Personalization of Power and Social Welfare Spending under Dictatorships

• In Event: Poster Session: Comparative Democratization

Abstract

Conventional wisdom posits that democracies have more social welfare expenditure than dictatorships, because electoral competition induces democratic leaders to adopt social policies in exchange for ordinary citizens’ political support. Yet, what left unanswered in the literature is why some dictatorships have more welfare spending than others. As elections are absent, unfree, and/or unfair in autocracies, why would some dictators have to allocate their lucrative rents to social welfare that benefit ordinary citizens?

In this paper, we argue that the structure of power-sharing determines the degree of social welfare spending under dictatorships. A lower level of personalist rule induces more social welfare expenditure. We develop our argument on the basis of recent theoretical and empirical efforts of studies on comparative authoritarianism. As social policy can be regarded as a tool of cooptation, authoritarian regimes with higher personalization of power would have lower expenditure of social welfare for two reasons. First, authoritarian leaders exercising a higher level of personalized rule are less constrained by institutions. Instead of sharing resources with other political elites and ordinary citizens, personalized leaders have a weaker incentive to adopt social welfare policies. They can just invest in their coercive power to repress their opponents when sustaining their authoritarian rule. Second, dictators who need to share more power with other political elites face more challenges within their ruling coalition. As social welfare spending can help dictators buy support from ordinary citizens, dictators facing more intra-coalition competition induced by more power-sharing are more likely to increase social spending as a way to prevent other elites from initiating coups.

Based on the recent datasets on social welfare expenditure (Lucas and Richter 2016) and comparative authoritarian regimes (Geddes, Wright, and Frantz 2018), we test our argument with time-series cross-sectional data in about 80 dictatorships between 1961 and 2006. Our key finding is that a high level of personalism leads to a lower level of welfare spending but a higher level of military spending in dictatorships. In addition, we find that more social welfare spending makes coups less likely to succeed in dictatorships. These findings advance our understanding of the political economy of authoritarian regimes.

Authors

• Wen-Chin Wu, Academia Sinica

• Ding-Yi Lai, National Chung Cheng University

• Jen-Der Lue, National Chung-Chen University