Political scandals and corporate political participation. Evidence from Brazil – 2018

Abstract – Curtailing corporate participation in electoral campaigns, in principle, improves electoral accountability by reducing the number of principals to whom politicians have to respond. Not surprisingly, the practice is widespread. Reforms in that direction, however, are not straightforward to implement and present clear conundrums with respect to corruption displacement. In settings with high-returns to political connections, corporate actors are unwilling to relinquish control over politicians and resources wilfully, and might oppose change, directly and indirectly. I approach the issue of campaign regulation effectiveness by focusing on the latter, specifically addressing the emergence of straw donor schemes – in which corporate actors circumvent restriction by employing private donors as intermediaries.
Empirically, I exploit a corporate donation ban in Brazil and study how it affected corporations’ ability to signal support by means of private-for-corporate donations of their employees. Employing a Difference-in-Differences strategy and lever- aging pre-intervention heterogeneity in firms’ donation patterns, I show that firms with a history of political participation orchestrate a clearer signal to politicians with respect to comparable firms without such record. Two measure of dona- tion orchestration are developed as adaptation of the Herfindahl-Hirschman Index: a within- and an across-candidates contributions concentration index. The papers shows that drastic campaign reform does affect the ability of firms to main- tain speech in the political arena. The study concludes that the all-encompassing ban was counterproductive, especially because implemented at the cost of electoral transparency and accountability.


Reforming the nepotist State? How familial and political networks undermine bureaucratic reform in Brazil (with F. Daniel Hidalgo and Gabriel Cepaluni) – 2018

Abstract – Rampant nepotism within government bureaucracies is possibly a cause of poor government performance. Politi- cians can degrade governmental performances through the discretionary hiring of their staff from a sub-optimal pool of candidates. Despite governments around the world have taken efforts to pass anti-nepotism laws, the effectiveness of ad- ministrative reforms to curb the practice is not well established. Presently, we evaluate one such intervention in Brazil: a universal ban by the Supreme Court affecting all levels of government. We use large scale administrative data on the universe of local bureaucrats in the country. Using a Regression Discontinuity (RD) design on the margin of victory at the municipal level, we –first – measure the extent to which winning candidates employ discretionary hiring compared to runner-ups, and – second – we evaluate the effect of the reform to tackle the issue. In the analysis we develop a minimal and scalable measure of nepotist links based on surname patterns and address its few shortcoming through econometric adjustments. We find that politicians who win local offices do appoint bureaucrats from their familial pool at a rate that is 15% higher than their counterparts. The reform partially effective, at face value, bringing this number down to 11%. However, we also show that the very same reform is partially offset due to the strategic reaction of local politician. Elected politicians manage to evade the reform by making deals with officials in neighbouring jurisdictions, and performing a relative exchange classifiable as ‘cross-nepotist’. The dynamic – identified through a similar strategy – emerges only at the onset of the reform, and is sustained only between close municipalities ruled by the same party. Cross-nepotism is a novel example of corruption displacement that should be brought to the attention of policymakers and political scientist alike. In light of the evidence produce, the piece initiates a discussion on the limits of top-down, all-encompassing mandates for reforming local democratically-elected governments.

Ideological camp effects on the mis-allocation of public resources. Evidence from a RD study in Italy – 2018

Abstract – Contingent political determinants of budgetary outcomes are overlooked in the political science literature when it comes to local level government. This study presents previously unavailable evidence of a heterogeneous degree of mis-management of public resources across party families at the local level. Municipal-level financial health is assessed by gauging the use of accounting tricks to defer spending constraints (i.e. overestimation of unsound receivables). Beyond the interesting factoid, the research presents preliminary evidence of patronage dynamics, and points at the way in which political class selection and its career prospects might determine policy and budgetary choices. The identification strategy exploits a classic data-driven Regression Discontinuity design on margins of victory at the municipal level, comparing Italian party families pairwise for the decade 2005-2014. The research piece has the potential to carry out implications going beyond the specific empirical context it considers. The construction of two novel data sets underpins the study.