We use evidence from detailed records of FOMC deliberations to argue that the theory of the time inconsistency problem provides a reasonable explanation of the Federal Reserve's excessively expansionary policy stance during the 1970-1978 period when Arthur Burns chaired the Board of Governors. The records suggest that the Fed perceived a Phillips curve tradeoff and political pressures that made it difficult to adopt disinflationary policies at any moment in time; the tendency toward excessively expansionary policy was exacerbated by the short-run planning horizon the Committee faced in each of a sequence of meetings. We further argue that comparative static predictions of the time inconsistency model are consistent with both the rise of inflation during the Burns years and its subsequent fall.
2
Did Time Inconsistency Contribute to the Great Inflation? Evidence from the FOMC TranscriptsThe time inconsistency problem is often invoked to explain an inflationary bias alleged to plague central banks. 1 The assumptions behind the reasoning are as follows.First, the central bank is assumed to choose policy actions on a period-by-period basis in each of its meetings; it is not constrained by rules, but instead uses discretion in each period. Second, public expectations of inflation for the upcoming period are viewed as given at the time of the meeting. Third, the economy can be characterized by an expectational Phillips curve; i.e., if inflation is higher than the public expects, unemployment will temporarily fall below its natural rate. Fourth, the central bank values a marginal reduction of unemployment below its natural rate, but is also averse to higher inflation. Finally, it is assumed that expectations are rational.With predetermined inflation expectations, the central bank sees an opportunity to lower unemployment via surprise money growth in each of its meetings. If expected inflation were zero, for example, the reduction in unemployment created by a money growth surprise would be "worth" the modest increase in inflation. However, a problem arises when public expectations are rational. The public, understanding the Fed's objectives, will correctly anticipate monetary stimulus, rendering the effort to reduce unemployment ineffective. Instead, the result is inflation. At the equilibrium level of inflation, the marginal gains from unemployment reduction are balanced by the added costs of additional inflation in the current period. This is suboptimal in comparison to a 1 Seminal contributions include Kydland and Prescott (1977) and Barro and Gordon (1983 Persson and Tabellini (2000), however, argue that these criticisms miss the point, since time inconsistency analysis does not predict that the Fed would want to generate policy surprises in equilibrium. Rather, in an inflationary equilibrium, the Fed's lack of credibility would cause a more restrictive policy to produce a recession. As a consequence, the Fed would refrain from pursuing a disinflationary policy. Some empirical evidence also supports the ...