In the mid- to late 1990s, the United States experienced low unemployment and
dramatically low inflation – an experience that is difficult to reconcile with the standard
Phillips curve, which predicts that unemployment and inflation are inversely related.
Writing in the April 2006 Economic Journal, Professor Bruce McGough shows that an
increase in productivity, while causing unemployment to decrease, also causes policymakers
to lower inflation quickly – an outcome that is consistent with the US data.
The mechanism behind the simultaneous fall in inflation and unemployment lies in policymakers'
perception of a Phillips curve. Understanding this mechanism requires some
background. In 1958, LSE economist Bill Phillips discovered an inverse relationship
between inflation and unemployment in the British data, a discovery subsequently
reinforced by Paul Samuelson and Robert Solow using US data.
The presence of this inverse relationship or Phillips curve altered the perception of the
trade-off faced by policy-makers when setting monetary policy: seemingly, the government
could choose to lower the unemployment rate simply by printing money and thereby raising
inflation.
This simplistic view was foreseen by the authors of the studies, who warned that any
attempt to exploit the trade-off may cause it to break down. The intuition behind their
argument is quite simple: printing money affects unemployment only to the extent that
people are unable to or don't know to alter their patterns of behaviour. If the government
tries to exploit the trade-off evidenced by the Phillips curve, people will eventually alter their
behavioural patterns and the trade-off will disappear.
The extent to which policy-makers listened to the authors' warning remains an open
question. In his book, The Conquest of American Inflation, Thomas Sargent takes seriously
the possibility that policy-makers did and still do believe that they face a trade-off when
setting monetary policy.
He finds that policy-makers set inflation high in an effort to exploit the trade-off; but because
people alter their patterns of behaviour, this policy has no impact on the unemployment
rate. Importantly, the trade-off doesn't even exist; however, the behaviour of inflation and
unemployment convinces the government that it does exist, and the subsequent policy
pursued by the government reinforces this belief. The resulting inflation bias is a wellknown
phenomenon, demonstrating that even a well-meaning government may employ
policies that have undesirable results.
Given the inflation and unemployment data since 1960, it is unreasonable to assume that
policy-makers still believe there is an unchanging inverse relationship between inflation and
unemployment. So Sargent modifies his analysis by assuming policy-makers recognise that
the trade-off may vary over time. This a priori simple modification has a dramatic impact on
the implications of the analysis.
Sargent finds that occasionally policy-makers will perceive that the trade-off they face has
worsened in the sense that raising inflation will no longer so significantly lower
unemployment. The worsened trade-off encourages policy-makers to reduce inflation,
which – and this is the important part – will reinforce their perceptions that the trade-off has
worsened. This reinforcement causes the inflation rate to drop quickly, an occurrence
deemed by Sargent an escape from the prevailing inflation bias.
McGough’s work asks whether events like productivity changes can cause policy-makers to
doubt the trade-off between unemployment and inflation and thus trigger escape-like
behaviour. He finds that precisely this occurs.
An increase in productivity lowers unemployment while having no simultaneous impact on
inflation. That unemployment falls while inflation stays the same causes the government to
doubt the magnitude of the trade-off between unemployment and inflation. This doubt
suggests a policy that lowers inflation, which, according to the perceived trade-off, should
put upward pressure on unemployment.
But because the trade-off doesn't really exist, there is no average upward movement in
unemployment. This reinforces the government's perception that the trade-off has
worsened and so they lower inflation further. Thus, through this reinforcement process, the
increase in productivity shocks the economy, causing it to escape its inflation bias.
ENDS
Notes for editors: ‘Shocking Escapes’ by Bruce McGough is published in the April 2006
issue of the Economic Journal.
McGough is Assistant Professor of Economics at Oregon State University.
For further information: contact Bruce McGough +1-541-737-1429 (email:
bruce.mcgough@orst.edu) or Romesh Vaitilingam on 0117-983-9770 or 07768-661095
(email: romesh@compuserve.com).