Tuesday, November 10, 2015

Measuring the Chilling Effect of Late-Term Abortion Limits

by Michael Dorf

Late last year I participated in a conference at the University of Chicago Law School on empirical work in constitutional law. In a post at the time, I provided very brief overviews of each of the papers. The papers have now been published in the NYU Law Review. They are well worth reading.

Here I will say a few more words about my paper, co-authored with Princeton political scientist Brandice Canes-Wrone: Measuring the Chilling Effect. After describing our paper, I'll offer some thoughts about how our findings might bear on the Texas and Mississippi abortion cases currently before the Supreme Court on petitions for certiorari.

Here is the abstract for our article:
Supreme Court doctrine grants special protection against laws that “chill” protected speech, most prominently via the overbreadth doctrine. The overbreadth doctrine permits persons whose own speech is unprotected to challenge laws that infringe the protected speech of third parties. The Court has not generally applied overbreadth and the other speech-protective doctrines to other constitutional rights even though other rights could also be subject to a chilling effect. The case law simply assumes that the chilling effect only acts on the exercise of speech, and that this justifies treating speech differently from other rights. 
We tested these assumptions with respect to abortion rights. By comparing abortion rates with state laws over a two-decade-plus period, we found a statistically significant correlation between laws forbidding late-term abortions and the reduction of not only late-term but also “near-late-term” abortions, i.e., abortions in the roughly one month before the period in which abortions are forbidden. That effect persists even after controlling for potentially confounding variables, such as the number of abortion providers and pro-life public opinion. Moreover, the effect is not limited to the year of enactment or associated with failed policy initiatives, suggesting that the impact is due to the law itself rather than associated publicity. These findings are consistent with, and strongly suggestive of, a chilling effect on abortion providers and/or women seeking abortions. This result undermines the implicit assumption that the chilling effect is unique to laws regulating speech and vindicates the general proposition that laws can chill the exercise of constitutional rights beyond their literal coverage. 
Because that is an abstract, i.e., a summary, it necessarily simplifies. The article itself makes clear that although the Supreme Court cases frequently say that overbreadth doctrine only applies with respect to freedom of speech, the Court in fact often applies something like overbreadth doctrine in a variety of contexts, including--sometimes--to abortion cases. We cite some of my own work, as well as work by Harvard law professor Richard Fallon and others that distinguishes between the Court's description of its practices and its practices themselves.

To the (quite considerable) extent that existing case law is unclear about whether laws restricting abortion are subject to something like overbreadth analysis, the empirical results of the Canes-Wrone/Dorf article may be helpful in addressing the normative question of whether something like overbreadth doctrine ought to be applied to laws restricting abortion. Our findings suggest that it should. The risk that overbroad speech-restrictive laws will chill protected speech plays a very important role in justifying overbreadth analysis of speech-restrictive laws; thus, our demonstration that overbroad abortion-restrictive laws chill constitutionally protected abortion should count as a reason to subject abortion-restrictive laws to overbreadth analysis.

To be sure, our findings do not bear directly on the questions in the Texas and Mississippi cases. We found that late-term abortion restrictions chill abortions in the "near-late-term," i.e., the period before the formal beginning of the late term. The Texas and Mississippi cases involve abortion restrictions--requiring doctors to have admitting privileges at nearby hospitals and imposing the standards of "ambulatory surgical centers" on facilities performing abortions--that make it more difficult for women in Texas and Mississippi to obtain abortions at all points in pregnancy. The issue is less that doctors will "self-censor" by denying women protected abortions out of fear that they will inadvertently cross the line than it is that doctors and facilities are unable to comply with the new laws and so women are denied access to abortion.

Nonetheless, our findings may be relevant to a broader question in the Texas and Mississippi cases: What, exactly, does the "undue burden" standard announced in Planned Parenthood v. Casey mean in practice? The Court has decided only three abortion cases since Casey in 1992, and none sheds much light on the issue. Two of those cases (Stenberg v. Carhart in 2000 and Gonzales v. Carhart in 2007) involved so-called "partial-birth" abortion restrictions that, at least so far as the Court was concerned in the latter case, did not infringe anyone's ability to obtain a constitutionally protected abortion. The other--Ayotte v. Planned Parenthood in 2006--ended up addressing only a remedial question about what courts should do if they find that an abortion prohibition is invalid.

In the more than 23 years since Casey, lower courts have issued a variety of conflicting opinions about the meaning of the undue burden standard. Judges sympathetic to the Court's abortion jurisprudence have tended to give the standard teeth, while those who are unsympathetic to abortion rights tend to cut the states a good deal of slack, as the Fifth Circuit did in the Texas case now on petition for cert. Undoubtedly, for many of the Justices, the question of how to flesh out "undue burden" is connected to the underlying merits of the abortion right.

