Historically, Black Americans have owned businesses at a lower rate than White Americans. While there is a vast literature on the factors that explain this disparity, machine-learning analytics have not been exploited to capture relevant predictive determinants of the formation of Black-owned Businesses. In this paper, we utilize the Brookings Institution 2022 Black Progress Index Data to explore, similar to the approaches in criminology, the attractors and generators of Black-owned businesses across US counties. We augment the Brookings data to account for the presence of a Historically Black College/University (HBCU) in the county, and deploy a rigorous least absolute shrinkage and selection operator—a machine-learning analytical tool—to identify the best predictors of Black-owned businesses in a county, and classify them as attractors and/or generators in regression specifications. Parameter estimates from several mixed effects regressions specifications reveal that HBCUs are an attractor/generator of Black entrepreneurship. Treatment effect matching parameter estimates suggest that the HBCU effects on Black entrepreneurship outcomes are causal. Overall, our findings suggest that HBCUs constitute an ecosystem for Black entrepreneurship as they are an important catalyst and driver of Black-owned businesses, and can serve a substantive role in reducing racial entrepreneurship disparities. Our results suggest that policies that engender private or public investments in HBCU capacity to enhance their Black entrepreneurship ecosystem can be effective in reducing Black-White disparities in entrepreneurship and business ownership.