Skip to content
Dominique Méda: "The link between changes in employment and the meaning of voting is too often ignored"
In a fiberglass reel manufacturing plant in Jiujiang, China, on July 15th. CHINATOPIX VIA AP

TheDo job transformations explain, at least partially, the results of the elections? Several recent American studies provide strong arguments in support of this thesis.

In the study "Importing Political Polarization? The Electoral Consequences of Rising Trade Exposure, David Autor, David Dorn, Hanson Hanson and Kaveh Majlesi analyze the effect of the penetration of Chinese imports into American states. Taking into consideration the results of the legislative elections of 2002 and 2010 and the presidential elections of 2000, 2008 and 2016 in 2 976 counties, they afford themselves the means to compare two to "Before and after China's integration into world trade â €" and thus highlight its specific effect on the counties where industries and jobs have been affected. They find that in the counties that were most exposed to Chinese competition, Republicans won votes and that the most moderate candidates were excluded at the polls. profit from the extremes. The authors even calculate that, had the penetration of Chinese imports been half as high, Michigan, Wisconsin, Pennsylvania and North Carolina would have eluded the candidate Democrat in place of Donald Trump.

Article reserved for our subscribers Read also The slow decline of the middle classes in the industrialized economies

The study builds on previous work that has highlighted that, contrary to mainstream economic theory, the idea of a creative destruction facilitated by the mobility of workers does not hold: the adjustments were on the contrary very slow, and the increasing penetration of imports from low-wage countries has disproportionately weighed on local labor markets historically specializing in labor intensive productions. In counties that are particularly affected by Chinese imports, many workers, far from finding a new job in better paid sectors, have simply lost their jobs, obtained lower quality jobs, or have experienced wage decreases.

These robots that made Trump read

Another study arrives at identical results by analyzing the effect of robotization: in Political Machinery: Did Swing Robots the 2016 US Presidential Election? AT"Carl Benedikt Frey, Thor Berger and Chinchih Chen analyze the penetration of robots in different constituencies to determine whether the sectors most exposed to automation in the years prior to the 2016 elections were more or less favorable to the vote in favor of Trump. They point to a positive link, which is explained by the fact that robotization has been accompanied by massive job losses or sharp falls in wages and / or quality. employment and calculate, as the authors of the previous study and using the same method, as Michigan, Wisconsin and Pennsylvania â € "states where the struggle was the oldest between the two candidates â € "would have voted in favor of Hillary Clinton if the number of robots had not increased.