Summary by Jeromin Zettelmeyer
The second Annual IMF Research Conference was held in Washington, DC, on November 29–30, 2001. The conference focused on the economic consequences of large devaluations and currency crises, stabilization policies in emerging markets, and the political economy of economic reforms and IMF programs. The Mundell-Fleming lecture was delivered by the Chief Economist of the IMF, Kenneth Rogoff, in honor of Rudiger Dornbusch, whose influential “overshooting model” of the exchange rate was published 25 years ago. The conference agenda and brief descriptions of the papers follow.
The Consequences of Large Devaluations and Currency Crises
Why Are Rates of Inflation So Low After Large Contractionary Devaluations?
Ariel Burstein, Martin Eichenbaum, and Sergio Rebelo
Discussant: Ilan Goldfajn (PUC Rio and Banco Central do Brasil)
Credit Stagnation in Latin America
Adolfo Barajas (IMF) and Roberto Steiner (Universidad de los Andes)
Discussant: Alejandro Werner (Banco de México)
Boom-Bust Cycles in Credit-Constrained Economies: Facts and Explanation
Aaron Tornell (UCLA) and Frank Westermann (University of Munich)
Discussant: Paolo Pesenti (Federal Reserve Bank of New York)
Output Response to Currency Crises
Poonam Gupta (IMF), Deepak Mishra (World Bank), and Ratna Sahay (IMF)
Discussant: Nouriel Roubini (New York University)
Cheap Labor Meets Costly Capital: The Impact of Devaluations on Commodity Firms
Kristin Forbes (U.S. Treasury)
Discussant: Jingqing Chai (IMF)
Economic Integration and the Exchange Rate Regime: How Damaging Are Currency Crises?
Axel A. Weber and Günter Beck (Frankfurt University)
Discussant: Charles Engel (University of Wisonsin)
Stabilization Policies and Economic Reforms
The Effectiveness of Fiscal Policy in Stimulating Economic Activity:An Empirical Investigation
Emanuele Baldacci, Marco Cangiano, Selma Mahfouz, and
Axel Schimmelpfennig (IMF)
Discussant: Alberto Alesina (Harvard University)
Interest Rate Effects on Output: Evidence from a GDP Forecasting Model for South Africa
Janine Aron and John Muellbauer (Oxford University) Discussant: Eswar Prasad (IMF)
Why Do Many Disinflations Fail?
Javier Hamann and Alessandro Prati (IMF)
Discussant: Holger Wolf (George Washington University)
IMF Programs: Who Is Chosen and What Are the Effects?
Robert Barro (Harvard University) and Jong-Wha Lee (Korea University)
Discussant: Tim Lane (IMF)
Budget Support versus Project Aid: A Theoretical Appraisal
Tito Cordella and Giovanni Dell’Ariccia (IMF)
Discussant: Craig Burnside (World Bank)
What Determines Individual Preferences Over Market Reform? Microeconomic Evidence from Russia
Stephanie Eble and Petya Koeva (IMF)
Discussant: Anders Åslund (Carnegie Endowment for International Peace)
What Explains the Success and Failure of Fund-Supported Programs?
Anna Ivanova (IMF), Wolfgang Mayer (University of Cincinatti), Alex Mourmouras and George Anayiotos (IMF)
Discussant: Patrick Conway (University of North Carolina)
Conditionality and Ownership in IMF Lending: A Political Economy Approach
Allan Drazen (Tel-Aviv University)
Discussants: Jeffry Frieden (Harvard University), and Mohsin Khan (IMF)
The papers presented in the first set of sessions examined the consequences of large devaluations and currency crises from four angles: inflation, credit, real output, and economic integration. Burstein, Eichenbaum and Rebelo ask why inflation after large contractionary devaluations has typically been much lower than purchasing power parity would predict. Their paper gives an answer that is consistent with the law of one price holding for pure tradeables. First, they point out the share of internationally tradeable goods in the CPI is much lower than typically assumed once one accounts for distribution services and goods produced locally for the domestic market. Second, they present an equilibrium model in which a devaluation generates low nontradeables inflation because the crisis tightens private borrowing constraints. Barajas and Steiner present a set of stylized facts on the credit slowdown in several Latin Amercian countries since 1998, and attempt to econometrically disentangle the causes of the slowdown for Colombia, Mexico, and Peru. Tornell and Westermann document the stylized facts of boom-bust credit and output cycles in emerging markets in the last two decades. They go on to present a tradeables/nontradeables model with asymmetric financing constraints and balance sheet effects that rationalizes these cycles, and use the restrictions implied by their model to estimate a structural vector autoregression.
