Expensive Statistical Analysis Doesn’t Justify the Systemic Discrimination of the State
Remember when Americans used to believe that people should be judged on their own merits, not on perceptions about groups to which they belonged? Remember when it was outrageous to come to specific conclusions about people and their worth based on obscure statistical analysis? Remember when there was nothing worse than a dream deferred?
That was back before the woke progressive resurgence.
Rhode Island’s Office of Diversity, Equity & Opportunity (ODEO) paid Mason Tillman Associates half a million dollars to investigate discrimination in the state’s procurement and contracting process, and you’ll never guess what the analysis found. Yes, it’s true… an expensive report commissioned by an office whose purpose is to push diversity and equity found that discrimination exists and the state must push diversity and equity. That finding is already being used to demand that the state implement discriminatory policies.
Yes, the Mason Tillman report does explain that the law requires government to consider race-neutral solutions for disparities before overtly discriminating, but based purely on statistics, it concludes that Rhode Island can go straight to discrimination. Note, in particular, that the discrimination is alleged in the private sector:
In the absence of a race and gender-neutral explanation for the disparities, the regression findings point to racial and gender discrimination that depressed business ownership and business earnings. Such discrimination is a manifestation of economic conditions in the private sector that impede minorities and Caucasian females’ efforts to own, expand, and sustain businesses.
Let’s be clear about the claim, here. Mason Tillman found statistical evidence (not factual evidence) of disparity (not discrimination) in business ownership and earnings, and that is being used as justification to have the state discriminate against white men in its own business operations. Fewer women and minorities own businesses in industries relevant to the state, so the state is going to make up for that by giving those that do exist an advantage over their competition.
To substantiate these findings, the report performs regression analysis, whereby one holds some variables constant and observes the effect on other variables. The problem — when looking at variables as broad and deeply cultural as race and gender — is that it isn’t possible to account for every benign explanatory factor, let alone unique circumstances that just happen to be the case. In essence, the analysts look at a handful of variables (age, education, resources, etc.) and attribute any remaining differences to bigotry.
Consider that the only significant disparities that Mason Tillman found in business ownership were in the construction industry for women, Blacks, Hispanics, and Portuguese and in goods/commodities/supplies for women (Table 9.7). One unsurprising fact about the construction industry is that people with advanced degrees are significantly less likely to own businesses in it (Table 9.3). Well, women have been more likely to collect advanced degrees for more than a decade.
So, women choose to pursue careers requiring higher education, which creates a disparity in the number owning construction businesses, and the State of Rhode Island is going to give an extra boost to them. This seems less like eliminating sexism in government contracts than ensuring that women have an advantage over men across the board.
And this is just an example to make a point. Other factors surely come into play, not only for women but for all of the demographic groups.
When it comes to the government’s own activities, the report simply takes a ratio of “available” businesses in each group measured against the dollar value of contracts awarded to businesses in that category. So, women represent 12.5% of the “available” construction industry in the area and received only 9.0% of the state dollars for large, formal contracts, so the report claims they “lost” $4,198,689.
But the number of factors that could legitimately explain such a “disparity” is huge. To return to our example, men are much, much more likely to go into construction than women, whereas women are much more likely to pursue higher education. Assuming even distribution of intelligence between the sexes, this would suggest that statistically the average man in construction is going to be more intelligent than the average woman in the same industry. Since the measurement of the report is minority and woman business owners, not employees, the so-called disparity would be even greater, because it’s a question of who among all the competent workers would also be competent to run the business.
Put differently, a woman whose competence and personality are apt to make her want to run a business is even more likely have gone into another industry. Meanwhile, the existence of government quotas creates incentive for minorities and women to rush (or be rushed) into management positions for which they may lack the experience or the skill in construction.
Of course, there are going to be exceptions, but if we’re making policy based on statistics, we can’t introduce the possibility of individual exceptions as a last-minute adjustment for a preferred outcome.
The end result is that among the large majority of people in the construction industry, the state is going to make it even more difficult for those with fewer advantages to succeed. The quota contracts aren’t going to be taken from the most advantaged (or politically connected) white men, but the least. What happens to them? They dry up like raisins in the sun.