This paper studies the current state of inflation dynamics through the lens of the Phillips curve and assesses the degree of anchoring of inflation expectations. I first estimate a Phillips curve model with both past inflation and a constant anchor as explanatory variables over the 1999– 2018 period for a variety of measures of consumer prices.
My results show that the Phillips curve has shifted away from an accelerationist form toward a level form, but that shift is incomplete, particularly for core inflation. I then turn to survey measures of professional forecasters’ and consumers’ inflation expectations and assess the degree to which those expectations are anchored. My analysis shows that although professional forecasters’ expectations have been well anchored, consumers’ expectations have not.
Further analysis using multiple empirical measures of inflation expectations suggests that in the context of the Phillips curve, consumers’ expectations have generally outperformed professional forecasters’ expectations in terms of explaining and forecasting the dynamics of inflation over the past two decades.
In addition to analyzing the form of inflation expectations in the Phillips curve model, this paper examines the slope of the Phillips curve, or the sensitivity of inflation to cyclical fluctuations in economic conditions. I follow Stock and Watson and estimate the Phillips curve for various components of aggregate inflation.
The results show that whereas price changes for many service categories (particularly shelter) and food remain largely procyclical, price changes for most goods have been either noncyclical or countercyclical over the past two decades. That finding helps explain the flatness of the Phillips curve on the aggregate level as well as the variation in cyclical sensitivity among the different aggregate inflation measures.
Chosen excerpts by Job Market Monitor. Read the whole story at Inflation, Inflation Expectations, and the Phillips Curve: Working Paper 2019-07 | Congressional Budget Office