is owned and operated by the Economic History Association
with the support of other sponsoring organizations.

Historical Political Business Cycles in the United States

Jac C. Heckelman, Wake Forest University

Macroeconomic Performance and Elections

Analyzing American presidential elections as far back as 1916, Ray Fair (1978) has shown that macroeconomic conditions consistently affect party vote shares. Specifically, the incumbent party is predicted to improve its vote share when economic growth is high and inflation is low. Using no information other than the growth rate, inflation rate, time trend, and the identity of the incumbent party, Fair was able to correctly predict the winning party for 15 of the 16 presidential elections from 1916-1978.

Given a strong connection between the economic environment and vote shares, incumbent politicians have an incentive to manipulate the economy as elections draw near. The notion that incumbents will alter the economic environment for their short-term political gain at the expense of long-term economic stability is referred to as generating a political business cycle. This theory of political business cycles has not generated much empirical support from myriad studies that concentrate on contemporary elections. Perhaps due to the lack of supporting evidence, and the belief that such manipulations were not possible before the advent of activist fiscal policy ushered in during the Keynesian revolution, there has been little attempt to test for political cycles in historical elections. There are, however, a few studies that do so, although their time samples and methodology differ widely.

National-Level Evidence on Historical Political Business Cycles

Adopting the standard procedure used in the empirical studies of contemporary political business cycles, Heckelman and Whaples (1996) test for cycles during the period after the Civil War and before the Great Depression. They find little evidence that either nominal or real GNP, or the GNP deflator, was significantly different than the expected level during the year of, or the year after, a presidential election from 1869-1929.

Davidson, Fratianni, and von Hagen (1990) employ a long time series from 1905-1984. They fail to find consistent evidence of a traditional political business cycle, or systematic differences by party control, of policy targets or policy measures during this time. However, they also test for alterations to the economy based on recent previous conditions and find that trends were significantly altered prior to elections only when macroeconomic outcomes in the recent past had been unfavorable to the incumbent: rising inflation, a rising rate of unemployment, a growing deficit, and a decline in monetary growth. In contrast, there were no changes in the dynamics when previous outcomes were favorable (p. 47), meaning, for example, that declining unemployment did not suddenly fall by an even larger degree just prior to the election. They find no electoral effects on the growth of real per capita GNP. They also present limited evidence that unemployment and inflation patterns differ by party control, but only following recent unfavorable outcomes in each, and the changes are further limited to the post-World War II period.

Klein (1996) takes a different approach. Instead of focusing on the actual values of the economic variables, Klein analyzes business cycle turning points, as identified by the National Bureau of Economic Research. He finds that 26 of the 34 presidential elections held from 1854-1990 were during an identified expansionary period. While expansions typically end in the period right after an election, he does not find that contractions are more likely to end in the period before an election. Thus, his evidence for political business cycles is somewhat mixed. Klein also finds that turning points differ by party control. Expansions are more likely to end following Republican victories, and contractions are more likely to end soon after Democratic victories. These partisan findings are much stronger after World War I.

It is perhaps not surprising that partisan influences on the economy are not stable during the long time series studies. In the earlier part of the Davidson-Fratianni-von Hagen, and Klein studies the Republicans, as the party of Lincoln and McKinley, had a large constituency base comprised of the industrial workers, and tended to support trade protectionism, the opposite of contemporary Republicans. It may still be true that significant differences in the structure of the business cycle occurred depending on which political party controlled policy, even in the period prior to the world wars, but since neither study examined these earlier time periods in isolation as they did for the later time period, that remains speculative.

Richard Nixon’s First Term

The strongest evidence for a political business cycle remains the first term of the Nixon administration. Some scholars have even argued this inspired Nordhaus’s (1975) early theoretical model of the political business cycle (Keech 1995, p.54) on which most empirical tests are based. Keller and May (1984) present a case study of the policy cycle driven by Nixon from 1969-1972, summarizing his use of contractionary monetary and fiscal policy in the first two years, followed by wage and price controls in mid-1971, and finally rapid fiscal expansion and high growth in late 1971 and 1972. They claim only the expansion portion of the cycle is evidence of electoral manipulation, and that the early contraction is merely consistent with modern Republican Party ideology. Although the latter is true, it does not disprove the conclusion of almost every other political business cycle scholar since it is not possible to pinpoint the motivation behind the policy change. Given, the abandonment of ideology displayed by Nixon in the second half of his term, it seems more likely the entire cycle, consistent with the predictions of a political policy cycle, was driven by electoral considerations rather than ideology. 1

State-Level Evidence

Little evidence has been accumulated for state-level political business cycles. An exception for historical gubernatorial elections is Heckelman (1998). Comparing the gainful employment rates across states with and without a gubernatorial election in the decennial years of 1870-1910, the evidence supports the notion of a political employment cycle for the states. This evidence is limited to the case of pooling all the years together, and may be driven by the strong result found for 1890. There is no further evidence of a federal employment cycle during the presidential election years of 1880 and 1900, or assistance directed at those states where the governor was of the same party as the sitting president.

