduced: exporters were paid a 4 percent cash grant on value of their exports. -9- III. Export Tax Rebates (a) Development of the scheme. Refund of indirect taxes paid on exported output has traditionally been one of main export incentives. Originally, tax rebates were paid individually for each exported product. In 1975, the system was simplified by consolidating all export on ten lists, each with its specific tax rebate rate. A special incentive was introduced for large exporters (defined at the time as those with annual exports in excess of $1.8 million) in form of additional tax rebates of five percentage points. After the June 1979 devaluation, the basic rebate rates were lowered, while the minimum for the application of the special rebate for large exporters was raised to $3.5 million.-1 The basic rebate rates ranged between 15 percent (on List 1) to 0 (on List 9) with specific rebates on List 10 items (packaging products). In May 1981 - simultaneously with the introduction of the floating exchange rate - tax rebates were raised, and all products eligible for export tax rebates arranged onto ten lists with rates varying between 20 (on List 1) and 0 percent (on List 9). The rates between the lists declined by 2.5 percentage points (e.g. List 1=20 percent, List 2=17.5 percent, List 3=15 percent, etc. with the exception of List 8 on which tax rebates are 5 percent as on List 7). List 10, as in the past, included specific export tax rebates (for example, tax rebates in absolute amounts). The decision on what list the product will be placed was, and is, based on the estimated amount of indirect taxes paid in the final and intermediate stages of production. However, in addition to that, several other criteria (e.g. the proportion of the domestic value added, products whose exported should be encouraged, etc.) are taken into account, with the result that the tax rebate 1/ See Turkey: Industrialization and Trade Strategy report (1982), pp. 56-59. - 10 - the product receives may only loosely be related to the actual amount of indirect taxes paid. The difference thus represent subsidy to exporters.-/ The decision about the shifts of the products between the lists presently lies with the TUB, and the Treasury and Foreign Trade Undersecretariat. Both bodies are under the Deputy Prime Minister. Prior to December 1983, however, a more formal government decree was required to effect a change from one to another list. In addition to the basic ten lists of tax rebates there are additional tax rebates given to large exporters. For all exports between $2 and $10 million, the exporter receives an additional marginal rebate of 6 percent; similarly, for all exports between $10 and $30 million, the additional marginal rebate is 12 percent. Thus, for example, if a firm exports $25 million worth of goods, it would be eligible for an average additional tax rebate of 9.12 percent.-/ If the value of exports exceeds $30 million, the additional tax rebate is 10 percent on the whole amount. Figure III.1 depicts the progressity of the additional tax rebates. At the end of 1983 Turkish government took the decision to gradually scale down export tax rebates. The decision, taken by the new government of Mr. Turgut Ozal, was motivated by several factors. First, it was thought that, in accordance with the policy of trade liberalization, an ever greater 1/ Also, to the extent that domestic producers cannot shift the whole of indirect taxes onto the final consumer, the subsidy to exporters (versus producers for the domestic market) is even greater. 2/ For the first two million of exports there is no additional tax rebate; for the next eight million, the additional tax rebate is $0.48 million, and for the further $15 million it is equal $1.8 million. This yields $2.28 million in total additional tax rebates or 9.12 percent of the value of exports. OOt0 0 g 09 0 a 0~ 00 D' Cl)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ C-) CL O 'O t 8 0 9 0*E D* O* cnc -4 I OpLLJ cr 0 CE v \0 O <~~~~~~~~~~~~~~L M n~~~~~~~~ cr_~ 0'-040 0 09 0 i 0 Z 0 0 S92LU83d XUI - 12 - emphasis should be placed on indirect tools for encouragement of exports (i.e. the exchange rate) and less on direct subsidies to exporters. Second, relatively high levels of tax rebates were responsible for emergence of the so-called fictitious exports. In effect, one part of what purported to be Turkish exports was actually never realized. Foreign exchange was transferred from the country of destination of exports (often by Turkish workers there, friends or business partners of the presumed Turkish exporter), while the actual transfer of goods never took place. The transfer of foreign currency entitled, however, the exporter to claim tax rebates. In such a way, he was eventually able to sell foreign currency at the premium (equal to the rate of tax rebate) which, in view of near absence of TL overvaluation since 1981, was fairly substantial./ The amount of fictitious exports was estimated at 5 to 10 percent of total Turkey's exports with a tendency to increase.-/ Almost all fictitious exports relate to exports to West Germany and Switzerland, both with sizable Turkish guest-worker population.-/ The third reason for the reduction of tax rebates lay in the domain of public finance, i.e. in the need to cut expenditures. The phased reduction of tax rebates took place as scheduled. On April 1, 1984 all rebates were lowered by 20 percent: for example, a product 1/ The premium could then be split between the presumed exporter and its partner abroad. 2/ In 1984, the value of fictitious exports was estimated at about $1 billion which was 14 percent of total exports. 3/ It should be mentioned, however, that from the balance of payments view ficticious exports are not subject for concern, and they may even result in greater inflow of foreign currency in Turkey. They simply represent the shift from workers' remittances into exports. But if all the money transferred via fictitious exports would not have otherwise come into Turkey, fictitious exports actually raise the amount of inflow. - 13 - which received a 20 percent tax rebate was to receive henceforth only 16 percent. Then on September 1, 1984, a further reduction of 25 percent of the original amount took place (with the product in our example receiving only 11 percent as tax rebate). In less than one year, rates have thus been almost halved. The same pattern of gradual reduction was also applied to additional rebates for large exporters.-Y (b) Distribution of the Tax Rebates. Table III.1 depicts the industry-wise distribution of the export tax rebates as of January 1985.-/ It shows the number of products belonging to a specific industry which receive a given export tax rebate, as well as the average rebate rate. 1/ The scale of the graph in Figure III.1 needs accordingly to be changed, although its progressivity and the relative additional incentive are large exporters, are the same. 2/ The classification of industries here, and throughout the text, follows the classification used by the SP0. - 14 - 0o o% O (C4 N co L cN -4 0% of 0 0 . .. c. c . . .c 4 4 4co 0% U _4 U% 4 co > P co U 0 .. -44 . of ciC - r o cI - 0~~~3 0 U Cs 0 C 44 00 NO 0 04 0 0 0 0 0 0 O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Lfs %0 00 0 0 0C r0 0 04 0 0 0 O O O- 14 - 0* 0 O~~~~~~~~~~~~C V O O Cn Ln 0 01 to es 1e -N 0% - -0 N 0° 0 0 11 o 0c 3 N 'ONO u4%0 0 40 0 . 1 _ s 0 1 _ O 0% 4 0 4ir~~ .4.0 0%f'.N .0 I~'~N N I'- . Wu 0 N I ,-4 . . . -' > 4 -> -4 .; - c1 N _ ~~~~1 tS L, o ) o Ai a 4j t o 4 co 4 's W o A W co >0" W W - 0 0 W Wr. c 0 Wc cti 0 4-4 _) s t O _ A O 0 00 0 O W 0i _ -4 _ _ e * ~~~~~0 -dl *i)Dwc Z,U4) zC 9-4 0~~A - . . d ~ go ._ 4 4 )) ns go n0 0*LAz c04 C 4 _ IN 0 c 0 ~ ~ C N 04 N- 0 .4 aW0 EQ SC^ W Ce @@ e W ^ ^C X_ - W^ XX dJ S .JQ0X ^eSXW >c0r 0 r c 1 r@00XC IXeOIQQQ C s >. e 0 %sccbbc : -W 4 )XX5AW Xo O O S @ @ K Z C: X- Q @ W Z X Z X Es ~ 0 Q.- . . . N -4 . -4 0j _S alu o o oo ooo o 4.- o °° 4) 0b0 - S 14 U .0 014 o4) _ _ _ _ - 15 - Figure III.2 shows the how the distribution of the itemSeligible for tax rebates has changed in the period 1981-1985. There was an increase in proportion of items receiving moderately high rebates (items on lists 3 and 4), and a decrease in proportion of items on lists 1 (high rebates) and 5 (relatively low rebates). The structure has thus remained firmly stable. Table II1.2 shows the evolution of the overall number of items eligible for tax rebates, and the average tax rebate. (The average tax rebate is calculated by multiplying the number of items on a given list with the tax rebate rate applicable to that list. We call this average-by-list tax rebate. It differs from the average actual tax rebate which is the ratio of tax rebates effectively paid out and value of eligible exports). We can see that the number of items eligible for tax rebates has steadily increased from 1980. While the scheme originally covered 373 items, it now extends over 648 items. The enlargement of the scheme has been continuous, and it embraced all sectors, with the single exception of consumer goods whose number slightly declined in 1985. The expansion has been most important for capital goods items whose number more than doubled in five years passing from 103 to 254. Table III.2: Number of items eligible for tax rebates and the average-by-list tax rebate January 1980 May 1981 December 1983 January 1985 Number of items Consumer goods 75 90 103 95 Intermediate goods 195 226 271 299 Capital goods 103 171 181 254 Total 373 487 555 648 Average-by-list rate Consumer goods 12.23 12.83 12.62 6.56 Intermediate goods 12.46 12.10 12.12 7.16 Capital goods 15.89 16.46 16.20 8.86 Total 13.39 13.77 13.55 7.79 Capital goods have also enjoyed the highest average-by-list tax rebate. It means that proportionally more than the average of capital goods F-I ~ LA0 LA 0S U) N N -a -Ln '- 0U 1-4 O o CO oi O C L oo 0 -9~~~~~~~~~U a,~~~~~~~~~~~~~~~~C >-4~ ~ ~ ~ ~ ~~~~~~~~~~~- C\3~ ~ ~ ~ ~ ~~~~~C - 17 - items have been placed on high tax rebate lists. The opposite is true for consumer and intermediate goods. The overall average-by-list tax rebate remained reasonably steady - at around 13.5 percent - until early 1984 when tax rebate rates, as explained above, began to be scaled down. The stability of the rates means that the new items which were regularly added to the lists were not disproportionally directed toward the high- or low-rebate lists. In January 1985, by which date basic rates of tax rebates were scaled down by 45 percent comparatively to early 1984, the average-by-list rate was only 7.8 percent, i.e. significantly below the level of one year ago (13.55 in December 1983). It may be noted, however, that the structure of items on the lists shifted slightly toward high-rate lists: in effect, if the distribution of the eligible items in January 1985 had been the same as in December 1983, a 45 percent across the board reduction of the rates would have lowered the average-by-list rebate to 7.45 percent, instead of the actual 7.79 percent. In terms of individual industries, proportionally more items were placed on high rebate lists in subsectors like iron and steel, transportation equipment, leather and fur, and both electrical and non-electrical machinery (see Table III.1). At the other end of the spectrum, chemicals and textiles have more than the average number of items belonging to low-rebate lists. This, indeed, does not necessarily mean that these subsectors were actually receiving higher (lower) tax rebates, as we abstract here from the relative importance (export value) of each product. To this issue we turn next. (c) Subsidy component of tax rebates. Table III.3 shows the relationship between total exports of an industry, its eligible exports (i.e. exports of products eligible for the tax rebates), and actual amounts of tax rebates paid in the period 1980-1984. 18 - 0 0 C .0 0~~~~~~S 0 14 1 C 40 en N4 .ll t 0 -0 40 .44 In 0, In 004 10 11 1 .44 H co N en 0 0 a -4 40 a, .44 _0 4 e1 I, 4 c bO co .0 'IN -4 -4 - -* ' - en .% co r- C4 0 -10 0 '0 0 t ' 4 t 40 - 5-4 _00 co _ _n 1- cn 0 -0 ..AU O 0 0 _ c0 C1 _ 4 4f U P 0 U N o0 4fl t . 44 4 0 P > _~ ~~~~~~~c Crv co a, In P- vD a, In - en un N v ao 40 40 N, C, 4 co 04 4 0 04 4f 04 0D CN CD 0a 044 0 4.0 4 O00 r' &o N '0 0 _ O co _,oo ax q N 00 .44 N C 0 04_1 cn A ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~A0.0 00 C0 co 'WI - c S 00 WA 0 -0 N co 414 0 Z 0% m 1% > 14 14e _ '0 N 0 co _ O I N :t 1 to a Q.4-4~. 4 r- _4 ON .4 4 It 0 en U'4 4 rN 4 ° 01 140 000_4e.. . . . . . . . . . . .CSr_. . 