Using a large panel of countries and crisis episodes, Sahay, Gupta, and Mishra find that the consequences of currency crises on GDP differ widely. About 40 percent of currency crises in fact seem to be expansionary. Crises tend to be contractionary if they are preceded by large capital inflows, occur at the height of a boom and in an open capital account environment, and if competitors also devalue. Forbes studies the effects of devaluations on the output growth and profitability of commodity firms between 1996 and 2000, and finds that devaluations tend to stimulate output and profits in the short run, but that long-run effects are mixed and depend on capital labor ratios and changes in real interest rates. Finally, Weber and Beck examine economic integration, measured by the degree to which consumer prices move in tandem across different locations. They find that the major currency crises of the 1990s had significant disintegration effects by this measure, both through increased “border effects” (i.e., disintegration across countries) and by increasing relative price dispersion within countries.
Four of the papers presented on the second day dealt with issues associated with stabilization policies and economic reforms: Aron and Muellbauer, who look at monetary transmission in a particular country case (South Africa), Baldacci, Cangiano, Mahfouz, and Schimmelpfennig, who explore the effectiveness of fiscal policy in responding to a recession in a broad selection of countries, and Hamann and Prati, who ask why some stabilizations are quickly reversed while others have more enduring success. The main finding of Hamann and Prati’s paper is that about 85 percent of dichotomous outcomes (success or failure) can be predicted without knowledge of poststabilization domestic variables. Exchange rate-based stabilizations appear more likely to succeed, and initial conditions such as the level of precrisis inflation, a history of inflation, and the structure of political institutions matter. Finally, Eble and Koeva look at the determinants of support for market reforms using a Russian longitudinal household survey. Their main finding is that economic self-interest goes a long way toward explaining why individuals are opposed to or in favor of market reforms; however, ideological factors also appear to be important in explaining the preferences of particular groups, such as pensioners.
The remaining four papers focused specifically on issues related to IMF (or other international financial institutions) lending and conditionality. Barro and Lee show that IMF stabilization loans have no effect on long-run growth once the endogeneity of the lending decision—which may depend on borrower characteristics which themselves affect growth—is properly dealt with (simple regressions show a negative relationship). Their analysis uses political and institutional variables, such as the size of the borrower country’s quota, its political proximity to the U.S. and other big shareholders as regression instruments. Cordella and Dell’Ariccia show that an international donor’s choice between conditional budget support and project aid should depend on the trade-off between two distortions: the fungibility problem associated with project aid, and the inability to observe all relevant government actions in the case of conditional budget aid. Ivanova, Mayer, Mourmouras, and Anayiotos analyze the determinants of IMF program implementation and find that domestic political economy factors, such as the strength of special interests, lack of political cohesion, and ethnic and linguistic divisions contribute crucially to program failure.
Finally, Drazen surveys the debate on the relationship between conditionality and ownership, pointing out some incongruences and suggesting how they could be resolved. He then develops the case in which there is full ownership on the side of the borrower country government, but conditionality is nevertheless desirable because it strengthens the position of the government vis-à-vis an interest group that opposes reforms.
Selected papers from this conference will be published in a special issue of IMF Staff Papers in 2002.
Proceedings of IMF conferences and seminars, including agenda and papers, can be obtained through the “Conferences, Seminars and Workshops” link at the Research at the IMF website at www.imf.org/research.