The Political Economy of Exchange Rate Policies In Latin America and the Caribbean Terms of Reference

Transcription

1 Banco Interamericano de Desarrollo Proyecto Red de Centros de Investigación The Political Economy of Exchange Rate Policies In Latin America and the Caribbean Terms of Reference I. Justification of the Study The exchange rate policies of Latin American nations were once quite similar: most pegged to the US dollar, a few had a crawling peg against the dollar. With the collapse of the Bretton Woods system, the region's currency policies diverged. Over the subsequent 25 years, they have continued to differ substantially, although some broad groups of experiences can be noted: * a fixed peg against an anchor currency, typically the US dollar * a target zone or band, of variable width, crawling or fixed, within which the exchange rate fluctuates * a floating rate, with various degrees of Acleanliness@, and no explicit target In addition, it seems clear that different national governments have pursued different policies toward the preferred level of the real exchange rate. Some countries appear committed to keeping the real value of their currencies Acompetitive@, that is relatively weak (depreciated). Others seem not to be particularly concerned about the impact of their exchange rates on export or import competitiveness. In some instances, this orientation has changed over time. Substantial economic analysis has been carried out to help clarify the impact of the choice of exchange rate regime, and of particular exchange rate policies, for economic growth. Fixing, floating, and bands -- and strong or weak currencies -- have been correlated with varying degrees of success in inflation reduction and control, export promotion, and adjustment, and a wide variety of other outcomes. In other words, economists have contributed substantially to our understanding of the impact of currency policies for aggregate macroeconomic trends. However, the study of the determinants of the exchange rate policies that governments actually adopt is embryonic at best. Indeed, it is striking that existing monetary authorities seem rarely to adopt exchange rate policies regarded by some analysts as optimal, or at least superior. This implies that macroeconomic optimality is only one of the factors affecting the choice of exchange rate policy, and suggests that there is room for investigation of other policy determinants. It is also clear that currency policies have a significant impact on other national economic policies. There is a direct relationship between the liberalization of cross-border financial flows, for example, and exchange rate policy: simple open economy macroeconomics implies that monetary policy in a financially open economy is fundamentally different than in a closed one. To take another example, trade policies and politics are often directly affected by currency movements: a strong real appreciation often leads to substantial pressures for trade protection. It thus seems appropriate to investigate the various economic, political, institutional, and other factors Anexo a p. 1

2 that affect the determination of national policies toward exchange rates. The proposed study is an attempt to develop both theoretical tools and empirical evidence to this end. II. Objective of the Study The objective of the study is to clarify the factors that affect national currency policy. Certainly a wide variety of potential explanatory factors can be adduced, but the hope is to narrow the range of explanations so as to allow some generalizations about determinants of national policies. It is anticipated that a major contribution of the project to the literature will be the explicit incorporation of political and institutional factors into the explanation of exchange rate policies, while almost all existing analyses have focused entirely on economic factors. There are several meaningful sorts of variation that need to be explored and eventually explained. Two dimensions of policy can be distinguished. First is the exchange rate regime, particularly the degree of fixity of the currency. As mentioned above, on this dimension Latin American experience runs the gamut from fully fixed to fully floating, with just about everything in between. Second is the general commitment to a particular level of the real exchange rate, from relatively strong (appreciated) to relatively weak (depreciated). The principal objective of the study is to explain why currency policies vary. Significant variation exists across countries, as Argentine policy differs from that of, say, Peru. Equally significant changes have taken place in the policy of particular nations over time: Argentine currency policies have varied substantially over the course of the 1970s, 1980s, and 1990s. The explicit introduction of political and political economy variables into explanations of currency arrangements is a major feature of the project. It is also to be hoped that the conclusions of the study will be relevant to analyzing current and future trends in exchange rates and exchange rate policies. There has been some talk about regional and sub-regional currency arrangements; a better understanding of how nations have arrived at past exchange rate policies and practices will help us understand what they might do in the future. III. Research Strategy The project involves both comparisons among different national experiences, and investigations of the evolution of particular national experiences over time. The time period under examination runs from 1960 until the present, though researchers who wish to draw lessons from a longer time horizon could also do so. The project will include a cross-national study of the determinants of exchange rate policies in the region in a broad context. This will involve a statistical analysis of the correlation of a variety of national-level economic, political, and other factors with national currency policy outcomes. The bulk of the project, however, involves detailed national case studies. These are not meant to be simple descriptions of exchange rate policymaking, although a degree of descriptive analysis is of course essential. Rather, the national studies are intended to be theoretically informed and analytically Anexo a p. 2

