Public Choice 73: 21-35, 1992.
© 1992 Kluwer Academic Publishers. Printed in the Netherlands.
Testing the satisficing version of the political business cycle
1905-1984"
LAWRENCE S. DAVIDSON
MICHELE FRATIANNI
JISRGEN VON HAGEN
Department of Business Economics and Public Policy, Graduate School of Business, Indiana
University, Bloomington, IN 47405
Received June 1990; accepted August 1990
Abstract. This paper develops a test of the satisficing version of the political business cycle. Previ-
ous tests have focused on maximizing models of political behavior and are not sufficiently general
to test for satisficing behavior. Using annual U.S. data for the period 1905 to 1984, we find evi-
dence supporting the satisficing version of the political business cycle model, but we reject the max-
imizing version. In accordance with the satisficing hypothesis, we find that increasing inflation or
unemployment and decreasing monetary base growth in the third year of a presidential term are
followed typically by reversals during the election year.
1. Introduction
Does knowing the date of the next Presidential election improve the forecast
of economic variables? According to Nordhaus (1975) and MacRae (1977) the
answer is 'yes' because incumbent politicians have a singular interest in gaining
reelection. These politicians have no misgiving in manipulating the economy
so as to create better economic outcomes in the sense of currently observed
phenomena before elections. Assuming a myopic public and an economic
structure described by a stable Phillips Curve, the incumbent is able to max-
imize reelection chances by choosing the vote-maximizing combination of
inflation and unemployment. 1 This is the essence of the political business cycle
(PBC). PBCs imply that knowledge about the dating of the next election can
improve forecasts of economic variables such as inflation, unemployment and
output. This political behavior also implies cycles in policy variables such as
the monetary base. Recently, Rogoff and Sibert (1988) and Laechler (1984),
among others, have shown that political business cycles can arise even if voters
are not myopic and form expectations rationally.
* We wish to thank Gordon Tullock and two anonymous referees for valuable comments.