Federal Appeals Court Responses to Supreme Court Precedent

Abstract

This manuscript examines how federal appeals courts respond to precedent, in this case, Central Hudson Gas & Electric Corporation v. Public Service Commission. This account includes a comprehensive theory that examines attitudinal factors that relate to Central Hudson, specific relevant legal factors that relate to the case, as well as strategic considerations. We additionally argue that federal appeals courts may reasonably ignore Central Hudson in certain specific instances (most notably when other highly relevant cases are available for lower federal court judges to use). Our results show partial support for several portions of our theory, including a lower propensity for federal appeals courts to positively treat precedent when ideological distance is high. We also find support for one of our factual-based hypotheses (regarding cases that involve drugs and attorney advertising, where other U.S. Supreme Court precedents are readily available for appeals court judges to use).

Keywords:

• U.S. Courts of Appeals
• precedent
• commercial speech

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 447 U.S. 557 (1980).

2 According to Clark, Lax, and Rice’s Supreme Court salience measure (2015), Central Hudson’s salience score (approximately 1.07) places it roughly two standard deviations above the mean Supreme Court case salience value in their dataset (which features both pre and post-decision coverage of cases in three newspapers); it was also the 13th most salient case from the 1979 term (out of a total of 156 cases from 1979 in their dataset). Although it is not politically salient according to the CQ list or Epstein and Segal’s (2000) salience measure, it is featured in constitutional law textbooks such as Chemerinsky’s Constitutional Law, suggesting a reasonably high level of salience examining multiple conceptions of salience. Furthermore, it has been cited and/or treated 42 times by the U.S. Supreme Court (as of January 20, 2022).

3 The three most recent U.S. Supreme Court precedents addressing commercial law are the following:Barr v. American Association of Political Consultants, 591 U.S. ——- (2020), which featured a 6–3 margin including the conservative bloc of justices but also Justice Sotomayor, Iancu v. Brunetti, 538 U.S. ——- (2019), which featured a 6–3 opinion, with Chief Justice Roberts, and Justices Sotomayor and Breyer partially dissenting, as well as Matal v. Tam, 582 U.S. ——- (2017), which featured a unanimous Court opinion authored by Justice Alito.

4 One large difference between Hinkle’s 2015 study and ours is the type of choice set used for comparisons, with Hinkle’s work using a large number of circuit court precedents that cite the Fourth Amendment, versus ours that focuses on the application of one specific precedent

5 Additional evidence for the fear of reversal conceptually comes from Randazzo (2008), which finds that federal district courts do show behavior consistent with fear of reversal when it comes to their federal circuit court superiors.

6 Thus, the only actual damage that an appeals court judge could suffer would be reputational damage in front of her federal appellate court peers or a loss of pride from having a case overturned by the U.S. Supreme Court.

7 425 U.S. 748 (1976).

8 A strict scrutiny test of commercial restrictions on speech would require federal judges to apply the following general standard to commercial speech restrictions: a) the restriction on commercial speech must meet a compelling interest, and b) the restriction on commercial speech must be narrowly tailored to that interest. By contrast, an intermediate scrutiny test, while still putting the burden of constitutionality on government, would use a lower standard: a) the restrictions on commercial speech must be for an important interest, and b) the restrictions on commercial speech must be substantially related to the previously cited important interest.

9 429 U.S. 190 (1976). The adoption of intermediate scrutiny replaced the prior lack of clear standard for gender discrimination cases in cases such as Reed v. Reed, 404 U.S. 71 (1971).

10 The four parts of the Central Hudson test, stated in full are as follows: first, does the speech being regulated address lawful activity and is not misleading; second, is the government interest in regulating said speech substantial; third, does the government’s regulation directly advance the government’s claimed interest; and fourth, is the government regulation not more extensive than needed for that interest. If a government regulation passes all four parts of the Central Hudson test, it is presumed to be constitutional, under the First Amendment of the U.S. Constitution.

11 One potential concern with using First Amendment cases in a study such as this one is the potential issue that some First Amendment cases are “weird” and do not feature ideological-based voting configurations. While this is true in some cases, Epstein and Segal (2006) note that ideologically-based vote configurations are at least somewhat common when a case involves solely First Amendment based rights, as opposed to the First Amendment with something else in addition.

12 The Court addressed attorney advertising in Bates v. State Bar of Arizona, 433 U.S. 350 (1977), and pharmaceutical advertising in Virginia State Pharmacy Board v. Virginia Citizens Consumer Council, 425 U.S. 748 (1976).

13 471 U.S. 626 (1985)

14 As one example, several court of appeals decisions in our initial search results used commercial speech as an example of an area where intermediate scrutiny is used when discussing the appropriate standard of review in an unrelated area.

15 While problematic for older state court citations (Kassow, Songer, and Fix 2012), Shepard’s Citations has been shown to be a highly reliable measure of precedent treatments when dealing with federal court opinions (Spriggs and Hansford 2000).

16 According to the coding rules from Shepard’s Citations, Neutral Treatments are those that explain or harmonize a precedent.

17 For our analysis, the IIA assumption would require that a courts’ evaluation of one choice for responding to Central Hudson (e.g. to ignore it) relative to another choice (e.g. just citing it) would be unaffected by the addition of another, irrelevant, category of treatment options.

18 Specifically, prior work has shown that estimates from a multinomial probit can appear to lack robustness due to model fragility that presents via instable coefficient and standard error estimates (see, e.g., Alvarez and Nagler 1998). Kropko (2007) illustrates this through simulations, showing that multinomial logit models tend to provide more accurate results than multinomial probit models in nearly all cases even when IIA assumptions are severely violated.

19 To test our speculation, we reestimated our primary model combining the Drugs/Attorneys variable and the Disclaimer (Post-1985) variable. The results for the alternative model showed that the combined variable has a statistically significant effect on the likelihood of ignoring versus just citing, and all other estimates were unchanged in any meaningful way. Repeating the simulations used to produce the first differences plotted in Figure 3 above, showed a slight increase in the substantive effect from just over 0.15 to nearly 0.2. While this small change is not conclusive evidence that our speculation is correct, it is consistant with what we would expect to see under the condition that our speculation was accurate.