Political bias in resource allocation during the first wave of the pandemic in Chile
Description
To determine the relationship between the distribution of aid from the central to the local level, we use panel data analysis and fixed effects for the country's 345 communes and for the months of March to September. The commune fixed effect controls for those commune-specific factors that are fixed over time, and the month fixed effect controls for those factors that vary over time but are common to all communes. In each of the models the dependent variable is the logarithm of current transfers at the per capita level, while the independent variables are divided into three dimensions, firstly, there are the political variables such as political party of the president and municipal victory margin, the first is a binary variable that is 1 when the mayor is of the same party as the President of the Republic and 0 otherwise. The second political variable corresponds to the percentage difference in votes between the elected mayor and the second candidate to obtain the highest number of votes, i.e. the margin of victory, which allows us to understand the degree of electoral competition in the territory. A second dimension is the variables associated with the pandemic, such as the number of days the municipality was under quarantine, the percentage of the population in the municipality below the poverty line, and the density variable, which is the population divided by the surface area of the municipality. While poverty is associated with the local socio-economic context and density is a good proxy for the shape and structure of the territory (OECD, 2013 2014), allowing to distinguish between metropolitan (centre or periphery), urban or predominantly rural areas. Finally, control variables associated with municipal financing were incorporated in the models, such as contributions from the Municipal Common Fund (FCM), Capital Contributions and Autonomous Income of the municipalities, the first indicating the level of vulnerability of the municipalities, since it is a formula-based transfer that is related to local vulnerability, the greater the dependence on the FCM the greater the vulnerability of the municipality. The second variable corresponds to transfers for the development of projects, mainly in the area of infrastructure and equipment; in order to obtain this transfer, the project must be technically evaluated by the National Investment System. Thirdly, there is autonomous income, 70 percent of which corresponds to income from land tax, economic activities and automobile circulation permits, so this variable allows capturing the wealth of the municipalities.