Presentation on theme: ". Catalytic IMF Finance in Emerging Economies Crises: Theory and Empirical Evidence by Giancarlo Corsetti and Nouriel Roubini."— Presentation transcript:
. Catalytic IMF Finance in Emerging Economies Crises: Theory and Empirical Evidence by Giancarlo Corsetti and Nouriel Roubini
Motivation for the paper Last decade: many crises in emerging markets: Mexico, Thailand, Indonesia, Korea, Malaysia, Russia, Brazil, Turkey, Argentina, Uruguay, Ecuador, Pakistan, Ukraine, Dominican Republic Emergence of external financing gaps in crisis even after current account adjustment as incipient capital flight and roll-off of short term claims. Thus, crisis resolution requires also official finance (bailouts) or private sector involvement (bail-ins) to fill the financing gap. Use of large catalytic IMF finance (“bailouts”) in some recent episodes. Roubini and Setser (2004) book on crises and its resolution addresses these policy options in some detail.
Policy debate If crises are mostly driven by panics and liquidity runs, an ILOLR may be optimal and eliminate liquidation costs of defaults caused by runs If crises are mostly driven by bad fundamentals and mismanagement (poor policies), moral hazard is an important issue and an ILOLR would lead to moral hazard. Thus, crisis resolution based on debt suspensions/standstills (bail-ins) may be better than bailout solutions.
Catalytic approach IMF/G7 current view: in between. Catalytic approach may work in these liquidity cases. Partial official support and policy adjustment may work to restore confidence, avoid roll-offs and return to market access Alternative view: only corner solutions of full bailout or full bailin can work. But can partial (catalytic) bailouts work or do you need bail-ins? Large catalytic finance used by IMF in most capital account crises of the last decade.
This paper We present a model of catalytic finance We present empirical evidence on effectiveness and success of catalytic finance based on the case studies of the eleven episodes of large catalytic finance in the last decade.
Empirical Evidence Based on Eleven Case Studies of IMF Catalytic Finance Mexico 1995 Argentina 1995 and 2000-2001 Korea, Thailand, Indonesia in 1997-98 Russia 1998 Brazil 1999 and 2001-2002 Turkey 2001 Uruguay 2002
Model of IMF’s Catalytic Finance. Main ingredients: Fundamentals and self-fulfilling panics play a role in the model. Thus, more an Allen and Gale model of fundamental banking crises than a D&D model of crises. Moral hazard is explicitly introduced and tradeoff is analyzed Global games (Morris and Shin…) allow to study both catalytic effects of IMF and moral hazard distortions (as you eliminate multiple eq.) Large player (the IMF) whose actions are endogenously determined based on preferences. This allows an analysis of catalytic finance. GNP as a measure of welfare, but government faces political costs to implement reforms (thus, basis for MH). So, objectives of government may differ from objective of country (GNP).
Main results Catalytic finance: Contingent liquidity support reduces the likelihood of a crisis. However, a partial bail-out can’t succeed if fundamentals are too weak. Catalytic effect depends on: Size of IMF liquidity support Precision of IMF info The sequence of moves in the game (i.e. if IMF moves before or simultaneously with private agents)
Moral hazard results Extent of moral hazard depends on the ex-ante likelihood of bad fundamental shocks: If fundamentals are likely to be relatively good (irrespective of government behavior), liquidity provision creates debtor MH (conventional view). If the probability of bad shocks to fundamental is high enough, however, an ILoLR actually provide an incentive for the government to implement beneficial but costly policies/reforms. I.e., IMF support may reduce, rather than increase, MH distortions.
Extensions IMF seniority (preferred creditor status). Optimal choice of size of liquidity support. Sequential games where the IMF moves first (IMF signaling to markets)
Implications of the model Large IMF programs should be “limited” – rather than being “unlimited” in the sense of covering the entire financing gap – to qualify as catalytic. Partial/limited IMF programs can be successful in catalyzing private capital flows to stay in rather than flee; however, the success of catalytic finance depends on the size of such programs. Large IMF programs may not necessarily induce debtors’ moral hazard; if they are limited and conditional, they may actually induce good policy behavior that would not have been induced in the absence of the financial support.
Implications of the model Catalytic finance is more likely to be successful if the economic fundamentals, while being weak, they are not excessively weak. IMF programs can signal to the market that an IMF with superior information about the country’s fundamentals, by deciding to lend to the country, is providing information that investors should stay in the country rather than rolloff their positions. An IMF program can signal that the crisis country is credibly committed to pursue painful economic reforms rather than reduce its policy effort, especially if the official lending does not lead to moral hazard but rather induces good policy effort.
Survey of the empirical literature Event case studies of crises and IMF programs Effects of IMF programs on private capital flows Effects of IMF programs on spreads
Empirical studies of recent capital account crises Eichengreen and Mody (2000) IMF (2002) Cottarelli and Giannini (2003) Hovaguimian (2003) Haldane (1999) Roubini and Setser (2004)
Empirical evidence Implications of forecast errors in IMF catalytic programs: they are more signals of program ex-ante optimism than catalytic failure. Issue figuring the appropriate counterfactual to assess the success of catalytic programs (growth, adjustment/reform, capital flows, deeper crisis). Catalytic finance may prevent broader unobserved runs on currency, bank deposits, domestic government debt. Are catalytic IMF programs partial or limited? Yes, but size does matter for success.
Empirical Evidence The effects of catalytic programs on short-run growth (not successful) and medium run growth (positive). Mixed experience with successes (Mexico, Argentina in 1995, Korea and Brazil in 1999) and failures (Indonesia, Russia, Argentina in 2001). Effects of catalytic programs on net private capital flows to the crisis country: stronger positive effects over the medium run. Speed of catalytic IMF loans repayments: rapid in success cases.
Empirical Evidence The reliance on augmentation of IMF programs is a problem with catalytic finance. Catalytic is consistent with soft forms of bail-in of the private sector. Catalytic programs did not generally lead to creditor and/or debtor moral hazard. They gave incentives to debtors to adjust and reform.
Empirical Evidence Catalytic is more likely to succeed after a move to a float, rather than in preventing the collapse of an unsustainable peg. Catalytic finance is more likely to succeed if fundamentals are weak, but not too weak. When debt levels are too high (signal of insolvency), catalytic tends to fail.
Summary of success and failures Clear successes: Mexico, Argentina 1995, Korea 1998, Brazil 1999. Clear failures: Indonesia, Russia, Argentina in 2001 Open cases: Turkey 2001, Brazil 2002, Uruguay 2002.
Lessons and conclusions 1.Large-scale catalytic financing works better when debt levels are low and the country’s commitment to reform is credible. 2.Large loans to countries with large debt levels are unlikely to be repaid quickly. 3.Rollover arrangements (appropriate bail-ins) can complement “catalytic” financing. 4.The implications of analytical models of catalytic finance are supported by the empirical experience. 5.The issue of whether and under which conditions the IMF should rely on catalytic finance remains highly controversial. This remains the most difficult issue in practice.