Public Works Investment and Electoral Data in Chile (1989-2018)

Published: 8 April 2021| Version 1 | DOI: 10.17632/y72v996tww.1


This dataset (dt_elecstds.dta) contains panel data by municipality on infrastructure investment by the Ministry of Public Works, electoral information and socioeconomic variables in Chile for the period 1989-2018. Electoral data includes: three dummies for municipalities' political alignment (a dummy coded one if the mayor belongs to one of the political parties of the central government coalition (m); a dummy coded one if the government coalition parties won presidential elections in the municipality (p); and a dummy coded one if the mayor belongs to one of the coalition parties and the coalition won presidential elections in the municipality (mp)); and two dummies coded one for local ballot years (ym0), and for national ballot years (yp0). The sample comprises seven local (1992, 1996, 2000, 2004, 2008, 2012 and 2016) and seven presidential elections (1989, 1993, 1999, 2005, 2009, 2013 and 2017). Socioeconomic data includes: population (log); regional GDP per capita; regional GDP growth rate in t and t-1; the percentage of new-borns measuring less than 50 cm; the percentage of new-borns weighting less than 3000 gr; and the percentage of mothers with more than three children. In the absence of poverty indicators for the whole sample series, the latter accounted for proxies since they are correlated with municipalities’ socioeconomic conditions (Mardones & Acuña, 2020). Population data is from the National Statistics Institute, GDP data from the Central Bank and poverty proxies from Mardones & Acuña (2020). This dataset has been used to test whether infraestructure investment has been distributed on electoral criteria or not. In particular, it allows to evaluate if the municipalities lined up with the central government coalition parties (m, p and mp dummies) have been systematically benefited, and if allocations increased during local and presidential ballot years (ym0 and yp0 dummies). Since the data goes from the first democratic elections after Pinochet's dictatorship (1989) to 2018, it provides the opportunity to analyse if electoral criteria have changed as democracy entrenched. In particular, it can be tested whether distributions favoured aligned municipalities and whether a political budget cycle persisted during the early years and later. We also include a code file ( with several estimations that show how these two electoral distortions evolved during the early years (1989-2005) and once democracy took root (2006-2018). These estimations are part of the paper entitled "Electoral incentives and distributive politics in young democracies: evidence from Chile".



Political Economy, Local Politics, Economic Analysis of Election