Policy Research WORKING PAPERS Trade Policy Country Eoonomics Department The World Bank November 1992 WPS 1040 Lessons of Trade Liberalization in Latin America for Economies in Transition Jaime de Melo and Sumana Dhar The reform packages of trade liberalization, stabilization, and supporting policies in Argentina, Bolivia, Chile, Mexico, and Uruguay offer lessons for the economies in transition in Eastern Europe and the former Soviet Union. PoJicyPachWckingPapudiminacttheSgsfiwofwkinPndpuug cnacegfichnge(efidasamengBanknaffand llothoined m develop i no mhesepapem distibutedbytheReseuchAdvisuy Stff,cany thonancloftheauthrdlca cdytheirviews,andddbeusedanciudocn3y.Thebmdingdintp,titm,andconclusionsaithcauthos'own.Theyshould not be aunluted to the Woerd Baik, its Bord of Diror, its mnagemat, or any of its member counetri Policy Researchj Trade Policy| WPS 1040 This paper- a product of the Trade Policy Divisic.., Country Economics Department - was prepared for the UNDPT'EP Conference on "World Experience of Trade Liberalization"held in Kiev, June 9-11, 1992. Copies of this paper are available free from the World Bank, 1818 H Street N'W, Washington, DC 20433. Please contact Dawn BaHantyne, room NIO-023, extension 37947 (November 1992,40 pages). Aft r four decades as prime examples of inward- * Rapid, far-reaching reform is possible in looking trade policies and import-substituting sectors that were subject to prolonged periods of industrialization, several Latin American coun- heavy protection. tries undertook comprehensive trade liberaliza- tion and macroeconomic adjustment in the * Sustained growth requires a comprehensive 1980s. De Melo and Dhar contend that the reform package, with supporting policies for experiences in those countries are relevant for labor, capital, and domestic product markets. the economies in Eastem Europe and the former Soviet Union in transition from socialism to * Liberalization of the financial sector re- market economies. quires investigating the links between commer- cial banks and private sector firms. In all of these Latin American countries, the move toward an outward orientation occurred * If trade liberalization is to succeed in the long run, it is important to study the evolution of * when the economy was facing a large the real exchange rate and measures to stabilize negative external shock because of faling terms it. of trade and rising debt payments; In the final section of the paper, de Melo and • after several decades of protectionism; and Dhar study the recent impetus toward trade liberalization through regional arrangements in * under severe macroeconomic imbalances. Latin America. The issue is relevant to countries in Eastern Europe and the former Soviet Union De Melo and Dhar study the reform package because they belonged to the CMEA, a regional of trade liberalization, stabilization, and support- trading arrangement, and because such arrange- ing policies in Argentina, Bolivia, Chile, ments are evolving anew among countries in the Mexico, and Uruguay. They conclude that for former Soviet Union. the economies in transition: - Rationalizing the foreign trade regime is crucial for the success of stabilization measures. The Policy Research Working Paper Series disseminates the fidings of work under way in the Bank. An orective of the sexies is to get these fmdigs out quiclcly, even if presentations are less than fuDy polished. The fndings, interpretations. and conclusions in these papers do not necessarily represent of ficial Bank policy. Pniduced by the Policy Research Disseniination Center Lessons of Trade Liberalization in Latin America for Economies in Transition Jaime de Melo Sumana Dhar Prepared for the UNDP/TEP Conference "World Experience of Trade Liberalization" held in Kiev, June 9-11, 1992. We thank Jacques Morrisset, John Nash and David Tarr for comments on an earlier draft. Table of Contents Page No. 1. Introduction 1 2. Trade and Growth in Latin America: A Comparative Perspective 3 3. Recent Trade Liberalization Episodes 10 4. Regionalism and Trade Liberalization 34 5. Conclusions 37 References 39 List of Tables Table 1: Macroeconomic Indicators 11 Table 2: Reforms in Argentina 13 Table 3: Reforms in Bolivia 21 Table 4: Reforms in Chile 24 Table 5: Reforms in Mexico 28 Table 6: Reforms in Uruguay 31 List of Boxes Box 1: Sources of Inefficiency in Highly Regulated Economies 5 Box 2: Experiences with Exchange Rate-Based Stabilization 15 Box 3: The Economics of Regional Integration 35 List of Flgures Figure 1: East Asia and Latin America Growth Rates: GDP, Exports, Imports 6 Figure 2: Relationship Between Total Factor Productivity Growth and Total Factor Input Growth 9 Figure 3: Openness and GDP Indicators 12 Figure 4: Real Interest Rate, Exchange Rate, Growth of GDP 16 Lessons of Trade Liberalization in Latin America for Economies in Transition 1. Introduction The literature on the relationship between trade regimes and economic performance has relied on Latin America as the example of the effects of inward-looking or import substitution industrialization (ISI) and East Asia as the example of outward-looking or export-led growth (ELG). Until the early eighties this was essentially the case. With the exception of Chile's successful trade liberalization carried out in the mid-seventies, Latin America still fitted the description of ISI at the time of the debt crisis in 1981. With the onset of the debt crisis, the shortcomings of the ISI strategy could no longer be hidden, and a move towards outward orientation took place in Latin America. In many ways the conditions under which trade liberalization took place (a large negative external shock due to the combination of falling terms of trade and rising debt payments) are quite similar in magnitude to those facing East European (EE) and and Former Soviet Union (FSU) countries. A review of how trade liberalization and macroeconomic adjustment were carried out is relevant for economies in transition. Slowly throughout the eighties, several countries carried out comprehensive trade liberalizations. Like previous failed attempts, these were undertaken from crisis situations with several macroeconomic imbalances reflected in large fiscal deficits and high inflation. Yet, unlike several previous attempts at trade liberalization which had been short-lived, this time there are no signs of reversal so far. Indeed, trade liberalization spread across the region and today, the majority of Latin American countries show strong signs of having overcome the legacy of four decades of ISI. At the same time, Latin America's privatization of State Owned Enterprises (SOEs) has been strong as it accounted for 37 percent of SOEs sold in the developing world between 1980 and 1991 (in 1980 SOEs accounted for 12 percent of GDP in Latin America). 2 Another dimension in which the recent experience of Latin America is relevant for economies In transition is that unlike East Asia, the move towards outward-orientation took place after many decades of high protectionism. And, unlike East Asia (but like Latin America), EE and FSU countries are liberalizing under conditions of severe macroeconomic imbalances. Besides being market economies rather than centrally planned economies, the major difference with the current situation in EE and FSU countries is that the recent move towards outward orientation in Latin America took place against a background of several failed attempts which lowered the initial credibility of the reform packages. The remainder of the paper is in three sections. To set the stage, section 2 draws some comparisons on aggregate performance between the two extreme cases: Latin America and East Asia. This comparison gives a feel for the long-term consequences of differences in trade orientation (and other supporting policies) on overall performance. Section 3 then describes briefly the recent trade liberalization measures that were carried out in five countries: Argentina, Bolivia, Chile, Mexico and Uruguay. In all cases, trade liberalization was carried out simultaneously with stabilization and other reforms. We describe briefly supporting policies, as the success of trade liberalization depends crucially on the design of the entire reform package. We close in section 4 with a discussion of the recent impetus towards trade liberalization achieved through regional arrangements. This should be of relevance to the current situation in some EE and FSU countries since the CMEA was in effect a regional trading arrangement and the issue of the trading arrangements among FSU countries is likely to be a continuing issue for debate. As a prelude, we give the main lessons from each country experience. Argentina's experience is an example of persistent macroeconomic imbalance (arge fiscal deficits and high inflation) leading, until recently, to several reversals of partial trade liberalization efforts. Bolivia is an example of how one can carry out an orthodox stabilization program (without resort to an income 3 policy and temporary wage and price controls) to wipe out hyperinflation while simultaneously carrying out trade liberalization. Chile is an example of the economic success story one would like to ..ee everywhere (in spite of a difficult period of adjustment and a delicate transition from authoritarian rule to democracy). Mexico illustrates how the prospects of regional integration with a stable and developed neighbor can help cement impressive stabilization and liberalization policies. And Uruguay illustrates the difficulty of carrying out trade (and other) reforms in an economy with a long tradition of rent seeking activities. At the outset of any exercise on lessons to be drawn from the Latin American experience with trade liberalization, a word of caution is in order. The region is known for its many policy reversals over the years. Therefore all the encouraging signs that are voiced in this paper should be interpreted cautiously, especially for Argentina, where reforms have only been in effect for three years. Also, unresolved losses of the banking system are significant in all five countries and could pose a threat for the success of the ongoing reforms. So could the loss of competitiveness through real exchange rate appreciation for the countries that have used the exchange rate to reduce inflation. 