3 rigorous. The methodology (described below) indicates the sorts of theoretically grounded factors that the national studies will explore. They will do this in such a way as to highlight how policy changed over time in response to alterations in the international and domestic environment. They will also explicitly refer to the broad cross-national issues discussed in the first paper (described above), and in other national studies. The purpose is to allow both for comparisons among countries, and for an understanding of trends over time. The project directors will of course be principally responsible for ensuring a common framework and terms of discussion, and introductory and concluding materials will be important in this regard. However, it is hoped that the national studies will incorporate broader comparative considerations into their own analyses. The result will be a set of national cases that help build up to broader generalizable conclusions about the determinants of exchange rate policies across a variety of settings. IV. Methodology Both the cross-national study and the national case studies will address a set of well developed potential explanatory factors. These variables come either directly or indirectly from a variety of analyses of the political economy of macroeconomic (or other related) policies. They can usefully be divided into several categories. In most, a useful benchmark is the possibility of a fully fixed exchange rate, and this is used as an analytical baseline. A. Economic structure. The welfare costs of surrendering the exchange rate as a policy instrument can be weighed against the benefits of currency stability and flexibility. A large literature has considered these costs and benefits, typically in the context of considerations of the formation of a currency union, but the general implications are similar for a fixed rate. 1 The more effective exchange rate policy would be in achieving national goals, the greater the costs of giving up this policy independence by fixing the rate. Typical characteristics of national economies that can be used to whether a fixed exchange rate implies forgoing an effective policy instrument include the following (in most instances, the relevant evaluation involves comparisons with potential anchor currency countries: * The degree of factor mobility, especially capital mobility, between the two countries, which reduces the scope for independent monetary policy * The degree to which the two countries are subjected to significant uncorrelated exogenous shocks, which increases the desirability of an independent monetary policy. This is usually closely related to the similarity of the countries= structures of industry and exports. * The degree of diversification of the economy, which reduces the impact of exogenous shocks. * The general exposure of the economy to external shocks, which may require real exchange rate adjustments. 1 The seminal papers are Mundell (1961) and McKinnon (1963). For useful summaries of the enormous subsequent literature, seede Grauwe (1994), Edison and Melvin (1990), Krugman (1990), Masson and Taylor (1993), and Tavlas (1993, 1994). Anexo a p. 3

4 * The structure of the financial system, including for example the pervasiveness of dollarization or indexation in financial contract. B. Problems of credibility. In recent years, it has been suggested that governments may benefit from fixing exchange rates in an effort to increase the credibility of their monetary policies. 2 The government of a country with a history of high inflation may be serious about fighting inflation, but face difficulties convincing economic agents of this. Such a government can, in this view, attempt to bind itself to reducing inflation by tying its currency to that of a low-inflation country. In this sense, a currency peg serves as a commitment mechanism. This suggests that governments more in need of some way to establish their anti-inflationary credentials are more likely to choose to fix the exchange rate: credibility is most valuable to those who least possess it. Of course, it is extremely difficult to measure credibility, so direct application of this principle is hard to imagine. However, one ramification is that, everything else equal, countries with relatively high inflation are more likely to use this Anominal anchor@ strategy than those that do not face major inflationary difficulties. C. Macroeconomic conditions. National and global economic conditions will be reflected in national currency policies. The impact will vary, but it is important to keep changing circumstances in mind both in comparing among countries, and in looking at a single country over time. Among the macroeconomic factors likely to have a substantial effect on exchange rate policy are: * the state of the business cycle, domestic and/or international. * substantial changes in resource inflows or outflows, whether driven by terms of trade shocks, global monetary and financial shocks, or capital movements D. Interest groups. Currency policies have a differential impact on groups in society. This is true both with regard to the fixity of the exchange rate and its real level. On both dimensions, distributional considerations are likely to play a role (Frieden 1994). 2 Giavazzi and Pagano (1989) present an early such argument in the European context; see also Weber (1991). Anexo a p. 4

