|
How
Do Currency Crises Spread? Macroeconomic
fundamentals are not enough to explain the currency crises of the 1990s.
These currency crises were regional because trade is regional, and
contagion tends to spread between countries with tight trade linkages.
This linkage is intuitive, economically significant, statistically
robust and the key to understanding the regional nature of speculative
attacks. This was the essence of the views expressed by Andrew
Rose (Haas School of Business, University of California, Berkeley,
and CEPR) at a CEPR lunchtime meeting in London on 2 October 1998. In
elaborating on these views, Rose noted first that the world had
experienced three waves of speculative attacks on fixed exchange rates
in the 1990s. The attacks on the European Monetary System in 1992–3
had forced a number of devaluations, with floatations of the Finnish
markka, sterling, the Italian lira and the Swedish krona, and,
eventually, the widening of the EMS bands to ±15%. The meltdown of the
Mexican peso in late 1994 had been followed by ‘Tequila hangover’
crises in Argentina and Brazil. And the collapse of the Thai baht in
July 1997 had been quickly followed by speculative attacks on the
currencies of Malaysia, the Philippines, Indonesia, Hong Kong and Korea. A noteworthy, indeed obvious, feature of all these currency crises was that they had been regional. Yet standard economic models did not predict that such crises would be regional, at least not without auxiliary features. Most economists thought about currency crises using one of two standard models of speculative attacks. ‘First-generation’ models directed attention to inconsistencies between an exchange-rate commitment and domestic economic fundamentals, such as an underlying excess creation of domestic credit, typically prompted by a fiscal imbalance. ‘Second-generation’ models viewed currency crises as shifts between different monetary policy equilibria in response to self-fulfilling speculative attacks. Common to both classes of models was their emphasis on macroeconomic and financial fundamentals as determinants of currency crises – and macroeconomic phenomena were not generally regional. Thus it was hard to understand why currency crises exhibited such a clear regional character, at least without an extra ingredient explaining why the relevant macro fundamentals were intra-regionally correlated. A second
feature of the crises was that they tended to be ‘contagious’. This
was more readily explicable, given that countries were linked by trade
– and, unlike macroeconomic phenomena, trade patterns tended to be
regional. Countries tended to export to, and import from, other
countries in geographic proximity. Prima facie, therefore, trade linkages seemed an obvious place to
look for a regional explanation of currency crises. It was easy to
imagine why the trade channel might potentially play this role. If
prices were sticky, a nominal devaluation would deliver a real
exchange-rate pricing advantage, at least in the short run. Countries
lost competitiveness when their trading partners devalued; they were
therefore more likely to be attacked – and to be forced to devalue –
themselves. Of course,
this channel might not be important in practice: nominal devaluations
did not necessarily result in real exchange-rate changes for extended
periods of time. Moreover, devaluations were costly and could be
resisted. Making the case for the trade channel was thus primarily an
empirical exercise. None the less, casual empirical observation was
highly suggestive. For example, once Thailand had floated the baht, its
main trade competitors (Malaysia and Indonesia) were suddenly at a
competitive disadvantage, and so were themselves likely to be attacked. Rose
proceeded, however, to present persuasive empirical evidence confirming
that trade linkages were the primary channel through which currency
crises spread. This evidence was derived from a recent CEPR Discussion
Paper, written jointly by Rose and Reuven Glick (Federal Reserve Bank of
San Francisco), in which they had systematically assessed the role of
trade linkages as a channel for contagion. Using data from five waves of
speculative attacks that had led to currency crisis episodes (in 1971,
1973, 1992, 1994–5, and 1997), Rose and Glick estimated equations
which predicted the probability of a crisis and the strength of pressure
on the exchange rate, as functions of trade variables and macroeconomic
variables. The
regression results showed that trade variables had a consistently
stronger effect than the macroeconomic fundamentals. These results were
consistent with the hypothesis that currency crises spread because of
trade linkages. The evidence also confirmed that countries that trade
and compete with the targets of speculative attacks are themselves
likely to be attacked, whatever their economic fundamentals. Such crises
therefore affect clusters of countries tied together by international
trade – a linkage that is important in understanding both the regional
nature and the order of speculative attacks. Thus, for example, once
Finland had floated the markka in 1992, Sweden – as Finland’s most
important trading partner – was next in line. After Sweden was
attacked, the crisis logically spread south in turn to Sweden’s
competitor, Denmark. A similar pattern had characterized the sequence of
events after the Thai baht was floated in July 1997. As Rose
pointed out, if it is true that countries may be attacked because of the
actions (or inaction) of their neighbours, who tend to be trading
partners merely because of geographic proximity, then this is an
externality with important implications for policy. The existence of
this effect would constitute a strong argument for international
monitoring. It would also warrant a lower threshold for international
and/or regional assistance than would be the case if speculative attacks
were solely the result of domestic factors. In seeking to model ‘contagion’ in currency crises, Rose did not rule out the possibility of (regional) shocks common to a number of countries; nor had he attempted to study the timing of currency crises. Instead, his research had been designed to show that, given the occurrence of a currency crisis, the incidence of speculative attacks across countries was linked to the importance of international trade linkages. In short, currency crises spread along the lines of trade linkages, after accounting for the effects of macroeconomic and financial factors. Indeed, macroeconomic factors consistently failed to offer much help in explaining the cross-country incidence of speculative attacks.
|