James P. Scanlan, Attorney at Law

Jones v. City of Boston

(May 5, 2015; rev. May 9, 2016)

Prefatory Note:  This is one of six subpages to the Disparate Impact page of jpscanlan.com.  Other subpage pages include the Four-Fifths Rule subpage (which addresses reasons why, contrary to the belief or many, the four-fifths rule of the Uniform Guidelines on Employee Selection Procedures does not provide a plausible measure of effect), the Less Discriminatory Alternative - Procedural subpage (which addresses an anomaly in a Supreme Court decision, which was subsequently codified by Congress, whereby a defendant’s liability would turn on events following the trial), the Less Discriminatory Alternative - Substantive subpage (which discusses the implications of the fact that reducing the prevalence of an outcome tends to increase relative differences in experiencing it as it bears on the identification of a less discriminatory alternative to a practice challenged on the basis of its disparate impact), the Fisher v. Transco Services subpage (which discusses a Seventh Circuit decision reflecting the mistaken view that the large relative differences in failure to meet a performance standard was a function of the stringency, rather than the leniency, of the standard), and the Bottom Line Issue subpage (which merely provides a reference to my 1985 treatment of that issue in the University of Michigan Journal of Law Reform).

***

In Jones v. City of Boston, 752 F.3d 38 (1st Cir. 2014), the court of appeals reversed a district court grant of summary judgment to the City of Boston on a challenge to the disparate impact of the Boston Police Department’s policy of the terminating (or requiring rehab for) police officers failing a hair follicle test for cocaine.  The case is of interest because, in a situation where adverse outcome rates were quite low,  and where the black rate was at least several times the white rate (e.g., in 2003, the black rate was 1.1% while the white rate was 0.2% ), the court discussed the differing perspectives as to the size of the disparity depending on whether one examined the relative difference in the adverse outcome or the relative difference in the corresponding favorable outcome.  In discussing this issue, while also discussing less discriminatory alternative, the court failed to recognize that any alternative that led to a general reduction in adverse outcomes would tend to increase relative differences in rates of experiencing the outcome.  Further, in discussing whether the disparate impact should be regarded as of sufficient “practical significance” to require that the city justify the practice causing  the impact, the court noted that “as a matter of theory,” it would not “expect to find any single measure of the size of the impact to determine its practical significance.”  The court also noted:  “Ultimately, we find any theoretical benefits of inquiring as to practical significance outweighed by the difficulty of doing so in practice in any principled and predictable manner.” 

As explained in Section E (at 27-32) “The Mismeasure of Discrimination,” Faculty Workshop, University of Kansas School of Law (Sept. 20, 2013), there is principled way of appraising the practical significance of the a disparity (or, at rate, the differences in the circumstances of two groups reflected by their rates of experiencing and avoiding an outcome, which might also be characterized as the strength of the forces causing the outcome rates to differ) and only one such way.  See also the discussion in Section B (at 15-23) of the paper and the discussion regard Table 5 (at 336) of “Race and Mortality Revisited,” Society (July/Aug. 2014).  The estimated effect size, as discussed in those papers, would be .59 standard deviations. [i]  

Jones v. City of Boston is also of interest because it involves Massachusetts civil services laws.  I have discussed in many places the fact that relative differences in adverse outcome tend to be larger, while relative differences in the corresponding favorable outcomes tend to be smaller, among comparatively advantaged population/subpopulations (where the adverse outcome are comparatively uncommon) than among disadvantaged population/subpopulations where they tend to be smaller.  I have made this point with respect to the Norway, Sweden, and Minnesota in “Can We Actually Measure Health Disparities?,” Chance (Spring 2006) (at 50) and “It’s easy to misunderstand gaps and mistake good fortune for a crisis,” Minneapolis StarTribune  (Feb. 8, 2014), and specifically with regard to Massachusetts in “The Mismeasure of Health Disparities in Massachusetts and Less Affluent Places,” Quantitative Methods Workshop, Department of Quantitative Health Sciences, University of Massachusetts Medical School (Nov. 18, 2015) (Abstract).  See also my letter to Stanford Center on Poverty and Inequality (Mar. 8, 2016)  (at 7) regarding a perception of “staggering health inequity” in Massachusetts in the Center’s 2016 Poverty and Inequality Report and my “The ‘Feminization of Poverty’ is Misunderstood,”  Plain Dealer (Nov 11, 1987), regarding the fact that female-headed families make up a larger proportion of white poor in Massachusetts and of the black poor in Mississippi (because the former is a wealthy state and the latter is a very poor state). 

But, apart from being a generally advantaged state in terms of things like income and health, Massachusetts tends to have a philosophy that leads to comparatively low rates of adverse actions pursuant to the state’s decision-making processes.  That will tend to result in comparatively large racial and other relative differences in adverse outcomes and comparatively small relative differences in the corresponding favorable outcomes rates.   See my letter to the Boston Lawyers’ Committee for Civil Rights and Economic Justice (Nov. 12, 2015) regarding perceptions about difference in suspension rates of black and white, and disabled and non-disabled, public school students.        

Similarly, with regard to many public employee matters Massachusetts can be likened to the U.S. Postal Service, where the extensiveness of guarantees of employee rights, which tend generally to reduce adverse employee actions, will tend to lead to large relative differences in rates of experiencing those outcomes but small relative differences in avoiding them.  See my “Getting it Straight When Statistics Can Lie,” Legal Times ( June 23, 1993).

As discussed in Section E of the Kansas Law paper and in my “Is the Disparate Impact Doctrine Unconstitutionally Vague?,”  Federalist Society Blog (May 5, 2016), there are situations where altering a standard may be deemed to increase or decrease a disparate impact.  But where meeting or failing to meet the standard entirely dictates whether one ultimately experiences the favorable or adverse outcome, there is no rational basis for maintaining that the altering of the standard affects the impact of the standard.  And policies of termination for rules violations like failing a drug test would seem to be of the latter category.  Thus, a policy of termination for two violations has the same impact as a policy of termination for one violation.  In any case, one must recognize that policies of termination for multiple violations of any rule will tend to show larger relative differences than policies of termination for a single violation.