Springer Nature is making Coronavirus research free. View research | View latest news | Sign up for updates

What influences the influence of U.S. Courts of Appeals decisions?


In this exploratory study, we develop models of factors that influence the citation or influence of judicial opinions written by U.S. Courts of Appeals judges. Prior studies of citation patterns in the U.S. Courts of Appeals largely focus on the judge’s career as the unit of analysis. Not surprisingly, this research suggests judge-level factors tend to influence the degree to which judges’ opinions are cited in subsequent decisions. Utilizing a dataset with a random sample of individual cases as the unit of analysis, we compare the effects of judge, panel, and case factors. Overall, while several case-level factors influence the number of citations a case receives, few judge- and panel-level variables affect citation rates. The findings suggest that opinion citation models based on judge- and/or panel-level attributes alone miss the influence of case attributes on citation rates.

This is a preview of subscription content, log in to check access.


  1. 1.

    Scholars have applied different nomenclature to measures of the number of citations to a case and/or judges’ opinions. Some view citation counts as “a measure of judicial quality” (Gulati and Choi 2007: 1302), while others use terms like “influence” (Landes et al. 1998). However, as Klein and Morrisroe (1999: 375) note, this is an “ambiguous” concept. For example, citation rates could be a function of the actual quality of the opinion, the novelty of the issue, the judicial entrepreneurship of the author, or the overall reputation of the author (Klein and Morrisroe 1999: p. 374–75). We use the term “influence” but recognize this limitation and suggest that it is just an indirect, proxy measure.

  2. 2.

    For readers unfamiliar with the U.S. Courts of Appeals, Appendix 1 is a primer describing the structure and role of the court.

  3. 3.

    Generally, U.S. Courts of Appeals cases are decided by panels of three judges randomly selected from all of the judges on the circuit. In a small number of cases, the entire circuit (or in the largest circuit, a subset of 11 judges) rehears the case in what is essentially an appeal of the decision of the three judge panel (Bowie et al. 2014).

  4. 4.

    We excluded en banc cases due to their idiosyncrasies. This also effectively controls for panel size since all non-en banc panels consist of three judges.

  5. 5.

    We know from studies of the U.S. Supreme Court that citations to precedent depreciate over time (e.g., Black and Spriggs 2013).

  6. 6.

    We used a 10-year period because it allowed us to identify the citation counts for the most recent cases in our dataset. In an unreported supplemental analysis, we estimated the same model using the citations in the subsequent 5-year period. The results were fairly similar. In the model of external citations, only two of the substantive variables were not significant using one period and significant using the other time period—and in one instance, the variable would be significant with a one-tailed hypothesis test. The internal citations models were slightly less consistent. Four substantive variables were significant for one-time period but not significant for the other period, though two of variables that were not significant had p values of 0.051 and 0.053, respectively (two-tailed). The similarities are not surprising given the 0.72 and 0.96 correlations between the 5- and 10-year period external and internal citation count measures, respectively.

  7. 7.

    For a description of WestlawNext KeyCite, see (last accessed June 23, 2015).

  8. 8.

    We logged age because of evidence of both outliers and positive skewness, and theoretical concerns regarding diminished marginal effects. For the same reasons, we logged several other independent variables (though we do not discuss this below).

  9. 9.

    Note that, based on Bhattacharya and Smyth (2001a), we also estimated the models using an untransformed age along with a squared term to test for a parabolic relationship. In both models, neither term was significant.

  10. 10.

    Retired Supreme Court justices and district court judges from that circuit may also sit as designated judges. However, none of the judges in our sample were former justices, and the designated judges almost never receive the assignment to write the majority opinion.

  11. 11.