0 X-r 4 co4 145a,-4 I -t T L *~ .44 N 04) 44 Cfl"4~~~~~~~~~~~~~~~~~~~C 0140 Ca 0 C4 44Q4b 4~ .4 0 .~ .44 C- N C- U4 04 N- (', 0-4 ~~~~~~~~C. _ 0 .44 4 00 40 r- 4 .4 4 4i' N I' -4 0 00 .14 E 0 04 ^4 4 N . en 0 e 0 0 C 4 N wo 0 . 00 00 -0 C- 04 oo, Wi 4 N 4 4 .4 .4 W 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ o _ . X. .44 0 '0 0 co 04 0 C- 4 en N o t- u c1 N N N * c 1 en f 40 00 0 C' 0en4 14 U1 Q4 ................00_ 0rw4 s u bU C1 X ~ ~ ~~ ~~4 .04 40 -4 04 $0 b~ 0u O 4 ' ' 4 4 ,00 04i O-. F' C' b4 0 ' 44 0 - 4 C u~ _ : 0 _1. C' N 4 N 0 C', I-. c' 1' 4O 0S N C' 0 _d o O . .4 0. C. 0 co . C .44 . N C. e' . 0 .. .. ~ ~ 4 04 -% X -4- i 5~ 4 U) rl C^14 4400 0 0 c S. 0~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~01 _A o Vco I oNo>x . o00N' 0 0 a0 VI% a, o N_ 00 0 40 en N Nj >. 0 13 _ 0. 14 _ 1- 4 0 .0 00 0 U az U b * .O 4 .4 ~ ~ ~ ~ . co 0 , ,:e:XX AUx co en 04 c a '0 In -4 0 -4 D. 4 o p. 4 0 g0 .0 140 I~~~~~~~~~~~~1 0 n e m- n P~4 * L WC 1-40 4 I 4 S 444. 0 ,.4 N X1 00 N sn4 loao O Q Q al ,C 0 04 U 9 Z 14 Z; Al b i ° 0 c~4. 0 00 14 0 14~0 c x~~~0- .44 '~0 '0 In a, 0 N C. 40 0 001400 w 0 0 co~~~0 N .44 co r 14' 00 p. 0 0 wCC 04 14- -A 0.4444 0 p4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~14 -4 CO -4 441 -4 00C N00 N -40.44 N~~~~r~N 1- C',4W 044 C' co00 co ~~~~~~~C. C-.4 N .4N0 .44 N '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0.0 0 U 0. 000~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. c - C-. C-. C-. t~~~~~~~~. '~0 N4 .44 4 N4 .4 4 40 4'- 04 0 00 Co r:~ W ' N 4 C' ' 4 4 i ' 0.0 .44 c U C--~~~~~r in 0 4' N 00 '0 0 c', 4 00 Wi '0 0 N4 to0 co 14 0 Jo 0 4-. '0 04~ '0 N4 Wi 4- Ws '0 4 C 14 0 I.4 C-. In 04 .4 4 N- 00 N .0 4 0 - 04 04 0011U '0 C, ' la 0 .4 ' - 4 ' 0 -.14 > 0 0 00a t 0 0 ~ ~~~~~~~~~ 0 40 0 0 ~~~~~~~~~.44 -4 '0 0 Co 0 00 U 0 .U 0 to1 1 0 0 U 14 A1 0 0 C 14 co0 0o .4 0 04 14.. 0 14 0 4 0 co 0 > w~~~00 . '4 Z * .44c 0 "0 ' 4C 0 0IV 0 0 000c CQ I0 0 00 0 U * 0 0. .4 U 0 -.4 0o0 0o '04 0to ' 0 0 4 ,~0 00- 0 0 U14 0 04 00 0 4 0 0 -4U 0 '~to0 0 co ' 44 0 0 0 - * 0 - *'Ai0 A 14 -44l0 00 0 'w .4 0 4 '0 4- U 01 .. .0 01 0 0 I 14 I 0 c 0.0 0 .0 004 0 .0 0 - 8 0 0) A 0 .4 14 0 0 0 -4 1 0 14 44 . 4 0 4 U 0 4-~4 z 14 z w 5- 2 -19 - co .0 . ci 0 '0 ci 0% N 0 It co PI N '' d D.N * Z-4 4- o M @ H N 4 '0 cr N - 0 0%' 0. 4 ci 0 NI xco% - C N .4 .- N N N N Nq N N NI I-' -0 ID '0 _4 00 40 ci U, 40 _4 U, O 0%S 0% 4 Ue N GI U " 4 u, 0 0 o 0 cO N c 0 co ei n N 0 U N C 0%MO ~~ ~~~0 ci N 0 U a, '0 a, 4 * 0% a, co.0 WN 0 0 N U, N4 N 0 ci 4 N ci 4z ci .10 _0 P4 D _ w _ _4 K 4 >1 co GDD. 0 . c ci4co 0 m' *., 4 c0% 4s In 4 o .4 ID N _ ' 0 0 N N N U _ b cn Z4 Y0. UN '0t 4" U N 0 N % _0 N U, _ U, -O U, ol c. .4 £ 'o1- °i -'o 4 .- o GO t 4 wo n ¢ o o 0 U 0 U Cfl-4 GDu£ ° ° ° V a 4 r, g4 C° o a( CO rl _OX0N4 r vh ec EC OC U0% i N ,o - N N* .4 CO 40 UN 4 01 _U*co N ,n -4 0% 4t en N 4 UA _ ,0 . - X 0 4 4 O 0 0 r 4 co 0 N i 4t N 0 'r4 co ci _ u-4-4~ 40 U, ci .-I n _4 N4 0 o N ci .-4 e n 0 F4 0D N ' a, e0 o _ 0 U~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 44.4 N U, 4 U,~ ~~~ ~~ ~ ~ ~ 00 0% 0% 4h c a, 40 0o t^ U 0 N 4 0% N 40 ci~~, .4 en N 0 ' % U 0 10 Ml0 0 0 % 0 OM0 N 0%r N '0 d .. W W N 0 ce 4 4 '0 _ 0o N -4 -4 r. 4.- G 0 4 0UO _, N -4 0S CS ci CS NS 4 O ^ 004 %O "0K cS N 3 cN N c 4 co en 0 0 co c o 1- A 0% o o4 CO N -4 CS _ CO 4 o _ N CO JO _ ..'4 N N .S° -A~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~G U_^4 4_ __ N .aU 0 C - t0 N cr N en cO, N _ In UN -1 0 'u -G U.. w% '4 _ i 0 _ , 0% 0 co '0 N ° i 0 U, ...4 o -.4 _o N N4 N D .4 N N N4 4 N NO N4 NO .0~ U0. 4 C0 o ^ cF e n r t ° C O d W 4- O Gc c D c D S r co 0 GD N~~r N, 40 4 N U,% '0 0 4 4 a, N (4 .4 . U} 4s N- _I 0 - N 0 4 '0 4 c N c CN '-4 U0 0 U, ' U U '0 a, > 4 ' 1 0% 0 Ci N ci .4 N- 40 PI -4 en Ci N 4 U s~~~~'1-4GD ci~~~e CI 0 K- 0 Cic co.4 U, U 0 N 4 .0 40 ,- 0 0% N 0% x Q~ F *4 O C i N N N N N 4 0 U, -4 U, 0 N F .1 44 ' U4 0 co C N 40 N CI co o L 4 0% NI 4-4 WO"-' '4~~~n 4 o 0 e C4 4, co0c - 4- N 4.eZ-4 s 0 -" '4 en GD 0% 0 -4 J X c) i 0 It 0 z '-4 '.4 COx 0.r a r n 0 I 4 4 .0~~~~ 0 U~~~~~~~~~~~~~~~~~~~~~~~0t K 0 ~~ ~~~~04 40 0 UN -4 40 40 4I 0 In - 4 40 0 OD~C N 40 0 '0 4h 0ID 4 N N 0 4 4 N. '0 U" - i C 40 '0 Ci .-4 Ci N0 4 4 .4 -.4 - .4 4 ci 4 .4 -4 ~~~ -4 0 .0 -4 -4 4~~~~~~~~~~~~~~~~~~~~~~~~% - ID 'A 6~~~~~~~~~~~~~~~~.4 ' 0. ~~~~~ 400 K 0~~~~~b N % 0 '0 Nco0 c N .. "4.46 ci N 4 N~~~~~t 0 N ri ' U '0 , N . '-'0 4 N Ci 0%~~~~~L 40 U 0 - No 0 U, '0 U,o U ".4 2'~~~~0 U, 4 40 4 0 i CA Ci 4 0 U 0 U 0 4 ~~ ~~~ ~~4 a0 U 4 % ' 0 N 4 4 N N 4 1-4 0% N~ ~~ ~~ 0. -4 N.- i 04 4 GO ) 1 (LP U0 0co Il0. di w. co. a I 0 GD 0 0 0 - 0. - u . GD - z W *-4 '-4 4 20 - 0 C4 coo .C 60 c Q X uo en en 1o co PI - 1t co W% NO N .- 00 r. a, c O f. 4 co cn o oN4 co _-4 1-4 C4 co 00 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 Q ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1. 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ uz m co r0 n o co co -4 1 C-4 Ns ct en ^ o a _ ,_ ~ ~ ~~~~, o 1^ o _ 4 c n O r, C- ON C- c o b _ X0 0 < > C4 CS4 a 0s LeN OD -0 U- ur% Lr -* 10 C rto1 .0 rv m 0N 40 -S4I - vt O 1 8 o X~~~~. n 4 0 41. 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 4~~~~~ 4i .14 0. 0AM 0 ~ ! N 0 ~ IA ' X~~~~4 Cl X 4 0 N ~~0 N 4LI) ~ 0' 0 F L - 4 00 -4 4 ~4 10 G)4.- 4 _N N -4 N co IA X 0 4.coJ,0'U- ~ I .4 o e o A 14 4_ F~~~~ 0 o c DnN4NHa rn4 Ch X t 4 VJ0N C O t H D v D cs X X co0 4 % '- 0 Co _i os . t0 o N D r0 1D D ^ c3 r, X a, 4 O 4. D O : O 4^D _ o o r Cf :s *-l a e *> D 04 `O e 04 N cO 4 oo -I VI% e1 Ai - 0. -n I.- 4 OID O .-4 N 0 P- 0 %l ON 0% t4 N440 OD X% '0 c 4 r-. 4 IA 4D D0 O 0 0% cr 0 O _ w N ct Xl r, :S en S XO N cS 4 -4 e - 4 14 C14 0 '4 44. 4.4~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~, OC~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~X ~4 n 0 ^ m~n 000 14n "4 co W 0 0-4 14 O4 CO 1- OO D ~ ~ ~ C4 m 1 D (S (n 2 D C1 n - °t% C4 P. 1D OD 1 CO OtC o) ~ r,r- c VI en 1 10 10 r. c 0 -4 ,4.'" U. 0 A U~Z 4-4C V O JC 0..41 b A 1 0 . 0% N4 IA 3 '- c0 toN1 b 0 . 0 1 j'n '0 L'0 CN ~4 -4 en .0 0% m $4 -~~~~~-4- ~ 4, - 0 h 4 tflC X U .P c.4 _ 0 " C 0 °° to 3 -4 '0 '0 cocco r. (4 U* C: _ 0 0 VI U) u " u a, 4 0 co 0 Aio0 c 0 4 w 4% . ., fi u co 0 u^ 6 D S U C I id 1 Q CX P O.4 N . 41I 1% C 4 . ( 14 ° 0 ° . Ez i (D 0 4 P. 0 ' ' 0 o 0 -4 0 0% P4 E- .4 . - 4 4 IA .0w Hz - 21 - The average tax rebate rate (proportion of taxes refunded against the eligible exports) for the manufacturing as a whole increased successively from around 9 percent in 1980 to 14 percent in 1981 and to 21 percent and 23 percent in 1982 and 1983 respectively. In 1984 it declined to 21 percent. The increase between 1980 and 1983 was not due to a changing composition of exports but to a general increase in tax rebate in all the industries. In effect, tax rebates rates increased between 1980 and 1981 and between 1981 and 1982 in all industries, by an average of 5 to 6 percentage points annually, and in some industries (leather and fur, papers, iron and steel) increases exceeded 8 percentage points p.a. In 1983 the rises, albeit much slower, continued in ten out of fourteen industries. Finally, in 1984 the trend reversed and tax rebate rates increased in only two industries, and declined in the other twelve. Coincidently with increasing tax rebate rates, the coverage of the tax rebate scheme (measured by the share of eligible exports in total exports) expanded from less than two-thirds in 1980 and 1981 about 80 percent in 1982 and 1983, and then to about 87 percent in 1984. Thus, in 1984, increase in the coverage has to a large extent mitigated the decrease in the average tax rebate rate and left the subsidy component of the scheme almost unchanged. Average tax rebates and coverage display a distinct pattern in terms of large groupings of industries. As Table III.4 shows, the average tax rebate in capital goods industries was until 1984 consistently higher than in consumer and intermediate goods. The differential, however, has narrowed since 1982 as the average tax rebate for intermediate and consumer goods increased sharply. In 1984, general reduction of tax rebates affected capital industries the most, and in a marked reversal from the past, they fell to the third place. General tendency of rising tax rebates - up to 1984 - clearly emerges from these data. In each year in the period 1980-83 the average tax - 22 - rebate for all three categories increased. Particularly noteworthy was the rise in intermediate goods industries, from b.6 percent in 1980 to about 23 percent in 1983 (the level at which it all but remained in 1984). Table II1.4: Average Tax Rebates Rates and Coverage, 1981 - 1984 (in percent) 1980 1981 1982 1983 1984 Average Tax Rebate Consumer goods 9.2 13.0 20.9 22.3 20.6 Intermediate goods 6.6 14.2 20.6 22.8 22.1 CapLtal goods 14.7 19.5 23.4 24.0 19.5 Coverage Consumer goods 68.0 61.1 70.3 74.9 82.8 Intermediate goods 44.b 67.5 81.6 79.5 85.7 Capital goods 60.9 86.6 110.8 117.5 122.4 Source: TUB (SP0), Central Bank. NOTE: Consumer goods industries include food and beverages and textiles. Intermediate goods industries are leather and fur, paper and paper products, chemicals, rubber and plastic, glass, cement, iron and steel, and non-ferrous metals. Capital goods industries are fabricated metal products, non-electrical machinery, electrical machinery, and transportation equipment. This classification is used throughout the text. - 23 - The comprehensiveness of the export tax rebate scheme was also greater for the capital goods industries than for the other two categories. Again, the coverage of the intermediate goods sector by the scheme expanded the most rapidly (from less than one-half of total exports in 1980 to 86 percent in 1984). The explanation for more than 100% coverage of capital goods sector lies in the different exchange rates used for actual exports (exchange rate at the time of exports), and the eligible exports (exchange rate at the later date when the tax rebates are refunded).l/ Industries with consistently high tax rebate rates included transportation equipment (with an average rate of 24 percent over five years), iron and steel (20 percent) and leather and fur products (19 percent). Textiles and metal products also enjoyed relatively high tax rebates (18 percent in the 1980-84 period). The industry with the lowest tax rebate was chemicals (less than 14 percent on the average). The subsidy element conferred by the export tax rebate scheme depends on (1) the share of the eligible in total exports, (2) the average tax rebate rate, and (3) the actual indirect taxes paid for which the scheme is supposed to compensate. Obviously, the higher the share of indirect taxes in total value of exports, the less would be, ceteris paribus, the incentive value of the export tax rebate scheme. As data on actual proportion of taxes in value 1/ Eligible exports in TLs, and the amount of tax rebates to be paid are calculated at that--higher--rate. - 24 - of exports for different industries are not available, we shall assume that in 1980 when the importance of the scheme was at its low point, tax rebates were merely sufficient to compensate for the actual incurred indirect taxes.l/ The subsidy component of the scheme is then derived as the difference between the tax rebate rates in years 1981-1984 and the tax rebate rate in 1980, multiplied by the share of the eligible in total exports. Table III.5 shows the estimated subsidy component of the export tax rebate scheme for the 14 industries, as well as for the groupings of consumer, intermediate, and capital goods industries. Table 111.5: Export Tax Rebate Subsidy Component (percent of export value) 1981 1982 1983 1984 Consumer goods 2.33 8.31 9.96 9.68 Intermediate goods 5.26 11.73 13.10 13.56 Capital goods 4.94 13.16 15.23 9.74 Food and beverages 2.60 6.42 5.90 7.24 Textiles and clothing 2.19 9.34 12.05 10.76 Leather and fur products 7.32 12.89 17.70 15.33 Paper and paper products 0.33 14.22 9.71 8.02 Chemicals 2.45 6.27 12.40 10.82 Rubber and plastic 3.71 11.96 9.37 10.13 Glass and glass products 2.47 10.89 6.80 9.59 Cement 10.70 18.80 16.09 13.82 [ron and steel 2.18 9.38 13.65 16.14 Non-ferrous metals 5.31 11.80 12.45 14.26 Fabricated metal products 19.52 62.04 79.04 50.54 Non-electrical machinery 2.73 9.74 17.79 6.75 Electrical machinery 3.54 10.90 13.21 14.87 Transportation equipment 3.96 6.32 4.57 3.61 Total 3.60 10.07 11.50 11.07 1/ This is also the assumption made in Industralization and Trade Strategy report (see p. 82). - 25 - The incentive value of the export tax rebates has substantially 'increased in the period 1981-1984: while in 1981 the value of the implicit subsidy in the scheme was equal to 3.6 percent of the value of total manufactured exports, it jumped to more than 10 percent in 1982, and then increased further to 11 percent in 1983 and 1984. The increase in the subsidy was across the board. In the consumer goods industries the subsidy element rose from a low 2.3 percent in 1981 to almost 10 percent in 1983 and 1984; in intermediate industries it went up from 5.3 percent to 13 percent, and in capital goods, in the same time, from 5 percent to 15 percent before falling to less than 10 percent in 1984. It is obvious from these results that the scaling down of export tax rebate rates in 1984 had less impact on the value of the overall subsidy than could have been thought solely from the announced reductions. This was due to a number of reasons: structure of exports shifted toward high rebate industries, the bunching of exports before the date when the tax rebate rates were to be lowered, and the increased coverage of the scheme. We can see from Table III.6 that if all of the elements had remained the same as in 1983, the subsidy value of the scheme would have been lowered by about 43 percent (from 10.5 percent of value of exports in 1983 to 6.6 percent). The change in the structure of exports (both within sectors, and between the sectors) toward the high tax rebate products is responsible for increasing the average subsidy rate by 3.84 percentage points;l/ greater coverage of the scheme and bunching of exports account for respectively 0.27 and 0.36 percentage points increase in the subsidy rate. All these effects have thus combined to make the reduction in the incentive component of the scheme much less than was anticipated in early 1984. 