Miguel Niño-Zarazúa on information externalities, programmatic spending, and voting preferences
WIDER Seminar Series
UNU-WIDER Research Fellow Miguel Niño-Zarazúa will present his work on information externalities, programmatic spending, and voting preferences at the seminar on 15 November.
Abstract – Information externalities, programmatic spending, and voting preferences: The case of Mexico’s Progresa-Oportunidades-Prospera programme
Conditional cash transfer programs (CCTs) represent one of the most important antipoverty policy innovations over the past two decades. A growing literature indicates that overall CCTs have produced positive impacts on various dimensions of wellbeing. Yet, the question of whether CCTs generate an electoral advantage to the incumbent remains an open question. A scant literature has provided conﬂicting results, with some papers reporting evidence of an ‘electoral bonus’, while other disputing those ﬁndings. This paper focuses on the Progresa-Oportunidades-Prospera (POP) programme, a pioneer CCT that was introduced in 1997, during the most important democratic transition that Mexico experienced over the past century. Here, we ask whether the incumbent has beneﬁted electorally from POP during the Presidential elections of 2000, 2006 and 2012; and if so, how, and to what extent.
To address these questions, we exploit the variation in the program expansion to compute Diﬀerence-in-Diﬀerences (DD) estimators in voter turnout and vote shares for the incumbent vis-à-vis the opposition. As robustness check, we also take advantage of a set of exogenous rules for program eligibility that are based on a ‘social gap’ index at locality level and an absolute poverty line at household level, to compute Regression Discontinuity (RD) estimators. Overall, we ﬁnd insigniﬁcant eﬀects of POP to the incumbent in the 2000 and 2012 elections but a negative net eﬀect in the 2006 election. We provide a theoretical rationalization for these puzzling results and argue that prospective expectations that POP generated among the poor in un-treated localities, together with information externalities from political campaigns, are the most plausible mechanisms that explain these seemingly inconsistent results.