The Hutchins Roundup brings the latest thinking in fiscal and monetary policy to your inbox. Have something you'd like us to include in the next Roundup? Email us and we'll take a look.
This edition was written by Elijah Asdourian, Alex Conner, Georgia Nabors, and Louise Sheiner.
Even expected inflation can impact household welfare because not all U.S. government programs and taxes are indexed to inflation. A bout of inflation can increase the real value of a household’s tax liability and reduce the real value of government benefits. Using data on household finances from 2019 and a financial planning tool that calculates eligibility for fiscal programs, David Altig of the Atlanta Federal Reserve and co-authors find that inflation reduces lifetime household spending much more for very wealthy households than for the rest of households. A permanent increase in the inflation rate from 0% to 10% lowers lifetime household spending by a median of 6.8%, while it reduces spending by 15.9% for households in the top percentile of resources. The most important cause of this heterogeneity is taxes on asset income, which is disproportionately earned by wealthy households. Taxes are levied on nominal returns rather than real returns, and thus skyrocket during periods of high inflation.
Régis Barnichon and Adam Hale Shapiro of the Federal Reserve Bank of San Fransisco show that the vacancy-unemployment ratio (V/U) and proxies for the cost of filling vacancies outperform other measures of economic slack, like the output or unemployment gap, in predicting inflation. They show that V/U and vacancy filling cost proxies have one-year-ahead forecast errors 10% to 30% below other common slack measures from 2005 to 2023. The authors argue that their measures perform well because they capture a key determinant of firms’ real marginal costs. Vacancy filling costs attempt to proxy marginal labor costs directly, while V/U captures changes in marginal labor costs through shifts in matching efficiency between job seekers and hiring firms. The authors decompose the changes in V/U into movements along the Beveridge curve and shifts in the Beveridge curve, showing that the latter, which are driven by changes in matching efficiency, explain almost all of the improved forecasting performance over the unemployment rate alone.
"I hear from some quarters the claim that the FOMC has become 'overly data dependent.' This is a phrase that honestly doesn't make much sense to me but is apparently supposed to mean that we are over-reacting to data and allegedly sending confusing messages about the stance of monetary policy," says Christopher Waller, member of the Federal Reserve Board of Governors.
"I don't see how that argument applies to the views of the FOMC, if one looks at the Summary of Economic Projections (SEP). Between the March SEP in 2023 and March 2024 SEP, the Committee median was relatively consistent in projecting around three rate cuts in 2024. This was in the face of some pretty dramatic shocks to the economy. There were the bank failures and wider stress in the financial system in the spring of 2023, when it was far from clear what the ultimate effects would be on the economy. There were significant fluctuations in inflation, which was hot in the first half of last year and then dramatically cooler in the second half. There was the revelation, over time and in different data, that a surge in immigration was augmenting labor supply and allowing a surge in job growth with very little upward pressure on wages and inflation. And then there were geopolitical developments, such as the threat that war in the Middle East might spread to become a wider conflict. Against this backdrop, the median FOMC participant only gradually reduced their expectation for the unemployment rate at the end of 2024 and essentially left inflation unchanged…This hardly seems like an 'overly data dependent' Fed to me."
Join us for an event
The Economic Studies program at Brookings invites you to attend Sanctions on Russia: What's Working? What's Not? The event will be held in person and online on May 28 from 10:30 AM - 12:00 PM EDT.
