December 2023


Predictable financial crises

State Street LIVE: Research Retreat offers a wide range of academic expertise and timely market insights.

Better analysis of credit market conditions could be key to more early and accurate predictions of financial crises, according Robin Greenwood, professor of finance and banking at Harvard Business School and a State Street Associates academic partner.

Speaking at State Street LIVE: Research Retreat 2023 in London this month, Greenwood said he and colleagues had been working on a model that questioned assumptions among policy makers that the timing of financial market crises, such as that of 2008, are largely unpredictable.

He cited comments from senior United States government and central bank figures at the time of the Great Financial Crisis – former Treasury secretaries Tim Geithner and Hank Paulson and Federal Reserve Chair Ben Bernanke – which, according to Greenwood, demonstrate the US has, historically, been good at “firefighting” crises once they occur but not at “policing them beforehand.”

He claimed there was an assumption in financial and economic circles that modelling for these kind of predictions was a bigger part of policy makers’ activities than it actually is.

“But then when we started looking at it, one of the things we realized was that, in fact, while there was a recognition in the policy community that crises were preceded by weak economic fundamentals, there was still the general belief that crises were largely unpredictable,” he added.

Greenwood’s research was based on analysis of financial crises in more than 40 countries going back to the 1950s and looked at data from “conventional sources,” specifically focusing on credit conditions, leading him to conclude that “financial crises are the by-product of a credit bubble.”

The reason predicting credit bubbles was difficult is they tend to build up in otherwise benign economic and non-volatile market conditions.

“We try to capture this idea of a credit bubble using the combination of two factors,” he said. “The first is: that you have a period of large credit growth. Then, the second is: that is occurring concurrent with growth in asset prices.”

Greenwood’s research explored the correlation between both private debt and asset growth (i.e., household debt and house prices) as well as corporate (business debt and equity values). He noted that “the combination of these two variables is remarkable at predicting crises.”

“When credit markets are overheated in the sense that we define it, the probability of a crisis rises from about 5-to-8 percent over a three-year window to about 40 percent, he said. “What we find is that you have about a two-to-three-year horizon for forecasting crises. In fact, it's much easier to forecast in a two-to-three-year horizon than it is at a one-year horizon.”

This leeway for seeing a significantly enhanced likelihood of a financial crisis gives policy makers time to prepare for them, potentially “putting on the breaks a little bit.”

The model also showed that when household and business debt and asset growth all happen simultaneously it is “a killer indicator,” and the likelihood of a crisis grows to approximately two thirds, although he cautioned this happens very rarely.

Turning his attention to the present day, Greenwood said credit growth in North America and Europe has been “very modest” over the past three years, whereas two or three years ago, the indicators were closer to suggesting a crisis, although “not over the top, like in 2005, 2006.”

Various Asian countries, like Hong Kong, China and Thailand, are currently closer to the “red zone” of a likely financial crisis in two-to-three years’ time, and “the warning signs are a little more concerning” there, he concluded.

Stay updated

Please send me State Street’s latest Insights.