The Phillips Curve is Back: Using Panel Data to Analyze the Relationship Between Unemployment and Inflation in an Open Economy

Working Paper: NBER ID: w7328

Authors: John DiNardo; Mark P. Moore

Abstract: Expanding on an approach suggested by Ashenfelter (1984), we extend the Phillips curve to an open economy and exploit panel data to estimate the textbook 'expectations augmented' Phillips curve with a market-based and observable measure of inflation expectations. We develop this measure using assumptions common in economic analysis of open economies. Using quarterly data from 9 OECD countries and the simplest econometric specification, we estimate the Phillips curve with the same functional form for the 1970s, 1980s, and 1990s. Our analysis suggests that although changing expectations played a role in creating the empirical failure of the Phillips Curve in the 1970s, supply shocks were at least as important.

Keywords: Phillips Curve; Unemployment; Inflation; Open Economy; Panel Data

JEL Codes: E31; F41; J64


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
supply shocks (E39)empirical failure of the Phillips curve (E31)
changing inflation expectations (E31)empirical failure of the Phillips curve (E31)
unemployment (J64)inflation (E31)

Back to index