Topics in Macroeconomics

National Bureau Of Economic Research Inc, Cambridge MA

Investigators

Abstract

This project seeks to improve our understanding of devaluation episodes. The first project studies the behavior of inflation in the aftermath of large contractionary devaluations. The second project investigates how the fiscal costs of banking crises that associated with currency crises are financed by the government. The third project characterizes the optimal time to abandon a fixed exchange rate regime in response to an increase in government spending that renders the peg unsustainable. The first project uses new data to understand how the prices of different goods behave in the aftermath of large devaluations. It examines a variety of episodes employing disaggregated Consumer Price Index (CPI) data together with new data obtained from a private provider that measures prices in different countries for a fixed set of brands. Some preliminary results suggest that there are two mechanisms important in shaping the behavior of prices in the aftermath of a devaluation. The first is the presence of distribution costs (transportation, wholesaling and retailing) that drive a wedge between the retail prices of tradable goods in different countries. The second mechanism is described as flight from quality in consumption. The large devaluations are associated with sizable recessions. The negative income effect associated with these recession induces agents to substitute away from imported good towards inferior local substitutes. This substitution can induce a significant downward bias in measured inflation rates. In principle the CPI measures the cost of purchasing a fixed bundle of goods. But in practice, in countries such as Korea and Thailand, the prices of imported goods are replaced in recessions by the prices of cheaper, inferior local alternatives in the CPl. This means that there is a sudden deterioration in the quality of the CPI consumption basket that is ignored in measuring the rate of inflation. The project documents the role played by flight from quality in consumption in influencing the behavior of CPI inflation. The second project explores the implications of different strategies for financing the fiscal costs of twin crises for inflation and depreciation rates. It uses a first-generation type model of speculative attacks which has four key features: (i) the crisis is triggered by prospective deficits; (ii) there exists outstanding non-indexed government debt issued prior to the crises; (iii) a portion of the government's liabilities are not indexed to inflation; and (iv) there are nontradable goods and costs of distributing tradable goods, so that purchasing power parity does not hold. The project shows that the model can account for the high rates of devaluation and moderate rates of inflation often observed in the wake of currency crises. The model and detailed fiscal and monetary data are used to interpret several twin banking/currency crises. Preliminary results have been obtained so far for the Korea 1997 and Mexico 1994 crises. The third project discusses the optimal time to abandon a fixed exchange rate regime in response to an increase in government spending that renders the peg unsustainable. The preliminary work considers two variants of an optimization-based first-generation speculative attack model. In the first variant there are fiscal costs of abandoning fixed exchange rates. These costs may represent a bailout of the banking sector, loss of tax revenues, difficulties in refinancing public debt, etc. The second variant incorporates a fiscal reform that makes the peg sustainable and that arrives according to a Poisson process while the exchange rate is fixed. In both cases it is shown that for moderate government expenditure shocks it is optimal to abandon the peg when international reserves hit a pre-specified lower bound. When the government expenditure shock is large it is optimal to abandon the peg as soon as the shock materializes.