It happened in one country after another during the 1990s: a large, contractionary currency devaluation followed by surprisingly mild rates of inflation.
But when Kellogg School Finance Professor Sergio Rebelo and his co-researchers looked closely at the aftermath of the devaluations, they observed a remarkably similar set of economic reactions.
Whether in Finland, Sweden, Mexico, Korea, Thailand, Malaysia, the Philippines, Indonesia or Brazil, the same patterns in pricing emerged. In each country, the prices that changed most were those of imports and exports at the dock. The prices of these goods in retail stores were more stable, Rebelo’s team found.
“To sell a good in a retail store, you need to use local labor, real estate and transportation services,” explains Rebelo, the Tokai Bank Distinguished Professor of International Finance. These distribution-associated costs can be quite large, he adds - up to 50 percent for the retail price of the typical consumer good. The presence of these local costs tends to stabilize retail prices.
At the other end of the scale were the prices for goods produced solely for local consumption. Rebelo and his co-researchers noticed that the price of the so-called “nontradable goods” - housing, education, health and transportation - seemed least affected by currency devaluations. These distinctions make it clear that aggregate measures of inflation - such as the Consumer Price Index - are insufficient to understand the effects of large devaluations, Rebelo says.
“Retail prices behave very differently from producer prices. Tradable good prices behave differently from nontradable good prices. And the prices of brands that are purely domestic behave differently than the prices of non-local brands,” Rebelo says.
“Once these distinctions are made, it is not difficult to understand why inflation, as measured by the Consumer Price Index, is often so low after large devaluations.”
Rebelo’s research associates include Ariel Burstein, an assistant professor at the University of California at Los Angeles; Craig Burnside, a professor at the University of Virginia; and Martin Eichenbaum, an economics professor at Northwestern University. Their latest findings will be published in upcoming issues of the Journal of Economic Theory and the Journal of Monetary Economics.
The team has worked together for the past several years to gain a deeper understanding of currency crises and how they unfold. Recent crises have occurred in Mexico in 1994, Asia in 1997, Brazil in 1999, and Turkey and Argentina in 2001.
Rebelo believes most currency crises stem from the inability - or unwillingness - of governments to collect enough taxes to finance their spending. “When this happens, the government, sooner or later, has to resort to printing money or using other financing strategies that tend to destabilize the exchange rate,” he says.
The Asian currency crisis, explains Rebelo, was a prime example. In the mid-1990s, many banks in Korea, Thailand, and Indonesia were close to bankruptcy.
“To avoid an exchange-rate depreciation, governments in these countries needed to increase taxes or reduce spending to finance the bailout of bank depositors,” Rebelo says. “This move was politically very difficult, and that made the crises unavoidable.”
Rebelo notes that currency crises tend to be dramatic events, often associated with severe financial distress. “But they are also times in which there are a lot of opportunities for companies that understand how these crises unfold,” he adds.
With a firm grasp on the behavior of prices during devaluations, Rebelo and his co-researchers are now turning their attention toward developing theories that explain this behavior.
“The hope is that our studies will improve our understanding of the economic effects of movements in exchange rates,” he says.