Uncertainty Shocks Can Trigger Recessionary Conditions

Discussion of increased uncertainty is widespread these days.Recent examples include the latest OECD Economic Outlook, titled “Steering through Uncertainty,” and commentary by Bloomberg columnist Mohamed El-Erian. Uncertainty is currently also an important topic among many Federal Reserve policymakers, as noted in their public remarks.For example, see recent speeches by Federal Reserve Bank of St. Louis President Alberto Musalem, Federal Reserve Chair Jerome Powell and Federal Reserve Gov. Christopher Waller. Many economists and policymakers who have commented on the issue contend that rising uncertainty importantly reflects the difficulty associated with estimating the scope, magnitude and persistence of recent increases in tariffs on goods imported into the United States—as well as the retaliatory tariffs imposed on exports of U.S. goods—and their potential economic effect on the U.S. economy.

Over time, monetary policymakers have always confronted some degree of uncertainty in their decision-making process—if for no other reason than consistently predicting the future is impossible. Indeed, then-Federal Reserve Chairman Alan Greenspan famously remarked in 2003 that “uncertainty is not just an important feature of the monetary policy landscape; it is the defining characteristic of that landscape.”

But what exactly is meant by uncertainty? Many people often use risk and uncertainty interchangeably. However, economists are careful to note that these are two different concepts. Risk can be thought of as measurable uncertainty; that is, an outcome with a well-known, or reasonably well-known, distribution of outcomes. For example, what is the risk (probability) that it will rain tomorrow? By contrast, uncertainty occurs when the distribution of outcomes is unmeasurable. For example, will a nuclear war begin tomorrow? This distinction was made by the American economist Frank Knight in the early 20th century.The distinction between risk and uncertainty is discussed on page 219 in Against the Gods: The Remarkable Story of Risk, by Peter L. Bernstein, published in 1998 by John Wiley and Sons Inc. Thus, when policymakers use the term uncertainty, they probably have in mind a range of outcomes—from well-defined risks to unmeasurable uncertainty.See Alan Greenspan’s 2004 article, “Risk and Uncertainty in Monetary Policy,” in the American Economic Review. The latter is regularly called “Knightian uncertainty.”

Much research shows that increasing uncertainty can affect decision-making and thus economic activity through a number of economic channels. For example, firms may delay investment and hiring during periods of high uncertainty. Households may exercise precautionary reductions in spending by increasing their saving rates in anticipation of possible changes in incomes or wealth. Financing costs may also rise if risk premiums embedded in interest rates increase.

Measuring Economic Policy Uncertainty and Its Economic Impact

Because uncertainty is unobserved, a key challenge is devising a proxy to measure it. Some economists use financial market volatility as a proxy for risk. One notable example is the Cboe Volatility Index, or VIX; this series measures volatility in equity (stock) prices. But probably the best known and widely used measure of uncertainty is the Economic Policy Uncertainty (EPU) index developed by Scott R. Baker, Nicholas Bloom and Steven J. Davis.See Baker, Bloom and Davis’ 2016 article, “Measuring Economic Policy Uncertainty,” in The Quarterly Journal of Economics.

Measuring economic uncertainty is one thing, but it is quite a different thing to measure the corresponding effects of increased uncertainty. Laura E. Jackson, Kevin Kliesen and Michael T. Owyang (JKO) devised one method of tackling this challenge. In a 2020 paper, they used the EPU index to assess whether large increases in uncertainty—termed an uncertainty shock—affect the future growth of key variables like personal consumption expenditures for durable goods and business fixed investment.See Jackson, Kliesen and Owyang’s 2020 article, “The Nonlinear Effects of Uncertainty Shocks,” in Studies in Nonlinear Dynamics & Econometrics. In contrast with most of the literature on economic uncertainty, the authors used a nonlinear framework to study the effects of uncertainty shocks. In this framework, uncertainty shocks have larger economic effects than what is typically found in linear models. The authors outlined their preliminary findings in a 2019 Economic Synopses essay.

