Is the observed correlation between current and lagged inflation a function of backward-looking inflation expectations, or do the lags in inflation regressions merely proxy for rational forward-looking expectations, as in the newKeynesian Phillips curve? Recent research has attempted to answer this question by using instrumental variables techniques to estimate "hybrid" specifications for inflation that allow for effects of lagged and future inflation. We show that these tests of forward-looking behavior have very low power against alternative, but non-nested, backward-looking specifications, and demonstrate that results previously interpreted as evidence for the new-Keynesian model are also consistent with a backward-looking Phillips curve. We develop alternative, more powerful tests, which find a very limited role for forward-looking expectations. JEL classification code: E31

In recent years, a broad academic consensus has arisen that favors using rational expectations sticky‐price models to capture inflation dynamics. We review the principal conclusions of this literature concerning: (1) the ability of these models to fit the data; (2) the importance of rational forward‐looking expectations in price setting; and (3) the appropriate measure of inflationary pressures. We argue that existing models fail to provide a useful empirical description of the inflation process.

Publication informationJournal of Money, Credit and Banking, 35 (1): 1-22 Publisher Blackwell on behalf of the Ohio State University PressLink to online version

Is the observed correlation between current and lagged inflation a function of backward-looking inflation expectations, or do the lags in inflation regressions merely proxy for rational forward-looking expectations, as in the newKeynesian Phillips curve? Recent research has attempted to answer this question by using instrumental variables techniques to estimate "hybrid" specifications for inflation that allow for effects of lagged and future inflation. We show that these tests of forward-looking behavior have very low power against alternative, but non-nested, backward-looking specifications, and demonstrate that results previously interpreted as evidence for the new-Keynesian model are also consistent with a backward-looking Phillips curve. We develop alternative, more powerful tests, which find a very limited role for forward-looking expectations. JEL classification code: E31