But assuming--as I think it is safe to assume absent a change in the Court's personnel--that there aren't five votes either to overrule Casey and replace it with Roe's strict scrutiny or to overrule Roe and eliminate protection for abortion as a constitutional right, a grant in the Texas and/or Mississippi cases will mean the necessity of applying Casey's undue burden test, which, in turn, should require clarifying it. And putting aside various doctrinal nuances, that will require the Court to decide whether to apply a more or less deferential test.

The findings of the Canes-Wrone/Dorf paper are relevant to that broad question. Although we only found a chilling effect with respect to late-term abortions, that is all we tested. Abortion rights may well be susceptible to a chilling effect across a variety of contexts. The existence of a chilling effect means that doctrine that is applied neutrally will, in many contexts, undercount constitutional violations: it will sustain laws that do not visibly violate constitutional rights but do chill them. An especially lax test will exacerbate that bias. Accordingly, our findings could be used to support a general approach that is either especially rigorous in finding undue burdens or at least neutral. A deferential version of the undue burden test would be especially difficult to justify.


David Ricardo said...

Two points

First of all it is admirable to see regression analysis and rigorous statistical methods used in a law review article.

Second, the results found in the research confirm that the results of late term abortion restrictions are exactly what those who enacted those limits intended. If I were a proponent of restricting abortion rights I would be very pleased with the research presented here. While I do not know if the intent of the authors was anything beyond intellectual curiosity and academic research, if their intent was to produce evidence that there was a chilling effect and hence further argument against late term abortion restrictions I would think their article might well have the opposite effect.

Joe said...

The USSC has three per curiams since Casey related to abortion as well:

Mazurek v. Armstrong (physician requirement)
Leavitt v. Jane L. (20 week law)
Dalton v. Little Rock Family Planning Services (funding)

On the merits, concur with first comment. As to the effects, "benefits" and "harms" is a matter of what side you are on. So, yeah.

Hashim said...

Mike -- isn't the "chill" from late-term abortion laws on near-late-term abortions qualitatively different from the chill involved in 1A overbreadth cases?

In the 1A context, the law *itself covers both* unprotected and protected speech, and the "chill" at issue is that protected speakers simply won't sue to vindicate their own rights; the court thus has allowed unprotected speakers to sue and vindicate third-party rights.

In the late-term abortion context, by contrast, the law does *not* purport to cover near-late-term abortions. So any "chill" is self-censorship based on misinterpretation of the law. Now, it may be that the self-censorship is induced by uncertainty concerning the line between late-term and near-late-term abortions. But that's not an overbreadth/chill concern -- it's a vagueness/chill concern, which *does* apply outside the 1A context (though often with considerably less rigor; but cf. Johnson).

Put differently, imagine an abortion statute that drew a clear and unambiguous line between the protected abortions and the unprotected abortions. In that case, neither vagueness nor overbreadth doctrine would allow those performing unprotected abortions to sue, notwithstanding any "chill" that may empirically be demonstrated due to self-censorship by those performing protected abortions.

Michael C. Dorf said...


1) W/r/t David Ricardo's second point: We undertook the study purely out of curiosity. We began by looking for free speech-restrictive laws but concluded we couldn't find reliable independent indicators of the volume of speech. I was surprised by our results. I thought the data would be too noisy to show anything, but the effect is fairly strong. As to who benefits, as I said, that wasn't my concern, but I don't think DR's worry is well-founded: Pro-life groups have ALREADY been enacting as many laws restricting abortion as they can.

2) Joe is right. I should have said "plenary" cases. Note, however, that none of the three per curiam cases involved application of the undue burden standard.

3) Hash is right as well, at least w/r/t some of the laws. In the paper--at pp 1101-1102 and more generally--we recognize that some of the chill here arises due to the inherent vagueness of both viability and number-of-weeks restrictions (due to uncertainty about when pregnancy began). Some of the abortion laws are overbroad in the classic sense because, by their terms, they forbid abortions prior to a point in pregnancy when viability is scientifically possible. Even w/r/t the post-viability prohibitions, however, we think that we have found something important because, as we explain, chilling effect is a broader phenomenon than overbreadth, and because of the close connection between overbreadth and vagueness in many contexts. Our main point in the paper is not that free speech overbreadth doctrine ought to be applied in abortion cases. Our point is simply that we found a chilling effect.

Joe said...

For completeness, there is also another per curiam: Lambert v. Wicklund (parental notification) (the one I meant to cite). I take the point as to "undue burden" but they are of mild interest on 'proper' regulations. As noted here (14-15), O'Connor and Souter had a chance to add a bit to the meaning of "undue burden" as well:


But, Kennedy didn't join in either statement, so the "light" they shined is unclear.