Emi Nakamura on Is the Phillips Curve Getting Flatter

Video & Slides

On Thursday, April 22, Emi Nakamura joined Markus’ Academy for a lecture. Nakamura is a Professor of Economics at the University of California, Berkeley, and a research associate and co-director of the Monetary Economics Program of the National Bureau of Economic Research. Nakamura’s presentation was based on joint research with Jonothon Hazell, Juan Herreno and Jon Steinsson, The Slope of the Phillips Curve: Evidence from U.S. States.

Watch the full presentation below and download the slides here. You can also watch all Markus’ Academy webinars on the Markus’ Academy YouTube channel.

Executive Summary

• The Volcker disinflation led to a dramatic decline in inflation which suggests the Phillips curve was steeper in the 1970s and 1980s than in more recent decades. For example, during the Great Recession, despite large increases in unemployment, there wasn’t a large change in inflation. Stock and Watson document this decline in the “Phillips correlation” by looking at the relationship between the unemployment gap (versus the natural rate) and the year over year change in inflation. There was a negative correlation in 1960-1983,  but the relationship has weakened since then. 

• The old Keynesian Phillips curve suggests that a weak economy with high unemployment has little incentive to raise prices. In contrast, in an overheated economy, there is pressure to raise wages and prices, leading to inflation. Friedman and Phelps critiqued the old Keynesian Phillips curve on theoretical grounds, arguing that firms would start to anticipate higher inflation–leading to an added expected inflation term. 

• The New Keynesian Phillips curve, inflation is driven both by demand and supply factors, and by inflation expectations.  A very flat Phillips curve implies a very “Keynesian” world, where prices respond sluggishly to demand pressures.

• The New Keynesian Phillips curve can be “solved forward”  to show that long-run inflation expectations may drive large movements in current inflation, even if the slope of the Phillips curve is low. Hence, movements in the Fed’s inflation target can be the major determinant of current inflation independent of the Phillips curve slope. 

• Long-run inflation expectations have been very stable since the late 1990’s but fell dramatically before then. This yields a potential explanation of the rapid fall of inflation in the 1980s arising from regime shifts as opposed to a steep Phillips curve. 

• Using regional data and analyzing the slope of a regional Phillips curve for non-tradables offers promise in distinguishing the roles of the Phillips curve slope and long-run inflation expectations. This approach “differences out” the effects of regime changes affecting all states (such as the Volcker disinflation). This analysis suggests that the slope of the Phillips curve was relatively small even in the 1980s. Most of the reduction in inflation in the early 1980s was associated with shifts in beliefs about the long-run monetary regime. The estimated cross-sectional slope is consistent with the relatively modest movements in aggregate inflation in recent recessions (i.e., no missing disinflation). 

• In conclusion, future research should focus on movements in the long-run long-run inflation target.