The first figure shows the JKO measure of uncertainty shocks since 1986. The uncertainty shock is calculated in the following manner: The value of the EPU in any current quarter is measured as the percentage change over the maximum value over the previous four quarters. If the change is negative, then the index value for that quarter is set equal to zero. Thus, only positive changes are shown in the chart. The figure shows the uncertainty shock that hit the economy in the first quarter of 2025 is historically large—surpassing the shock associated with the COVID-19 pandemic.

Economic Policy Uncertainty Shocks

DESCRIPTION: A line chart looks at the current quarter’s percent change from the maximum value of the economic policy uncertainty index in the previous four quarter. From the mid-1980s to 2019, most increases were in the range of about 5% to 25%, with the highest increase being 30% around the 1990-91 recession; many increases were also not associated with a recession. With the COVID-19 recession, the change reached 48%. In the first quarter of 2025, the change rose to 68%.

SOURCES: Author's calculations.

NOTES: Only positive changes are shown; negative changes are assigned a value of zero. The last observation is the first quarter of 2025. Gray shaded areas represent quarterly recessionary periods.

The modeling in JKO suggests shocks of this magnitude could lead to recessionary conditions. Further, it suggests the economic slowing is made worse in those cases because of nonlinear effects. That is, when uncertainty has recently been high, a further increase in uncertainty causes stronger contractionary effects, leading to larger reductions in economic activity—but also lower inflation—for several quarters after the shock. Under the assumption of nonlinear effects, a one-standard deviation shock in the EPU would be expected to reduce real GDP growth and employment by a little more than 1% over the next six to eight quarters, and inflation would be expected to decline by about 0.4% over the next 12 quarters, according to the modeling. The shock in the first quarter of 2025 was 3.8 standard deviations, which suggests larger responses would be expected based on this model.

Forecasts Continue to Point to U.S. Economic Growth in 2025

At this point, though, the consensus of professional forecasters surveyed by the Federal Reserve Bank of Philadelphia in February 2025 is that the economy will continue to expand in 2025 and 2026, with continued low unemployment and a modest further slowing in inflation. However, as shown in the second figure, many economists have recently been increasing their estimates of the probability of a recession over the next 12 months as of the end of March 2025; elevated economic policy uncertainty is probably a key factor.

The Probability of a U.S. Recession Sometime over the Next 12 Months

SOURCE: Bloomberg Survey of Professional Forecasters.

NOTES: The last observation is March 2025. Each observation is the median probability estimate for that month. The gray shaded area represents a recession. Unconditional probability is calculated by taking the number of months from October 1945 to October 2023 that the economy was in recession (124) divided by the number of months it was expanding (827).

Notes

1. Recent examples include the latest OECD Economic Outlook, titled “Steering through Uncertainty,” and commentary by Bloomberg columnist Mohamed El-Erian.
2. For example, see recent speeches by Federal Reserve Bank of St. Louis President Alberto Musalem, Federal Reserve Chair Jerome Powell and Federal Reserve Gov.Christopher Waller.
3. The distinction between risk and uncertainty is discussed on page 219 in Against the Gods: The Remarkable Story of Risk, by Peter L. Bernstein, published in 1998 by John Wiley and Sons Inc.
4. See Alan Greenspan’s 2004 article, “Risk and Uncertainty in Monetary Policy,” in the American Economic Review.
5. See Baker, Bloom and Davis’ 2016 article, “Measuring Economic Policy Uncertainty,” in The Quarterly Journal of Economics.
6. See Jackson, Kliesen and Owyang’s 2020 article, “The Nonlinear Effects of Uncertainty Shocks,” in Studies in Nonlinear Dynamics & Econometrics. In contrast with most of the literature on economic uncertainty, the authors used a nonlinear framework to study the effects of uncertainty shocks. In this framework, uncertainty shocks have larger economic effects than what is typically found in linear models. The authors outlined their preliminary findings in a 2019 Economic Synopses essay.