5 The exchange rate regime (fixed, floating, or something in between) can give rise to costs and benefits that vary among socioeconomic actors. Those heavily involved in cross-border economic activity presumably are, all else equal, most favorable to a fixed rate, as it increases the predictability of their international commercial and financial relations. On the other hand, economic actors with interests solely in one national economy -- especially producers of nontradable goods and services -- gain little by stabilizing exchange rates, and see governments forced to surrender monetary autonomy that they may value. So domestically oriented economic actors are unlikely to be so enthusiastic about fixing the exchange rate as those that are globally or regionally oriented. 3 In addition, where fiscal policy has come to depend on the inflation tax, the exchange rate regime might be implicated should it threaten such preexisting political arrangements. The level of the real exchange rate is also of distributional significance. Generally speaking, tradables producers benefit from a relatively weak (depreciated) rate, while nontradables producers benefit from a relatively strong (appreciated) currency. This is tempered, especially for tradables producers, by such factors as the importance of imported inputs in production, and the degree of Apass-through@ (i.e. the responsiveness of product prices to exchange rate movements). In many Latin American countries, private firms or individuals have also accumulated substantial foreign-currency liabilities whose burden would be made heavier with a real depreciation. Again, there may be a political tie with fiscal policy, should the currency's value affect the government's income or expenses. In this sense, the principal economic interests in society can have a significant impact on national currency policy. Policymakers may be loathe to pursue exchange rate initiatives -- even if they can be justified on aggregate social welfare grounds -- if they imply substantial losses to sectors of the economy, and of society, that are economically, or politically important. The sorts of interests most likely to be relevant include: * Firms and sectors with major cross-border financial or commercial interests, whether as net creditors, net debtors, or intermediaries * Firms and sectors heavily involved in price-sensitive export and import competition E. Institutions. Economic and other interests are only relevant to policymaking to the extent that they are mediated through social and political institutions. Such institutions are of practically infinite variety and significance, but some general characteristics, with implications for the making of international monetary policy, can be identified. 3 It should be noted that these considerations must be tempered with much detail about the specifics of sectoral and other interests. For example, for some internationally oriented agents, the existence of broad and deep forward markets eith er for their products or for their currencies may temper concerns about currency volatility. Anexo a p. 5

6 Societal institutions include the organization of business, labor, and other sectors of society. Disorganized or poorly organized groups are likely to be less relevant to policymaking than those that are cohesive and concentrated. The structure of the private sector -- such as relations among banks and industrial firms -- may well affect the character of business interests on the currency (Henning 1994). Some scholars also believe that the ease of adopting socially desirable policies -- such as inflation reduction and a stable exchange rate -- increases with the degree to which labor market institutions are centralized and wage bargaining is carried out at the national (as opposed to firm or industry) level. 4 Political institutions start with the national electoral system, which can have significant effects on the representation of interests. 5 Many scholars believe that representatives of larger constituencies are more likely to attend to broad social concerns, while those from smaller constituencies are more prone to reflect particularistic interests. Similarly, where the Executive is elected separately from the legislature, it may face different electoral concerns than parliamentarians -- typically more focused on broad national pressures than narrower local or sectoral ones. Closely related to the electoral system is the party system. Typically, analysts regard strong national parties as more likely to focus on broad policy concerns than weak, regionalist or personalistic political organizations. So too may the ideological, organizational, or regional base of political parties affect their impact on currency policy. Characteristics of electoral competition may also play a role in currency policy. Just as politicians can manipulate monetary policy for electoral or partisan purposes (Alesina 1989; Grilli, Masciandaro, and Tabellini 1991), so too might stages in the electoral cycle affect currency policy. Incumbents might use the exchange rate in order to stimulate the economy in the runup to an election, or after elections to reward their supporters. Policymaking institutions include, most prominently, the agencies assigned particular tasks related to exchange rate policy. These typically include the central bank, whose independence has frequently been associated (although less so in the developing world) with low and stable inflation; this implies that more independent central banks may contribute to currency stability and the viability of a fixed exchange rate (Cukierman 1992). The finance minister is typically directly responsible for exchange rate policy, and the character of ties between him, the central bank, and other ministers may also affect currency policies. 6 4 Calmfors and Driffill 1988 is a seminal statement; Garrett and Way (1995) provide an application in which public employee unions, of especial significance in Latin America, play an important role. 5 A very useful survey of electoral and partisan characteristics of the industrial countries is Lijphart (1994). 6 Although there is very little work on this with regard to exchange rate policy, some of the literature on relations among ministers in the fiscal process (see von Hagen 1992 and von Hagen and Harden 1994 for early examples) may be relevant. Anexo a p. 6