1/ Difference between lines (1) and (2), plus difference between lines (3) and (4) in Table III.6. - 26 - Table III.6: Subsidy from Tax Rebates, 1984 (percent of export value) (1) - Actual subsidy 11.07 (2) - Without change in subsectoral structure of exports* 7.39 (3) - (2) + without change in coverage 7.12 (4) = (2) + (3) + without change in sectoral composition of exports 6.96 (5) = (2) + (3) + (4) + without 'bunching' of exports 6.60 * i.e. if the structure of exports within a sector were the same as in 1983, and the average tax rebate rates accordingly decreased as scheduled (2U percent after April 1, 1984, and 45 percent after September 1, 1984). The individual industries which enjoyed the highest tax rebate subsidies were fabricated metal products (about 42 percent of the value of exports on the average in the 1980-84 period), cement (12 percent), and leather and fur products (11 percent). The industries where the subsidy elemaent was small include food processing and transportation equipment (about 4 percent). - 27 - IV. Export Credits (a) Export credit and general short-term credit. The incentive value of export credits resides in the difference between the rate of interest charged on general short-term (ST) credits and a (lower) rate of interest on export credits. Figure IV.1 shows the rates for the period January 1980 - January 1985. As can be observed, the differential which at the beginning of 1980 was 17 percent increased to about 30 percent in 1981; since then it has consistently declined. The average differential between the general and the export credit was 20.3 percent in 1982, 18.9 percent in 1983 and 17.0 percent in 1984. Finally in January 1985 the government decided to eliminate subsidized export credit. The decreasing differential between the two rates since 1981 is essentially the reflection of a fast increase in the rate of interest changed on export credits. Thus, for example, in January 1980, at the inception of the stabilization program, the effective rate of interest on export credits-I was estimated at 8.25 percent, while inflation was running at more than 100 percent p.a. However, by the end of 1984 - i.e. before the subsidized credit was entirely eliminated - the effective rate of interest reached 50.5 percent, not much below the ruling rate of inflation 1/ The effective rate adjusts the rate charged by the commercial banks for all government's contributions (subsidies) passed over to the borrower. It does not, however, take into account counterbalancing funds which many borrowers were required to deposit in order to get a credit. Counterbalancing funds existed both for general and export credit and, hence, the calculated differential between the two effective rates is probably not much influenced by their omission. Rates of interest paid on counterbalancing funds by the commercial banks were low. Amount of the funds - as the proportion of the credit - is difficult to gauge, both because the operation is illegal and also because the proportion varied (significantly) as the function of the borrower and the commercial bank involved in the operation. However, estimates put these funds at 20 to 40 percent of the credit amount. -28- 0 0) 0 OD 0 0 0 C) 1- 1- --_ M C)O -} I 0° ci -4~~~~~~ CL-~~~~~~~~~~~~~~~~~~~~~~~I D! D Z~~~~~~~~~~~~~~~~~~~~~~~~ OU _ CD~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ I-,s CD CD 0 C-) m~~~~~~~~~~~~~~~~~~~~- cn~~~~~~~~~~~~~~~~~~~~~c 0 0 04 0 0 - 29 - (56 percent p.a.). Figure IV.2 shows the relationship between the effective rate of interest on export credits and the rate of inflation.l/ We see that the real rate of interest on export credits shows a continuous improvement (increase) from the first quarter of 1980 (when it was -49.0 percent) to the last quarter of 1982 (when it reached +4.8 percent). Effectively, in each quarter during these three years the differential between inflation rate and effective cost of export credit was reduced. There was some retranchment from that policy in 1983, as most of the time the real rate of interest on export credit hovered around zero percent. Furthermore, in 1984, brisk acceleration of inflation failed to be entirely offset by increased nominal rate, thus leading to an average negative real rate of interest of approximately 2 percent per annum (see Table IV.1). After January 1985, however, elimination of special export credits increased the level of interest on these credits by approximately 15 percentage points, i.e. set them on a par with interest rates on general purpose ST credits. The real rate of interest on export credits was thus about 10 percentage points above the rate of inflation. Tabie IV.l. Interest Rates on Export and General Short-term credit (X p.a.; Average Yearly Values) 1980 1981 1982 1983 1984 (1) Effective nominal interest on export credits 15.1 20.3 27.2 28.0 46.1 (2) Effective nominal interest on general ST credit 38.3 50.2 47.5 46.9 63.1 (3) The differential; (2)-(l) 23.3 29.9 20.3 18.9 17.0 (4) Effective real rate of interest on export credit -44.6 -12.8 +1.1 -1.8 -2.3 1/ The shaded area shows periods of positive real rate of interest. -30- o~~ C~~~~ m s O CD~~) DV U~~~~~~~~~~~ ( 0~~~~~~~~~~~3 1-4 C- FZ4 01 CD~ 0~~~~~~~~~~~~~~~~~~~~~~0 Co I - 31 - The difference between the interest rate on general and export credit stems from three factors: (i) a higher "scheduled" 1/ (and discount) rate for the general credit; 2/ (ii) a positive contribution to the Interest Rate Rebate Fund (IRRF) on general credits versus a net subsidy received from the Fund on export credit; and (iii) exemption from the transaction tax for the export credits. Elements (ii) and (iii) represent the most obvious subsidies (lower discount rate, reflected in the element (i) is another). General purpose credits were charged in the period 1980-end 1983 between 10 and 15 percent of the base rate as contribution to the IRRF. On the other hand, tax on export credit was (until the end of 1982) 10 percent of the base rate and the subsidy - from the IRRF - between 35 and 25 percent. When the tax on export credits was eliminated in January 1983, the subsidy was also reduced from 25 to 15 percent. Thus, throughout the period 1980-end 1983 export credits were receiving a net subsidy of about 15 percent from the IRRF vs. a net contribution of 10-15 percent to the same fund from general ST credits. Similarly, the transaction tax charged on general purpose credits was 25 percent of the base rate between January 1980 and April 1981; it was then reduced to 15 percent, the level at which it remained until the end of 1983. 1/ The base (or scheduled) rate of interest is determined by the "gentlemen's agreement" between the major commercial banks (and is largely influenced by the Central Bank's discount rate). It is with reference to this rate that different contributions and subsidies are calculated, so that the cost to the borrower is equal to the base rate plus (minus) contributions (subsidies). 2/ Quite exceptionally, in the first half of 1980 the base and discount rates for export credits were higher than for the general ST credit (e.g. between March and June discount rate for export credit was 15, and for the general ST credit 14 percent p.a.). The final cost to the borrower for the export credit was less because of government's subsidies. - 32 - Export credits on the other hand were exempt from the transaction tax throughout. The Ozal government, in its general attempt to simplify the incentive system, took in December 1983 important measures whereby subsidies attributable to factors (ii) and (iii) were sharply reduced. The subsidy from IRRF to exporters was eliminated. Simultaneously, net contribution to IRRF from general purpose credits was halved: from 15 percent it was reduced to 7.5 percent. Also the transaction tax was lowered from 15 percent to 3 percent. The impact of these measures on the interest rate differential was, however, somewhat offset by a steeper increase in the base interest rate on the short-term general credit than on the export credit. Both the general and export credits thus become more expensive (see Figure IV.1), but the differential was only slowly reduced in 1984, until in January 1985 preferential financing of export credits was entirely discontinued. Table IV.2 shows outstanding export credits of the Central Bank (CB) the and the commercial banks. The share of the Central Bank's export credits in Central Bank's total outstanding credits clearly displays an increasing trend up to the end of 1983, speedily rising from around 6 percent in most of 1980 to almost 15 percent in the fourth quarter of 1983. On the average, export credits accounted in that period for little over one-tenth of total CB's outstanding credits. Starting from the early 1984 there is, however, a sharp decline in their importance, so that by the end of the year - i.e. shortly before export credits were eliminated in January 1985 - their share dropped to less than 6 percent. As for the relative shares of the Central bank and commercial banks' export credits, the evolution was somewhat different. In effect, throughout the whole period export credits extended by the Central Bank have failed to keep pace with the volume of export credits extended by the commercial banks, and their share in total outstanding export credit has accordingly declined. - 33 - TABLE IV.2: Total Outstanding Export Credit (TL Billion (1) (2) (3) (4) Central Bank Outstanding (1) As percentage (1) As percentage of Outstanding Export Credits of Total CB Credits total export credit Export Credit Of Commercial Banks (1):(1)+(2) 1/80 26.0 17.6* 6.0 59.6 11/80 28.7 21.4 5.9 57.3 III/80 34.8 30.5* 6.1 53.3 IV/80 48.1 43.2 7.3 52.7 1/81 47.7 96.9* 7.0 33.0 11/81 73.9 128.7 9.9 36.5 III/81 94.1 177.6* 12.0 34.6 IV/81 113.2 206.9 12.2 35.4 I/82 94.7 235.9 11.1 28.6 11/82 81.3 275.2 9.5 22.8 III/82 94.5 310.7 11.3 23.3 IV/82 100.6 379.2 11.0 21.0 I/83 98.o 391.5 11.2 20.1 11/83 108.o 446.4 12.1 19.6 III/83 116.5 505.0 12.1 18.7 IV/83 181.6 547.2 14.7 24.9 1/84 108.1 510.7 9.6 17.5 II/84 107.9 567.9 8.5 16.0 III/84 70.2 536.5 5.8 11.6 Note: Outstanding credit at the end of the period. Sources: Quarterly Bulletin of the Central Bank, data supplied by the Central Bank. * Estimates. - 34 - In the beginning of the period under study the Central Bank's export credits represented more than 50% of total outstanding export credits; their share continuously declined to reach, by the end of 1984, very low 11.6 percent. Commercial banks' export credits expanded. The attractiveness of export credits for commercial banks (despite lower interest they carried) stemmed from several factors.l/ Firstly, commercial banks were receiving direct subsidy from the IRRF when extending export credits from own resources. The subsidy was so calibrated as to make the return on general purpose and export credits almost equal. For example, when the base lending rate for general ST credits was 50 percent, and for export credits 40 percent, the rebate paid to the commercial bank was 8 percentage points. Secondly, commercial banks could hold smaller required reserves on export credits (20%) instead of the normal (25%) requirement. Thereby, the banks could earn approximately 1 percentage point of additional interest.2/ Thirdly, commercial banks normally expected that the exporter would deposit his foreign exchange earnings with the bank which financed his exports. (If he did not, he would have more difficulty obtaining export credit next time).3/ The return to the bank on foreign exchange was the difference between the selling and buying price of the foreign exchange, as well as the commission on the letter of credit (when the foreign exchange is sold to importers). These two commissions added 3 to 4 percentage points to the return to commercial banks. Similarly, in condition l/ Commercial banks were obliged, in addition, to allocate 15 percent of total credits to export credits for industrial products. 