    The measure is static since it is based on the policy preferences of the key actors involved in the appointment of the judges—the President and the Senators from the state in which the vacancy occurred (see Appendix 1 for a discussion of the selection process). The constructs are generated from the Poole and Rosenthal common-space NOMINATE scores of the appointing President and the judge’s home state Senators from the President’s party (which can be found at Additionally, the measure implicitly assumes those involved in selecting the judge prefer to appoint judges with similar preferences, and they can accurately predict the nominees’ preference. There is reason to question the latter assumption (Szmer and Songer 2005). The measure also treats ideology as one dimension, which may not be realistic. These are clear limitations. However, since we only have small samples of judge’s votes we are unable to generate ideal point estimates. As such, the GHP scores are by far the standard measure of judicial ideology for the U.S. Courts of Appeals judges and have been used in articles published in top political science journals like the American Journal of Political Science (e.g., Beim et al. 2016) and the Journal of Politics (e.g., Zorn and Bowie 2010); top law and courts journals like the Journal of Law, Economics and Organization (Choi et al. 2012) and the Journal of Law and Courts (Szmer et al. 2016); and the most recent books on the U.S. Courts of Appeals (e.g., Bowie et al. 2014; Epstein et al. 2013; Haire and Moyer 2015).

  12. 12.

    For readers unfamiliar with the U.S. judicial system, the U.S. President nominates federal judges, along with the advice and consent of the Senate. In practice, this means the Senate votes to confirm nominees. During this process, the ABA, the primary national professional associate for U.S. lawyers, rates the qualifications of potential nominees.

  13. 13.

    The measure is the natural log of the total number of opinions in the Multi User Database authored by the judge in the previous 2 months, adjusted by circuit caseload, and then standardized by circuit mean.

  14. 14.

    In our previous study (Christensen and Szmer 2012), we measured disposition time using the days to decision since the final brief was filed. In this study, due to data availability, we measure disposition using the days to decision since the case is docketed. We nevertheless observe that these are highly correlated measures of deliberation (Pearson correlation coefficient is over 0.80 for the years in which we have both measures).

  15. 15.

    The majority of District Court opinions are not published in a reporter.

  16. 16.

    The test statistic is two times the difference of the log likelihoods of the negative binomial and Poisson regressions (Long and Freese 2014).

  17. 17.

    The presumed preeminence of the D.C. Circuit is reflected by the backgrounds of the U.S. Supreme Court justices. Presently, three of the nine justices served on the D.C. Circuit, as did the recently deceased Justice Antonin Scalia. By comparison, none of the other sitting justices on the Court served on the same court of appeals.

  18. 18.

    We used the cutoff defined in Fox (1991): Cooks’ distance divided by degrees of freedom.

  19. 19.

    Specifically, disposition time was still significant, and most of the significant variables were either case or opinion author measures. There were some changes with respect to specific variables (e.g., an additional case variable was significant; one panel level variable was significant and another was no longer significant; and one opinion writer variable was only significant at the 0.063 level) but they did not change the conclusions at all. Given that, we chose to present the model with the outliers.

  20. 20.

    At present it is not feasible to measure clerk traits because judges typically have three clerks. This makes clerk contribution/authorship difficult to identify.

  21. 21.

    Presently defined as a dispute over at least $75,000 U.S. dollars.

  22. 22.

    This was intended to minimize the potential favoritism toward an in-state litigant.

  23. 23.

    The Courts of Appeals can also hear certain types of appeals from specialized trial courts, as well as habeas corpus cases drawn from the state legal system.

  24. 24.

    A thirteenth circuit, the Federal Circuit Court of Appeals, has geographic jurisdiction over the entire country but a much narrower subject matter jurisdiction.


  1. Allison, P. (2009). Fixed effects regression models. Thousand Oaks, CA: Sage.

  2. Allison, P., & Waterman, R. (2002). Fixed-effects negative binomial regression models. Sociological Methodology,32(1), 247–265.

  3. Beim, D., Hirsch, A., & Kastellec, J. (2016). Signaling and counter-signaling in the judicial hierarchy: An empirical analysis of en banc review. American Journal of Political Science,60(2), 490–508.

  4. Berger, J. A., & George, T. E. (2005). Judicial entrepreneurs on the US Courts of Appeals: A citation analysis of judicial influence. Vanderbilt Law and Economics Research Paper (05-24).

  5. Bhattacharya, M., & Smyth, R. (2001a). Aging and productivity among judges: Some empirical evidence from the High Court of Australia. Australian Economic Papers,30(2), 199–212.

  6. Bhattacharya, M., & Smyth, R. (2001b). The determinants of judicial prestige and influence: Some empirical evidence from the High Court of Australia. Journal of Legal Studies,30(1), 223–252.

