Using Electoral Signaling to Measure Health Good Clientelism in Brazil

• In Event: Poster Session: Health Policy and Politics & Information Technology and Politics
  In Poster Session: Poster Session: Emerging Research in Health Policy and Politics

Abstract

In low capacity states with unmet demands or long wait times for public health services, opportunities for politicians to provide health services in a targeted fashion for electoral gain abound. Electoral budget cycles for health personnel have been observed, for example, in less economically-developed regions of Italy (Stolfi and Hallenburg, 2015) and qualitative evidence suggests services like tubal ligations may be used in a targeted fashion to garner political support in the poorest region in Brazil (Caetano and Potter, 2004). While health clientelism is recognized as a problem in popular discourse in Brazil, it can be hard to systematically demonstrate. Nichter argues that Brazilian health clientelism does not lead to single election-year returns to those who express political support, but rather to continuous post-election access to targeted health services for those who make pre-electoral declarations of support (2018). If this is true, electoral health budget cycles would be hard to observe in Brazil. In this paper, I first verify this claim by examining whether election-year cycles are identifiable in health personnel expenditures or in counts of various high-complexity services in Brazilian municipalities. I then elaborate an alternative measurement strategy that may systematically capture a given municipality’s propensity toward health clientelism. Ballots in Brazil have a built-in observable signaling mechanism for candidates who want to identify themselves as having a comparative advantage in the clientelistic distribution of health goods. Candidates may identify themselves on the ballot by nicknames and many choose medical pseudonyms like “Dr Tonho,” “João da Ambulância” (John the Ambulance Driver), or “Enfermeira Rita” (Nurse Rita). I treat this mechanism as a proxy indicator of the prevalence of health clientelism in a municipality, observing the frequency with which city council candidates use medical pseudonyms in municipal elections. I use this metric to evaluate which municipal and political indicators best predict high levels of medical signaling. I examine factors such as municipal capacity (measured as completeness of death certificate documentation), the adoption of civil service tenure for healthcare workers (compared to contracted labor which is easier to politicize), and the percent of the municipal population that lives below the poverty line. I also examine whether candidates from certain political parties are more inclined toward medical signaling than others. The contribution of this work is that it proposes a novel method for measuring health clientelism in Brazil, which may allow for a more systematic evaluation of policies and parties that best reduce clientelistic targeting of decentralized public health services.

Author

• Marika Danielle Csapo, Loyola Marymount University