2/ The rate of interest on required reserves was, obviously, lower than the opportunity cost of funds (i.e. rate of interest banks could charge their customers). 3/ About 90 percent of foreign exchange earned through exports financed by a large commercial bank was deposited at the same bank. - 35 - of real depreciation of TL, foreign exchange conveyed a real gain to the holder. Hence, even if the IRRF subsidy to banks fell somewhat short of the difference between the lending rates on general short-term and export credit, other factors made export credits fairly profitable for commercial banks. When the government's effort in financing of exports began to slacken (in 1983), commercial banks were able to step in, and keep the level of export financing, both as a ratio to actual realized exports and in real terms (see Table IV.2) unchanged, or even slightly higher. As data in Table IV.3 show, the ratio of total outstanding export credit (both CB's and commercial banks') to exports remained at an average level of 170 to 180 percent (outstanding export credits as percentage of quarterly exports). Total outstanding export credits have gradually risen as percentage of exports until the end of 1983. This tendency is clear except for seasonal spurts of exports, toward the end of the year, with the resulting decline in the ratio. However, beginning with 1984 the ratio between outstanding export credits and exports markedly declined. Only partly was it the result of a brisk increase in exports in 1984. The main factor was a rapid withdrawal of CB's export credits, whose amount, as shown in Table IV.I, was more than halved in nominal terms in three quarters of 1984. This was only imperfectly compensated by the increased commercial banks' lending. The outstanding credit/quarterly exports ratio dropped to 100 percent (in III/1984), the lowest level in four years. In real terms (adjusted for the wholesale price increase) export credits rose almost six-fold between the early 1980 and the end of 1983 (the average real rate of growth of almost 80 percent p.a.). The most dramatic increase took place in 1981 when the December 1981 stock of real outstanding export credit exceeded its level one year ago by 2.8 times (see Table IV.3). The growth was, as explained above, sustained mostly by expansion of commercial banks' credits. Export credits extended by the Central Bank, after - 36 - TABLE IV.3: Export Credits and Exports (TL Billions; at the end of the period) (1) (2) (3) (4) Total Outstanding Total (1) as X of (2) Total Real Export Credit Exports Outstanding Export Credit* I/80 43.6 50.9 85.7 100 II/80 50.1 44.7 112.1 104.9 III/80 65.3 45.0 145.1 130.0 IV/80 91.3 91.2 100.1 158.6 1/81 144.6 96.5 149.8 236.9 11/81 202.6 97.7 207.4 307.8 III/81 271.7 124.3 218.6 396.0 IV/81 320.1 218.5 146.5 445.0 1/82 330.6 185.6 178.1 410.8 II/82 356.5 191.0 186.6 423.4 III/82 405.2 218.6 185.4 457.7 IV/82 479.8 352.0 136.3 529.3 I/83 490.1 266.4 180.4 474.4 II/83 555.0 280.0 198.2 514.4 III/83 621.5 290.0 214.3 546.4 IV/83 728.8 484.9 150.3 573.9 I/84 618.8 580.1 106.7 439.4 II/84 675.8 592.7 114.0 396.0 III/84 606.7 602.9 100.6 339.1 1980 91.3 231.8 39.4 158.6 1981 320.1 536.1 59.7 445.0 1982 479.8 947.2 50.7 529.3 1983 728.8 1321.3 55.2 573.9 1984** 606.7 1775.7 34.2 339.1 * 100level in the first quarter of 1980. ** September 1984. Note: Outstanding credit at the end of the period. Exports during the period. Sources: quarterly Bulletin of the Central Bank; data supplied by the Central Bank. - 37 - increasing in real terms throughout 1981 (see Figure IV.3) grew in the following two years, at best at the rate commensurate with the rate of inflation. From the early 1984, total outstanding export credit (in real terms) began to decline. Real growth was thus abruptly reversed in 1984 as both CB's and commercial banks' credits steeply declined: commercial banks' credits decreased by 30 percent in real terms between December 1983 and September 1984 while Central bank's credits fell in the same period by more than 70 percent (in real terms). (b) Subsidy component of export credits. As in the case of export tax rebates, the importance of the subsidy element contained in the export credits will be studied for Turkey's manufactured exports. Table IV.4 shows the relationship between extended export credit to manufacturing 1/ and manufacturing exports as well as the subsidy component of the credit. The overall level of the subsidy continuously declined from about 16 percent in 1980 to 6.5 percent in 1983 and, then to a mere 1.1 percent in 1984. The decrease in the subsidy came as both the amount of extended credits comparatively to manufactured exports, and the interest rate differential between general and export credits, declined. Thus, the ratio between the extended export credit to manufacturing, and manufactured exports which, in 1980 reached almost 70 percent fell to only 12 percent by 1984.2/ Retranchment of the export credit to manufactured exports paralleled lower availability of export credit for all exports. Simultaneously, the interest rate subsidization of export credit went down from more than 20 percentage / The previous analysis dealt with the outstanding (and not extended) export credit to all exports (and not to manufactured exports only). 2/ This ratio is different from the one used above, where we compared outstanding export credits ( a stock) to exports (a flow). Here, both are flows as we have data on gross extended export credits in a period. -38- O 0a002 oeose ooooe oti o1,ooi oeos oo0 0 /~~~~~~~~~~3 0 n~~~~~ $_4 ~ ~ ~ ~ ~ ~ ~ ~~- Z N 0~~~~~~~~~~~~~~ MCT) c zz: C~~~~~~~~~ H C.aJ C 3_ X:ZI ' PL4~ ~ ~ ~ ~ ~ ~ ~ .4 CD: 00002 OlOSe O0oo ST 'os 001 009S 0O %-I - 39 - points p.a. in 1980 and 1981 to 17 points in 1984. Beginning with 1984 another element contributed to a further reduction in the subsidy: maturity of export credits was shortened from 1 year to 6 months thus directly cutting by half the already dwindling subsidy component. The maturity of export credits was reduced in order to maintain a reasonably good availability of export financing in conditions of overall scarcity of funds (and stringent credit ceilings as part of the IMF stand-by arrangement), and to reduce the amount of "leakages", i.e. use of export credits for other purposes.l/ Finally, as mentioned above, in January 1985, preferential export financing was entirely eliminated. Table IV.4; EXTENDED EXPORT CREDIT TO MANUFACTURING AND MANUFACTURED EXPORTS (TL millions) 1980 1981 1982 1983 1984 1. Total export credit to manufacturing 55060 105233 180690 269695 220305 2. Manufacturing exports 80372 250334 508537 788916 1750222.5 3. (1) : (2) (%) 68.5 42.0 35.5 34.2 12.6 4. Average interest rate differential (percentage 23.3 29.9 20.3 18.9 17.0 points) 5. Subsidy as % of exports 15.93 12.56 7.23 6.47 1.07 Source: TUB (SPO). 1/ It was alleged in effect that producers often applied for export credits determined to use them for other purposes and essentially uninterested in exports, so that when they had to fulfill the export pledge (made in order to get the credit) they were ready to accept lower prices just to sell goods. This practice was alleged to be partly responsible for decrease of Turkish export prices. Turkish Government was thus subsidizing foreign buyers. - 40 - The credit financing of export was generally higher in capital than in consumer and intermediate goods industries (see Table IV.5). Accordingly, the subsidy component of export credits was higher in capital goods industries: in 1980 it amounted to almost 37 percent of value of exports, and subsequently declined to 22 percent in 1981 and to little over 10 percent in 1982 and 1983. In 1984 it was sharply reduced to 2.4 percent. The subsidy component in exports of intermediate goods went down from 11 percent in 1980 to only 1 percent in 1984. In consumer goods exports the decline was even steeper: from 15.2 percent in 1980 to less than 1 percent in 1984. Table IV.5: Export Credits and Subsidies by Different Groups of Industries 1980 1981 1982 1983 1984 Export Credit as Proportion of Exports (%) Consumer goods 65.2 41.0 30.1 27.5 9.6 Intermediate goods 47.4 33.8 38.5 40.4 14.1 Capital goods 157.1 74.2 52.5 53.4 27.9 Subsidy as Proportion of Exports (x) Consumer goods 15.2 12.3 6.1 5.2 0.8 Intermediate goods 11.0 10.1 7.8 7.6 1.2 Capital goods 36.6 22.2 10.7 10.1 2.4 Source: Calculated from data supplied by SPO. A similar pattern emerges when we study the subsidy component for the 14 industries (see Table IV.6). Industries with a high subsidy component include metal products (about 34 percent over the five years), electrical and non-electrical machinery (26 and 15 percent) and non-ferrous metals (16 percent). On the other hand, industries with the lowest subsidy component were leather and fur (4.5 percent) and cement (6 percent). - 41 - c n ' 4 1 C- 4 O r- '0 0C s 0 1- in% .0b4 C.Cq- 4 . - 0 0' ' m &J % r, 0 0 o _ 4 N 0 10 -4 oi' O' 0 O O a, 1 a,C 41%CY, - U .%-4 en .-4 1% In- 00 4 r' 0 r..c' r-. .0 . . . . . . . CO -4 O- O4 s _ r r Ci 1 % LPN C" P 4 U 00 CS 0 I .4 4 -X4 41 v *E CS _ o o b o % C-. 0 0 N 0' a N o O0 00 4 0s _ 0 -4 -4 i _% C- r1 41 0 _1 N ~~0 *~~00 XL m 0% r- co U- co 0 C4 It # 4 on C I _ u 4 0I IT _ - -4 a, l O r' Cl -- ' 0 0 C_ LO Inc4 _4 4 n -4 4.1~~~~~~~~~~~~~~~~~~~~~~~~C a)O~~~D . 0 W" U I co C- ' -0 Cli C N .1% * en 4 0 C0.4, 14.4 0 , 0 4 0 N 4_ N co ON' 0 . 0 CDO~~~~~~.. 00 00~~~~~~~c - co0 -4 0 U,, .-4 a, In In 4- 0 N 4 ' C Ca) 0% LA 0 4 '0 0 N '0 N _ N '1% 4M 0' C- '0 0 6 ~ 1 NJ 41% ' -4 0 .. s "4 .-4 - li~4e W04 "4 4 < - N O X 00 41% N C- '0 en CI CO C" 0 0 0 CD 1% Nt 0 C 4 0e .0 C) '0 C N N CO .-4 -4 .4 -4 _N N -4 c_ N 4 '0~~~~c -4 O0 _% 0 C O .O - 1 N N 00 N 0O 0 - C w ,4 O. It on _0 .4 N '0 - c 0> In Cq a) 00 0%~~~~~~~~~~~~~~7 0 C-. r- t co cl C-. 0 0' . 0 4z 00 Nl '0 '4 0 CD O D X )- N u _ _-4 "4 4 C N 0 10 N 'P co 4 0 co O~~~C1 4 41 4 41% 41 00 0' '0 0 00 4 Cl '0 o _ D . . . .I..... 0n - C- C- '% 4 0' 41 -4 '0 4 0 0, N- PC C4 N 4 O N '0 Cl C- N -4 CC 4-.In .. _4 U CU qz~ ~ ~~a a)4zC w 4~~~~~~D . U~ U i 0 04 1 -4 a) 4 - S _ 0 0 . _ D X . 4-' 0 4' U U S. 0 ID 4 0 a) O4 0 CD .-4 a) U t: &. ) 4 4 0. a) 0 ~ " .0 4. .4 a) CD 0 4 a) 4.4 o J o x CD 0. a) .0 IV 0 . a) . 4. 0o a a C . 