  7. Binder, S. A., & Maltzman, F. (2009). Advice and dissent: The struggle to shape the federal judiciary. Washington, DC: Brookings.

  8. Black, R. C., & Spriggs, J. F., II. (2013). The citation and depreciation of U.S. Supreme Court precedent. Journal of Empirical Legal Studies,10(2), 325–358.

  9. Boyea, B. D. (2007). Linking judicial selection to consensus: An analysis of ideological diversity. American Politics Research, 35, 643–670.

  10. Bowie, J. B., Songer, D. R., & Szmer, J. (2014). The view from the bench and chambers: Examining judicial process and decision making on the U.S. Courts of Appeals. Charlottesville, VA: University of Virginia Press.

  11. Burger, W. (1972). Report on the problems of the judiciary. Supreme Court Supplement,93, 3.

  12. Cameron, C., & Trivedi, P. (2009). Regression analysis of count data. New York, NY: Cambridge University Press.

  13. Choi, S. J., & Gulati, M. (2006). Ranking judges according to citation bias (as a means to reduce bias). Notre Dame Law Review,82(3), 1279–1309.

  14. Choi, S. J., & Gulati, G. M. (2008). Bias in judicial citations: A window into the behavior of judges? The Journal of Legal Studies,37(1), 87–130.

  15. Choi, S. J., Gulati, M., & Posner, E. A. (2012). What do federal district court judges want? An analysis of publications, citations, and reversals. Journal of Law Economics and Organization,28(3), 518–549.

  16. Choi, S. J., Gulati, M., & Posner, E. A. (2013). How well do measures of judicial ability predict judicial performance? International Review of Law and Economics,33(1), 37–53.

  17. Christensen, R. K., & Szmer, J. (2012). Examining the efficiency of the U.S. Courts of Appeals: Pathologies and prescriptions. International Review of Law and Economics,32(1), 30–37.

  18. Cohen, J. M. (2002). Inside appellate courts: The impact of court organization on judicial decision making in the United States Courts of Appeals. Ann Arbor, MI: University of Michigan Press.

  19. Corley, P. C., Collins, P. M., Jr., & Calvin, B. (2011). Lower court influence on U.S. Supreme Court opinion content. Journal of Politics,73(1), 31–44.

  20. Cross, F. B., Spriggs, J. F. I. I., Johnson, T. R., & Wahlbeck, P. J. (2010). Citations in the U.S. Supreme Court: An empirical study of their use and significance. University of Illinois Law Review,2010(2), 489–575.

  21. Dimitrova-Grajzl, V., Grajzl, P., Slavov, A., & Zajc, K. (2016). Courts in a transition economy: Case disposition and the quantity–quality tradeoff in Bulgaria. Economic Systems,40(1), 18–38.

  22. Epstein, L., Landes, W., & Posner, R. (2013). The behavior of federal judges: A theoretical and empirical study of rational choice. Cambridge, MA: Harvard University Press.

  23. Federal Judicial Center. (2012). Federal court cases: Integrated data base, 1970–2000; 2001; 2002 [Computer file]. Conducted by the Federal Judicial Center. (Study #8429). ICPSR Version. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [producer and distributor].

  24. Fowler, J. H., Johnson, T. R., Spriggs, J. F. I. I., Jeon, S., & Wahlbeck, P. J. (2007). Network analysis and the law: Measuring the legal importance of precedents at the U.S. Supreme Court. Political Analysis,15(4), 324–346.

  25. Fox, J. (1991). Regression diagnostics. Thousand Oaks, CA: Sage.

  26. Fronk, C. R. (2010). Cost of judicial citation: An empirical investigation of citation practices in the federal appellate courts. Journal of Law, Technology, & Policy,2010(1), 51–90.

  27. Giles, M. W., Hettinger, V., & Peppers, T. (2001). Picking federal judges: A note on policy and partisan selection agendas. Political Research Quarterly,54(3), 623–641.

  28. Grossman, J. B. (1965). Lawyers and judges: The ABA and the politics of judicial Selection. New York: John Wiley.

  29. Gryski, G. S., Zuk, G., & Goldman, S. (2004). A multi-user database on the attributes of U.S. Courts of Appeals judges, 1789–2004. NSF Grant # SBR-9810564.