2. Trade and Growth in Latin America: A Comparative Perspective It is now widely accepted that, controlling for other factors, more open economies - in the sense of less distortions in their foreign trade regimes - grow faster.' More open economies both discriminate less against tradable activities than closed economies, and within tradable activities, they T'he evidence of higher growth for open economies based on country studies and on cross- section regression analysis is overwhelming. For a recent entry point in that literature, see e.g. Dollar (1992) and references therein. 4 avoid creating what is referred to as "home market bias", that is, they avoid discriminating against exporting activities and in favor of import substituting activities that cater to the domestic market.2 Why do more open economies grow faster (after controlling for other factors)? Here the arguments rely on microeconomic analysis at the firm level and for which it is much harder to get conclusive evidence, even though it is widely perceived that the facts support the reasoning. It is generally agreed that liberalization in highly distorted economies improves existing resource allocation across sectors, across flrms in a sector, and within firms in a sector. In a dynamic sense, liberalization is also intended to improve the efficiency of investment by allocating it to activities that are profitable under incentives free of egregious distortions (for a further elaboration of the sources of inefficiency arising out of a distorted incentive structure, see box 1). Thus, it has often been observed that more open economies have both higher GDP growth rates, and higher growth rates of total factor productivity (TFP). TFP is defined as the difference between output growth and a factor share weighted growth of inputs (this difference is also known as the "residual" - a measure of the unexplained source of output growth).3 Figure 1 compares the evolution of GDP, export, and import growth for the Latin America and East Asia regions (all figures are weighted by country shares in their respective regions). Over the period 1970-89, GDP growth was consistently higher in East Asia. In particular the gap in 2 Tradable activities are those activities that are potentially exchanged internationally. Nontradables, on the other hand, are activities that are not exchanged internationally because of high transportation (or storage) costs. Examples of nontradables activities are construction, utilities and services like haircuts. Typically, about one half of GDP is classified as nontradables. The distinction between the two types of activities and the evolution of their relative price between the two (known as the real exchaunge rate or relative price of nontradables) is very important for following the evolution of a country's external balance. 3 For a further discussion of the link between trade policy and TFP, see Havrylyshyn (1990) and Tybout (1992). 5 Box 1: Sources of InefficMency In Highly Regulated Economies' To name a few of the effects of excessive regulation, finns become less than efficient, as they are restricted in the choice of their optimum output and input mix. To begin with, these frms are constrained in their input markets. Adjustments in their work forces are impeded by labor legislation. Management of inventories Is complicated by uncertainty about whether and when the firms will be able to import spare parts and raw materials. Nor are enterprises sure about how much capital they will have access to - and at what cost. Firms in such economies also operate in restricted product markets. There are barriers to the entry of foreign companies. There also are barriers to the entry of new domestic firms, for even if the cost of entry Is not a problem, access to capital may ba - because the limited pool of capital is parceled out according to long-standing ties between existing producers and their creditors. Barriers to entry, of course, benefit existing firms as they have more leeway in their pricing policies. The firms thus suffer from some of the restrictions and benefit from others. Restrictions in highly regulated economies include made-to-measure prctection. This situation guarantees the coexistence of firms with a wide dispersion of efficiency within a sector, since the level of protection is chosen to make the least efficient firms profitable. As a result, many firms can operate at a profit with dated machinery and equipment. In such economies, most firms are inefficient. They use outmoded technology and use it badly (technical inefficiency). Too many of them operate at too small a scale (scale inefficiency). They do not equate marginal products with factor costs (allocatdve inefficiency). Because consumers have nowhere else to go, the leading firms can charge higher than normal mark-ups, with other firms following suit. Firms therefore price their products above the marginal cost of production (price inefficiency). Under such conditions the economy becomes characterized by idle capacity and unproductive work forces. In addition, the incentive system is greatly distorted because protection has been tailored for individual products to ensure the survival of firms already operating. The made-to-measure protection usually includes quantitative restrictions that result in exceedingly large rents and induce economic agents to expend much of their effort appropriating rents rather than engaging in activities that would be more socially productive. Thus quantitative controls result in rents which in turn give rise to rent-seeldng and lobbying activities to capture these rents. Resources that would otherwise be productively used get diverted to chasing these rents with potentially large welfare losses which could add up to the value of the rents. 1. Source: Adapted from V. Corbo and J. de Melo (1985) 'Introduction". 6 Flgure 1 East Asia and Latin America Growth Rates: GDP, EXPORTS, DMPORTS (3 Year Moving Average) 10 GDP 2\1 §~~~ 1'7 \7 IIX If 18~ 12 EXPORTS , I 2 -41 - 70 I 1172 1n 47 1 igis 1 tu7a ivao I :laa Il f I1 hS I §u 151 111 *1*5 *I? 1171 1121 1X8 *tUb 16~ ~ ~~~67I" 2 n 12 'a IMPORTS -2 -4 .-4 -:2 "| 18701 :87 § a1a1:'4 1#188 tusl11218111512 1ou1ce:73 1Wor It 1Tbe a18 18 ISO II Source: World Tables V . C 7 performance between the two regions was staggering during the 1980s. Taking into account population growth rates, average per capita GDP growth during the 1980s was 1.6 percent in Latin America and 7.8 percent in East Asia. Both regions felt the effects of the oil price shocks (1973, 1979) and the debt crisis (1981), but the more flexible East Asian economies bore the brunt of the shock much better. In terms of the above discussion, the widening gap in performance throughout the seventies and eighties is attributable to the more flexible economic structures associated with open economies that had less distorted incentive structures.4 Export growth rates were also consistently higher in the East Asian region. Finally note the particularly strong correlation between import growth and GDP growth for the Latin American region. Tbis is what one would expect of economies where two-thirds or more of imports are for intermediates and capital goods not produced at home: when a foreign exchange crisis hits the economy, essential inputs for production can no longer be purchased from abroad, and output declines sharply. The terms of trade loss for Latin America during 1982-86 (compared with 1978-81) was between 3.8 percent and 4.5 percent of GDP (see Faini et al. 1991, table 2). Interestingly this estimate is about the same as the estimate of the terms of trade loss (4 percent) experienced by Poland, Hungary and Czechoslovakia when they shifted to world prices. Figure 1 indicates that Latin America's average GDP growth during 1982-86 was 2.33 percentage points below 1978-81 (which were boom years because of unusually large capital inflows). Of course, the entire loss in output growth cannot be imputed to the effects of this terms of trade loss, since there was strong private capital outflow during the debt crisis. ' A more flexible economy adjusts better to an external shock than a more rigid economy because the costs in terms of idle resources during adjustment is less. 8 What accounts for the much higher growth of East Asia? Figure 2 plots the contribution of factor input and TFP growth to GDP (or value-added) growth for a sample of developed and developing countries. TFP growth is measured on the vertical axis and total factor input growth on the horizontal axis. Each isoline in figure 2 corresponds to a different rate of GDP growth with isolines further from the origin indicating higher GDP growth rates. The figure Is divided into three clusters: cluster A consists mostly of developed countries in which TFP growth is the most important contributing factor to growth; cluster B has some developed countries and a number of Latin American countries in which growth is mostly due to factor accumulation; and cluster C with East Asian countries, Japan and a few others where the high growth rates is due both to large contributions of factor input growth and to TFP growth. The contribution of TFP growth to GDP growth cannot be overemphasized even though it is difficult to pinpoint its causes since it includes not only the effects of differences in policies, but also the acquisition and mastering of existing technology, technological progress and other improvements in production activities. For example, in figure 2, Brazil and Korea had similar factor input growth rates (around 6 percent) but Brazil grew at about 7 percent while Korea grew at close to 10 percent. (The differences between Venezuela, and Hongkong and Taiwan are even more striking). In sum, these differences in long-run growth performance are due to a number of factors including more stable macroeconomic policies, less distorted factor markets, and better functioning institutions. But it is generally agreed that less distorted foreign trade regimes resulting in more open economies are an important contributing element to superior economic performance. 