Inflation, following four decades of dormancy, has reemerged as the biggest problem facing the US economy today. In comments at last month’s Jackson Hole Symposium, Federal Reserve Chair Jerome Powell put it bluntly but accurately, saying “without price stability, the economy does not work for anyone.” In the same speech, Chair Powell warned that reducing inflation “will also bring some pain to households and businesses.”
In a recent conference paper, economists Laurence Ball, Daniel Leigh, and Prachi Mishra identify the proximate causes of today’s high inflation, and quantify the costs of bringing inflation back down. The authors conclude on a note even more sobering than Powell’s. “Reducing inflation,” the authors contend, “is likely to require higher unemployment than the Fed anticipates.”
Economist Jason Furman recently summarized the paper by Ball, Leigh, and Mishra in a Wall Street Journal editorial, calling it “the scariest economics paper of 2022.” Furman’s own analysis, based on the paper’s findings, suggests that “to bring price increases down to 2 percent, we may need to tolerate unemployment of 6.5 percent for two years.” Scary, indeed. But a closer look at the details of Furman’s analysis also reveals what policymakers can do to raise the odds of less-scary outcomes.
Ball, Leigh, and Mishra’s analysis is organized around a Phillips curve equation that links inflation (p) to three proximate determinants. First, there is expected inflation (pe). Second, there is labor market tightness, defined as the gap between the natural rate of unemployment (un) and the actual rate of unemployment (u). Third, there is a shock (e) reflecting disruptions to supply chains, global commodity markets, and so on. Thus, the equation reads:
p = pe + a(un – u) + e,
where the slope coefficient (a) measures the sensitivity of inflation to labor market tightness.
Ball, Leigh, and Mishra’s empirical implementation of this Phillips curve innovates in a number of ways, but three are most important when considering possible policy implications of their work.
First, Ball, Leigh, and Mishra acknowledge that “a central tenet of mainstream macroeconomics is that the inflation rate depends strongly on expected inflation.” This is why pe enters their equation with a “unit coefficient,” implying that shifting inflationary expectations affect actual inflation by the same amount.
Likewise, Ball, Leigh, and Mishra recognize a second tenet of contemporary macroeconomic theory: that the natural rate of unemployment may vary considerably, even over short periods of time. In fact, their most important contribution is to propose a new way of accounting for this variation in practice. Specifically, they measure more accurately the gap between the natural and actual employment rates by taking the ratio of posted job vacancies (v) to the unemployment rate (u). That this approach to measurement makes sense can be understood by observing that the continuing very high level of posted job vacancies signals that today’s labor market is considerably tighter than the unemployment rate alone would suggest.
Finally, they note that supply-side shocks like those to supply chains or oil markets have not only a direct and immediate effect on inflation, but additional “pass-through” effects on wages or other input prices that appear only with a lag. Thus, the shock (e) appearing in the Phillips curve may be more persistent than previously thought. Ball, Leigh, and Mishra implement this idea by including an average of current and lagged supply shocks in their estimated equation.
Ball, Leigh, and Mishra’s main conclusions are as follows. Their estimated model decomposes the 7 percentage-point increase in inflation since the end of 2020 into a 0.5 percentage-point increase due to higher expected inflation, a 1.0 percentage-point increase due to tighter labor markets, and a 5.5 percentage-point increase due to supply shocks and their pass-through effects. The estimated model suggests scenarios where the Fed is able to bring inflation back down while holding unemployment below 4 percent are exceedingly optimistic. Instead, as Furman emphasizes, unemployment will more likely rise to 6.5 percent. These results show that Chair Powell, though correct to note at Jackson Hole that disinflation will be costly, continues to provide a public assessment of those costs that is far too sanguine.
Ball, Leigh, and Mishra’s analysis, fortunately, provide clues to what Chair Powell and other policymakers can do to make the prospect of disinflation less scary. Firstly, the key role played by expected inflation in the paper’s Phillips curve implies that, to the extent that the Fed can bring expected inflation back down, actual inflation will follow without hardly any cost in higher unemployment. The best way for the Fed to do this would be to adopt and announce a specific monetary policy rule – or at least spell out a coherent multi-year plan – that would describe more fully to the public how the Fed plans to adjust its federal funds rate target to stabilize inflation under various scenarios, benign and more toxic. Secondly, since labor market tightness, not the unemployment rate alone, governs inflationary dynamics, leaders in Congress and the Biden administration should look for ways to help lower inflation by reducing the natural rate of unemployment. Fiscal and regulatory policies that currently discourage firms from creating jobs, and potential workers from looking for jobs, should be replaced by policies that favor job creation and search instead. Finally, as Ball, Leigh, and Mishra suggest supply constraints have persistent effects on measured inflation, policymakers should take every opportunity to ease those constraints. They might focus first, for example, on reversing recent policy actions that have stymied domestic oil and gas exploration and production.
Economists are correct in warning the public about the likely costs of disinflation. But leaders within and outside the Fed should not sit back and take these costs as given. The costs are “endogenous,” as economists also like to say. Policymakers can coordinate to make disinflation much less scary.