Unemployment has become a chronic problem of the Turkish economy and has remained at high rates without showing any significant improvement despite many structural changes implemented. This situation has been a source of motivation for the empirical analysis of unemployment in Turkey through three different approaches underlined by the theoretical literature on the course of unemployment. The first of these approaches is known as the natural rate hypothesis, the structuralist approach that supports the natural rate hypothesis when structural breaks are taken into account, and the third one is the hysteria hypothesis. Based on these approaches, in this study, it was investigated which of the natural rate, structuralist approach and hysteria hypotheses of unemployment in Turkey for the period of 1923-2020 acted in accordance with. For this purpose, in addition to traditional unit root tests, structural break unit root tests and Fourier unit root tests were used. Thus, the possibilities that each type of unit root approach might miss and give misleading results have been taken into account. The empirical results show that unemployment in Turkey follows a course that supports the hysteria hypothesis. The empirical results show that unemployment in Turkey follows a course that supports the hysteria hypothesis. Empirical results indicate that political shocks will cause permanent effects on unemployment in Turkey, there will be no natural rate for unemployment, and shocks will create changes through permanent effects on the natural rate.