Dornbusch's Overshooting Model After Twenty-Five Years, The Mundell-Fleming Lecture by Kenneth Rogoff, Economic Counselor and Director of the IMF Research Department

IMF News, November 29, 2001

Source details

Canonical URL: Dornbusch's Overshooting Model After Twenty-Five Years, The Mundell-Fleming Lecture by Kenneth Rogoff, Economic Counselor and Director of the IMF Research Department

Other formats

Bibliographic details

Published: November 29, 2001

Introduction

Lecture given at the Second Annual IMF Research Conference, Washington, D.C., November 29-30, 2001; Mundell-Fleming Lecture delivered November 30, 2001 (revised January 22, 2002).
Purpose: (1) convey why Dornbusch's "Expectations and Exchange Rate Dynamics" (Journal of Political Economy, 1976) has been so influential; (2) discuss empirical evidence for and against the model and sketch the model; (3) touch on competing notions of overshooting.

Why Dornbusch (1976) Matters

Characterized as "elegant" and "path breaking"; revived and reformed the Mundell-Fleming framework by incorporating sticky prices with rational expectations.
Practical relevance: remains a quick policy tool for assessing monetary policy effects on exchange rates; applied beyond exchange rates to issues such as "Dutch disease," exchange rate regimes, commodity price volatility, and disinflation in developing countries.
Conceptual insight: highlights interaction of sluggish goods markets with fast-clearing asset markets — a core idea retained in modern international macroeconomics.

Empirical Influence and Pedagogical Reach

Citation and teaching measures (1976–2001):
- 917 published articles citing Dornbusch (1976) in the social science citation index (major economics journals).
- 42 separate articles in International Monetary Fund Staff Papers cite Dornbusch (1976).
- Roughly 40 percent of the issues of Staff Papers published between 1977 and 2001 included at least one article citing the Dornbusch model.
- Peak citation years 1984-86: article received over 50 citations per year.
- Late 1990s: still receiving over 25 citations per year.
Popular and policy salience:
- Economist magazine: 14 articles including "overshooting" and "exchange rates" during 1999-2000.
- Financial Times: eleven references to overshooting during 2001.
- Cited in speeches by prominent central bankers (examples in 2001: Alan Greenspan in June 2001; Jean-Claude Trichet in May 2001).
Teaching: Dornbusch (1976) appears on virtually every advanced international finance course reading list surveyed.

The Basic Overshooting Mechanism (Model Intuition)

Core relationships:
- Uncovered interest parity: domestic nominal interest rate i equals foreign interest rate i* plus expected rate of depreciation Et (et+1 - et); e is log of exchange rate (home currency price of foreign currency).
- Money demand: real money balances depend on money supply m, price level p, and output y (logarithms), with interest sensitivity lowering money demand and output raising transactions demand.
Key assumptions driving overshooting:
- Domestic price level p is sticky (does not move instantaneously).
- Output y is exogenous in the short run (or adjusts sluggishly).
- Money is neutral in the long run: a permanent rise in m leads to a proportionate rise in e and p.
Thought experiment:
- A one-time unanticipated permanent increase in m with p fixed raises real balances m - p.
- To equilibrate money market, domestic interest rate i must fall.
- By uncovered interest parity, expected future appreciation is required; because long-run effect is proportional depreciation, initial nominal exchange rate depreciation must exceed the long-run depreciation — hence overshooting.

Theory: Formal Structure and Dynamics

Rational expectations introduced to international macroeconomics: expectations must be model-consistent.
Aggregate demand relates negatively to the real exchange rate; price adjustment follows a forward-looking sticky-price rule (Mussa-style).
Dynamic system reduces to two simultaneous difference equations for the real exchange rate and price adjustment, producing saddle-path stability under typical parameter values (< 1 in Rogoff's exposition).
Graphical solution: initial post-shock exchange rate lies at intersection of 45-degree line and saddle-path; initial nominal depreciation exceeds long-run depreciation (overshooting).

Empirics: Evidence For and Against Overshooting

Broad empirical consensus: difficulty systematically explaining short-run exchange rate movements for major currencies (Meese and Rogoff, 1983; Rogoff, 2001).
Model performance:
- Captures major turning points in monetary policy (examples cited: Volcker-era U.S. deflation, Thatcher-era U.K. deflation).
- Less successful at explaining many other large exchange rate swings.
Frankel (1979) generalization: under assumptions where monetary shocks predominate, generalized Dornbusch predicts a positive correlation between the real exchange rate and the real interest differential (high real interest rates bid up the real exchange rate).
Forward vs. spot comovement:
- Flood (1981) prediction: spot should move by more than forward in a one-time unanticipated money supply change; empirical series for yen/dollar and mark/dollar show movements in spot and forward series that are almost indistinguishable, with forward rate volatility slightly lower than spot volatility in sample — differences small.
Caveats and refinements:
- Empirical testing requires handling endogenous money supply, interest rates, real shocks, and alternative shock processes.
- Model may describe major monetary regime shifts but not all short-run volatility.
- Modern versions: when output is endogenous or money demand depends on consumption, overshooting need not occur.

Undershooting and Parameter Sensitivity

Parameter regimes matter:
- Rogoff presents case with parameter < 1 producing upward-sloping saddle-path and overshooting.
- If parameter > 1 (money demand sensitive to output and aggregate demand sensitive to real exchange rate), saddle-path can be downward sloping and the model can predict undershooting.
Modern models with money demand dependent on consumption make undershooting more plausible.

Competing and Complementary Overshooting Mechanisms

Alternative approaches contemporaneous with Dornbusch:
- Kouri (1976) and Calvo and Rodriguez (1977): overshooting driven by slow adjustment of national wealth/current account dynamics rather than sticky prices.
- These models emphasize wealth/current account channels and medium-term real exchange rate effects.
Integration:
- Dornbusch-type monetary sticky-price mechanisms mainly drive short-run dynamics.
- Kouri/Calvo-Rodriguez wealth/current-account mechanisms can be central to long-run real exchange rate changes.
- Unified frameworks (e.g., Obstfeld and Rogoff, 1995) can incorporate both channels.

Lasting Contributions and Policy Relevance

Main legacies:
- Revitalized Mundell-Fleming framework and sustained its policy relevance for decades.
- First mainstream marriage of sticky prices with rational expectations in international finance.
- A model of notable theoretical simplicity and pedagogical clarity that continues to inform policy thinking.
Policy takeaway:
- Dornbusch (1976) remains a useful, intuitive benchmark for analyzing monetary policy impacts on exchange rates, especially around major regime shifts, while acknowledging empirical limitations and the need for richer modern specifications.

Lecture: "Dornbusch's Overshooting Model After Twenty-Five Years," Kenneth Rogoff, Mundell-Fleming Lecture, Second Annual IMF Research Conference, November 30, 2001 (revised January 22, 2002).