7 F. Ties to other economic policies. As already indicated, there are many instances in which exchange rate policy involves implicit or explicit tradeoffs with -- or more broadly, relationships with -- other policy areas. The choice of a strong currency may imply higher levels of trade protection than a weak currency. Opting to liberalize cross-border financial flows may affect the difficulties the authorities face in defending a fixed exchange rate. A fiscal contraction may be complicated by the relative-price and interest-rate effects of the exchange rate regime. In ways that will vary over time and across countries, links with other policies may well be worthy of close attention. G. Regional agreements. Although such discussions are embryonic at best (with the exception of some Caribbean countries), there is reason to believe that growing regional integration may encourage regional cooperation on monetary matters as well. This has certainly been the experience of the European Union, in which it is commonly held that regional commercial and financial integration has tended to reinforce trends for monetary integration, and vice versa (for a survey see Eichengreen and Frieden 1994; see also Frankel and Wei 1995) Whether Mercosur, NAFTA, the Andean Pact and other such regional arrangements will have explicit monetary concomitants remains to be seen. It is already clear, though, that regional trade integration increases the importance of currency movements among integration partners: it is not uncommon to hear complaints from one member of a regional trade agreement that another member is engaging in a Acompetitive depreciation@ that threatens the commitment to the trade agreement. This may seem a daunting array of potential factors of importance to national currency policy. It should be remembered, however, that their potential relevance may be bounded. For example, for studies that focus on change within one nation over time, many of these variables will be constant or nearly so: economic structure, interest groups, political and policymaking institutions. The task here is primarily to present the national characteristics in such a way as to highlight their continuing impact (especially in contrast with those of other nations), then to emphasize those factors that do in fact change over time. On the other hand, studies of a more comparative cross-national nature will tend to downplay passing changes over time, so as to emphasize enduring contrasts among countries. This might take the form of examining how different nations respond to a common shock -- the debt crisis of the early 1980s, the Mexican peso crisis of Or it might involve exploring longer standing cross-national differences -- Mexican and Venezuelan commitments to a fixed exchange rate over several decades, as opposed to Brazilian and Chilean use of a variety of floating and crawling currency arrangements. Generally speaking, it is expected that most or all papers will share some common experiences. In the 1970s, most of the region=s nations experienced favorable terms of trade shocks and capital inflows, and currency policies responded to these conditions in different ways. In the early and middle 1980s, the debt crisis again had a major impact on policy -- although, once more, policy divergence was the rule. And since the late 1980s, economic reform and the resumption of capital inflows have presented the region with new exchange rate policy problems and opportunities. Most papers will include consideration of national responses to these (partially) exogenous conditions. However, it is obvious that not all papers will necessarily address all of the factors mentioned above. However, they are all worthy of consideration both in explaining differences among Latin American national policies, and in analyzing divergent exchange rate policy responses to developments in the Anexo a p. 7

8 region. V. Anticipated results The country studies prepared for the project will, of necessity, vary in line with differences among the countries themselves. However, they are expected to share several common characteristics, as follows. A. A general description of the country=s exchange rate experience, including both choices of currency regimes and policies toward the level of the exchange rate. B. Presentation of the general economic and political setting within which exchange rate policy has been made, and analysis of how this setting has affected the broad trends of policymaking. C. Where possible, a systematic (statistical) analysis of aspects of exchange rate policy. This might include, if appropriate, analysis of legislative debates and votes on important currency issues; structured analyses of the impact of exchange rate movements on different industrial sectors and their trade policy preferences; and other available and relevant factors. D. Where relevant, a careful analysis of important episodes in exchange rate policymaking. This might have to do with the response of currency policy to particular exogenous shocks or crises, or important changes in political regime. VI. Contents of the proposals To participate in the project, research institutions should submit a research proposal of roughly 10 pages detailing: A. Why the proposed country presents an interesting case for the project, B. A discussion of the potential determinants, both political and economic, of exchange-rate policy that will be explored in the study, and of the hypothesized impact of these variables on currency policy. C. A discussion of the data and methodology that will be used to test the alternative hypotheses of interest. VII. Responsibilities and schedule This project will be coordinated by Jeffry Frieden (Harvard University) and Michael Gavin (Office of the Chief Economist, IDB). The deadline to submit proposals is September 5, The evaluation committee will complete the selection of the winning proposals by September 19, 1997, and written agreements with the institutions will be signed. An initial seminar will be held in the first week of October with the directors of the projects and the coordinating committee for the purposes of discussing the methodology and establishing more detailed commitments. The research centers will then be expected to begin the research phase of the project. First drafts of the paper will be due Anexo a p. 8

