Racial Impact Statement Laws in New Jersey and Elsewhere
On February 27, 2017, the New Jersey Assembly Law and Public Safety Committee reported favorably on Senate Bill No. 677 (with amendments). The original bill, passed by the New Jersey Senate in June 2016, required racial and ethnic impact statements for any legislative measure that affects pretrial detention, sentencing, probation, or parole policies concerning adults or children. The version recently reported out of committee somewhat expanded these requirements.
Racial impact statement laws have previously been enacted in Connecticut, Iowa, and Oregon, and similar legislation has recently been introduced by lawmakers in Arkansas, Florida, Mississippi, and Wisconsin.
The New Jersey legislation, or at least some of the support for it, seems to have been prompted by a 2016 Sentencing Project study titled “The Color of Justice: Racial and Ethnic Disparities in State Prisons,” which found that New Jersey, with a black incarceration rate 12.2 times the white rate, had the largest racial disparity in incarceration rates in the nation. As reflected in a Star-Ledger editorial, the finding was apparently a source of consternation because New Jersey considered itself to be a leader in reducing prison populations and had incarceration rates below the national average.
Discussion of the subject, and the legislation itself, accord with the near universal belief – shared and promoted by the United States Department of Justice (DOJ) – that generally reducing adverse criminal justice outcomes will tend to reduce (a) relative (percentage) racial differences in rates of experiencing the outcomes and (b) the proportion racial minorities make up of persons experiencing the outcomes.
But, as I have explained here several times – including in “Things the President Doesn’t Know About Racial Disparities” (Aug. 5, 2016) and, most recently, in “Compliance Nightmare Looms for Baltimore Police Department,” (Feb. 8, 2017) – reducing any outcome in fact tends to increase, not reduce, (a) and (b) as to the outcome. That is, as illustrated in Table 1 of the February 8 post for example, lowering a test cutoff, while tending to reduce relative differences in rates of passing the test (the increasing outcome), tends to increase relative differences in rates of failing the test (the decreasing outcome). Correspondingly, lowering a cutoff, while tending to increase the proportion the lower-scoring group makes up of persons who pass the test, tends also to increase the proportion the lower-scoring group makes up of persons who fail the test.
Similar patterns will tend to appear when any outcome is increasingly restricted to those most susceptible to it – not in every instance, of course, but a good deal of the time.
In the case of state-by-state comparisons like those in the Sentencing Project study, the inverse correlation between the size of incarceration rates and the size of relative racial differences in incarceration rates will likely be a limited one. Many factors that vary from state to state also play importantly into racial differences. But it nevertheless warrants note that Table 3 of the Sentencing Project report (at page 8), which lists the ten states with the largest ratios of the black incarceration rate to the white incarceration rate (ranging the 12.2 figure for New Jersey to 8.4 for Nebraska), shows that New Jersey has the lowest incarceration rates for both blacks and whites. Given that fact, and notwithstanding the factors just noted, a comparatively high black-to-white incarceration ratio is something to be expected rather than to be surprised by.
One may observe a variety of cross-state patterns regarding general incarceration patterns and relative racial differences in incarceration rates. But, subject to the qualifying factors noted in the above-mentioned posts, any substantial reduction in incarceration rates in a particular state is very likely be accompanied by an increase in the black-white incarceration ratio and the proportion African Americans make up of persons incarcerated. See my Mired in Numbers,” Legal Times (Oct. 12, 1996), regarding the way that changing a three-strikes law to a four-strikes law is virtually certain to increase the proportion African Americans make up of persons adversely affected by such a law; see also my “Things DoJ doesn’t know about racial disparities in Ferguson,” The Hill (Feb. 22, 2016), regarding the way that increasing the number of missed court appearances to trigger an arrest warrant is virtually certain to increase the proportion African Americans make up of persons against whom warrants are issued. Though I have not seen discussion of the situation in New Jersey before the general reductions in incarceration, I assume that prior to those reductions the black-to-white ratio was lower than it is now.
