Abstract Using data developed for, and presented to, the US District Court, this study evaluates the performance of women-, African-American-, and Hispanic-owned firms vis-a-vis firms owned by neither women nor minorities in three different South Florida markets: architecture, structural engineering, and civil engineering. After controlling for the influence of experience, size, and location, three of the most commonly used determinants of firms' earnings, the empirical results suggest that the three markets do not convert firms' characteristics into economic outcomes in the same manner for women and Hispanic owners as for owners who are neither women nor minorities.
Keywords: affirmative action, discrimination, earnings, ethnic disparities, gender disparities
"Affirmative action" is a widely misunderstood, misused, and maligned term. It has been utilized in public-sector procurement, small business development, hiring and promotion, housing, lending, education, provision of healthcare services, drawing the boundaries of electoral districts, and numerous other contexts. From a positive perspective, it is heralded by some as an impartial mechanism rendering criteria that prevent and denounce the formation of unwarranted barriers to opportunity for historically disadvantaged groups; from a negative prism, others portray it as an inflexible device providing unfair prerogatives to chronic underperformers, stigmatizing, in the process, all members of the affected group(s) and squandering society's scarce resources in a distribution scheme that is far from optimal (Edley 1996). The arguments and counterarguments are so contradictory that the viability of policy decisions often has to be decided by the courts.
In 1982 the Board of County Commissioners of Miami-Dade (then Dade) County, Florida enacted a Black Business Enterprise Program which was amended in 1994. Also in 1994, a Hispanic Business Enterprise Program and a Women Business Enterprise Program were enacted. All three programs established set-aside and participation goals designed to secure for women- and minority-owned firms a certain percentage of the County Government's procurement and contract dollars. The idea was to foster small businesses operated by affected groups via opening income sources that the affected groups probably would be unable to approach by themselves, with the purpose of increasing their long-term comparative advantage and capability to compete effectively in both the public and private sectors, creating jobs for women and minorities, and reducing gender and ethnic inequalities in various forms of market accessibility, sales, and earnings (Bates 1998, 2001; Rice 1993).
Recently the US District Court for the Southern District of Florida, Miami Division, heard the argument (Hershell Gill Consulting Engineers, Inc. et al. v. Metropolitan Dade County, Florida et al.) that Miami-Dade's three remedial programs should not apply to procurement in architecture and engineering. Three distinct markets--architecture, structural engineering, and civil engineering--were identified as meeting the narrowly tailored criterion of the US Supreme Court (Day 2001). (These three markets account for the lion share of the Miami-Dade County architecture and engineering procurement business.)
The purpose of this study is to explore whether gender and/or ethnic disparities in marketplace earnings exist in each of the three markets, and, if disparities do exist, determine if they can be attributed to discrimination. Specifically, the study seeks to detect the presence of significant inter-gender and inter-ethnic firm earnings differentials after adjusting for key earnings determinants using data generated and analyzed for the Court. Evidence of such differentials persisting after controlling for universal earnings determinants would imply that the markets do not convert firms' characteristics into economic outcomes in the same manner for owners of both genders and/or different ethnic groups, and would suggest discrimination (Mason 1995). …