Who S at Risk for the Next Financial Crisis

WHO’S AT RISK FOR THE NEXT FINANCIAL CRISIS?

by Steve Nadis [October 2011] HKS Research IMPACT

Faculty Researcher

Jeffrey Frankel, James W. Harpel Professor of Capital Formation and Growth, HarvardKennedySchool

Paper Title

Can Leading Indicators Assess Country Vulnerability?

Evidence from the 2008-2009 Global Financial Crisis (June 2011)

Coauthor

George Saravelos, Deutsche Bank

Observers tend to assume that every financial crisis is different. According to this perspective, the world is in so much flux

that the lessons gleaned from, say, the international debt crisis of 1982 or the East Asia/Russia financial crisis of 1997-1998 had little bearing on the global economic recession of 2008-2009. There is some truth to this viewpoint, in that the different crises had different causes and originated in different parts of the world. However, Kennedy School Professor Jeffrey Frankel and his co-author George Saravelos—a currency strategist for the Deutsche Bank and recent KennedySchool graduate--wondered whether some lessons might have endured across these historical episodes. Can the leading economic indicators that characterized which countries were impacted by one shock help in predicting which countries are most vulnerable to the next shock? Their recent paper finds that the answer is “yes,” implying that we don’t have to reinvent the wheel each time disaster strikes.

The 2008-2009 recession, the worst financial crisis since the Great Depression of the 1930s, was devastating to many people who lost their homes, jobs, or savings, or saw their businesses collapse. Yet because this event was, as Frankel and Saravelos put it, “uniquely broad and relatively synchronized across the global economy,” it a good test case for assessing the usefulness of key economic variables in signaling a country’s susceptibility to large shocks.

Frankel and Saravelos reviewed more than 80 studies from the pre-2008 literature to find out which variables were found most often to be the “leading indicators of crisis incidence.” Based on their meta-analysis, two variables stand out in this regard: central bank holdings of reserves, which are major foreign currencies (such as dollars, euros, and yen) and past movements in the real exchange rate (shifts in a country’s overall exchange rate relative to its long-run equilibrium). Countries with the lowest reserves and most overvalued currencies were systematically the hardest hit in the past. Other useful indicators include countries’ rates of national saving.

Frankel and Saravelos found that these same two variables-- reserves and the real exchange rate—were the two most important indicators for the 2008-2009 crisis as well. They reached this conclusion, which others had overlooked, by introducing three innovations. First, their analysis relied on five measures for how badly a country was hit, whereas earlier studies generally looked at a single measure such as currency devaluation or GDP losses. Second, whereas some researchers focused on the characteristics of the 2008-2009 event itself, trying to determinewhat was special about it, Frankel and Saravelos took the opposite approach, trying to see whether an early warning indicator that worked in past crises also applied to this event. It did, they concluded.

Finally, they tried to be more precise in spelling out when the crisis began (around the time of the Lehman Brothers collapse in September 2008) and when it ended (in the middle of 2009). Earlier studies had defined the crisis period as the entire year 2008, even though it had not gone global until September. Others included all of 2009; but that can be misleading if the episode ended in mid-2009. “If you’re trying to pick outthe fastest entry in a horse race, you see which horse got to the finish line first,” Frankel explains. “You don’t see how far a horse wandered around after the race is over, because that information is irrelevant.”

Even though he and his coauthor have concentrated on periods of major economic havoc, Frankel finds some reassurance in their overall finding. “It may be a mistake to assume the world is changing so fast that we have nothing to learn from the past,” he says. “It turns out that if you take a long-term perspective and see what has gotten countries into trouble in the past, that can be useful in determining which of them may be vulnerable to the next shock—whenever the next shock may come.”