An explanation of the new anti-bias law de Blasio and Quinn are fighting over
Yesterday, the City Council overrode two mayoral vetoes to create an inspector general for the NYPD and to make it easier to sue the police department for biased policing.
The inspector-general bill is fairly straightforward, and had relatively broad support on the Council.
The anti-bias bill, which Public Advocate Bill de Blasio frequently refers to somewhat reductively as a "ban on racial profiling," attracted the bare minimum of support it needed on the Council to override a mayoral veto, and is more complicated.
Council Speaker Christine Quinn, who supported the inspector-general bill, opposed the anti-bias bill, which she considers redundant, since racial profiling is already illegal, it is already possible to sue the department and she doesn't want excessive court oversight hampering her own control of the NYPD. De Blasio, who is vying with Quinn for one of two spots in a likely Democratic mayoral runoff, has focused on her opposition to this bill to illustrate the difference between his opposition to stop-and-frisk as currently practiced by the NYPD and Quinn's.
(Former comptroller Bill Thompson, the other top-tier Democratic mayoral candidate, also opposes the NYPD's stop-and-frisk program, but opposed both of the Council's police-oversight bills.)
Given the starkly different assessments of the anti-bias bill among declared opponents of stop-and-frisk, and given what seemed to me to be the considerable overlap between that bill and existing law, I sought a assessment from N.Y.U. constitutional law professor Roderick Hills about what the new legislation actually does. The conclusion he led me to: Nobody knows yet.
Here is a condensed version of our email exchange, which we conducted back in July.
Me: Rick, do you have a good grasp of the Council's racial profiling bill? I've had the bill's supporters explain it to me, but I'm still not clear on what it actually does.
Hills: (1) It authorizes private persons to sue either or both the police department or individual officers (section § 14-151(c)(1), as amended by section 2 of the Act) for two very different claims: Section 14-151(c)(1) authorizes claims for what we lawyers call “intent-based discrimination, providing that individuals can bring claims to enjoin either (i) the department’s “intentional bias-based profiling of one or more individuals” or (ii) an individual officer’s “intentionally engag[ing] in bias-based profiling of one or more individuals.” This provision adds zero, nada, zip, to existing law: Any individual can ALREADY sue the department or individual cops for intentionally targeting them based on their race. (Indeed, this is what the case before Judge Scheindlin is largely about.)
So the interesting provision is § 14-151(c)(2)(i), which authorizes what we lawyers call a “disparate impact” claim. Specifically, the law authorizes individuals to bring claims to enjoin “a policy or practice within the police department regarding the initiation of law enforcement action” that “has had a disparate impact on the subjects of law enforcement action on the basis of characteristics [listed in the statute].”
The problem with disparate impact claims is that they tend to attack a lot of ordinary law enforcement policies that most people would regard as a good idea. Suppose, for instance, that the NYPD has a “policy or practice” of more aggressively stopping and frisking people in high-crime zones where crimes have recently been committed with firearms. Such a general enforcement technique will certainly have a disparate impact on black neighborhoods, because black neighborhoods tend to have higher rates of firearm-related violence than non-black neighborhoods. But it would be odd to say that the fact that a firearms-related crime has recently been committed in a neighborhood is not a good reason to get a little more aggressive with stopping and frisking for firearms.
Intro. 1080 guards against such an over-broad rule in two ways.
First, § 14-151(c)(2)(ii) allows the NYPD to vindicate policies that have a disparate impact on racial minorities by “plead[ing] and prov[ing] as an affirmative defense that each such policy or practice bears a significant relationship to advancing a significant law enforcement objective…” This is actually the customary way in which disparate impact claims work in employment and housing discrimination: On a showing of a racially disparate impact, the burden of proof shifts to the party imposing the impact to show that the impact is justified by some race-neutral purpose (“business necessity” in employment). The problem with such an affirmative defense is that it invites judges to engage in a fairly mushy policy-based inquiry into the law-enforcement merits of NYPD’s law enforcement policies. I can imagine that few judges would take up this invitation and might very well rubber-stamp a lot of policies that have some tenuous relationship to good law enforcement. But maybe not: The consequences of this provision are really unpredictable, depending entirely on how aggressively a judge would enforce it.
Second, § 14-151(c)(2)(iii) provides that the mere existence of a statistical imbalance between the demographic composition of the subjects of the challenged law enforcement action and the general population is not alone sufficient to establish a prima facie case of disparate impact violation unless the general population is shown to be the relevant pool for comparison, the imbalance is shown to be statistically significant and there is an identifiable policy or practice or group of policies or practices that allegedly causes the imbalance.
