MNI: Stickier Prices Amplify Fed's 'Last Mile' Problem -Paper

New research on the tradeoffs between unemployment and inflation by academic and Federal Reserve economists finds the "last mile" of bringing inflation down to 2% much costlier than the road traveled thus far, co-author Virgiliu Midrigan told MNI.

Analyses of CPI microdata during Covid-19 and in the Great Inflation of the '70s and '80s show firms adjust their prices more often when price levels are rising quickly in the economy. That in turn exerts more upward pressure on inflation, meaning that the shape of the Phillips curve charting the relationship between unemployment and the rate of increase in prices can become steeper than assumed in standard macroeconomic models, said Midrigan, economist at New York University.

"We call that the inflation accelerator, and it is largely responsible for the steeper slope of the Phillips curve in high inflation periods," he said in an interview. Conversely, he added, the curve flattens out as inflation cools, meaning that greater labor market tightening is required for further deceleration of prices.

"When inflation is high and the slope of the Phillips curve is steep, it doesn’t cost you anything to bring inflation down. If inflation is not very high, then the slope of the Phillips curve is flat. As others have said, it’s easy to bring inflation down from 8% to 3%. It’s really hard to bring it down from 3% to 2%."

FIRMS ADJUST QUICKLY

Before the pandemic, firms typically updated their prices every 10 months. That interval began to shrink substantially in early 2021 as inflation climbed. By the start of 2022, price changes were occurring on average every five months, according to a study by Hugh Montag of the Bureau of Labor Statistics and Daniel Villar of the Federal Reserve Board of Governors.

"The frequency of price increases seems to be driving inflation and how firms are reacting to inflation," Montag said in an interview. "If firms have very sticky prices, there would be large real effects on the economy. In contrast, if they can adjust quickly, the real effects on the economy are smaller."

"This has been the biggest episode of elevated inflation we've seen in decades and firms have reacted significantly to it in ways important for propagating inflation," he said.

Price-setting frequency slowed to seven months on average in the third quarter of last year, the latest set of numbers crunched by the authors, still above pre-pandemic levels.

BIGGER SACRIFICE

Midrigan's paper, written with Andres Blanco of the Atlanta Fed, Corina Boar of NYU and Callum Jones of the Federal Reserve Board and not yet published, estimates that a doubling of the frequency of price updates knocks down the slope of the Phillips curve by a factor of four.

It also calculates the sacrifice ratio, or the drop in output suffered to reduce inflation by 1 percentage point over the course of one year, to be 1.4% during periods of lower inflation -- more than twice as much as the 0.6% loss needed when inflation was at its peak in 2022.

Midrigan cautioned a number of other factors, such as supply side improvements, may ultimately cause inflation, output and unemployment to behave differently in the real world. But this research is the first to incorporate observed price setting patterns on the part of firms into a model that can be used for forecasting and analysis of policy options.

"We kind of know how things work. The nonlinear Phillips curve makes a lot of sense. The hope is the model we propose can be used down the road to quantify those issues."