A Simulation of Counter-cyclical Intervention: Lessons for Theory and Practice

Abstract

This paper provides an Excel-based simulation of counter-cyclical interventions to highlight important issues surrounding the theory and practice of counter-cyclical government interventions. The simulation is very simple with deviations from full-employment output following an AR(1) process. This relatively atheoretical approach has several advantages. First, the simple model produces output patterns that are realistic without burdening students with complex IS-Phillips curve discussions; even principles students can access the discussion. Second, the assumed process is simple enough that it is possible to explicitly work out an optimal policy. Finally, by staying at an abstract level, the simulation is equally applicable to discussions of fiscal and monetary policy. After presenting the simulated economy, the paper discusses how student interaction with the simulation can motivate a wide array of issues taken up in the theoretical and empirical literature. First, through policy lags built into the simulation, students are lead to understand why economists have debated so energetically about the speed at which economies return to full-employment equilibria. Having introduced the idea of policy lags, the paper then points out the important role that many economists feel is played by automatic stabilizers. The paper extends this discussion even further by connecting the concerns resulting from policy lags with the structure of democracies because the form of government institutions (parliamentary vs. republican, unicameral vs. bicameral) affects the speed of response. Having acquainted themselves with the simulation, students are now prepared to consider what an optimal policy would look like. Most will initially make the size of their intervention roughly proportional to the size of the current deviation from full employment; if GDP is low hit the gas, if it is high hit the brakes. Students who aggressively pursue this policy inevitably produce cycles of increasing intensity. They have just discovered the political business cycle. The paper shows that the truly optimal policy responds to today's shock - not today's level. This is the inherent danger of turning economic policy over to elected officials and leads nicely to a discussion of central bank independence. Even once the optimal policy is identified, students will see that much economic variation is simply unavoidable. Moreover, even an optimal policy will sometimes exacerbate inflationary or recessionary episodes. After all, is only optimal on average; over short time frames bad luck may lead to bad outcomes. Students are encouraged to remember that we must resist the temptation to conclude that an intervention strategy has been a success or failure based on short time series. Ultimately, the antidote to such bad experiences may be found in better forecasting. The paper concludes by showing how we can calculate the degree to which the Fed's forecasting work may improve economic performance. Throughout, the paper includes references to accessible empirical papers which show how economists are currently researching these very issues. These can be used to supplement the discussion of the simulation or as resources for the energetic, advanced student.