Skip to main content
. 2021 Apr 20;78:102462. doi: 10.1016/j.jhealeco.2021.102462

Table 2.

Coefficients of OLS regressions of Covid-related mortality on being a candidate in 2020 in the whole sample of candidates in 2014.

Whole country
Hotspots
(1) (2) (3) (4) (5) (6)
Dep. Var.: Probability of death between February 26 and April 28, 2020
Candidate in 2020 0.00109*** 0.000790*** 0.000521** 0.00108** 0.000573 0.0000694
(0.000202) (0.000201) (0.000217) (0.000522) (0.000519) (0.000594)
Age 0.000181*** 0.000178*** 0.000319*** 0.000312***
(0.0000219) (0.0000217) (0.0000637) (0.0000630)
Rank w.r.t. last elected 0.000942*** 0.00164**
(0.000248) (0.000646)
Constant 0.00248*** 0.0101*** 0.00988*** 0.00344*** 0.0186*** 0.0182***
(0.000157) (0.00144) (0.00143) (0.000358) (0.00425) (0.00421)



Observations 227,369 227,369 227,369 41,742 41,742 41,742

Notes: This table presents OLS coefficients of regressions of the probability of dying between February 26, 2020 and April 28, 2020 on the fact of being a candidate in the 2020 municipal election, the age of the candidate, and its rank in the list relative to the last candidate elected. The sample is composed of all male candidates in the 2014 municipal election born before 1960 and alive on February 25, 2020. Robust standard errors clustered at either the county (département) level or the city level (in the hotspot-only samples) are reported in parentheses. * p<0.1, ** p<0.05, *** p<0.01.