9 by the end of March, 1998, and a second seminar will be held to discuss the results at the end of April, After the seminar research centers will have approximately 6 weeks to refine the study and submit a final research paper. Studies that are considered of good quality will be published as working papers and potentially included as chapters in an edited volume. The proposals and research papers may be submitted in English or Spanish. VII. Financing The objective of the second Regional Research Center project of the IDB is to strengthen policy research capabilities in the Latin America and the Caribbean. To this effect, the Bank will provide as a contribution to the total budget of the study the equivalent in local currency of US$35,000. Funds will be disbursed in the following manner: a. 30 per cent within 30 days of signing the formal agreement between the IDB and the respective research center. b. 40 per cent within 30 days following the presentation and approval of the first draft of the research paper. c. 30 per cent upon approval by the Bank of the final research paper. Anexo a p. 9

10 References Alesina, Alberto (1989). "Politics and business cycles in industrial democracies," Economic Policy 8, pp Calmfors, Lars and John Driffill (1988). "Bargaining Structure, Corporatism, and Macroeconomic Performance," Economic Policy 6, pp Cukierman, Alex (1992). Central Bank Strategy, Credibility, and Independence: Theory and Evidence. Cambridge: MIT Press. De Grauwe, Paul (1994). The Economics of Monetary Integration 2nd edition. New York: Oxford University Press. Edison, H. and M. Melvin (1990). AThe Determinants and Implications of the Choice of an Exchange Rate in W.S. Haraf and T. D. Willett (eds.), Monetary Policy for a Volatile Global Economy (pp. 1-44). Washington: The AEI Press. Eichengreen, Barry and Jeffry Frieden (1994). "The political economy of European monetary unification: An analytical introduction" in Barry Eichengreen and Jeffry Frieden, eds., The Political Economy of European Monetary Unification. Boulder: Westview Press Frankel, Jeffrey A. and Shang-Jin Wei (1995). "European Integration and the Regionalization of World Trade and Currencies: The Economics and the Politics," in Barry Eichengreen, Jeffry Frieden, and Jürgen von Hagen, editors, Monetary and Fiscal Policy in an Integrated Europe. New York: Springer. Frieden, Jeffry (1994). AExchange rate politics: Contemporary lessons from American history@, Review of International Political Economy, 1. Garrett, Geoffrey and Christopher Way (1995). "The Sectoral Composition of Trade Unions, Corporatism, and Economic Performance," in Barry Eichengreen, Jeffry Frieden, and Jürgen von Hagen, editors, Monetary and Fiscal Policy in an Integrated Europe. New York: Springer. Giavazzi, Francesco and Marco Pagano (1989). AThe Advantage of Tying One's Hands: EMS Discipline and Central Bank Credibility@, European Economic Review 32, No. 5. Grilli, Vittorio, Donato Masciandaro, and Guido Tabellini (1991). "Political and monetary institutions and public financial policies in the industrial countries," Economic Policy 13 (October), pp Henning, C. Randall (1994). Currencies and Politics in the United States, Germany, and Japan. Washington: Institute for International Economics. Krugman, Paul (1990). "Policy Problems of a Monetary Union," in Paul DeGrauwe and Lucas Papdemos, eds., The European Monetary System in the 1990s. New York: Longman. Lijphart, Arend (1994). Electoral Systems and Party Systems: A Study of Twenty Seven Democracies Oxford: Oxford University Press. Masson, Paul and Mark Taylor (1993). "Currency unions: a survey of the issues," in Paul Masson and Mark Taylor, eds., Policy issues in the operation of currency areas. Cambridge: Cambridge Anexo a p. 10

11 University Press. McKinnon, Ronald (1963). "Optimum Currency Areas," American Economic Review 53, pp Mundell, Robert (1961). "A Theory of Optimum Currency American Economic Review 51, pp Tavlas, George (1993). "The 'New' Theory of Optimum Currency Areas," The World Economy 16, No. 6 (November). Tavlas, George (1994). "The Theory of Monetary Integration," Open economies review 5. von Hagen, Jürgen (1992). "Budgeting Procedures and Fiscal Performance in the European Communities." Economic Papers 96 (October). von Hagen, Jürgen and Ian Harden (1994). "National Budget Processes and Fiscal Performance." European Economy Reports and Studies 3, pp Weber, Axel. (1991) "Reputation and Credibility in the European Monetary System," Economic Policy 12, pp Anexo a p. 11