Rational expectations and stabilisation policies


It could be argued that if agents1) hold rational expectations then traditional economic stabilisation policies will be ineffective. If correct, that stabilisation policies become ineffective in the face of rational expectations, there would be no reason for their implementation by governments. However, governments certainly attempt to regulate the economy though the use of macro-policy tools. In light of this it might seem easy to conclude that such an assertion is false, or why would governments persist in the use of monetary and fiscal policies. The issue therefore raises further questions; whether agents have rational expectations, and how realistic is it to claim policy ineffectiveness if agents do hold rational expectations.

Modelling future expectations is important, and attitudes to what influences agents during the decision making process have changed over the years. The initial statement suggesting ineffectiveness can certainly be justified in theory. The continued use of regulatory policies brings together the arguments.

All agents make choices that will reflect their attempts to maximise the utility they get from their actions. When making these choices they have to weigh up influencing factors in order to make a choice that will optimise the satisfaction received. Economists have often tried to incorporate the decision making process undertaken by agents into economic models. This would enable them to understand and model how agents react to different stimuli. The subsequent models could then be used by policy makers in decision making. Through an understanding of agent's reactions the job of policy makers should theoretically be made easier; in understanding how agents will react they can more easily select appropriate policies to achieve desired effects.

Prior to 1970, the theory of adaptive expectations (AET) was used to model agent's future expectations. This theory claimed that future expectations are based on current and past data. If future expectations of inflation are used as an example, then under adaptive expectations the forecasts of future inflation will be a function of past and present inflation rates, with a portion of the forecast error included to increase accuracy. The change in the error term indicates the adaptive nature of this theory. The problem with this approach is that forecasts will often and consistently lag behind real levels.

The limitations of adaptive expectations theory led to the development of rational expectations theory (RET)2). RET claimed that agents will not continue to base future expectations on persistently wrong forecasts. The result was that a new approach was required. The difference centred on RET also taking future expectations into account in the forecasting process. This change means agents have more information available to them when making future predictions, theoretically making them more accurate. The transition between these two theories is evident in the development of the Philips curve, and later the expectations augmented Phillips curve. This history also demonstrates the importance or rational expectations to policy makers.

The traditional Phillips curve suggested that there is a trade-off between inflation and unemployment within the economy:

Short-Run Phillips Curve before and after Expansionary Policy, with Long-Run Phillips Curve (NAIRU)

This implied that policy makers could reduce unemployment through the use of expansionary fiscal and monetary policies. Such policies would take the form of increased government expenditure, reduction in taxation, or exchange rate manipulation. This would shift the aggregate demand curve to the right, which would in turn cause a rise in prices. Under this scenario the result is increased output and inflation as shown by the traditional Philips curve.

The RET argument is that under the assumption of rational expectations agents and firms will also take relevant economic theory into account when reacting to government policy decisions. In the same scenario, RET argues that if a government employs expansionary policies, firms will not necessarily employ more labour, as they will understand the inflationary consequences of policy. Agents revise their expectations of future wage rates and prices as expansionary monetary or fiscal policies are implemented. This revision takes place such that agents do not experience a fall in their real wage levels.

The outcome is that under rational expectations there can be no trade off between unemployment and inflation. The result is vertical long run Philips and aggregate supply curves with no temporary reduction in unemployment through the use of stabilisation policies. This could suggest that stabilisation policies cannot be effective under rational expectations theory, as policy actions will be anticipated in advance of their implementation.

It's evident however that Governments continue in attempting to control the economy through the use of fiscal and monetary policy. This might infer some constructive policy application, or a means of avoiding the rational expectations effectiveness trap. One might deduce that if policy makers wish to ensure the effectiveness of policy, they must be unexpected. Shock measures would leave agents unable to incorporate policy change into their future forecasts, as no agent can use information they don’t posses. But this leaves complications in making and implementing surprise policy. Surprise policy announcements may succeed on occasion but attempting to repeat the procedure will become increasingly difficult. Intelligent agents holding rational expectations will not be consistently fooled. Thus an inability to smooth out business cycles with diminishing returns nevertheless.

While it’s not logical to discard all stabilisation policies by implication, it's apparent that there are perceived benefits to be gained from their use. Despite RET assumptions of a perfect markets framework, markets are imperfect so knowledge is not total and actions are not instantaneous. Market imperfections have implications for policy makers on two levels. The first relates to the potential success of policies as regulatory tools, and the second to the political popularity policy makers can provide for governments.

Assuming imperfect market time lags, or between announcement and final resulting inflationary observation, expansionary policies can effect incomes in the short term. This may not imply the rational expectations argument is incorrect. Agent's expectations cannot always be correct, resulting in scope for governmental operation to achieve objectives. However restrictive policies are not popular with the electorate. Conversely short term expansionary policies and resulting temporary wage increases may raise government popularity. Time lags may allow governments to implement a combination of monetary and fiscal policies according to objectives without compromising their political position.

Thus the idea that agents holding rational expectations must lead to ineffective traditional stabilisation policies may be theoretically correct, but in real world application it becomes more complex. Market prices for example may reflect policy decisions many months in advance of their implementation. Likewise stock prices may reflect expected valuation growth but zero current profits. Objectives may be achieved in certain circumstances, and policy necessarily includes a political dimension which may trump any broad economic claim. Rational expectations is a tool or framework within which to implement and appraise choices.


An agent as an actor or decision maker in the economy, or a model of the economy
For example, John Muth (1961) “Rational Expectations and the Theory of Price Movements”

QR Code
QR Code rational_expectations_and_stabilisation_policies (generated for current page)