While the New Jersey Senate bill requires statistical analyses to identify the racial impact of proposed legislation, it does not specify how such impact will be measured. Almost invariably there will be some racial impact of criminal laws. Further, the same groups will be disproportionately affected by a law regardless of how one measures the impact. It is when one endeavors to quantify that impact for purposes of determining whether it should be deemed large or small, or whether a modification to a law increases or decreases the impact, that issues I have raised here and elsewhere come into play. And the quantification problem is a complex one that few observers have yet even glimpsed, as I discussed here in “Is the Disparate Impact Doctrine Unconstitutionally Vague?,” (May 6, 2016) (PDF version). But it is not possible to rationally address the matter without fully understanding the ways measures tend to be affected by the frequency of an outcome, and certainly not when proceeding on an understanding of such effects that is the opposite of reality.
So if the New Jersey bill becomes law, the state will face some challenges in implementing it in a manner that makes any sense. The same holds for similar laws already enacted (or to be enacted) by other states, none of which is likely to understand better than New Jersey that generally reducing adverse criminal justice outcomes tends to increase the most common measures of racial disparity in criminal justice matters. And in all places having such laws there should be ample puzzling over why general reduction in incarceration rates have left relative racial/ethnic differences larger than before, just as, all across the country, school administrators have been puzzling over why relaxing of discipline standards was accompanied by increased, rather than decreased, relative racial/ethnic differences in discipline rates. See page 27 of my Comments for the Commission on Evidence-Based Policymaking (Nov. 14, 2016).
The February 8 post discussed the proposed consent decree covering Baltimore police practices and the compliance problems arising from the mistaken premise of the decree that generally reducing adverse interactions between the police and the public would tend to reduce (a) relative racial and other demographic differences in rates of experiencing those interactions and (b) the proportions racial minorities and other more susceptible groups make up of persons experiencing them. In the post, I discussed that the court in the case intended to accept public comment on the proposed decree, as well as prospects that the comment process might lead to a better understanding of the issues.
On February 15, the court issued an order establishing a procedure for submitting written comments until March 7 and setting a public hearing for April 6. The comments I submitted make points similar to those in the February 8 post. They also discuss the obligations of the DOJ attorneys handling the case, now fully informed of the measurement issues, to address those issues with the new leadership of the agency. So the decree approval process may provide an opportunity for the DOJ finally to understand this matter.
An opportunity for DOJ to educate itself on the matter also exists as a result of its issuance, on January 13, 2017, of an investigative report finding that the Chicago Police Department had engaged in a pattern or practice of unconstitutional use of force, and, on the same day, the execution of an agreement in principle with the City of Chicago calling for negotiation of a consent decree to address the finding. I am uncertain whether the report contains a specific determination that unjustified use of force had a disparate on racial minorities. But among its more provocative findings was that African Americans, who make up about a third of the city’s residents, made up 76 percent of persons subjected to police use of force (83 percent in the case of children) and 80 and 81 percent of persons subjected to police use of firearms or tasers.
Possibly, in the course of deciding whether it wishes to follow through with the contemplated decree, the DOJ will come to recognize that restrictions on police use of force that the findings suggest Chicago should implement will tend to increase each of these percentages. Once DOJ understands the issue, it may be able to provide guidance to states contemplating laws requiring racial impact statements.
And, along with the Departments of Education and Health and Human Services, DOJ can explain to states and other jurisdictions that previous guidance to the effect that relaxing public school discipline standards will tend to reduce relative racial and other demographic differences in discipline rates was not correct. Also, along with the host of other federal agencies enforcing fair lending laws, DOJ can explain to lenders that, contrary to guidance dating back more than two decades, relaxing lending standards tends to increase, not reduce, relative racial/ethnic differences in rates of failing to meet the standards.
But, as I discussed here in “Will Trump Have the First Numerate Administration?” (Jan. 4, 2017), there is much that many arms of the government need to learn about measuring demographic difference in adverse and favorable outcomes.
Questions? Reactions? Email firstname.lastname@example.org with your response to continue the conversation. Select comments may be published on the blog.