Empirical research of political business cycles (PBCs) may suffer from endogeneity bias when incumbent governments have discretion to call for an early election. Using an instrumental variable (IV) routine on data from Japan and the U.K., we find strong evidence to support the notion that election timing is a function of the economy rather than the macroeconomy being driven by elections as assumed in PBC. In single-equation regressions, no evidence of political cycles are found, but Hausman tests suggest elections are endogenous in our regressions. A monetary cycle in Japan and an inflation cycle in the U.K. are uncovered through IV estimation.