This provision mimics the doctrine governing disparate impact employment discrimination cases. Basically, the provision uses the idea of a “relevant pool” to make it harder for a plaintiff to shift the burden to the NYPD by showing a racially disparate impact. The plaintiff could not show such a disparity merely by proving up that a racial disparity between NYC’s general population and the population subject to more aggressive policing (e.g., the folks in those high-crime neighborhoods). Instead, the plaintiffs would have to show a disparity between policing in COMPARABLE neighborhoods – say, high-crime predominantly white neighborhoods and high-crime predominantly black neighborhoods.
Depending on how the comparable “pool” is defined, the disparate impact test could be a severe limit on the NYPD, or it could be just another way to show racially discriminatory intent. Again, no one quite knows what “relevant pool” means.
Me: I have been told that, generally speaking, it is harder to bring cases in federal court, versus state court, and therefore that this provision will make it easier for New Yorkers to sue the department or individual cops for biased policing. I suppose that is, on its face, a meaningless question, because one could not bring this sort of suit in state court prior to this legislation. ... Is there any truth to the idea that, broadly speaking, federal courts present a more formidable threshold for suits of this sort?
Also, I'm not sure how to put this nicely, but what do you think of the idea that the state courts are typically populated by judges of a less impressive pedigree than federal courts are? And that in a world where a court can enjoin the NYPD from continuing certain policing practices, one would prefer that a federal judge have that authority rather than a state judge?
Hills: Who told you THAT? Any federal claim for unconstitutional racial discrimination can be brought in state court or federal court, at the plaintiffs’ option: State courts have jurisdiction over federal claims, and plaintiffs can use section 1983 (a federal remedial statute) in state court to seek damages, etc. So one does not need this provision to bring a disparate treatment (aka discriminatory intent) claim in state court: The plaintiffs in the Floyd, Davis, etc., cases could have filed in state court, had they thought that the Manhattan supreme court was so great.
Until you tell me that some veteran trial lawyer has said that there has been some difficulty in filing federal constitutional claims (or, for that matter, state constitutional claims) in state court, I will be unwilling to believe that this charter amendment does anything whatsoever insofar as intentional discrimination is concerned. Maybe I’m missing something here – it would not be the first time – so you might ask Burt Neuborne, former legal director of the ACLU, my colleague, and general ace trial lawyer about whether the intentional discrimination provision in Intro. 1080 would make any practical difference in ease of suit. I doubt it, personally, but I don’t mind Burt schooling me on trial stuff.
(Neuborne, who is now a professor of civil liberties and legal director at N.Y.U.'s Brennan Center for Justice, emailed me the following reaction after reading the new law: "It's a swamp. First - no damages, so it's toothless. Second - while, as a matter of theory, injunctive relief is possible on the basis of disparate impact, not merely intentional discrimination, proof of disparate impact under the law is almost impossible because you can't use statistical disparity to prove it. I may lack imagination, but I don't see how disparate impact can be proved without statistics. I don't think it changes much - but, if I were challenging the stop and frisk practices, I'd certainly use the statute for its atmospherics, if nothing else.")
2) There is a rich literature on the differences between state and federal judges. Again, my colleague, Burt Neuborne, wrote the seminal piece (“The Myth of Parity,” 90 Harv. L. Rev. 1105 (1977), arguing in favor of the idea that state judges just as not as reliable in protecting individual rights. Since 1977, there have been follow-up work suggesting that (a) yes, state judges’ pedigree is not as distinguished but (b) that might be a good thing for plaintiffs seeking to vindicate controversial rights, because a distinguished pedigree can make one more stodgy and conservative. Dan Pinello has some empirical evidence that state judges actually vindicate rights more vigorously than federal judges (see http://papers.ssrn.com/sol3/papers.cfm?abstract_id=270185 ). And Bill Rubenstein, longtime director of the ACLU’s gay and lesbian rights project, has argued that certain characteristics of state judges make them more sympathetic to certain minority rights (see http://papers.ssrn.com/sol3/papers.cfm?abstract_id=205228 ).
But all of this is beside the point. The bottom line is that this charter provision is not necessary to get any plaintiff into state court. Plaintiffs could always bring disparate treatment claims in state court under either or both the state and the federal constitution. Until Burt tells me differently, I am inclined to think that the disparate treatment/intentional discrimination provision is just a lot of noise to make Brad Lander’s constituents feel good.
The disparate impact provision of 1080 is another thing altogether: It could have a big effect, but sidestepping the very difficult proof of discriminatory intent. But this has nothing whatsoever to do with the court in which the plaintiff files, state or federal. A plaintiff could bring a charter-based claim in federal court, for that matter, asserting supplementary jurisdiction. (They’d just have to assert a substantial federal question as well.)
UPDATE: Councilman Brad Landers' policy director Michael Freedman-Schnapp writes to point out something I've mentioned in previous coverage but not in this article, which is that "intentional discrimination" piece of the bill "expands the protected classes covered by profiling laws" to explicitly include sexual orientation, gender identity, age, immigration and disability status, which weren't in the previous profiling ban.