by Alexander Zentefis
Economic uncertainty, to the extent that it can be measured, appears to be countercyclical, moving in opposite directions from the business cycle. Uncertainty tends to go up when the economy declines in a recession and wanes as the economy expands in a boom.
Given this observed pattern of elevated uncertainty at times of economic stress, a natural question arises: Does economic or policy uncertainty cause the stress, or does economic decline cause uncertainty? Or does something else cause both?
Visiting Scholar Nicholas Bloom presented his recent work exploring the sources and impact of uncertainty at the Becker Friedman Institute’s Macroeconomic Lecture Series on Dec. 4.
This is an active area of research, especially in the aftermath of the 2008 global financial crisis and, more recently, during the strained U.S. congressional talks over the country’s fiscal situation. “A lot of theory talks about uncertainty generating recessions, but there are plenty of theories that talk about recessions generating uncertainty,” Bloom noted.
Bloom, a professor of economics at Stanford University, defined economic uncertainty as variability in the potential values of forthcoming but indeterminate economic outcomes, such as prospective stock prices or gross domestic product growth. Unlike volatility (standard deviation), which measures historical variability and is essentially backward-looking, uncertainty is concerned with the future.
The challenge is that economic uncertainty is not directly observable. “Ideally, we would have an uncertainty barometer, measuring different levels of uncertainty. But unfortunately, it doesn’t exist in reality,” Bloom noted. (Bloom, Stanford Colleague Scott Baker, and Steven J. Davis of the University of Chicago Booth School of Business have developed an index to attempt to measure uncertainty.) However, there are a number of indirect proxies for uncertainty, including dispersion in aggregate stock market returns, productivity, and sales growth that can be tracked through time.
Collectively, these proxies suggest that uncertainty at the macroeconomic level rises when the economy slows and drops when it booms. Interestingly, this countercyclical behavior characterizes uncertainty at the microeconomic level as well. Dispersion in the growth rates of industries widens during economic downturns. “Some industries, as it happens, do very well in recessions, while many others do badly,” said Bloom. These differences in growth rates between industries are less drastic during normal economic times. The same can be said of the dispersion in industry stock returns: It swells during recessions and recedes during booms.
Uncertainty Runs Deep
This pattern in uncertainty pervades even deeper levels of the microeconomy. Firms within an industry and factories within a firm also show greater variability in sales growth and productivity during recessions, even after controlling for firm entry and exit. Still more remarkably, individual products see the range of their price changes expand during these periods. “Every level you look at, uncertainty goes up [in recessions],” Bloom said.
Some research has suggested that recessions may be good times for firms to experiment with their prices, in order to learn more about demand, which broadens the dispersion of individual product price changes, as observed in price data. And the wider dispersion in firm-level productivity could be an artifact of a greater number of unproductive firms that abound during recessions. This is in contrast to periods of economic expansion, when only the most productive firms enter the market and survive, compressing the dispersion in productivity.
Bloom’s own research relates to the opposite causal direction: how a spike in uncertainty can lead to a recession. Adopting insights from the literature on “real options,” which emphasizes that many investment and hiring decisions are partially irreversible, Bloom described how a sudden rise in uncertainty would induce firms to postpone new hiring and spending until the economic outlook became clearer. This is because adjustments to production plans are costly, and the chances of making a mistake are higher in periods of greater uncertainty. Firms optimally choose to sit tight for the time being.
This extra caution and delay causes aggregate investment, employment, and output to drop precipitously, leading to a recession. However, these uncertainty “shocks” tend to be short-lived; once they wear off, economic activity quickly rebounds. “Uncertainty works well as a driver [explaining] recessions, but it probably only accounts for one-third of the movement. You need something else to account for the rest,” explained Bloom.
A Perfect Storm
The exact causal direction between uncertainty and recessions may be difficult to ascertain, and it could be that the two are mutually reinforcing. But the importance of further study of uncertainty is clear. Elevated uncertainty can have real negative effects on the economy. These effects, Bloom suggested, tend to be notably more resistant to policy interventions designed to soften the impact of recessions.
When firms are less certain and therefore more cautious, they also tend to be less responsive to monetary and fiscal policy. Thus the traditional tools of policymakers are dulled in these periods. “It’s like a perfect storm,” Bloom noted. “Your policy instruments—interest rates and fiscal policy—are particularly ineffective when you want them to work, which is right when the recession hits and uncertainty is large.”
It’s equally troubling if the policymakers themselves add to the uncertainty by sending mixed messages on the future direction of regulatory, tax, or stimulus policies, Bloom concluded. So-called “policy” uncertainty could also induce firms to delay expenditure and hiring plans, contributing to a slow-down in the economy.