Political Business Cycles, Institutional Structure and Budget Deficits in Turkey?

This study tests the existence of political business cycles and effects of various instituitonal factors on budget deficits in Turkey. For this purpose, the annual (for 1960-1996), quarterly (for 1983:QI-1997:Q2) and the monthly data (for 1990:01-1997:6) are used. Results with annual data show that as the number of parties in a coalition government and the number of fiscal authorities increase, the budget deficits also increase while elections do not significantly affect the budget deficits. However, the coalitions, the military coups, petroleum shocks, the Cyprus war and terrorism have significant and negative effects on budget deficits. Quarterly data shows that elections, number of parties in a coalition government and the number of fiscal authorities have some effects on sub items of the budget though they do not affect the budget deficit significantly. On the other hand, monthly data show that especially elections have significant and negative effects on the budget deficit and on all of the sub items of the budget except the investments. However, monthly data do not indicate any significance for the effects of number of fiscal authorities and number of parties in the coalition governments. The analysis with monthly data of budget subtotals clearly indicate the existence of political business cycles for the period 1990-1996. Therefore, while the analyses with the annual data disguise the effects of elections, the analyses with quarterly and monthly data reveal the existence of political business cycles. Moreover, quarterly data indicate that the political power dispersion increases the transfer expenditures. We stress some policy implications of these analyses. The emphasis is on the unification of fiscal authorities, and reducing off-budget expenditures.