9 Figure 2 Relatboship Between Total Factor Productivity Growth and Total Factor Input Growth 8 Gv 8 Gv 10 Gv 12 7 c \ ~~~~~ong 6 Gv6 4 ai 2\> ltl Germany Israel orway a _ AS w oi = jjtal..nc 3r a Netherl\ I- 3 ede I Uniited Staa\ Argentn \ Venezuela\\ 2 3 4 5 6 7 8 Total factor input growth (percent) Notes: G, refers to value-added growth. Total factor input growth includes the contribution of labor and capital. Periods for calculations vary from country to country. In most cases the period covers 1955 - 15. Key: m Developed country A Developing country 13 Hongkong, Israel, Korea, Spain, Taiwan Source: Chenery, Robinson and Syrquin (1986, figure 2.2) 10 3. Recent Trade Liberalization Episodes5 We now turn to a brief description of the recent trade liberalization episodes in Argentina, Bolivia, Chile, Mexico and Uruguay. The emphasis is on conveying a sense of the magnitudes of the trade reforms and, of the country's initial macroeconomic conditions at the time of the trade reforms, and of the accompanying policies (e.g. privatization of SOEs, labor and capital market deregulation, etc.). We give two sets of indicators on a comparative basis. For background information, table I gives three year average values for standard macroeconomic indicators for the period 1980-90.1 Three year averages remove some of the volatility of the series but masks the turning points following the reforms. Figure 3 compares the evolution of a GDP index (centered with 1980 = 100 for all countries) and a measure of openness (the ratio of exports to GDP) for the period 1970-91. Argentina (table 2). Argentina is the classic case of a succession of failed attempts at trade liberalization. In all attempts, trade liberalization was accompanied by stabilization policies to reduce fiscal deficits and inflation. In several failed attempts, inconsistencies between monetary and exchange rate policy (which was used as an anti-inflationary device) led to sharp real exchange rate appreciation (a rise in the relative price of nontradables) that was not sustainable because of the sharp loss in competitiveness of the traded sectors. Often, these attempts ended with quantitative restrictions (QRs) to stem growing balance of payments deficits and wage-price freezes to try and curb inflation. s In most cases, events are covered until 1991. For more details on Argentina, Chile and Mexico, see Nogues and Gulati (1992). 6 Indicators for the size of the fiscal deficit and unemployment, are probably the least reliable, and should be interpreted with extreme caution. 11 Tablek I M Indis (average per perio) Argentina Bolivia Cbile Mexico UmgUuay 1980/ 19W 1988 19801 19W 1988 1980/ 19841 19881 1980/ 19W 1988/ 19801 1984t 1988 Years 83 87 90 83 87 90 83 87 90 83 87 90 83 87 90 I Growth rateof GNP/ - 4.0 - 0.02 3.8 * 5.8 - 4.1 - 0.S -4.0 4.1 4.3 - 0.1 -1.6 2.2 - 3.9 3.6 0.2 capita n Gross domestic 17.8 12.4 9.7 12.8 9.1 12.7 16.2 14.7 18.9 24.6 19.8 17.4 14.3 9.0 9.S investment/GDP (%) m Government deficite - 5.9 -9.3 -8.9 -21.1 -10.3 - 1.12 2.9 -0.3 - 4.3 -11.6 -10.6 - 7.3 - 3.6 - 2.2 - 1.54 (-) or rplus/GDP (5) IV Unemployment rate 4.0 4.7 6.9 10.1 18.4 19.0 14.4 12.1 6.9 5.1t 4.7 3.0 10.3 11.9 9 1b V Inflation (rate of 178.3 380.1 1932.1 119.6 3274.0 16.1 21.1 20.1 20.5 54.0 85.2 53.6 41.4 66.9 85.0 change in CPI) VI Total external debh/ 68.3 74.7 82.4 113.8 160.4 105.3 67.7 130.8 82.5 45.5 69.3 49.9 33.4 72.1 50.0 GNP (%) [50.71 (65.51 138.31 (45.3) [45.71 (40.01 (58.41 (47.41 (25.51 [52.41 [49.51 (37.9] (23.3] 137.1] [36.4] (Total debt service/ exports of goods and services (%)] b. 1988. c. 1981/83. d. For Argentina and Mexico, figures are for the public sector borrowing requirement (PSBR). Source: World Tables, World Development Report, Internatonal Fdancial Statistics, World Debt Tables, Yearbook of Labor Statitics. 12 Figure 3 Openness and GDP Indicators 1970-1991 ARGENTINA BOUVIA 30 *1S0 30.._10 140 -4 25S t130g. 1U0 20- ~~~~~~~~~13 ¶20-D 110 Q &10Q 15. .000 10s-,;,.-9 3 : 1, 1001 O O~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~O 700 1~~~~~~~~~~~~~~~~~~~~-0 tO70 1 92174 1978198M I 1982 194 196 I 98 1990 1 70 19721974 1975 1978I 190196I 194 198M iim 1 w0 YEAR YEAR CHILE MEXICO 30 16f0 ao30 -18 I40 140 0- t5 < /100 - 15C .120 z gZE g100 5 70 o / *70 o~~e f0*e 1 9701972 1974 176 197190 192 16 988'1990' * ; 1970 1972 1974 1976 t9781#10 1982'16419681981990' w YEEM YEAR URUGUAY 30_ 10 140 25 130 -110 19 70 1972 1974 '1978 '1978 196 0 '1982' 1964 im i 1990 YUR -OPENE6S * GOP INDE Source: World Tables 13 Table 2 Reforms in Argentina Period Growing contros Fmaneiu1 reforms Restrictions Expot Deregulation Acceierated 1950-1976 and panial 1982 -1914 promotion 1988 - 1989 libralization liberalization 1985- 1987 1990- 1976-1981 _ _ _ _ _ _ _ Ovcrall policy Impor substiuting Military tkever in Debt erisis 1982 Austral Stabilisation ( Failure of Primevera Privatization; environment industrialisation (ISI) Marcb 1976 Foland War 1982 7 months): Stabilisetion (I fiscal sector Democracy restored 1983 devaluation, wage- year); adjusamenl price freeze, upward hyperinflation loan inflation; adjusment of public sector prices _ _ _ _ _ _ _ _ External factors Trade policy Imports Maximum tariff rate of Maximum tariff lowered In 1985 import licenses cover 52.1 % of In 1988 tenporary Specific tariff; Specific tariff 200%; to 55% from 100% in production of tradcables; 10% surchargc; prohibitions removed; maximum tariff rate abolished; 3 tier 19s2; average tariff was prohibition and prior approval for 15% surchargc; 22%; surcharge tariff (22%, 11%, 30.2%; impott licenses impons. import licnses cover abolished; O%); few QRs cover 4.3% of 32% of production of remaia (on cars); production of trdeables. tradeabers; unweighted mean Boom-bust cycle (see tariff rate is 15% box 2 and figure 3) in 1992; Exports Targeted export Programas Especiales de Exportacion for Special schemes for Phasing out expot diversification; export taxes nontraditional exports (In 198087 US export promotion subsidies; S36S3m paid as export subsidies). continued Exchange rate Active erawl (1978-81) Foreign exchange rationing Fxed exchange rate Fied, then flxible 1990; raed exchange rate exchange rate; since March 1991 flexible exchange ratc Main results 14 Examples of failed attempts include the "open import liberalization" policy during 1976-82 under the military, when it was believed that pressures from foreign competition under the exchange- rate-based stabilization program would help reduce oligopolistic pricing behavior. In the end the stabilization program had to be abandoned as the use of the exchange rate to reduce inflationary expectations (the government was pre-announcing, up to six months in advance, future rates of devaluation at less than the differential between domestic and foreign inflation) had resulted in too much loss of confidence in the exchange rate. Capital flight (outflow) set in when the private sector realized that the loss of competitiveness and resulting current account deficit was no longer sustainable. The inconsistency of exchange rate policy and fiscal policy also played an important role in the failure of all Argentine stabilization plans. In Latin America, the underlying cause of rising inflation was almost always large increases in the fiscal deficit so that reducing fiscal deficit reduction is then a prerequisite for eliminating inflation.' Yet, eliminating inflation has often proved very difficult. Figure 4 shows the outcome of the exchange-rate approach to reducing inflation in the case of Argentina, Chile and Uruguay when it was tried for the first time around in late 1978 (since then it has been used many times). It also illustrates the difficulty of coordinating stabilization and liberalization policies (for details of the episode, see box 2). The use of the exchange rate to reduce inflation in the three countries was started at the end of 1978. Initial conditions were different (see box 2), but the pattern of GDP growth, real interest rates and real exchanges was strikingly similar. Three phases are apparent. In a first phase, the real 7 Of course fiscal deficits are not always the result of autonomous increases in spending. Often they occur because of difficulties in reducing components of spending when there is an unanticipated rise in debt service payments and it is difficult for political and economic reasons to raise tax revenues. 15 Box 2. Experiences with Zxchange Rate-Based Stabilizatioo Ladn American countries have typically liberalized trade while simultaneously trying to reduce inflation. Figure 3 shows the outcome of stabilization during 1978-82 for Argentina, Chile and Uruguay while they simultaneously used a pre-announced exchange rate to reduce inflationary expectations. This box gives further details about the outcome of this experiment. At the start of this policy, Argentina's public sector deficit was still almost 10 percent of GDP; Chile still had a wage indexation mechanism that was bound to result in a real appreciation of the exchange rate; and Uruguay's fiscal deficit increased substantially in 1981. The anti-inflationary programs in all three countries were flawed. As noted in the text, domestic demand grew faster than output in the three countries with the gap filled by foreign finance. The demand pressures on nontradables lead to a real exchange rate appreciation and limited the effectiveness of anti-inflationary policies. In Chile, the annual inflation rate was reduced from 50 percent in 1978 to 20 percent in 1981 and to zero in early 1982, but the accumulated real appreciation of the peso was large. Argentina's inflation rate only fell from 175 percent in 1978 to 101 percent in 1980. Inflation eventually fell to 63.5 percent in 1980 and 34.0 percent in 1981. In all three countries, increased imports and loss of export competitiveness combined to raise the current account deficit. In Chile, the deficit rose from 5.6 percent of GDP in 1977-78 to 9.1 percent in 1979-81; Argentina moved from a current account surplus equal to 2.1 percent of GDP in 1976-78 to a deficit of 1.8 percent in 1979-80; and in Uruguay the deficit increased from 3.2 percent of GDP in 1977-78 to 5.4 percent in 1979-8 1. Because all three economies were booming, the average unemployment rate was reduced from 14.2 percent to 13.6 percent in Chile, from 12.4 percent to 8.4 percent in Uruguay and from 3.4 percent to 2.2 percent in Argentina. As real exchange rate appreciation continued so doubts grew about the sustainability of the exchange rate policy. These doubts were reflected in growing interest rate spreads despite the shrinking (Chile) or absence (Argentina and Uruguay) of impediments to short-term capital flows. Real interest rates rose sharply, adding to the difficulties of the tradable goods sectors. Companies were doing more and more borrowing to stave off bankruptcy and