| Abstract: | The 1997 Economic Census data on minority and women-owned businesses reveals distinct cross-state and cross-county differences in ownership characteristics. The breakdown of this data into racial groups provides the opportunity to study the geographical features of black enterprise. As will be seen, both in aggregate and across states and counties, significant differences exist in the prevalence and magnitude of black and white enterprise. I propose a "legacy of slavery" hypothesis to explain these observed differences. Preliminary analysis indicates that this hypothesis has considerable explanatory power. A significant negative relationship exists between the geographical concentration of slavery in 1840 and the current prevalence and magnitude of black-owned enterprise. At the state level, this relationship remains even after controlling for the normal business climate as indicated by white enterprise levels. County level regressions with state fixed-effects yield similar estimated coefficients. In total, I find that the legacy of slavery has reduced the number of black-owned businesses by at least 71,009 and sales by $27.3 billion per year. |
