Much of what drives the policy choices of Ben Bernanke and the Federal Reserve is a belief in the ability to trade higher inflation for lower unemployment, known within the economics profession as the “Phillips curve.” But does this trade-off actually exist?

While its true that many have found a negative correlation between inflation and unemployment prior to 1960, looking at U.S. data, this relationship appears to have broken down in the mid-1960s, just about the time policy-makers thought they could exploit it (Lucas critique anyone?).

It is hard, looking at the graph, which displays the annual change in consumer prices over the previous year and unemployment, to see much of a relationship. In fact, since 1960, the correlation between changes in CPI and unemployment has been positive. We have generally seen rising unemployment along with rising inflation. Of course, one might be concerned that the stagflation of the 1970s is driving this result. But looking at the data since 1980, there still remains a positive correlation between inflation and unemployment. While I am not arguing that inflation causes unemployment (after all, correlation is not causation), it should be clear from the data that there is not some exploitable trade-off that policymakers get to choose.

The Richmond Fed also has a great history of the Phillips curve that is well worth the read. Perhaps Fed President Jeff Lacker should bring copies to the next FOMC meeting.