awaiting a bailout after devaluation. The inconsistency in the three countries' economic policies became apparent in late 1980 in Argentina and in early 1982 in Chile and Uruguay. In Argentina, with an externally financed public sector deficit of over 10 percent of GDP and no prospect of fiscal reform, doubts about the sustainability of the exchange rate regime began as early as the first half of 1980. They were increased by the collapse in April 1980 of the largest commercial bank which prompted a 25 percent increase in the money supply in a single month. In Chile, despite a fiscal surplus in 1979-81 equal to 2.1 percent of GDP, the current account deficit reached 14.6 percent of GDP in 1981. Doubts about the sustainability of the exchange rate started to set in, with inflows of private capital decreasing from US$1.6 billion in the second half of 1981 to only US$900 million in the first half of 1982. The monetary contraction that followed resulted in high interest rates and a sharp recession. In Uruguay, the fiscal deficit (which had fallen continuously through 1980) started to increase in 1981 with an underfunded social security reform. Meanwhile, the real exchange rate rose by 27.4 percent between 1978 and 1981 - and even more 'tive to Argentina, after the latter's stabilization attempt collapsed. As doubts grew about the sustainability of the tanlita (the pre-announced devaluation schedule), so outflows of private capital started to increase in 1981 (de Melo 1987). Thus capital flight started in Argentina and Uruguay (and to a lesser extent in Chile) before their economies were hit by the adverse external developments of the early 1980s. * See figure 4. 16 FIgure 4 Real Interest Rate, Exchange Rate, Growth of GDP 1977-1982 Arienuina Chile Urugua 140 lw ~~N 1977 1962 1977 1982 19r7 1982 Left scale: - ~~- Real interest rate (five-quarter moving average of annual lending rate in percent) Real GDP (fourth quarter 1978 = 100) Right scale: Real exchange rate (fourth quarter 1978 = 100) Sore Corbo and de Melo (1987). 17 interest rate (in pesos) fell, stimulating aggregate demand (accelerating GDP growth) and a real exchange appreciation. It was hoped that as inflation fell both the real interest rate and the real exchange rate would move to their long run equilibrium values. In the second phase, a continuing real exchange rate appreciation was observed with real interest rates starting to rise, partly reflecting higher expectation of devaluation. The continued real exchange rate appreciation stimulated aggregate demand (and GDP growth), while the increase in real interest rates stimulated short term capital inflows. With competitiveness deteriorating, the current account deficit widened, while domestic borrowers found it preferable to shift to dollar-denominated debt. Finally, in a third phase, when loss of credibility set in and the private sector no longer believed in the pre-announced exchange rate policy, capital outflow set in, triggering a foreign exchange crisis, the abandonment of exchange rate based disinflation, and a return of inflation. With the return of democracy in 1983, economic policies temporarily took second place and Argentina continued to have large fiscal imbalances, high inflation, high tariff protection (maximum tariff of 55 percent), export taxes (amounting to 14 percent of export revenues) and import licensing covering over 50 percent of imports in 1985. Between 1980 and 1989, average yearly inflation was 395.4 percent and Argentina, with its long history of inflation belonged to the category of "chronic inflation" countries where stabilization is especially difficult to carry out because of widespread indexation. Another attempt at stabilization took place in June 1985 with the "Austral" plan (the Austral was the new currency that was introduced as part of this heterodox stabilization package that included a wage-price freeze with a large devaluation upfront). The fixing of the exchange rate and the wage- 18 price freeze was meant to be temporary and to stabilize prices, nQi to, fix prices.8 For seven months, the government managed to refrain from financing the deficit by printing money, and trade liberalization took the form of intensive export promotion coupled with a reduction in licensing requirements for imports (only 32 percent of the tariff positions remained subject to prior approval) and all temporary restrictions on imports had been removed. However, when once more, the government failed to achieve fiscal restraint, tariff surcharges on imports (up to 15 percent) were applied and by 1988 weighted average protection was 40 percent. Currency overvaluation and hyperinflation set in once more with a peak annual inflation rate of 20,000 percent in early 1990, and a staggering annual average rate of consumer inflation of 450 percent for the decade ending in 1990. During that decade, income per capita fell by more than 20 percent and the investment rate slumped from 23 percent to 8 percent. With a new government in 1989, stronger support for trade liberalization led to a progressive reduction of export subsidies and removal of import licensing. The surcharge on imports was abolished and the maximum tariff rate on imports was reduced from 50 percent to 22 percent. At the beginning of 1991 there was further progress towards trade liberalization with the abolishment of specific tariffs and most non-tariff barriers to trade. Quantitative restrictions still exist in the automobile sector and industrial promotion regimes. Tariff rates of 22 percent, 11 percent and 0 percent were applied to all imports and a schedule for the progressive phasing out of export and industrial subsidies was adopted. Average protection is now 9 percent and the only remaining QR is on autos. s The experience of the Austral plan bears similarities with the Polish stabilization plan of 1990 when a fixed exchange (with a large upfront devaluation) was introduced. As in many Latin American experiences with exchange-rate-based stabilzation there was initial overshooting (probably necessary to reduce inflation expectations) that resulted in large export increases and a large reduction in imports. Because of inflation inertia, the exchange rate eventually became overvalued, export and import growth trends were reversed, and the zloty was devalued in May 1991. 19 The cornerstone of the latest attempt at stabilization was the reduction of the fiscal deficit and the adoption of supply-side reforms to increase competition and efficient functioning of critical enterprises. The telephone network, the state airline, railway lines and two television stations have been privatized and the government intends to sell off at least 51 percent of all SOEs.9 At the same time pay bargaining was decentralized, and barriers to entry in markets were removed while the VAT base was widened and tax compliance increased. As in the case of Chile discussed below, the key to success so far has been widespread reforms accompanying the trade liberalization. It might be added that the needed political support for these tough reforms materialized because inflation had become intolerable and the public was ready for change after years of economR; decline. In April 1991, in another extremely important macroeconomic measure, a Convertibility Act required by law that the monetary base be 100 percent backed by gold and foreign currency reserves fixed the exchange rate to the dollar and banned indexation of wages and contracts. As of the middle of 1992, there is considerable optimism about the success of this approach to stabilization and widespread popular support for the wide ranging market reforms. Bolstered by the success so far (annual inflation of about 30 percent and GDP growth of about 5 percent per year forecasted for 1992-93), in January 1992 the government returned from the Austral to the peso now worth $1. There remain, however, some fears that the fixed exchange rate will have to be abandoned if there is, once more, a sharp deterioration of the current account caused by real exchange rate appreciation, if current capital inflows are not sustainable over the medium run.' Tbere is concern that the loss in 9 About half of the fiscal deficit was caused by SOEs, with losses amounting to 9 percent of GDP in 1989. The proceeds of these recent privatizations amounted to 1.4 percent of 1990 GDP. '° In 1991, gross private capital inflows to Latin America were estimated at US$40 billion, equivalent to 4 percent of GDP. There is renewed concern that these inflows are, once more, eroding the competitiveness of the export sector, as they did during the first phase of the boom-bust cycle of 1978-81 (described in box 2 above). The private capital inflow to Argentina was US$5.2 billion in 1991. Fifty percent of this amount was foreign direct investment and portfolio capital. 20 competitiveness of domestic industry may be beyond the levels of competitiveness that could be supported after restructuring has raised domestic firm efficiency. Whether this stage has been reached or not is an open question. However, insofar as this time the fiscal deficit had been quasi eliminated and monetary expansion automatically controlled, an exchange rate adjustment would not necessarily have to result in a new inflationary spiral, as it did invariably did in the past, since credibility would be much greater than in the past. Bolivia (table 3). Bolivia is a small landlocked economy with an enclave in mine-related activities. Its experience is interesting for some of the small landlocked FSU countries which have few opportunities in the short-run to export anything else than natural-resource-based products. It is also interesting because, unlike the other Latin American economies that reached hyperinflation, Bolivia was never a chronic inflation economy. As a result, the economy did not have widespread indexation, and an orthodox (i.e. no control of prices and no incomes policies) approach to stabilization worked. A typical boom-bust cycle took place in the late seventies when favorable international markets for minerals attracted foreign capital that led to the discovery of hydrocarbon deposits and capital inflows that averaged 50 percent of exports during 1975-78. At the time, the mining, hydrocarbon and energy sectors were nationalized and key industrial plants in manufacturing were in the hands of the public sector. Domestic markets were ridden with price controls