0 a 44 0 0 0 0 e E-4 - a 0.. C. e si -4 _ z e2 i- . C - 42 - V. Foreign Exchange Allocation and Duty-Free Imports. Exporters who hold TUB's Export Encouragement Certificates or letter of credit can apply to get foreign exchange for their import needs and/or receive the right to duty-free imports up to the amount of the allocated foreign exchange. The latter, until the end of 1983 could not normally exceed 60 percent of the pledged export value 1/. After December 1983 this proportion was reduced to 40 percent. For exporters requesting more than 60 (or, later, 50) percent of export value as foreign exchange allocation, special screening was made. In that case also, the tax rebates were not paid on the whole amount of exports but only on the net foreign exchange earned. In view of importance of tax rebates for exporters, the measure discouraged high claims on foreign exchange. The incentive value of the foreign exhange allocation scheme is twofold. First, it gives right to duty-free imports of intermediate and raw materials. The amount of duties thus saved represents the subsidy. Second, foreign currency generally commanded a premium over its official rate, which could be appropriated by the recipient. However, as the overvaluation of the 1/ It does not include the exporter's own foreign exchange which might have been retained from his previous exports earnings. After the liberalization of foreign exchange accounts in December 1983, allowing Turkish citizens residing in Turkey to hold them, all foreign exchange - no matter how acquired - owned by the exporter could be used for imports (provided, of course, import regime restrictions, e.g. if the imported good requires a license or is banned, are cleared). - 43 - TL gradually decreased the premium diminished. It is estimated that the premium which in 1980 exceeded 12 percent, declined to less than 7 percent in 1981, and then 4.5 percent in 1982. In 1983 the premium again increased (to an average of about 13 percent), and then was virtually eliminated in 1984. Also, it became more difficult to estimate the premium since the "flex-price" regime whereby exporters could transfer their foreign exhange to their suppliers, and presumably charge them the full market price, was replaced in 1982 by the system where the exporters could use the foreign exchange only for their own input needs. If a firm could not obtain foreign exchange, the premium theoretically became highly specific, e.g. equal to, say, loss of output due to inability to acquire imported inputs. The foreign exchange premium was thus highly individualized whereas, across the economy as a whole, it became much smaller. In our calculations the subsidy element due to the foreign exchange premium is assumed nil. Table V.1 shows the evolution of the foreign exchange allocation scheme for the manufacturing sector. The estimated average subsidy element (the value of the subsidy measured against the total exports) remained between 4 and 5 percent in the 1980-1983 period and then dropped to 3 percent in 1984. The foreign exchange allocated as a proportion of exports increased from between 15 to 17 percent in 1980-82 to 25 percent in 1983. It was not accompanied by a similar increase in the subsidy element because the structure of the allocated foreign exchange shifted toward the industries with relatively lower import tariff rates, where the subsidy conferred by the saved duties, is smaller. The decline in the subsidy component of the scheme in 1984 occured both because foreign exchange allocations were less (16 percent of export value vs. 25 percent in 1983), and also because import liberalization lowered tariff rates by, on the average 20 percent, so that the worth of saved duties was also smaller. -44 - * N~~~~~~~~~~~~~~~~~~~~ N 04. N - N U N 0 U) ~ ~ N N ) 4 .0j . 0~~~~~~~~~~~~~~~~~~~~~~~~~0 410 - 01 .30z 'a N 40 0 ID 4 v 0 0 ~ ~ ~ ~ ~ ~ ~ ~ ~ %'a= . - ~ ~~~~~mN co 4M .0 1.3 ~~~~~~~~~~~~~~~~~~~~~~~~~L0 C a 4 0 - N ) 9*0 bo n 04. * 40 NWmO N 0 N - N ) 0 0 0. 4 U)~ ~ C. @4 N 0 4 N N - ) 0 U @44CM -~~L N ) N N 4 4 U ) N N0 0 N 0 @ 0 4 NI -j C 0~ ~~~~j ENN N6 ZaM - 45 - In terms of individual industries subsidies are heavily concentrated in four industries: non-ferrous metals, metal products, electrical machinery, and transport equipment. These four industries receive by far the largest subsidy (about 20 percent of the export value over the five years), although their share in total allocated foreign exhange is broadly commensurate with that of other industries. But some of them (e.g. metal products) have low exports so that the allocated foreign exchange (and the subsidy) when expressed in terms of exports become quite substantial. In effect, foreign exchange allocations to the metal products industry exceeded its exports in 1981 and 1983. The foreign exchange allocation subsidy component is significantly greater in capital goods industries (it was relatively low in 1984, 12.4 percent, but exceeded 26 percent in 1980), than in the consumer (about 2 percent) and intermediate goods industries (between 4 and 6 percent). It is due both to a greater allocation of foreign exchange to capital goods industries (see Table V.2), and to higher tariffs on inputs used by them. All major industrial groupings show the same pattern in terms of the behavior of subsidy rates as the manufacturing sector as a whole: slight decline in 1981 and 1982, and an increase in 1983, bringing the subsidy conferred by the scheme to the level at which it was in 1980. In 1984 the subsidy and foreign exchange allocation as share of exports declined in all of them. Table V.2: Allocation of Foreign Exchange and the Subsidy Component (as % of exports) 1980 1981 1982 1983 1984 Foreign Exchange Allocation Consumer goods 8.97 9.32 8.98 12.39 7.86 Intermediate goods 19.06 16.63 24.16 40.69 26.44 Capital goods 64.20 44.68 33.63 47.81 35.32 Subsidy Consumer goods 1.98 2.14 2.17 2.75 1.43 Intermediate goods 6.57 4.35 4.23 6.07 3.37 Capital goods 26.43 17.44 14.16 20.17 12.43 - 46 - VI. Real Exchange Rate Figure VI.l depicts the evolution of the real exhange rate of the Turkish lira since the January 1980 devaluation 1/. It clearly shows the three periods characterized by a rapid real devaluation (devaluation is shown by an increase in value). The first begins with the daily adjustment of TL by the Central Bank in May 1981, and ends by the close of the year. Between these two dates TL depreciated in real terms by about 18 percent. The second period, of a somewhat less sharp devaluation, extends from April 1982 to August 1982: during that time TL depreciated by 7 percent in real terms. This period coincided with the calming down of inflation in Turkey (reduced to about 22 percent per annum), while the rate of nominal depreciation continued more or less as in the past. For a whole year afterwards the real exchange rate remained stable: practically no real devaluation occured between August 1982 and July 1983. Beginning with December 1983 real exchange rate began to depreciate fast again. In January 1985, in particular, the real exchange rate depreciated by about 7 percent. This was due to the change in the exchange rate regime whereby commercial banks were allowed to trade foreign currencies at the spread of 6 percent below or above the daily rate fixed by the Central Bank. (The spread was later increased to 8 percent). As the banks tended immediately to push the rates to the upper bound of the margin, this provided for a one shot devaluation of the lira. Since April 1984 the process was reversed, and there was a steady real appreciation: by December 1984 the real exchange rate of TL had appreciated - comparatively to its level one year ago - by about 4 percent. 1/ The real exhange rate adjusts the nominal rate for the inflation differential between Turkey and the five countries (USA, UK, France, Germany, and Switzerland). The weights used are trade shares of each of the countries in total Turkish trade in 1981. -47- 0 0 ~~~~~~0 S ) 5-4 0 0 C) * ~ - *- 0* *GO C" 00 CD 0D U ~~~~~~~~~~~~~~~~a: Cn -O M E- ) CD cDal * ~ r N )c _ _ _ CD 0)~~~~~~~~~~~~~~~C CD~~~~~~~~~~~~~~~~~~~~C C~~~~~~~~~~~~~)~~~~C Cii C) 0 C3 0 03 0 0 0) GO0 9-4 -~~~~~~ -4 -4 - 48 - VII. The Combined Export Subsidy: Optimal and Actual (a) Total export subsidy. Total subsidy element to exports provided by the three incentive schemes discussed above (tax rebates, preferential credit, and foreign exchange allocation for duty-free imports) ranges between 23 (in 1983) and 15 (in 1984) percent of the value of exports (see Table VII.1). In effect, this proportion was fairly stable, at 20-21 percent for three consecutive years (1980, 1981, 1982), and then increased slightly, to 23.4 percent in 1983. In 1984, however, for the reasons explained above, the subsidy declined by more than a third to 15 percent of the export value. The subsidy shows a seasonal pattern: it was often high in the first quarter of the year, and relatively low in the last quarter. This was due to a seasonal behavior of both subsidies and exports. Turkish exports in volume terms show a tendency to accelerate in the last quarter of the year. This has, combined with a depreciation of the lira, which was also particularly pronounced toward the end of the year, led to a significant increase in TL value of exports. The nominal value of subsidies on the other hand, has tended to be the lowest in the last quarter due to the end of the year stringencies. These include the need for the Central Bank to remain within the credit ceilings and thus check the expansion of the discounted credits, as well as shortage of funds to pay for tax rebates. It has thus led in 1983 and 1984 to the build-up of important arrears in both tax refunds due to exporters and interest rate rebates for export credits due to commercial banks. The exception to this seasonal pattern was 1980 (when as the different incentive schemes expanded the subsidy to exports increased in every quarter), and 1984 (when a very sharp decline in tax rebates occured in the third quarter). - 49 - Table VII.1: Total Subsidy Rate (TL Bill) (1) (2) (3) Total Subsidy Total Exports Total Subsidy Rate (1):(2) (%) 1/80 2834.29 15o88.29 18.07 II/80 3387.5b 17716.79 19.12 III/80 3823.89 164b3.17 23.23 IV/80 7677.94 30504.54 25.17 1980 17723.68 80372.79 22.05 1/81 9235.11 37269.38 24.78 11/81 10966.64 51069.47 21.47 III/81 13824.22 61952.99 22.31 IV/81 17293.84 100042.38 17.29 1981 51319.81 250334.22 20.50 I/82 23114.85 90452.79 25.55 II/82 29997.51 115124.95 26.06 111/82 21316.15 122900.43 17.34 IV/82 30365.47 180058.86 16.86 1982 104793.98 508537.02 20.61 I/83 4U512.17 149259.83 27.14 II/83 43125.56 181045.50 23.82 IlI/83 42567.47 180772.43 23.55 LV/83 58325.63 277838.80 20.99 1983 184530.83 788916.55 23.39 I/84 64663.80 401985.86 16.09 11/84 77436.91 418905.55 18.49 III/84 33734.24 409004.98 8.25 IV/84 88001.4b 520326.12 16.91 1984 263836.41 1750222.50 15.07 - 50 - Table VII.2 shows the evolution of the structure of the export subsidy. It is characterized by an increase in importance of tax rebates and a reduction in importance of export credits. The value of export tax rebates in overall subsidies increased from nil in 1980 and less than 18 percent in 1981 to about 50 percent in 1982 and 1983. In 1984, despite scaling down of the scheme, its relative inportance further increased to almost three quarters of total subsidies given to export. It was, of course, due to an even faster decrease in importance of other subsidy components, and particularly preferential export credits. The subsidy element contained in preferential credits sharply decreased from representing 75 and 60 percent of the overall value of subsidies in 1980 and 1981, to only 31 percent in 1982, and then to 28 percent in 1983. In 1984 preferential export credits accounted for only 7 percent of the overall subsidies to exports. The value of duty-free access to imports has in relative terms remained at a fairly stable level, around 20 percent of the total subsidy to exports. - 51 - Table VII.2: The Structure of the Subsidy Tax Rebate Export Credit Duty-free Imports percentage percentage percentage percentage percentage percentage of export of total of export of total of export of total value subsidy value subsidy value subsidy 1/80 0.0 0 12.10 67.0 5.96 33.0 II/80 0.0 0 12.60 65.9 6.52 34.1 III/80 0.0 0 18.69 80.5 4.53 19.5 IV/80 0.0 0 20.11 79.9 5.06 20.1 1980 0.0 0 16.60 75.3 5.45 24.7 1/81 3.12 12.6 16.49 66.5 5.87 20.9 II/81 3.71 17.3 11.99 55.8 5.77 26.9 III/81 3.82 17.1 13.45 60.3 5.04 22.6 IV/81 3.58 20.7 10.72 62.0 2.98 17.2 1981 3.60 17.6 12.51 61.0 4.39 21.4 I/82 10.85 42.6 10.25 40.1 4.46 17.4 II/82 10.62 40.7 10.63 40.8 4.81 18.5 III/82 9.96 57.4 3.68 21.2 3.70 21.4 IV/82 9.41 55.8 3.50 20.8 3.95 23.4 1982 10.07 48.9 6.36 30.9 4.18 20.3 I/83 11.26 41.5 9.65 35.6 6.23 23.0 II/83 11.32 47.5 7.18 30.2 5.31 22.3 III/83 11.48 48.7 6.46 27.4 5.61 23.8 IV/83 11.75 56.0 4.30 20.5 4.94 23.5 1983 11.50 49.2 6.47 27.7 5.42 23.2 I/84 11.09 69.0 1.48 9.2 3.51 21.8 II/84 14.94 80.8 0.73 3.9 2.81 15.2 III/84 3.82 46.3 1.25 15.1 3.18 38.6 IV/84 13.65 80.7 0.87 5.1 2.40 14.2 1984 11.07 73.5 1.07 7.1 2.94 19.5 -52- * * ** * O ° Lf C U) C) C) LA C -L - L) tl] I I - ,1. _ ,,, v Gi -.2 -.sJa E- .-"-' o' ( x,,~~~~~~ .1 O .~~ ~ ~~~~~-' / X P .'< ts-- VV * t: . __ > n n~~~~~~~~~~~~~r cr G~~ ~~ ~ ~~~~~~~~~~~~~~~~ la a)_ |_~~~~~ / LO z~~~~~~~~~~4 . 0~~~ .* * .. * - * A IA LL]~~~~~~~~~~~~~~~~~~~~~c CD 5 -/~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~% - S.- * *~~~~~~~~~~~~~~~~~~4)C C: ..C.