Fiscal Policy in American Presidential Elections: A Simulation

In this paper a simulation technique borrowed from civil engineering is applied to American presidential elections to explore the key relationship between federal spending and incumbent reelection, represented by the fiscal model. On the one hand, as Machiavelli would have understood, an expansionary fiscal policy militates against incumbent reelection but a cutback policy facilitates it. That is the `demand' side of the model. There is also a `supply' side: the longer the incumbents have occupied the White House, the more likely they are to implement fiscal expansion. We simulated 1,000 elections under conditions that replicated the values of the predictor variables of the fiscal model over the 1880—2004 and 1932—2004 periods of American history. The latter period deserves attention in its own right, because starting with the 1932 election, the federal share of gross domestic product broke out of the 2—3% range for the first time since World War One. This marked a qualitative change in the role of government in the United States of America. The simulated series allow patterns that are weakly detected in the historical data to emerge more clearly for observation and analysis. The results of the simulations not only confirm the empirical findings from the historical data, but suggest that the American political system is stable, maintaining alternation between political parties in the White House, a characteristic of democracies, and keeping fiscal policy within bounds of what the majority of the voters will support.