After a Fiscal Crisis, Economic Pain Fades Fast

The conventional wisdom says that financial crises cause severe and long-lasting economic damage, but that view may be much too simple, according to new evidence from University of California, Berkeley economists Christina Romer and David Romer.

Analyzing financial distress in 24 countries over the last forty years, they found that the downturns that followed were highly variable. On average, economic output declined only moderately and often temporarily — at least in modern, developed economies.

The findings are based on a dataset the husband-and-wife research team developed to avoid possible weaknesses of previous studies on the impact of financial crises. David Romer offered some background on this work when he and Christina Romer, a former chair of the Council on Economic Advisers, visited the Becker Friedman Institute in 2014–15.

First is the issue of deciding what constitutes a financial crisis, and when it starts and stops. As Romer and Romer describe in their paper,

… the existing chronologies sometimes differ substantially from one another. For example, both Reinhart and Rogoff (2009a) and the International Monetary Fund (IMF) Systemic Banking Crises Database (Laeven and Valencia, 2014) identify a crisis in Norway following the collapse of house prices in the late1980s; but Reinhart and Rogoff date it as occurring from 1987 to 1993, while the IMF dates it as running from the second half of 1991 to 1993.

What’s more, picking financial crisis dates retrospectively risks selection bias, Romer noted. “There’s the worry that if you identify the crisis after the fact, you know it was coming, and pick your date because bad things happened ….”

Another problem is that most previous research relied on a binary classification: you’re either in a financial crisis or you’re not. At best, Romer said, researchers used three categories: crisis, no crisis, or systemic crisis. That’s not the way the world works, and such classifications hide the severity and progression of financial problems, he noted.

To avoid these issues, they set out to build a richer measure of financial distress, derived from real-time narrative accounts of country conditions prepared by the Organisation for Economic Co-Operation and Development (OECD). Their sample includes semiannual reports from 24 advanced economies from 1967 to 2007. They write,

Our measure of financial distress is designed to capture a rise in the cost of credit intermediation. Using a detailed reading of the relevant sections, we look for discussions in the OECD Economic Outlook of such factors as perceived funding problems and rising loan defaults, which could reduce the willingness of banks to lend at a given safe interest rate. In this way, we focus on disruptions to credit supply, rather than on broader conceptions of financial problems. We classify the degree of financial distress on a scale from 0 to 15. Compiling a continuous measure allows us to take into account the severity and duration of financial distress, and to analyze how crises emerge and progress.

“We tried to focus on narrow question: if you wake up one day and your country is in a financial crisis, how is your country likely to do economically?” David Romer explained.

The results suggest the answer is that there will be some pain, but for many countries it doesn’t last long. In a moderate crisis (rated 7 on the scale of 0 to 15), industrial production and real GDP fell by 3 to 4 percent, fairly quickly. Romer and Romer’s data show that on average, industrial production begins to bounce back after six months, but GDP is depressed for at least five years. However, the persistence of the GDP effects is likely skewed by Japan, which since the 1990s has experienced a long period of financial distress and a very large slowdown.

“In the non-Japan sample, the distress is relatively short-lived. The economy falls hard, then the effect goes away, and in a couple of years, things are back to normal,” David Romer said. “This study suggests that a financial crisis isn’t inevitably a disaster, and so it moves you away from a focus just on prevention to also thinking about how to respond if a crisis occurs.”

He stressed that the study does not identify the causal links between financial crises and economic distress. But the close examination of historical data to draw conclusions from evidence is a common theme in the couple’s work together.

In that, they are inspired by Milton Friedman and Anna Schwartz’s monumental monetary history of the US, which identified the cause of the Great Depression.

“Friedman and Schwartz are heroes to us. Their monetary history is beyond seminal, and was life-changing for us. To bring together evidence to find causation for economic events is the theme of the work we’ve been doing together for 25 years.”