and marketing boards. Foreign exchange was allocated by non-market mechanisms to specific sectors which also received subsidized credit and the foreign trade regime had a complex system of tariffs and quotas. By mid-1985, the public sector deficit reached 25 percent of GDP and the inflation rate was 28,000 percent. Fixed investment had fallen from 14 percent to 7 percent of GDP. The stabilization program of August 1985 included a massive devaluation, the unification of the official and parallel market exchange rates and an auction to determine the competitive equilibrium exchange rate. Thus, 21 Table 3 Reforms in Bolivia Periods Rcvolution Favourable Crisis yea Refoms Aflenmath Recent measures 1952-1958 trade climate 1978-1985 1985-1986 1987-1989 1989-1991 1959-1978 Oveall policy Agmran rcnrm; Key industries in No coherent economic Reform of the finial sector, Decree of pivsation for environment Nationalistion of public sector, price policy; labour markt, tax ansctu, publc sector industries Mines controb; polical iility Removal of price controls; subsidised credt, reduction of goverment political stabiliy 1971- expenditure 78; External factors Favourable World Debt build up Decine of tin (1986) Lack of Market for Mineash; revaluation of debt and natural gas prices infastructur servicing and expoit (1987) Unainity about the Discovery of capacity by external permanen c of the ._______________ hydrocarbon deposits lenders reforms Trade policy Impods CompleCX system of Maximum tariff of 20%; Maximum taif of 10%; tariffs and quotas; licensing abolished; tariff of 5% on capital goods; few QRs rmn (on sugar, edib oils, wheat flour); unweighted mean tariff rate is 8% in 1992; the range of taiff is between 5 to 10%. Exports 10% rebate to nontadiionl Scheme for Duty drwbak wheme exportrs noditional Allocation of foreign Overvaluation of expods; Exchange rate exchange exchange rate Massive devaluation; unification and market detemination of exchange rate Main results High infation; Large investments of Fail in real GDP of Fiscal dericit reduced; inflation Stagnation of Growing at 2-digi kvels GDP gmrth <2%; exterl capital; GDP 10% in 1980-1985; brought under control invcstment and Nonuaditional exports; income redistribution growth 5% very high inflation rate exports, pubLic sector deficit and budge deficit; GDP grwth of 2.7% reduced to 3.3%; shaup fal in exporb inflation rate 17% am invcstment GDP gowth of 4.0% in 1991 22 unlike the other stabilization programs described in this paper, the stabilization did not rely upon a fixing of the exchange rate. Concurrently with the foreign exchange auction, interest rates were freed and controls on capital flows were lifted. Government expenditure reduction, tax reform, increases in public sector prices, elimination of price controls and subsidies were the complementary measures that made the stabilization successful. By 1986, the inflation rate was down to 66 percent and the average yearly rate of inflation since 1987 has been 18.7 percent. Quite remarkably, Bolivia has not reversed its policies in spite of a collapse of the mineral markets in 1986 and of hydrocarbons in 1987. At the same time, Bolivia entered a comprehensive trade liberalization program. Import licensing was abolished, all QRs were eliminated (except for sugar and edible oil) and in 1985 the maximum tariff rate was lowered to 20 percent. By 1991 the maximum tariff rate was 10 percent, and the 10 percent rebate for imported inputs provided to nontraditional exporters was replaced by a duty drawback system. While Bolivia's stabilization program was highly successful, and the Bolivian government carried out all the structural adjustment programs that are usually advocated, growth performance has been mixed. GDP growth has averaged 2.7 percent between 1987 and 1990, and it was only in 1991 that GDP growth reached 4 percent. Also investment response has been slow with private share stagnating at 4 percent of GDP till 1989. The private investment rate grew at 4 percent in 1990. The relative slow supply response to reforms in Bolivia is an example of the time that it can take for reforms to bear fruit. Even though it took time for investment and GDP growth to pick up, the Bolivian reforms resulted in a strong performance on the export side. Private sector mining grew rapidly, compensating for the decline in public sector mining, and non-traditional exports have grown at double-digit levels since 1987. Thus there was a strong supply response to the reforms in Bolivia 23 even though the pace of recovery was somewhat slower than in the cases of Argentina, Chile, or Mexico. aCle (table 4). Chile is heralded as the success story of trade liberalization as it has the most transparent and open trade regime in Latin America and carried out sweeping reforms (removal of price controls, widespread privatization, stabilization), while simultaneously liberalizing its foreign trade regime." Following a takeover by the military when the economy was in chaos (an average inflation of 150 percent and an average fiscal deficit of 16 percent of GDP during 1971-73), the government carried out its bold reforms. In September 1973, at the time of the takeover, tariffs ranged up to 750 percent with an average weighted tariff of 105 percent and prohibitive prior import deposit requirements. By June of 1979, virtually all quantitative restrictions had been abolished and a uniform tariff of 10 percent was established. During the 1983 financial crisis that followed the poorly-designed exchange-rate-based stabilization policy (see box 2), the uniform tariff was raised 20 percent, then again to 35 percent in 1984 (a higher rate was not adopted as Chile had joined the GATI), before being brought progressively to 11 percent by June 1991.12 Unlike many other Latin American countries, Chile has not used export subsidies to try to counteract the bias against export activities that occurs when protection is high. The approach of a low uniform tariff and no export subsidies has the advantage of transparency and of avoiding incentives to try and obtain subsidies (e.g. overinvoicing of exports). Chile has, however, introduced some export promotion policies in recent years. These include: (i) an extension of the drawback "1 For a detailed description and thorough interpretation of the 1973-83 period, see Edwards and Edwards (1987). 32 Chile has adopted tariff surcharges as antidumping policy for a few products, and there has been some use of reference prices. The protective effects of these measures have not been measured yet. 24 Table 4 Refonns in Chile 1950-70 1970-73 1974-76 1977-2 1983-1991 Overall Import substitution. Populist policies Military takeover Labor union activity Return of democracy in Stable democracy Sale of public enterprises suppressed since late 1973 1988 (0) Land reform (500 sold by 1979) now allowed; full wage indexation External Lae terms of trade loss Global limits on external Improvemet in terms of (exteral shock of 15% brrowig prgresssively trade starting in 1987 of GDP) elimited Trade policy Imports High effective Quotas; average protection Removal of QRs, Witdxrawal from Andean Average protection up to protction to of 105% maximum tariff of 35% Pact. Tariff of 10% by 35% then down to 10%; manulacturing (360% in June 1979 unweighted mean tariff 1967); QRs rate is I1I% in 1992, QRs removed. Exports Exchange rate Multiple exchange rate Crawling peg 1978-82 preannounced Some export promotion system exchange rate. policies. Copper stabilization fund Crawling peg Main results GDP growth of 3.89% Average fiscal deficit of Negative growth; Growth picks up; huge Financial crisis in 1983. 16% and 150% inflation. reduction in fiscal deficit capital inflows leading to Collapse in 1982-83 and inflation large real exchange rate (159% decline in GDP). appreciation and crisis in Strong growth starting in eady 1982 1985. Large FDI flows. 25 system introduced in 1988; (ii) exemption from the stamp tax for export business; (iii) delayed payments of duties on capital goods; (iv) the introduction in 1983 of a temporary admission regime; (v) a 10 percent of export value reimbursement for small exporters since 1985; and (vi) the creation in 1987 of a small fund for export financing. Of the five countries, Chile opened up its economy the most to foreign trade. The effects of this trade liberalization on the openness of the economy (measured by the ratio of exports to GDP) and on growth is impressive, especially when compared with the other countries (see figure 3). The cornerstone of Chile's success reflected in higher growth and booming exports has been the ability to reduce the fiscal deficit and provide a stable macroeconomic environment for exports through a sustained an stable real exchange rate that guarantees the profitability of exporting. This, of course, also means a competitive real wage. At the same time, an open import regime was also very important for the success of Chile's trade liberalization, as were the accompanying reforms that led to well-functioning labor markets. The end results have been impressive with substantial increases in TFP growth and more competitive product markets (as would be suggested by the discussion in box 1 and figure 2). Some of the effects of the reforms on manufacturing-sector performance are revealed by comparing concentration and profitability ratios in two census years: 1967 and 1979. In 1967, the average price-cost margin for manufacturing - a measure of oligopolistic behavior - was 48 percent. In 1979, when the trade liberalization has been completed and uniform 10 percent tariff had been adopted, the price cost margin had fallen to 32 percent. At the same time, many firms had exited, and concentration rose sharply. Further analysis showed that the reforms had resulted in contestable markets in manufacturing and that there was an "import-discipline" effect, namely that, after controlling for other factors, sectors which experienced the highest import penetration rates were those that experienced the largest decline in price-cost margins. Finally, during the early 1980s, the 26 process of rationalization engendered by trade liberalization (and other reforms) led to the exit of the least efficient firms and to an increase in TFP growth."3 A relevant aspect of the Chilean experience for FSU countries was the large-scale privatization that took place in the early years of the military takeover. In 1973, 39 percent of GDP was in the hands of SOEs. By 1989, only 12 percent of GDP was produced by SOEs, and the gross proceeds of privatization are estimated at 12 percent of 1989 GDP. However, the rapid privatization of the mid-1970s was accompanied by large-scale bankruptcies during the debt crisis of the early 1980s. While faulty macroeconomic policies carry part of the blame, the lack of a proper regulatory framework led to the build-up of a few large conglomerates ("grupos") involving interlocking ownerships with commercial banks that delayed adjustment substantially [see Galvez and Tybout (1986)].14 Of the five countries, Chile experienced the highest rate of transitional unemployment. During the 1960s, the "natural" rate of unemployment was around 5-6 percent, and it was only towards the end of the 1980s that unemployment returned to these historical levels. Thus there was a relatively long period of high unemployment in the 15-25 percent range."5 While there is disagreement about the specific numbers, it is generally agreed that, prior to the 1982-84 crisis, a large part of the unemployment was attributable to the effects of contractionary stabilization policies and then of layoffs following the closing up of public enterprises. However, about 3.5 percentage 13 For more details see de Melo and Urata (1986) and Tybout (1992). 