Incentives to Produce Race-related Research

An established literature has studied potential biases in the economics publication process based on traits of authors. We complement such work by studying whether the subject matter of study relates to publication outcomes. We do so in the context of race-related research : work that studies economic well-being across racial/ethnic groups. We investigate the implicit career incentives economists have to work on such topics by examining paths to publication for a corpus of 22,056 NBER working papers (WPs) posted from 1974 to 2015. We use an algorithm to classify whether a given WP studies race-related issues. We then construct paths to publication from WPs to data on published articles, and compare paths for race-related WPs to various counterfactual sets of WPs. We document that unconditionally, race-related NBER WPs are less likely to be published in any journal, in an economics journal, and more likely to publish in lower tier economics journals. Once we condition on observable characteristics including field and author affiliations, differences in paths to publication largely disappear, and such work is actually slightly more likely to publish in top-tier economics journals. Consistent with unconditional differences in paths to publication being salient to researchers, we find evidence of ex ante selection into WPs studying racerelated issues in that they are of higher readability than other WPs. To understand the interplay with selection of researchers, we compare results to paths to publications for 10 306 CEPR WPs posted from 1984 to 2015. We conclude by discussing implications for economists’ incentives to contribute to debates on race and ethnicity in the economy JEL Codes: A11 ; B41