Rational Expectations
Making a mystery of money

The rational expectations hypothesis has to a large extent transformed the debate about monetary policy, even though it has not gained universal acceptance. In Discussion Paper No. 104, Research Fellow Patrick Minford examines its implications for monetary policy.

Minford begins with a reassessment of the effects of monetary shocks on output. The conventional 'Phillips-curve' accounts of such relationships are based on variations in the expected real wage in response to a monetary shock. These explanations are undermined by rational expectations, because workers' price expectations respond immediately to available information and the expected real wage thus remains unchanged. These difficulties have led to a search for 'nominal rigidities' that might explain the observed relationship between money shocks and output, but Minford argues that none of these alternatives are entirely satisfactory.

Minford then analyses the relationship between fiscal and monetary policy in the light of rational expectations. The literature on the 'government budget constraint' draws attention to the instability which could arise if monetary and fiscal policy are 'inconsistent', the standard example of this being where a fiscal deficit is permanently bond-financed. Under rational expectations such instability becomes not a long-run but an immediate problem: it is impossible to define an equilibrium path for the model. Thus, deficit policy implies bounds on the range of feasible monetary policies. Furthermore, these future possible paths of monetary policy have effects on current output and inflation.

Finally Minford examines the role of monetary policy in stabilizing output fluctuations. Again, rational expectations has changed the nature of the debate. Previously, there was no question that, given slowly moving price expectations, monetary policy could and should stabilize output. This assumed that the central bank had up-to-date information at least as good as the private sector's, a good model of the economy with which to forecast the effects of policy, and that it was efficient in implementing required policy. Opponents of activist stabilization policies argued that these assumptions were implausible and concluded that activist policy would be at least as likely to increase as to dampen fluctuations. The debate has now widened, because under rational expectations people incorporate knowledge of the central bank's reactions into their expectations. Under certain conditions this can neutralize the effects of monetary policy on output, and in general it complicates the economy's responses to stabilization policy.

Minford concludes that stabilization policy will in general be effective in rational expectations models but that such policies may be undesirable. The Lucas critique adds force to Milton Friedman's original complaint that our models are not good enough to guide stabilization policies. Policies designed to maximize objective functions based on output and unemployment are unsatisfactory, Minford argues. Policy should be designed to remove distortions such as those which give rise to unemployment, rather than to 'stabilize the economy'. But the most serious problem with 'activist' policies is time inconsistency. 'Reflationary' policies involve reneging on previous counter- inflationary commitments, and models based on government's concern for its reputation do not yet suggest that this can be prevented without constitutional limitations. Minford concludes that rational expectations have made monetary theory 'more of a mystery than ever before'.

Patrick Minford discussed these issues at greater length in his article in CEPR Bulletin No. 14.


Rational Expectations and Monetary Policy
Patrick Minford

Discussion Paper No. 104, April 1986 (IM)