Ferreira, José LuisHidalgo-Pérez, ManuelRubio-Castaño, CarmenUniversidad Carlos III de Madrid. Departamento de Economía2018-09-242018-09-242018-092340-5031https://hdl.handle.net/10016/27445Economic performance has been historically better under Democrat presidents compared to Republican ones. This gap has not yet been fully explained appealing to better management or luck. In fact, the economic cycles under one group of administrations or the other are quite similar. Blinder and Watson (2016) provide the best attempt so far at solving the paradox, but can explain only half of the gap. Drawing from them, and using a different method to account for the initial conditions of each presidential term, we are able to show that the phase of the economic cycle at the different elections are correlated to the party of the president. We also find ample evidence suggesting that there is a subtle causality: when the unemployment is high, the probability of a person voting for a Democrat president increases, thus causing Democrats being elected more often at the end of a recession and the beginning of a recovery. This, and not the difference in competence dealing with the economic cycles, is enough to close the gap.text/xmlapplication/pdfengAtribución-NoComercial-SinDerivadas 3.0 Españaworking paperopen accessDT/0000001626