Political Business Cycles in Türkiye: A Chaotic Perspective

This paper has the goal of analyzing the chaotic structure of political business cycles in Türkiye for the period from 1986 March to 2020 January. To investigate the existence of uncertainty, the chaotic dynamics of the analyzed variables was examined using the Lyapunov exponent, Shannon entropy, and Hurst entropy. The Fourier VAR model was applied to determine the effects of unemployment, inflation, economic growth, and the minimum wage on voting behavior in Türkiye. The Granger Causality approach was also used to assess the direction of the causal relationship between macroeconomic factors and political party groups. Granger causality analysis findings of this paper reveal a bi-directional Granger causality link between the center-right party and economic growth, inflation, unemployment, and the minimum wage. Furthermore, there is also a uni-directional causality relation from inflation, unemployment, and economic growth to the center-left party and a bi-directional causality relation between the minimum wage and the center-left party. In this regard, one of the most remarkable evaluations of this paper is that the sensitivity findings of party groups to macroeconomic factors in Türkiye do not align with the core assumptions of the Partisan Theory concerning the right-left party division. Therefore, contrary to the related theory, the analysis results indicate that it is not eligible to make a sharp and distinct distinction between right-wing and left-wing parties in Türkiye. Additionally, Lyapunov exponent, Shannon entropy, and Hurst entropy findings point to the existence of chaotic structure and uncertainty in political cyclical fluctuations in Türkiye. Accordingly, it is figured out that the existence of chaotic structure and uncertainty in party groups and analyzed variables allows only short-term and not long-term forecasting.