14 Interlocking ownership in itself should not be a cause of concern, if ownership of firms by commercial banks is properly supervised, as in Germany and Japan. Is The figures in table 1 understate the level of unemployment as they do not include people in the public employment programs. When these are included the average, the average unemployment during 1980-83 was 22.4 percent and, during 1984-87, 19.5 percent. See Meller (1991, table 9). 27 points of unemployment has been attributed to the restructuring that accompanied trade liberalization."6 While the success of Chile's trade liberalization is the classic example of what is to be expected from a widespread trade liberalization, these transition costs should not be underestimated. Between 1975-81, while the economy was adjusting to the new system of incentives, open unemployment more than doubled to about 13 percent, more than doubling again during the 1982-83 crisis (it was still 14 percent in 1987) before returning to about 6 percent with the recovery of the late 1980s (Meller 1991, table 9). While the transitional high unemployment was largely due to the concurrent stabilization and privatization policies, it is likely that the removal of protection was accompanied by a transitory loss in jobs. Mexico (table 5). With a population of 86.2 million (1990), Mexico is the largest country in the sample. The relatively large internal market helped sustain a high growth through ISI (6.4 percent average GDP growth between 1950 and 1974), and moderate inflation. Protection was given through domestic regulations to key industries, most of which were public enterprises. Between 1977 and 1982 oil price rises sustained the economy, and by the time of the crisis in the early 1980s, import licensing and QRs combined with reference prices and ad-valorem tariffs to give very high protection. 16 Part of the high unemployment during the late seventies was due to full wage indexation on past inflation, so that real wages were rising when inflation was falling. On the other hand, the boom-bust cycle during 1979-81 would have been much less (and hence employment changes would have been dampened) had capital inflows been controlled and the exchange rate not been fixed. For further discussion, see Corbo, de Melo and Tybout (1986) and Condon, Corbo and de Melo (1990). This estimate of the unemployment due to trade liberalization is from Edwards and Edwards (1987, p. 122). It is worth noting that the removal of manufacturing protection led to strong job creation in agriculture and to a migration from urban towards rural areas. An alternative estimate by de la Cuadra and Hachette (1991) finds a small net employment creation from the trade reforms. However, their simulation methodology ignores the role of relative price shifts, assumes no effects of trade liberalization on nontradables, and does not account for the unusually strong effect of capital flows on expenditures during 1979-81. Their estimates are therefore, difficult to interpret. m - = * = I I. I U I a if SI ¶F1[ ij*i I I; I 111 LI ij iIJ**' 1* P U H.  0-. -a."--.3 1U  aft .9 LQ 0 **if*U3*if if90. 3 ;u.:  j -.4 - 1w --. a. u.  .-J U ci . 100 ,.l  -. St.B 2* =.- ri 1:1. .1  B if 3m.. a g* -* ii  I 3 i. r I. I' WI 11111 11.11 0 0 Fi 'U . - -- - 3 2ge2, =3 if4 ow  3 LA 'I 29 When the price of oil tumbled in 1982 and foreign interest rate increases raised the external debt service payments, foreign banks refused to roll over short term debt. A crisis ensued, as Mexico suspended interest payments on its external debt. The primary fiscal deficit reached 7 percent of GDP, and by 1985, the inflation rate was 57 percent. Starting in 1985, Mexico carried out wide ranging reforms while simultaneously starting to stabilize.'7 It joined the GATT in 1986, thereby binding its tariff structure and reduced its maximum tariff from 100 percent in 1985 to 20 percent in 1988. At the same time, import licensing was dismantled and the import licensing coverage ratio fell from 92 percent to 20 percent of non-oil tradable production. Also, in accordance with the Customs valuation code of the GAIT, the use of reference prices for valuation purposes (which had covered up to 25 percent of tradable production in 1985) was discontinued after 1988. Export subsidies were also dismantled in 1985, and import and duty drawback regimes for exporters were started in 1985. Some export controls remain for agriculture and agro-industry exports.18 With the collapse of the oil market in 1986, Mexico experienced a second sizable shock and inflation jumped to 159 percent in 1987. The economy has recuperated strongly since then (see figure 3). The Economic Solidarity Pact among the government, labor and private business accelerated strucural reform and resulted in an elimination of the fiscal deficit by 1989. Since 1989, inflation is below 30 percent and falling. Since the initiation of the reforms, non-oil exports have doubled their share in merchandise trade. The reforms have also been successful in restoring growth (GDP growth of 3.1 percent in 1989 and 3.4 percent in 1990). 17 A major component of the stabilization program was the sale of over 400 SOEs. Subsidies to SOEs were cut down in half (to 1.5 percent of GDP) during 1984-88, and the proceeds from privaization amounted to 3.S percentage points of 1990 GDP. 11 The progressive approach to trade liberalization in Mexico in which sizable import (and export) controls remain, bears resemblance to the Hungarian trade liberalization. 30 Mexico also benefitted from the "announcement effect" in 1990 of impending negotiations to form a free trade area (FTA) with the US. The developments are interesting for the recent negotiations between the EC with Czechoslovakia, Hungary and Poland that resulted in the Europe Agreements of 1992.19 The negotiations have largely been led by the Mexican desire to achieve an improved and more secure access to the US market. Clearly another major reason was to use the trade agreement to cement domestic policy reform and to attract foreign direct investment (FDI). As a result of Mexico's own trade liberalization measures and the prospects of an FTA, trade with the US more than doubled between 1985 and 1990. And FDI flows from the US into Mexico in 1992 were double the 1990 levels. Like Argentina's, Mexico's success with the simultaneous pursuit of stabilization and liberalization shows that the two can be combined rather than carried out in the more traditional sequence of stabilization followed by trade liberalization and domestic reforms, which was the path followed by several of the successful East Asian countries. In spite of sizable adverse external shocks, Mexico has also taken advantage of the opportunity to enter into an FTA with the US. The prospects of an FTA have also helped lower inflationary expectations, but there is some concern that the use of the exchange rate to lower inflation might lead to a real exchange rate overvaluation and a loss of competitiveness for Mexican non-oil exports. Uruguay (table 6). Starting in 1955, this country - once dubbed the Switzerland of Latin America because of its high standard of living and progressive social policies - embarked on a period of twenty years of stagnation that culminated in an economic and political crisis that brought military rule in 1974. In many ways, the sclerosis in Uruguay resembled that of some East European 19 Essentially the agreements stipulate that Czechoslovakia, Hungary and Poland must open their markets within ten years to EC goods while in return EC will open immediately its markets for "non- sensitive" goods but taking four to six years to remove tariffs and QRs for "sensitive" goods. Czechoslovakia, Hungary and Poland will also have to adopt EC competition policy even though their goods will still be subject to EC's anti-dumping policy, at least for some time to come. 31 Table 6 Reforms In Uruguay Period PROTECIION REFORMS CRISIS MILD REFORMS 1955-1974 1974-1981 1981- 1984 1984 - 1990 Overall pdicy ISI Military takeover in 1974. Return of democry in 1985. environment Development of a large Wide-ranging reforms Fuiancial liberalization welfare state. (financial libealization; preserved. Strong power of fiscal reform: price labor unions restored. decontrel) 1979; exchange rate based stabilization program; gradual price decontrol; and interest subsidies External factors Closing of EC market for Collapse of Argentina (1981) and crisis in Brazil. beef exports Trade policy Imports Capital goods imports Removal of prior deposit Maximum tariff rate reduced to 45 percent from Increase in minimum duty on subject to prior approval; requirement, import 75 percent in 1982 imported raw materials; few widespread tariffs and QRs; licensing and QRs; Referenc prices (anti-dumping measure). QRs remain (on agricultund progressive reduction and Minimum export prices 1983 goods); unweighted mean tariff unification of tariffs; rate is 15% in 1992; 3 tier tariff (10%, 17%, 24%). Exports Taxes on traditional Removal of export taxes; exports. tax rebate and interest subsidies for nontraditional exports until 1978 Exchange rate Dual foreign exchange 'Tablita' or preannounced Crawling peg to maintain a competitive real Reform of reference prices and market; periodic change in exchange rate exchange rate minimum export price system devaluations (1978-82) Main results Stagnation Increase in export share of GDP decline of 14.7 percent in 1982-83. Capital Sluggish growth; resturn of the Large fiscal deficit and high GDP; GDP growth 4.1 flight and debt crisis. welfare state. High inflation. inflation percent; appreciation of by 1974. real exchange rate, and Rising social conflicts. collapse in 1982. 