-C. -- C* * cf) *~~~~4f SD - 53 - Figure VII.1 shows the importance of each of the three subsidy schemes in the period I/1980-IV/1984. As observed earlier, the total subsidy granted to exporters of different types of goods has differed: it has consistently favored exporters of capital goods over exporters of intermediate, and in particular, consumer goods (see Figure VII.2). The subsidy component has thus been inversely proportional to the importance of the type of goods in total Turkish manufactured exports. As Table VII.3 shows, consumer goods which have accounted for between 53 and 63 percent of the total manufactured exports received on the average only about 45 percent of the total subsidies. On the other hand, capital goods whose share in manufactured exports was on the average, less than 10 percent, received about 20 percent of the total subsidies. The effect was that in the period 1980-83, the yearly average subsidy in terms of export values ranged from a high 37-65 percent to exporters of capital goods, to a moderate 18-27 percent to exporters of intermediate goods, and to relatively low 16 to 18 percent to exporters of consumer goods. In 1984, subsidies to all types of goods declines, although the ranking between the main groups of industries remained unaffected: capital goods still received the highest relative subsidy (about 24 percent of export value), followed by intermediate (18 percent), and consumer goods industries (12 percent). The range obviously was even wider when measured for individual industries (see Table VII.4). Exporters of metal products, electrical machinery, and non-ferrous metals which combined represent 4-5 percent of the total manufactured exports, received by far the highest subsidies, averaging approximately 94, 62 and 49 percent of export value respectively. On the other extreme, food products and textile exporters, whose exports account for more than one-half of manufactured exports, received on the average exports subsidies of about 11 and 19 percent only. - 54 - Table VII.3: Exports and Subsidies (in percent) 1980 1981 1982 1983 1984 Consumer Goods Share in total manuf. exports 62.6 55.5 52.9 57.2 56.5 Share in total subsidies 50.0 44.7 40.7 43.8 44.7 Average rate of subsidy 17.6 16.5 15.9 17.9 11.9 Intermediate Goods Share in total manuf. exports 28.4 34.1 36.5 33.6 35.8 Share in total subsidies 23.4 32.8 40.5 38.4 43.1 Average rate of subsidy 18.2 19.7 22.9 26.7 18.1 Capital goods Share in total manuf. exports 9.0 10.4 10.6 9.2 7.7 Share in total subsidies 26.6 22.5 18.8 17.8 12.2 Average rate of subsidy 64.7 44.0 36.6 45.3 23.9 -55- o to o o o o o CD (f)D CM C CD Li: GC(r) XC,E. @' '0 t = ( a _ LLCD . . .~~~~~~ .5- Foo COE > L.JI * -, CE:~~I a) E-4 0 C) 0) 03 C; Ui 63 S OD 0o 04 00 56 - 4 -e t C4 t- . ,- Ns Ns o 4 uo 0 X $4O - 0% 0 e.n cn N Cn -.'. N -4 0 N N4 N 004. N1 fn ,-4 P% ID '-4 C.- N N4 -4 - % .' r 0.00c C 10 a, In 0 '0 0 - L^ 0% N 0% U% Il E40CI '- 4 - --4 C N N '0 -4 N N - $4COI 0% _. t% N c_ N 4. _ N N N 0 N N . o to4 ' C N r C C- 0s '0 _O 00 $4, S o e4 * N I 0 s cs o e% N 0j to X~ _ ~ i -O 444.J col '0 0 L0% 0% C4 0 % 0 * U-~ 0% C.. 00 o 14 C: 4 0 14 _n _ od ~ ~ 4 -0 CN 'D _n a ,¢ N 'N ID 0 CO r, -.1 c~ ~ o 4 a, o l _ f uo It ID _ 0 coX s _ ol co It cn 00 ID E n a: _4 Cs CS 4 N C.' C'0 -4 -4 0 0 o n N CO .rl - - a 0 r, , C1 cn ID os e- C- Z o e o ,> uo - 0 o 0 > F4 (k) ' . 0 N 4 U- '0 _ 0 : Is 4 o% 44 440 N N -4 N 0 j N N u O 4 0 N 4 Cl O X fi 0.00 NS 00 0 :o '0 o% C 0 N .- /% l _ o n~~~~~m CLc _~~~t C 0 C 4 ~ ~~~~~~ _% -n r- r; C1 z 0 , 50 44~~~~~~~~~~~~~~~~4i co O4 to VS _4 Ch 4 I i n VI en a - a, 1- 0 4 0 - N4 C '0 _ 0__ N Cl rl . 00 $40 C0 O O - l N 4 Cl 4 0% _ O4 0 S _l . O S.0A N Xl tN 4 *C 0 lN no _4 -) 'D : N 00 CN 1- 'D 00 0 N C 00 N 00 0 401 I l 4 U1 C4 4 4 N C- _- 4 0 NS 0O O.0 0 0 % 40 .0 '0 4 '0 'D CN e Cn '0 PI N FQ~~~~4 c __DNe>o r. 8 's 4.4 ~~~~~~U 0 0C bo 0 _ 0 $4 Q 0 .0 -- $, :.J .,4 0 -4 U O $44 44 0 UA 0 0 0 > u u0 e Q Q : 0 0 _- 0 - C-4 0 0 - $ 4 C 04 0 ^ 0 0 0 ' 0 O A 0 'i 4. 0 e4 U U CQ 0 x $ 0f .O e O Ai 0 0 0 O 14 0 0. 0: .0 0 GZ 0 O : 0 0 4 X4 -H 04 :> Z) 4; Z 1: Z :£ $-4 P - 57 - (b) Optimal and actual subsidies. The lopsided structure of export incentives (e.g. directed toward industries with relatively low exports) may yet be the optimal one if it encourages the industries which may possess a comparative advantage. The optimal structure of export incentives should be such as to stimulate exports of products which are just on the verge of the comparative advantage and/or which in the future seem likely to develop a comparative advantage 1/. In other words, the policy should be concentrated on industries where the export subsidy can make a difference between an export success and an export failure and should deemphasize support to industries which are either very competitive (as to be able to sell internationally on their own) or very inefficient (so that almost no amount of subsidy is likely to help them move towards the more competitive range). Table VII.5 thus divides all the manufacturing sectors into four groups. First, industries where additional strong export support may be justified, i.e. those with Domestic Resources Cost (DRC) between 1 and 1.4. Second, those where a medium export subsidy may be needed to reinforce the comparative advantage, or offset subsidies that foreign competitors may be receiving (those with DRC between 0.7 and 1). Third, industries which should either be able to fend it off alone (with DRCs less than 0.7), or those which are fairly inefficient (with DRCs between 1.4 and 3) so that in both cases only a very modest export subsidy may be warranted. And, fourth, industries which are highly inefficient with the result that any export support may be deemed wasteful (with DRCs in excess of 3). Compared to that scheme of "optimal" subsidies is a classification of industries according to actual subsidies received. 1/ The assumption is that the industries in the "grey area" of the comparative advantage may, by judicious export incentives, be helped to secure a "bridgehead" on the foreign market, and then, by expanding output, reduce their unit costs, and thus establish, on a more permanent basis, their export competitiveness. - 58 - Table VII.5: Relative Level of Export Subsidies Acc. to comparative Actuals advantage 1980 1981 1982 1983 1984 (1) Food and beverage Low Low Low Low Low Low Textiles Medium Medium Medium Medium Medium Medium Leather and furs Low Low Medium Medium Medium Medium Paper Nil Medium Medium High Medium Low Chemicals Nil Low Medium Medium Medium Medium Rubber, plastic Low High Medium High Mediu High Glass Low Medium Medium Medium Medium Medium Cement Low Medium Medium Medium Medium Medium Iron and Steel Nil Medium Medium Medium High High Non-ferrous metals Nil High High High High High Metal products High High High High High High Non-electrical machinery High High Medium Medium Medium Medium Electrical machinery Medium High High High High High Transportation equipment High High High High Medium High Absolute Discrepancy 0 13 15 17 16 16 Note: For Column 1 (the optimals): High subsidy for industries with DRCs between 1 and 1.4; medium if 0.7 DRC 1, low if 1.4 DRC 3 or if DRC 0.7, nil if DRC 3. DRCs refer to 1981 and are taken from F. Yagci, Production and Incentives in Turkish Manufacturing. For actuals: The level of export subsidy is calculated in terms of the average annual export subsidy for the whole manufacturing (in order to adjust for variations in the average rate of subsidy and concentrate instead on the relative export incentive given to various industries); high subsidy if 30 percent and more above the average subsidy: medium subsidy if between 30 percent below or above the average (i.e. between 0.7 and 1.3 of the average), low if more than 30 percent below the average. Absolute discrepancy shows the number of levels by which the actual subsidy differs from the "optimal." The ranking is High-Medium-Low-Nil. If the "optimal" subsidy is high, and the actual is low (or vice versa) two "discrepancy points" are assigned; if the "optimal" is medium, and the actual high, once discrepancy point is assigned; if the rankings coincide no discrepancy point is assigned. - 59 - Table VII.5 illustrates the following points: (1) The level of the discrepancy between the actual relative export subsidies to industries, and the subsidies based on the estimate of the comparative advantage (the "optimal" subsidies), increased up to 1982, and then decreased marginally in 1983 and 1984. It went up from one-third of the maximum distance from the optimal structure of subsidies to almost one-half 1/. (2) The industries where the departure from the optimal level of subsidy is the largest are, in descending order, non-ferrous metals, iron and steel, paper products and chemicals . They all received export subsidies in excess of what seems to be warranted on the basis of their estimated comparative advantage (DRCs in all of them are in excess of 3). 2/ The industry which, on the other hand, should receive larger subsidies is non-electrical machinery. The discrepancy between the actual and the optimal level of subsidies by industries is also shown in Table VII.6. The plus (minus) sign in the first column indicates that the industry has enjoyed more (less) incentives than suggested by the comparative advantage analysis. As mentioned above, the industries which seemed most favorized were non-ferrous metals, iron and steel, paper and chemicals. The second column in Table VII.6 gives the average rates of growth of exports (in value terms) in the 1980-1984 period. 1/ The maximum distance from the optimal structure is 36. 2/ This conclusion must be somewhat qualified for chemicals where there are products in which Turkey does have a comparative advantage. One would need to know a more disaggregated structure of export subsidies in order to determine what proportion of them went into the comparative advantage area. For non-ferrous metals, and iron and steel industries, the Production and Incentives in Turkish Manufacturing by F. Yagci (1984) study found production in most lines to be inefficient at world prices. - 60 - We see that three of the four most favorized industries display a higher than average rate of export growth (iron and steel, non-ferrous metals, and paper). It thus suggests that the level of subsidies given was sufficient to compensate for lack of export competitiveness. Table VII.6 Discrepancy between Actual and Optimal Subsidies and the Growth of Export Value by Industry Average rate of growth (1980-84) Discrepancy (X p.a.) Food and beverages 0 31.9 Textiles 0 34.5 Leather and furs +4 50.3 Paper +10 48.6 Chemicals +9 18.9 Rubber, plastic +8 37.0 Glass +5 28.6 Cement +5 -2.1 Iron and steel +11 70.7 Non-ferrous metal +15 40.5 Metal products 0 13.7 Non-elect. machinery -4 38.5 Elect. machinery +5 53.0 Transp. equipment -1 20.5 Total +67 35.4 Note: Average rate of export growth calculated by the least squares method. - 61 - Table VII.7 compares the relative levels of subsidies by industries (relative with respect to their "optimal" level determined on the basis of comparative advantage) and export growth in the 1980-1984 period. We see that for four industries (leather and fur, paper, iron and steel, and non-ferrous metals) high subsidies "paid off" in terms of higher than average export growth registered by these industries although DRC calculations show them to be non-competitive. For electrical machinery the situation is somewhat different as the industry was found to be fairly efficient (DRC = 0.9), and, thus, it might be assumed, high growth of exports could have been achieved with less subsidies. On the other hand, chemicals, glass, and cement received incentives in excess of what seemed to be economically warranted, while their growth was below the manufacturing average. These industries are thus prime candidates for a cut in the level of incentives. Conversely, non-electrical machinery recorded a respectable export growth although the relative level of incentives was low. This would suggest that an increased support of that industry could pay relatively high dividends in terms of export growth. Table VII.7: Relative Level of Subsidies and Export Growth Relative level of incentives/ Too About Too Export growth Low Right High Slow Metal Prod. Chemicals Transp. Equip. Glass Cement Average Non-elect Food Rubber, plastics machinery Textiles Fast Leather, furs Paper Iron, steel Non-ferrous metals Electrical machin. Note: Relative level of subsidies is determined from the discrepancy between actual and optimal levels in Table VII.6: all industries where the divergence is greater (smaller) than +4(-4) are considered as receiving too high (too low) incentives. Export growth is considered fast (slow) if the rate of growth exceeds 40 percent p.a. (is lower than 20 percent p.a.). Data are taken from Table VII.6 - 62 .