Once the empirics are locked down, ELS scholars then typically turn to the legal implications of their empirical findings. In an interesting essay, Statistical Inequality and Intentional (Not Implicit) Discrimination, Michael Selmi (GWU) engages with the issue that while "not all statistical disparities are the product of discrimination but some are, and central to achieving greater racial equality will be determining how we can tell the difference--how we know when an observed statistical disparity should be equated with discrimination?" Two specific contexts frame Selmi's discussion: police automobile stops and school discipline.
Troubling racial disparities, according to Selmi, likely produce "one of three responses, what I will refer to as (1) 'Yes but . . . ,' (2) 'Racism is everywhere,' and (3) 'Tell me more.' Those in the 'yes but' category are discrimination skeptics who resist concluding that discrimination underlies the observed inequities and they seek alternative explanations. Those in the second group see discrimination everywhere there are disparities and may go so far as to cast racial aspersions on those in the first group. The third group wants to know more about the inequities, wants to know more about why wage gaps persist even at the highest levels, or why affluent African Americans tend to live in neighborhoods that are less affluent than their white cohorts.” In the end, Selmi "urges policymakers to treat repeated patterns of behavior as intentional, as opposed to implicit, discrimination, and offers a critique of the recent turn to implicit bias."