32 countries. Very inefficient industrles protected by a complex structure of administrative protection, tariffs and an array of QRs that closed the country to imports so that low quality domestic products were all that was available. Powerful labor unions had essentially stifled private investment and during the period 1955-74, average yearly GDP growth was 1 percent. The initial conditions at the time the reforms were initiated in 1974 were similar to those in Argentina and Chile: a large flscal deflcit and high inflation. The trade liberalization included a removal of prior deposit requirements, import licensing and a progressive reduction in tariffs. These measures were accompanied by the removal of export taxes and strong export incentives for non- traditional exports (in the form of tax rebates and interest rate subsidies that reached 15 percent of export value). At the same time, the peso was made fully convertible, and a fiscal reform simplifled the tax system. While the adoption of an exchange-rate-based stabilization program in 1978 was poorly conceived (see box 2), and Uruguay could not isolate itself from the instability in Argentina and Brazil, the reforms led to a marked relative improvement in performance and to a noticeable increase in the export share in GDP (see figure 3). But in the end, after one takes into account the crisis years of the early eighties, the improvement was mild and average yearly GDP growth during 1974-89 was only 1.1 percent which is about equal to the historical average. Uruguay is a small economy. With only 3 million people, it needs to export to overcome a small internal market. In spite of a succession of tariff reductions resulting in a maximum tariff level of 4S percent for final goods (down from 75 percent in 1982), barriers to foreign products remain high because of the extensive use of reference prices for the application of tariffs. Very often the base for the reference prices is high, and calculations for some 500 products in 1989 show that the tariff reductions since 1986 had been more than compensated for by the protection granted through the use of reference prices with the application of reference prices boosting protection on average by 33 18 percentage points. High rates of protection through administered protection thus continued to act as a tax on exports. Uruguay is also potentially interesting for FStJ countries because of the extensive lobbying and rent-seeking that has taken place to try to appropriate the rents created by the complex system of protection. Rama (1991) reports on the number of lobbying activities carried out in Uruguay during 1925-83. He counted the specific protective measures that were carried out to protect particular interests (e.g. a higher tariff for a particular good) rather than general protective measures (e.g. a change in the maximum tariff level). Resolutions with a specific promoter amounted to 1849 measures over the 1925-83 period.' Adding the measures that concerned a small number of products to his list resulted in a total of 3973 measures over the period. In ftrther analysis of the data, Rama finds a significant positive correlation between the intensity of rent-seeking and sectoral output growth with a three year lag, suggesting that the rents can only be appropriated after investing in the favored sector. Furthermore, the positive correlation disappears after the fifth year and eventually becomes significantly negative with the stagnation of the 1960s reflecting the rent-seeking peak of the 1940s. Interestingly, rent-seeking activities did not diminish with the reforms in the 1970s. Lobbying shifted towards exporting activities as incentives were being provided to non-traditional exports. In spite of notable reforms starting in the middle of the 1970s, in comparison with the other countries considered here, Uruguay has continued to be governed by extensive regulations including complicated rules for wage bargaining with labor unions. Reforms have had some effect in reversing the stagnation that had set in since the mid-fifties, but not enough has been done to turn the economy X He finds for instance that 71 foreign-trade regulations werd approved for the benefit of a single textile firm, and 39 in favor of one firm in the rubber sector. 34 around from the ISI strategy. The lesson is that as long as extensive regulations prevail, lobbying and rent-seeking activity will be widespread, and resource use will be inefficient. 4. Regionaltsm and Trade LUberalizatione In the 1960s, Latin America engaged in a series of regional arrangements aimed at promoting growth and industrialization through the reduction of barriers to intra-Latin American trade. The agreements ranged from FTAs (e.g. Andean Pact and LAFTA) to commnon markets (CACM). It is generally agreed that this regional approach to trade liberalization was a failure. First, reductions in trade barriers were not across-the-board (as in the European cases of EC and EFTA) but on a product-by-product basis. This approach resulted in many exceptions. Second, high rates of protection were maintained (or erected) against third countries. Third, there was little scope for efficiency gains as the economies had little to trade with one another. And, for manufactured goods, when trade was created as a result of the preferential tariff reduction, partner countries were generally very high cost suppliers when compared with non-partners. As a result, the trade among regional partners reduced rather than increased welfare. For FSU countries during the transition, it can be argued that some sort of preferential trading arrangements would be warranted to avoid a sudden (rather than gradual) collapse in trade. However, the lesson from regional arrangements in Latin America during the 1960s is that they lowered, rather than raised, welfare. After two decades of no progress, recently there has been a sharp revival in regional integration. For this new wave of regionalism, the outlook is more favorable than for the first wave of regionalism in the sixties. This is so because the countries have on their own unilaterally reduced 2 For further discussion, see Nogues and Quintanilla (1991) and de Melo, Panagariya and Rodrik (1992). 35 Box 3 The Economics of Regional Integration Under almost all circumstances, a unilateral trade liberalization in which trade barriers are lowered against all trading partners is superior to the discriminatory approach followed by regional arrangements such as an FTA. This is because imports will always be supplied by the lowest cost supplier when there is no discrimination among trading partners, whereas this may not be the case under a regional arrangement, as a partner may displace imports from a low-source third country supplier and the government loses tariff revenue since partner imports enter duty-free. Regional integration usually starts with trade liberalization and most arrangements are either Free Trade Areas (FTAs) or customs unions (CUs). Common markets are established when there is not only a common external tariff (CET) as in a CU, but also free mobility of labor and capital across borders. Full economic integration occurs when other policies (e.g. tax policy regulations, and standards) are harmonized. One can show that one can always design a partial trade arrangement (involving less than full tariff elimination) that improves welfare for union members while leaving unaffected the welfare of third countries. One car. also show that there are benefits from being a large negotiator (as in the case of the EC) in multilateral trade negotiations as one is likely to strike better terms of trade than a single member in isolation. Finally one can also show that if the world divides into a few trading bloc each of which erects high barriers against countries (with free trade for members in the bloc), then the regional approach to trade liberalization which guarantees market access to one of the blocs can be superior to unilateral trade liberalization if the barriers imposed by each bloc are sufficiently high. Past regional integration schemes have usually been successful for developed countries but not for developing countries. This is because trading partners in developed country integration schemes were usually low cost producers, whereas this was not the case for developing countries. For developing countries more success usually accompanied the regional integration arrangements that were less ambitious and encouraged cooperation (e.g. common infrastructure, education, etc.) rather than trade liberalization. A number of recent regional trading arrangements have been between developed and developing countries. These arrangements have been sought by the developing country partner not only to secure market access and cement reform measures (e.g. as in the Mexican case), but also to acquire the institutions of the developed country partner (more stable macro policies, higher standards, etc.). 