- VIII. Subsidies, Real Exchange Rate and Volume of Exports. The relationship between the two forms of incentives to exporters (direct subsidies and real exchange rate devaluation) is shown in Table VIII.1. A broadly unchanged level of direct subsidies (until 1984) was accompanied by a steady, although staggered in time, real devaluation of TL. In 1984 real exchange rate of TL has devalued by about 13 percent comparatively to 1980. The level of total direct subsidies (from tax rebates, preferential credits and foreign exchange allocations) has, between 1980 and 1982, remained more or less unchanged at between 20 and 22 percent, before increasing to 23.4 percent in 1983. In 1984 it declined to 15 percent. The real effective (or export) exchange rate (which is equal to the real exchange rate augmented by the rate of direct subsidy to exporters) declined in 1981 by about 7 percent from its average level during 1980.1/ In the next two years it increased by respectively 8.5 and 5.5 percent (see Table VIII.1), before declining by 3 percent in 1984. During all these three years the level of the real export exchange rate remained above its level in the beginning of the period (1980). Yet the downward movement in real export exchange rate begun in the second half of 1984 (reduction of direct subsidies was accompanied by a real appreciation of the TL) meant that by the fourth quarter of 1984, the overall level of direct and indirect export incentives was about the same as in the first quarter of 1980. Figure VIII.1 shows the level of the real effective exchange rate, and the breakdown between indirect (exchange rate devaluation), and direct incentives (the subsidies). 1/ Decrease in the real (effective) exchange rate implies lower level of incentives for exports (See note at Table VIII.l). - 63 - Table VIII.1: Direct Subsidies, Real Exchange Rate and Volume of Exports (per quarter) (1) (2) (3)=(1)+(2) (4) Total Direct Real Exchange Real Effective Volume of Exports Subsidy Rate Exchange Rate (in million (% of exports) (1/1980=100) 1980 US$) 1980 I 18.07 100 118.1 228.31 II 19.12 91.0 110.1 233.83 III 23.22 92.6 115.8 205.29 IV 25.17 86.4 111.6 338.04 1980 22.05 92.5 114.5 1005.47 1981 I 24.78 82.5 107.3 366.57 II 21.47 83.3 104.8 441.26 III 22.31 87.4 109.7 469.06 IV 17.29 95.4 112.7 688.86 1981 20.50 87.1 107.6 1965.75 1982 I 25.55 94.6 120.2 560.44 II 26.06 93.6 119.7 660.47 III 17.34 97.0 114.3 630.61 IV 16.86 99.4 116.3 861.11 1982 20.61 96.2 116.8 2712.63 1983 I 27.14 96.1 123.2 666.22 II 23.82 98.5 122.3 746.11 III 23.55 101.4 125.0 667.58 IV 20.99 103.4 124.4 899.86 1983 23.39 99.9 123.3 2979.77 1984 I 16.09 110.3 126.4 1101.91 II 18.49 105.1 123.6 1005.96 III 8.25 102.4 110.7 889.76 IV 16.91 99.8 116.8 1036.90 1984 15.07 104.5 119.5 4034.53 Note: Increase in real exchange rate implies real devaluation. Increased incentives to exporters thus translate in higher real (and effective) exchange rate. -64- o c}o o o U S S 5 ~~ ~~ ~~~~~~~~~~~~0 0 03 o Zo o s a a - 0 0 0N 0 0 0 C- 0-O - - -a a) aD l: l *_ . . . . . . .01 IL] I~ ~ ~~~~~~ IL]~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~r Q:f CE:U C-)~~~~~~~~~~~~~~~~~~- nj~~~~~~~~~~~~~~~~~~c w ) / C.)~~~~~~~~~~N M e IV V N _ - C) 0) OD N _4 _ 4 _ 4 _ 4 -4 - 65 - Table VIII.2 depicts the relationship between the real exchange rate and volume of manufacturing exports. The equations relate volume of exports (of all manufacturing, consumer, intermediate, and capital goods) to the level of the real effective exchange rate for the corresponding category. These results show: (a) a close correlation between the volume of manufactured exports and the level of incentives; (b) that a one percent increase in the level of incentives (coming either through a real exchange rate devaluation or an increase in direct subsidies) is associated with an increase in volume of exports in the same quarter of 2.8 percent; (c) that the responsiveness of manufacturing exports to incentives is higher in consumer than in intermediate and capital goods industries. This suggests that an increase in the level of direct export subsidies in consumer goods industries (where they are the lowest) may be viewed as a relatively efficient way to further export growth. Table VIII.2: Relationship between the Level of Incentives and Volume of Exports Ln (Exports) = B. Ln (Real Effective Exchange Rate) B1 R2 (SE) Total Exports 2.78 0.999 (1) (0.46) Consumer Goods Exports 2.69 0.990 (1) (0.42) Intermediate Goods Exports 2.55 0.998 (1) (0.55) Capital Goods Exports 2.20 0.997 (1) (0.66) Note: Real Effective Exchange Rate = Real Exchange Rate plus Direct Subisides. Real depreciation of TL translates as an increase in real exchange rate, so that incentives to exporters go up when either TL devalues in real terms or direct subisides increase. Export volume per quarter in thousand of 1980 dollars. The equations are fitted through the origin on the assumption that at an extremely overvalued exchange rate (when the real exchange rate approaches zero) exports would be nil. Numbers in parantheses show the level of significance of t-values. - 66 - IX. Subjective Perception of the Importance of Incentives The previous sections provided an estimate of the importance of different direct and indirect incentives to Turkish exporters. However, it is indispensible to compare "objective" importance of the incentives (calculated as the subsidy component of total value of exports), to the perception of their importance by the industrialists. This essentially raises the following problem: if the perception of particular incentives differs significantly from their calculated "true" incentive value, the policy based purely on an estimate of the "true" incentive value is bound to be fairly inefficient. It is true, that in the long-run, the "objective" worth of an incentive and its perceptions would tend to coincide, but wide discrepancies may exist in the short-run, the period with which the economic policy is mostly concerned. As perceptions, rather than the "objective" calculations of the averages, determine the behavior, it seems useful to complement calculations of the "true" value of incentives by interviews assessing how different incentives are perceived by those at whom they are aimed. Table IX.1 shows schematically how the export incentives policy was viewed by a sample of six exporting firms. The firms in the sample are export-oriented: they exported about 40-50 percent of their output in 1984 when the interviews were made. Their total exports in 1984 were estimated at about $170 million (2.5 percent of total Turkey's exports). All of the firms registered fast export growth in the period 1980-84; in effect, for all of them the shift toward exports occured around 1980 and 1981. By far the most important "export shift factor" (cited by all the interviewed firms) was depressed domestic demand after the lauching of the stablization program in 1980. The firms were thus clearly compelled to look abroad for outlets to their production. This, in particular, since most of the firms completed - 67 - significant investment programs (financed with relatively cheap foreign exchange) in the mid- and late-1970, so that both their capacity and the overhead costs were relatively high. Conditions of depressed domestic demand were brought to bear on the firms even more through unavailability and/or high cost of the general short-term credit. The opening up of Middle Eastern markets provided, for some of the firms, a timely respite from the stringencies of the austerity policies at home. - 68 - X~~~~~~~I r V~~~~~~~~~~~~~~~~~ 0~~~~~~~~~~~~~~~~~~~~~~0 X o 2 _coC c 0 U 4) K .C b v° rU @ a ~ ~ ~~o _ &,, V 4)4) ~~0 '-4.4 4co4 4 A ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~W - _ W W la _ a E > ~~~~~~~~~co W b 11 U a _ f W go 44 .0 0 V .ou~~~~~~~~~~M M~ .44C:4 4) 10 ) * c 0 0 - - O4 bU 4U Z O g 0 1 Oa 0. 40 . ° 0 b b Y C X .^ 0 _e ~~~~~~~~~~~t U 3 I la #C 4)4W oo 4))coW Or ~~~~~~~~~~~~~~~~~4 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ " U *40 4) C 0 "4 .4 .~~~~~~~~~~~~~~~- V4 4) a *o X e c ° .44°e° 0 V 0 4)40 * .4 be 4 14 0 c 4) v e 340 oF 44) u.4 4440~ U o) U) 0 o C3 In W E " 4) u4 140 . 0 u .1 0 0 O a ~~~~14 ~~~~ 4) ~ ~ ~ ~ ~ K4)0 40~04 4 _ 0 O '0 _ O E 1-' >- U ::1 O K 4) 14Co~.V "4 ) 0r I co 1r -44 40 0.14 404)~ ~~~~~~~~ 4)4). u 4) 4)4) e14 . 4W 0 0U e.40 1 > .4 C4 W ( ' 4-4 U. - D"W~ ~ I0 co x f _4 0 4 E to 4o W W co aa eX e 8 ^rC^ .0 0 * .4~~~~~~~~~~~~~~ ~~~~4).F. x .4 * e 1> 0 4) e V .4 0 X' " 'o W . 3SX -M u M la I, A i@ X X~~~~~W > c co M.......oc>o-o.. C.44 0 . ~~~~~~~~~f, 0 A )0 to 4 4) A 040 Ai.c 40v 4) )0. 40 W . 0 M 2 ~~~~~~~~~ C u W il E o 0 0.1.0 0.14 OV.0~~~~~~~C 0 14 a44 .4 ~ ~ ~ ~ ~~~~~a co u ba 0 £ ° < o o yco 4)~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 4 "4 0 0 1 V 0 X *X X E 0 o 4) 4 4)a .4 s 0 2 .444 04044W u - 69 - When it comes to individual export incentives, the interviewed firms were generally concerned (three out of the six) by the reduction of export tax rebates. Firms which were not affected by their gradual removal are in the agro-processing where tax rebates were relatively low even before the reduction. Cost of export credit was perceived to be high as the differential between the cost of the general and the export credit was being reduced. It is a high value of that differential which was, as we have seen, responsible in several instances for a shift toward exports and away from the domestic market. Yet the decrease in the subsidy (differential) alone did not seem likely to lead to a retranchment of exports. Finally, most of the firms enjoyed duty-free imports of inputs used in production for export. In the ranking of export incentives by their perceived importance, export tax rebates come on the top in four cases out of five (availability of credit is cited once as a most important incentive). The perception of importance of tax rebates closely matches the "objective" importance of the scheme. It may be mentioned in addition, that exporters often tend to regard the overall amount paid of the tax rebates (cash payment) as the incentive without deducting the indirect taxes incurred. This view thus implicitly assumes that all of indirect taxes are borne by the producer when selling at the domestic market. Although this may be an extreme position, it is nevertheless quite plausible that one part of these taxes is borne by the producer, in which case the incentive value of the scheme is greater than simply the difference between the tax rebates received and indirect taxes incurred. Export credit is mentioned twice (one first, and one second place) as an important incentive. The fact that export credit (or, equivalently, high cost of general credit) was viewed as an important factor in reorienting - 70 - enterprises toward exports is not inconsistent with a much lower importance assigned to it at the time of the interview (1984): in effect, the subsidy element contained in export credits was significantly less in 1984 than in 1980-81 when most of the firms shifted their production toward exports. Duty-free imports were found relatively unimportant, probably because the interviewed firms already enjoyed them for a while, and thus tended to take them for granted. Direct income tax deduction were often mentioned: in two instances as the second, and in two instances, as the third most important incentive. Finally, exchange rate policy was quoted three times as an important incentive for exports: twice as the second most important, and once as the third. The ranking of export incentives is summarized in Table IX.2, and then compared to the "true" calculated value of different incentives in 1984. Although, the interviewed sample differs from the overall population of exporters we observe significant similarities between the "objective" worth of incentives (i.e. their subsidy component) and the perception of their importance. Export tax rebates and direct tax deductions were both perceived and "objectively" found to be the two most important export incentives. Perception of the real devaluation of TL as an impetus to exports also coincides with the actual developments in 1984. Export credits and duty-free imports, however, display the inverted ranking comparatively to the one according to their subsidy component in 1984. The reasons for that seem to be