Ask the Author: More on the Impact of Pauper Status

In the comments to two previous posts (here and here), many people have expressed interest in knowing more about IFP petitions, only a handful of which the Court grants each term (despite their comprising the majority of petitions filed), and to which we don’t devote nearly as much coverage as we do to paid petitions. In the comments section, I linked to this paper on the topic entitled “Paupers as Litigants: Does the Supreme Court Provide a Forum for the Economically Disadvantaged?” which provoked further discussion.

For more on this issue, we turned to the co-authors of that paper, Christina Boyd and Ryan Black, who have generously answered several questions on this subject. Ryan and Christina are both Ph.D. students in political science at Washington University in St. Louis and associates in the Center for Empirical Research in the Law at Washington University School of Law, and their work focuses on judicial politics.

Below is part 1 of our discussion; part 2 will run tomorrow.

Q: The central claim in your paper is that, “justices on the Court are, all else equal, less likely to support petitions filed by paupers than they are paid parties.” Can you talk a little bit about what “all else equal” means in terms that non-statisticians can understand? Can your methodology correct for what is likely a higher quality of “lawyering” that goes into paid petitions?

By accounting for other factors that might be influencing why a justice might vote to grant or deny cert in a particular case, we attempt to isolate the impact that a paupered status petition has on the review of cert. Perhaps the easiest way to think about “all else equal” is to think about two petitions that might come before the Court. Suppose these two petitions will look effectively the same – e.g. they fall in the same issue area, they allege the same level of circuit conflict, they have the same number of amicus briefs – except that one will be an IFP petition and one will be a paid petition. Our discussion below regarding hypothetical petitions should hopefully provide some insight on this.

Our study really relies on the pool clerks to draw our attention to the quality of lawyering for a particular petition. By this we mean that because we are using the certiorari memos written by the pool clerks in the 1992 term to provide us the information for our study, it is really necessary for the clerks to note the poor or high quality of lawyering for us to be alerted to it. This happens on occasion (e.g. “the lawyering is poor,” “a remarkably lucid pro se brief,” or “the petition is incomprehensible”), but on the whole the clerks limit their discussion to the substance of the case before them and its worthiness for plenary review.

In your paper, you create four hypothetical petitions with various combinations of factors that have been proved to influence cert.; you then hold those factors constant and suppose that a petition was filed either on a paid or pauper basis. When you run the numbers, Petitions B and C had particular sets of facts such that whether or not that petition was granted depended heavily on whether it was filed on a paid or pauper basis [see table on pg. 30 and graph on pg. 22 of their paper]. What was it about those arrangements of factors that meant that these two hypothetical petitions showed such a gap between the likelihood of a grant when the petition was filed on a paid vs. pauper basis? Why did other sets of facts seem to create a situation where the difference between a petition filed on a paid basis and one filed IFP did not play as prominent a role in determining whether the petition was granted? Why might a justice discount certain types of petitions because they come from paupers but not others?

For petitions with a reasonably large probability of receiving a vote to grant (i.e. a baseline between 30-75 percent), the impact of IFP status costs about 20 percent. That is, holding other factors constant and only altering IFP status reduces the likelihood that a justice votes to grant cert for the petition by about 20 percent. If you make the baseline likelihood of review either very large or very small (e.g. 99 or 1 percent), then the impact of IFP status becomes much smaller. Our intuitions from cue theory tell us that in the high baseline case of 95 percent, the other facts in the petition (e.g. presence of real circuit conflict, participation of organized interests, etc.) “overwhelm” the effect of the negative IFP cue. Conversely, in the low baseline case, there are already so many other negative cues that adding yet another makes very little substantive difference.

Were you able to calculate a difference in voting preferences between Justices that either did or did not participate in the cert. pool?

In 1992, despite numerous rumblings about reforming the system, Justice Stevens was the only non-participant in the cert pool. Though we did not control for this in our initial model, we were able to re-estimate our statistical model with a separate variable to account for Justice Steven’s votes. This variable failed to achieve conventional levels of statistical significance, so we cannot confidently say that Justice Steven’s votes were systematically different than those of the other justices in the cert pool. [Note: Ryan and Christina recommend that you see this article by Tony Mauro for more discussion about the cert. pool (in the context of Robert’s decision to join it)].