36 most of their barriers to trade. Under these circumstances, the scope for inefficient preferential arrangements is greatly reduced as partners are no longer high cost suppliers. Also, these arrangements are viewed as an alternative to the slow process of multilateral trade negotiation in the GATT and as a way to dilute the resistance to domestic lobbies that would likely oppose more strongly opening the doors to imports from all sources than to regional neighbors. In addition to the Economic Initiative of the Americas launched by the U.S. in 1991, which aims at launching a hemispheric FTA following substantial trade liberalization by Latin American countries, three regional arrangements have established ambitious targets for the next few years. The CACM has been revived in 1990 with the aim of establishing a CET structure of 5, 10, 15 and 20 percent by early 1993 and to harmonize all instruments of trade policy such as rules of origin and anti-dumping regulations. The Andean Pact has also been renewed with two new members, Ecuador and Peru, Joining Bolivia, Colombia and Venezuela. A CET structure of 5, 10, and 15 percent is to be implemented by 1994 (Bolivia will maintain her lower tariff structure and autos and agricultural products will be exceptions), Finally, Argentina, Brazil, Paraguay and Uruguay signed a Treaty in 1991 to establish MERCOSUR which is to result in free trade in goods and services and capital and labor mobility by 1994. However, a CET remains to be agreed upon for the MERCOSUR. It is too early yet to tell whether the agreed upon reduction in tariffs will be carried out even though so far the initial announced tariff reductions to intra-regional trade have taken place. Although the start seems promising, in the end the success of this regional approach to trade liberalization will depend upon limiting the scope for interim exceptions, as lobbies are likely to use the opportunity to oppose further liberalization. Likewise, rules of origin in the case of FrAs should put low limits to domestic content requirements and CETs in the case of CUs should also be kept low, preferably to the tariff levels of the partner with the lowest protection. If such pitfalls are 37 avoided, the new wave of regional integration in Latin America will at least avoid the negative efficiency effects that took place following the preferential arrangements of the 1960s. 5. Conclusions The experience of Latin America shows that trade liberalization can be carried out successfully at the same time that stabilization measures are implemented to reduce inflation and the real exchange rate is realigned to correct for external deficits. Indeed, it can be strongly argued that a rationalization of the foreign trade regime is essential for the success of stabilization measures since an open trading regime imposes discipline on the economy's activities and can be a powerful complement to stabilization. The experience of the five countries examined is consistent with this view, as all successful stabilization episodes were accompanied by trade liberalization. Second, far reaching reforms can be carried out fairly rapidly. This is especially relevant for EE and FSU countries which also have in place economic structures that have been heavily protected from international competition for a long time. Even though the extent of protection is likely to be higher in many FSU and EE countries, the Latin American countries examined here all had highly protected manufacturing sectors for three to four decades. lhird, success requires a comprehensive reform package. In the sample of five countries, Uruguay reformed the least and has, so far, the fewest signs of sustained growth. To get a sustained supply response, trade liberalization must be accompanied by the appropriate supporting policies, e.g. the removal of price controls in domestic product markets, of constraints In labor markets through institutional rigidities for labor use, and of constraints in capital markets through non-market allocations of credit. 38 Fourth, prudential regulation was overlooked during the banking sector reforms of the late 1970s in Argentina, Chile and Uruguay and in the extensive enterprise privatizations in Chile. This resulted in interlocking ownerships that resulted not only in wealth concentration but also in risky portfolios in the banking system. And the absence of fair deposit insurance contributed to excessively high deposit rates by commercial banks experiencing repayment difficulties on the part of borrowers. Thus financial sector liberalization should be accompanied by appropriate prudential regulation and fair deposit insurance, and the links between commercial banks and private sector firms should be checked. Fifth, countries that stabilize while opening up should carefully follow the evolution of the real exchange rate (and real interest rate), which is crucial for the long-run success of the reform. Too often, Latin American economies reversed liberalization policies because the real exchange rate was allowed to appreciate while the authorities were trying to reduce inflation. Indeed, the specter of real exchange rate appreciation is still today a potential threat for the recent reforms in Argentina and, to a much lesser extent, in Mexico. Long-run success of trade liberalization requires a competitive and stable real exchange rate. 39 References Chenery, H., S. Robinson and M. Syrquin (eds.) (1986). Industrialization and Growth: A Comparative Stldy, Oxford University Press. Condon, T., V. Corbo, and J. de Melo (1990). "Exchange-Rate-Based Disinflation, Wage Rigidity and Capital Inflows: Tradeoffs for Chile: 1977-81," Journal of Development Economics, vol. 32:113-31. Corbo, V. and J. de Melo (1987). "Lessons from the Southern Cone Reforms", The World Bank Research Observer. Corbo, V., J. de Melo and J. Tybout (1986). "What Went Wrong with the Recent Reforms in the Southern Cone," Economic Development and Cultural Change, vol. 34, no. 3, pp. 607-40. Corbo, V. and J. de Melo (eds.) (1985). Scrambling for Survival: How Firms Adjusted to the Recent Reforms in the Southem Cone, WBSW No. 764, World Bank. de ia Cuadra, S. and D. Hachette (1991). "Chile" in D. Papageorgiou, A. Choksi, and M. Michaely (eds.) Liberalizing Foreign Trade, Basil Blackweil, London. Dollar, D. (1992). "Outward-Oriented Countries do Grow More Rapidly: Evidence from 95 Countries" Economic DeveloDment and Cultural Chan2e. Edwards, S. and A. Edwards (1987). Monetarizing and Liberalization: the Chilean Experiment, Ballinger, Cambridge, Mass. Faini, R., J. de Melo, A. Senhadji-Semlali and J. Stanton (1991). "Growth-Oriented Adjustment Programs: A Statistical Analysis" World Development, vol. 19, no. 8, pp. 957-67. Havrylyshyn, 0. (1990). "Trade Policy and Productivity Gains in Developing Countries: A Survey of the Literature" The World Bank Research Observer. Meller, P. (1991). "Adjustment and Social Costs in Chile During the 1980s" World Development vol. 19(11), pp. 1545-62. Melo, Jaime de, A. Panagariya and D. Rodrik (1992). "Regional Integration: An Analytical and Empirical Overview" (mimeo) World Bank, Washington, DC. Melo, Jaime de and S. Urata (1986). "The Influence of Increased Foreign Competition on Industrial Concentration and Profitability" Tnternational Journal of Industrial Organization, 4:287-304. Nogues, J. and R. Quintanilla (1992). "Latin America's Integration and the Multilateral Trading System" (mimeo) World Bank, Washington, D.C. 40 Nogues, J. and S. Gulati (1992). "Economic Policies and Performance Under Alternative Trade Regimes: Latin America During the 80s", Report No. 16 Latin American and the Caribbean Technical Departnent. Rama, M. (1991). "Rent-Seeking in Foreign Trade Regulations: A Time-Series Approach" (mimeo), CINVE, Montevideo. Tybout, 1. (1992). "Researching the Trade Productivity Link", The World Bank Economic Review. Policy Research Working Paper Series Contact Title Author Date for paper WPS1019 How Effective are Directed Credit Anita M. Schwarz November 1992 M. Raggambi Policies in the United States? 37664 A Literature Survey WPS1020 Another Look at Population and Nancy Birdsall November 1992 S. Rothschild and Global Warming 37460 WPS1021 Measuring Welfare Changes from Jonathan R. Coleman November 1992 D. Gustafson Commodity Price Stabilization in Chris Jones 33714 Small Open Economies WPS1022 A New Approach to Evaluating Trade James E. Anderson November 1992 M. T. Sanchez Policy J. Peter Neary 33731 WPS1023 Tariff Index Theory James E. Anderson November 1992 M. T. Sanchez 33731 WPS1024 An Exact Approach for Evaluating Will Martin November 1992 D. Gustafson the Benefits from Technological Julian M. Alston 33714 Change WPS1025 Openness and Economic Kazi M. Matin November 1992 D. Ballantyne Performance in Sub-Saharan Africa: 38004 Evidence from Time-Series Cross- Country Analysis WPS1026 Financial Liberalization and Paul D. McNelis November 1992 A. Maraiion Adjustment in Chile and New Zealand Klaus Schmidt-Hebbel 31450 WPS1027 Lessons from Bank Privatization Guillermo Barnes November 1992 W. Pitayatonakarn in Mexico 37664 WPS1028 Socioeconomic and Ethnic Harry Anthony Patrinos November 1992 L. Longo Determinants of Grade Repetition George Psacharopoulos 39244 in Bolivia and Guatemala WPS1029 Controlling Tropical Deforestation: Robert T. Deacon November 1992 P. Pender An Analysis of Alternative Policies 37851 WPS1030 Measuring the Effects of Urban Alan J. Krupnick November 1992 P. Pender Transportation Policies on the 37851 Environment: A Survey of Models WPS1031 Measuring the Possibilities of Robert Bacon November 1992 P. Pender Interfuel Substitution 37851 WPS1032 East Asia and Pacific Region, Eduard Bos November 1992 0. Nadora South Asia Region Population My T. Vu 31091 Projections, 1992-93 Edition Ann Levin Policy Research Working Paper Series Contact Title Author Date for paper WPS1033 Latin America and the Caribbean My T. Vu November 1992 0. Nadora Region (and Northern America) Eduard Bos 31091 Population Projections, 1992-93 Ann Levin Edition WPS1034 Revising Financial Sector Policy David H. Scott November 1992 K. Waelti in Transitional Socialist Economies: 37664 Will Universal Banks Prove Viable? WPS1035 How Heavy Protection Affects the Wendy E. Takacs November 1992 D. Ballantyne Philippines' Motor Vehicle Industry 37947 WPS1036 Output Decline in Hungary and Simon Commander November 1992 0. del Cid and Poland in 1990-91: Structural Fabrizio Coricelli 35195 Change and Aggregate Shocks WPS1037 Vocational Secondary Schooling, Ana-Maria Arriagada November 1992 C. Cristobal Occupational Choice, and Eamings Adrian Ziderman 33640 in Brazil WPS1038 Determinants of Expatriate Workers' Ibrahim A. Elbadawi November 1992 A. Maraiion Remittances in North Africa and Robert de Rezende Rocha 31450 Europe WPS1039 Education, Extemalities, Fertility, Martin Weale November 1992 PHREE and Economic Growth 33680 WPS1040 Lessons of Trade Liberalization in Jaime de Melo November 1992 D. Ballantyne Latin America for Economies in Sumana Dhar 37947 Transition WPS1041 Family Planning Success Stories in Moni Nag November 1992 0. Nadora Bangladesh and India 31091 WPS1042 Family Planning Success in Two Jane T. Bertrand November 1992 0. Nadora Cities in Zaire Judith E. Brown 31091