That’s the main topic of a labor market paper by Sahil Chinoy of Harvard University. Here is the abstract:
We study political filtering in the labor market and examine its sources. Combining online voter file and degree data to create a panel of 34.5 million people, we show that Democrats and Republicans choose different career paths and employers. This leads to significant segregation at work: a Democrat or Republican co-worker is 10% more likely to share their party than expected. We then ask whether discrimination is caused by jobs shaping workers’ politics or workers’ politics shaping their career choices. For the first study, we used an experimental design similar to that of work shift time. We find that unaffiliated workers take their workplace politics into account, but not workers who are already registered as Democrats or Republicans. The average effect is too small to produce the classification we write. For the second study, we measure the intensity of employees’ preferences for politically relevant jobs using two exploratory experiments motivated by observational data. Here, we find that the median Democrat or Republican will trade 3% in annual wages to get the corresponding version of the same job idea. These preferences are strong enough to produce classifications similar to the observed levels.
It is co-authored with Martin Koenen, also a candidate in the labor market from Harvard. Some of Koenen’s papers, at the link, look very interesting as well.
Source link