twofold. First, the interviews were conducted in the first half of 1984 when the shortening of maturity export credits was too recent to have been fully felt, and the new level of subsidy contained in export credits to be more accurately perceived. The interviewees probably tended to think of the worth of the export credit in the time of its greater importance in 1983, and earlier years. Second, all of the interviewed enterprises enjoyed duty-free imports - 71 - of inputs for a significant period period of time, and tended to take these rights for granted, which probably made them discount heavily saved import duties. Table IX.2 - Comparison between Perceived and "True" Value of Different Export Incentives Subjective Importance* "True" Value* Tax rebates 18 11.1 Direct tax deductions 10 8 Export credits 7 1.1 Exchange rate 7 4.6 Duty-free imports 1 2.9 * Calculated from Table IX.1 rankings. Top incentive given 4 points, and each following incentive 1 point less (minimum = 1 point). ** For tax rebates, export credits and duty-free imports subsidy as percentage of export value in 1984; data taken from Table VII.2. For the exchange rate, percentage of real devaluation in 1984. For direct tax deductions, value of average tax deductions for incorporated enterprises (see p. 8). On balance, the analysis suggests that the correspondence between the ''true" and the perceived value of different incentives is fairly close, although a long and continuous existence of one incentive tends to blunt exporters' awareness to its worth. This is particularly the case when the incentive scheme itself does not contain a very high subsidy element. In other words, it is more difficult to conceive that even if export tax rebates were kept unchanged for as long a period as the one during which the interviewed enterprises enjoyed duty-free imports, their importance would be discounted by exporters. This is so both because tax rebates are more important, and there is an explicit cash payment to the exporter, which does not exist in import duty exemptions. With these caveats in mind, broad coincidence between the real value of incentives and their perception is comforting because it represents an essential precondition for the effectiveness of economic policy. - 72 - X. An assessment of the role of direct export incentives Growth of Turkish exports in the period 1980-84 has been nothing short of spectacular. At the time when the overall world trade contracted by about 1 percent per annum Turkish exports registered an average annual growth in excess of 25 percent p.a. Particularly noteworthy was the performance of industrial exports which grew at the rate of 49 percent per annum and whose share in total export doubled in four years. It is true that Turkey's interest have been served favorably during this period by a few exogeneous developments, in particular by the Iranian revolution and the outbreak of the Iran-Iraq war (respectively in February 1979 and September 1980). The revolution cut Iran from its major suppliers, and necessitated a reorientation of its trade. The war, thanks to geographical proximity of Turkey, and the adopted neutral stance toward the belligerants, increased demand for Turkish exports by both countries. Total exports to Iran and Iraq thus passed from $220 million in 1980 to almost $1.7 billion in 1984 (rate of growth of 66 percent per annum). Imports from Turkey accounted for an estimated 6 percent of total Iranian 1984 imports, and 9 percent of total Iraqi imports (jumping from 0.8 and 1.1 percent respectively in 1980). Although it is probable that Turkish export performance would have been, in the absence of these two developments, less impressive, it is nevertheless true that without an export policy, as followed in the 1980-84 period, this opportunity could have easily gone unused. Devaluation of the TL, introduction of direct export incentives, repressed domestic demand, and a general emphasis on exports after 1980, have encouraged (sometimes even obliged) industrialists to look to - 73 - exports as a way out of difficulties. It is only because the overall context was propitious to exports, that industrialists were able to seize upon the opportunities offered by the Iranian revolution, and the Iran-Iraq war. To quote Pasteur, "Le hasard ne profite qu'aux esprits prepare's". The main questions that one would hope to be able to answer after a study of export incentives are the following: "Were export incentives instrumental in effecting a change toward greater export orientation", and "Will the export orientation persist later even in face of declining direct subsidies". Ideally, if the answer to the two questions were in the affirmative, namely, that subsidies were necessary to change the orientation of the economy, and that once this is accomplished they can be safely withdrawn, a strong case for a policy of temporary export subsidies could be established. Turkish evidence to date suggests that this is indeed the case. First, introduction of an explicit export encouragement policy - with the level of incentives, both direct and through exchange rate devaluation, much higher than in the past - produced a definite change in the mentality of industrialists. It was the period best characterized by the "export or perish" dictum. Industrialists started, some of them for the first time, to look seriously at foreign markets. A casual evidence, such as the number of Turkish businessmen encountered in places where none of them could be met only a few years ago strengthens this impression. From the government point of view, the success of the export drive became almost synonomous with the success of the stabilization program. The government thus remained strongly - 74 - and persistently committed to encouragement of exports. This, in turn, strengthened industrialists' emerging orientation towards foreign markets, as they gradually grew more aware both of exports' preferential treatment and of the fact that the government's policy in this respect was not likely to change. Furthermore, the first export successes reinforced the confidence of the economy that it could effectively compete internationally. The combination of all these self-reinforcing effects produced the "export miracle" of 1980-84. It would seem that the three clearly identifiable elements were necessary to effect this change: (a) government's strong commitment to exports, and the general perception that it would not slacken, (b) increased export incentives, and (c) relatively favorable external developments. This is diagrammatically shown in Figure X.l. Figure X.1 Exogeneous Government Industrialists Favorable Strong Government Increased export external export commitment incentives developments Export growth 1 Depressed domestic demand Perception that the Government committment is Greater knowledge of not transitory international markets Perception that industry can compete successfully - 75 - Direct subeidies clearly played a major role in the element (b), and although their existence alone could not have been sufficient to effect the reorientation of the economy, it is probably fair to aver that, alike the other two elements, they were a necessary condition for success. To go to our second question, What are the prospects for Turkish exports in the years ahead when direct incentives are to be gradually phased out? To answer this question, one needs first to complement the picture of export success drawn above by considering also tensions that the preferential treatment of exports, and the export-orientation in general, have produced in the economy, as well as the quarters from which the backlash against that policy may come. Obviously, producers for the domestic market, unable to compete internationally, were hurt first. This occurred through a change in their relative position vis-a-vis exporters (both through greater incentives to exporters and depressed domestic demand), and by a liberalization of imports which the government undertook with particular force in 1984. There is thus little doubt that their relative position deterioriated. They were also the ones most affected by persistently high interest rates on general credits (the rates for which preferential credits to exports were partly to blame). Also the policy of export encouragement was not uniform towards all exporters. As we have seen, it consistently favored producers of capital goods over consumer and intermediate goods industries, and, in addition, gave special incentives to large exporters. Additional tax rebates for large exporters (see Section III) have existed since the inception of the scheme. In 1984, with the introduction of the concept of Foreign Trade Corporations, that is, large trading firms with the volume of exports in excess of $3U million per annuml, and paid-up capital of at least TL 500 billion 1/ Later raised to $50 million. - 76 - (t1.7 million at January 1984 exchange rate), this aspect was reinforced. Only Foreign Trade Corporations, among private finns, were allowed to engage in often profitable trade with countries with the nationalized foreign trade (essentially CMEA countries). For the purpose of export incentive policy they were treated as industrial firms, although they were essentially commercial establishments. Credit allocations to Foreign Trade Corporations were made separately (i.e. outside the allocations by industries, since their exports covered a broad range of products). Additional tax rebates paid to large exporters often gave rise to special preying of large exporters over the small: the latter, instead of selling goods themselves to a foreign company which they identified, would sell goods to the foreign trade corporations with which they would later split the additional tax rebate. Although both sides benefitted by such a clever use of the government regulations, these transactions - where the large companies were seen by the small exporters to be receiving a commission for no work, and solely because of their size--often alerted small exporters to the inequity of the government's policy. Enterprises which could normally be expected to hail government export policy were thus often lukewarm toward it (or rather toward its specific manifestations). It seems at present unlikely that the business interests, fully or partially affected by the export-encouragement policy as pursued by the government since 1980, will be strong enough to produce a reversal of that policy. The pressure for a reversal may even be lessening as the government reduces (as it is currently doing) most of direct export incentives. Elimination of preferential export credits, and a reduction in inflation should also produce a decrease in nominal interest rates, and assuage the opposition of most heavily-indebted firms. The differential treatement - 77 - granted to large export firms is less likely to disappear, as the government views centralization in export activity as the surest way to increase exports, and to reap benefits of the economies of scale which may exist in international marketing, and entrepreneurship.l/ An extrapolation of current developments then seems to suggest that in the years to come, some of the antagonisms between the gainers and the losers from the government policy may be reduced (by a combination of the effects of stronger domestic demand, and smaller direct export incentives), while the export growth will not be adversely affected. Although the unique combination of circumstances in the period 1980-84 will not reoccur, and the export growth may slow down (which in view of a very low export base from which it started in 1980 is no surprise), the prospect for the immediate future is at least the consolidation of export gains so far achieved, and most likely a moderate expansion of Turkey's share in the world trade. The return to status quo ante is difficult to imagine, which is probably the best testimony of the correctness of the policies. 1/ Foreign Trade Corporations accounted for 36.8 percent of total exports in 1984, up from 29.1 percent which was the share of these companies in 1983. I The World Bank Headquarters European Office Tokyo Office 1818 H Street, N W. 66, avenue d'l6na Kokusai Building Washington, D C. 20433, U S A 75116 Paris, France 1-1 Marunouchi 3-chome Telephone (202) 477-1234 Telephone (1) 47.23.54 21 Chiyoda-ku, Tokyo 100, Japan Telex: WUI 64145 WORLDBANK Telex: 842-620628 Telephone (03) 214-5001 RCA 248423 WORLDBK Telex. 781-26838 Cable Address. INTBAFRAD WASHINGTONDC ISSN 0253-2115/ISBN 0-8213-0677-4