Racial/ethnic misrepresentation of and bias against minority executives

Abstract

We reassess whether and to what degree the hiring, development, and promotion decisions of S&P 500  companies has led to misrepresentation of and bias against their minority executives. Instead of the US population benchmark that has conventionally been used to measure misrepresentation, and from such misrepresentation attribute the presence and magnitude of racial bias and discrimination, we measure misrepresentation in US executives using the benchmark of the racial/ethnic densities (RAEDs) of their college cohort peers. Our key result is that the differences between US executive RAEDs and the RAEDs of their college peers are far smaller than those found using the US population, typically by an order of magnitude or more. Whereas under the US population benchmark, Black and Hispanic S&P 500  US executives are reliably greatly under-represented by-9.1% and-15.5%, and Asians and Whites are reliably over-represented by 1.3% and 24.4%, respectively, we find that Asians and Blacks are statistically at their college-peer benchmark levels, with college-peer benchmarked misrepresentations of just-0.4% and 0.1%, respectively. White executives are reliably slightly above their college-peer benchmarks by 1.9%, while Hispanics are reliably slightly below by-1.2%. Our study highlights the importance of the benchmark used to measure the signs and magnitudes of racial misrepresentations, and the sensitivity of inferences as to the presence or absence of racial biases to the choice of benchmark.

Keywords: Executives, race, ethnicity, benchmarks, misrepresentation, racial bias JEL classifications: A13, J01, J15, J29, M12 Data availability: The datasets analyzed are available from mainly public sources

JEL Classification: A13, J01, J15, J29, M12