Policy Cycles

Empirical studies of contemporary political cycles have turned more attention recently to policy, rather than business, cycles since policy instruments would need to be manipulated in order to affect the economy. Lack of evidence of political business cycles would be consistent either with no attempted manipulation, or policy cycles that did not have the desired effect due to other exogenous factors and the crudity of macroeconomic policy. There does appear to be strong evidence of modern policy cycles even when political business cycle evidence is weak or non-existent. (See for example Alesina, Roubini and Cohen 1999.) With the exception of the well-documented Nixonion policy cycles, there has been no attempt to document the occurrence of historical policy cycles. This remains the largest gap in the empirical literature and should prove a fertile ground for exploration.

New Deal Spending

There is, however, a related literature which examines New Deal spending from a political angle. Beginning with Gavin Wright’s (1974) study, scholars have generally concluded that allocations of spending across the states were directed more by Roosevelt’s electoral concerns than by economic need (Couch and Shughart 1998), since a disproportionate share of federal spending under the New Deal went to the potential swing states. Anderson and Tollison (1991) find that spending was also heavily influenced by congressional self-interest. In contrast, Wallis (1987) presents evidence that both political interest and economic need were important by noting that payments to Southern states were lower in part due to their reluctance to take advantage of federal matching grants. Most recently, Couch and Shughart (2000) test the matching grant hypothesis on one component of New Deal spending, namely the Works Progress Administration (WPA). They find that federal land ownership, political self-interest, and state economic need were all contributory factors to determining the allocation of WPA spending across the states. Wallis (1998) also showed that much of the prior empirical analysis of New Deal distributions depended critically on the inclusion or exclusion of Nevada, a state unique in its low population density and large proportion of federal land. The political aspects of New Deal spending are also summarized in the Fishback’s (1999) review. Fleck (2001) and Wallis (2001) provide the most recent exchange on this subject.


Alesina, Alberto, Nouriel Roubini, and Gerald D. Cohen. Political Cycles and the Macroeconomy, Cambridge, MA: MIT Press, 1997.

Anderson, Gary M. and Robert D. Tollison. “Congressional Influence and Patterns of New Deal Spending.” Journal of Law and Economics 34, (1991): 161-175.

Couch, Jim F. and William F. Shughart. The Political Economy of the New Deal, Cheltenham, UK: Edward Elgar, 1998.

Couch, Jim F. and William F. Shughart. “New Deal Spending and the States: The Politics of Public Works.” In Public Choice Interpretations of American Economic History, edited by Jac C. Heckelman, John C. Moorhouse, and Robert Whaples, 105-122. Norwell, MA: Kluwer Academic Publishers.

Davidson, Lawrence S., Michele Fratianni and Jurgen von Hagen. “Testing for Political Business Cycles.” Journal of Policy Modeling 12, (1992): 35-59.

Drazen, Allan. Political Economy in Macroeconomics. Princeton: Princeton University Press, 2000.

Fair, Ray. “The Effects of Economic Events on Votes for the President.” Review of Economics and Statistics 60, (1978): 159-173.

Fishback, Price V. “Review of Jim Couch and William F. Shughart II, The Political Economy of the New Deal.” Economic History Services, June 21, 1999. URL: htp://

Fleck, Robert K. “Population, Land, Economic Conditions, and the Allocation of New Deal Spending.” Explorations in Economic History 38, (2001): 296-304.

Heckelman, Jac C. “Employment and Gubernatorial Elections during the Gilded Age.” Economics and Politics 10, (1998): 297-309.

Heckelman, Jac and Robert Whaples. “Political Business Cycles before the Great Depression.” Economics Letters 51, (1996): 247-251.

Keech, William R. Economic Politics: The Costs of Democracy. New York: Cambridge University Press, 1995.

Keller, Robert R. and Ann M. May. “The Presidential Political Business Cycle of 1972.” Journal of Economic History 44, (1984): 265-71.

Klein, Michael W. “Timing Is All: Elections and the Duration of the United States Business Cycles.” Journal of Money, Credit and Banking 28, (1996) 84-101.

Nordhaus, William D. “The Political Business Cycle.” Review of Economic Studies 42, (1975) 169-190.

Wallis, John J. “Employment, Politics, and Economic Recovery during the Great Depression.” Review of Economics and Statistics 69, (1987): 516-520.

Wallis, John J. “The Political Economy of New Deal Spending Revisited, Again: With and without Nevada.” Explorations in Economic History 35, (1998): 140-170.

Wallis, John J. “The Political Economy of New Deal Spending, Yet Again: A Reply to Fleck.” Explorations in Economic History 38, (2001): 305-314.

Wright, Gavin. “The Political Economy of New Deal Spending.” Review of Economics and Statistics 56, (1974): 30-38.

1 See also Drazen (2000, pp. 231-232) for a brief discussion of Nixon’s manipulation of taxation policy and Social Security payments.

Citation: Heckelman, Jac. “Historical Political Business Cycles in the United States”. EH.Net Encyclopedia, edited by Robert Whaples. August 14, 2001. URL