  30. Gulati, M., & Choi, S. J. (2007). Ranking Judges According to Citation Bias. Notre Dame Law Review, 82, 1279.

  31. Haire, S., & Moyer, L. (2015). Diversity matters: Judicial policy making in the U.S. Courts of Appeals. Charlottesville, VA: University of Virginia Press.

  32. Haire, S. B., & Moyer, L. P. (2008). Advocacy through briefs in the US Courts of Appeals. Southern Illinois University Law Journal,32, 593–609.

  33. Hettinger, V. A., Lindquist, S. A., & Martinek, W. L. (2006). Judging on a collegial court: Influences on federal appellate decision-making. Charlottesville, VA: University of Virginia Press.

  34. Hume, R. J. (2009). The impact of judicial opinion language on the transmission of federal circuit precedents. Law and Society Review,43(1), 127–149.

  35. Hurwitz, M. S., & Kuersten, A. K. (2012). Changes in the circuits: Exploring the courts of appeals databases and the federal appellate courts. Judicature,96(1), 23–34.

  36. Judicial Caseload Indicators—Federal Judicial Caseload Statistics. (2014). Accessed 10 June 2015.

  37. Klein, D., & Morrisroe, D. (1999). The prestige and influence of individual judges on the US Courts of Appeals. The Journal of Legal Studies,28(2), 371–391.

  38. Kuersten, A. K., & Haire, S. B. (2007). Update to the United States Courts of Appeals Database. Created by a Grant from the National Science Foundation Grant (SES-0318359).

  39. Landes, W. M., Lessig, L., & Solimine, M. E. (1998). Judicial influence: A citation analysis of federal courts of appeals judges. The Journal of Legal Studies,27(2), 271–332.

  40. Landes, W. M., & Posner, R. A. (1976). Legal precedent: A theoretical and empirical analysis. The Journal of Law and Economics,19(2), 249–307.

  41. Lindquist, S. A. (2007). Bureaucratization and balkanization: The origins and effects of decision-making norms in the federal appellate courts. University of Richmond Law Review,41, 659–706.

  42. Long, J. S., & Freese, J. (2014). Regression models for categorical dependent variables using Stata. College Station, TX: Stata Press.

  43. Maltzman, F., Spriggs, J. F., & Wahlbeck, P. J. (2000). Crafting law on the Supreme Court: The collegial game. Cambridge, UK: Cambridge University Press.

  44. Melcarne, A., & Ramello, G. B. (2015). Judicial independence, judges’ incentives and efficiency. Review of Law & Economics, 11(2), 149–169.

  45. Pelled, L. H., Eisenhardt, K. M., & Xin, K. R. (1999). Exploring the black box: As analysis of work group diversity, conflict, and performance. Administrative Science Quarterly,44(1), 1–28.

  46. Posner, R. A. (1985). The federal courts: Crisis and reform. Cambridge, MA: Harvard University Press.

  47. Slotnick, E. E. (1983). Federal trial and appellate Judges: How do they differ? Western Political Quarterly,36(4), 570–578.

  48. Solberg, R. S., Emrey, J. A., & Haire, S. B. (2006). Inter-court dynamics and the development of legal policy: Citation patterns in the decisions of the US Courts of Appeals. Policy Studies Journal,34(2), 277–293.

  49. Songer, D. R. (1996). The United States Courts of Appeals data base: A multi user data base created by a grant from the National Science Foundation (SES-8912678).

  50. Songer, D. R., Johnson, S. W., Ostberg, C. L., & Wetstein, M. E. (2012). Law, ideology, and collegiality: Judicial behavior in the Supreme Court of Canada. Montreal: McGill-Queen’s University Press.

  51. Songer, D. R., Sheehan, R. S., & Haire, S. B. (2000). Continuity and change on the United States Courts of Appeals. Ann Arbor: University of Michigan Press.

  52. Staats, J., Bowler, S., & Hiskey, J. (2005). Measuring judicial performance in Latin America. Latin American Politics and Societies,47(4), 77–106.

  53. Steigerwalt, A. (2010). Battle over the bench: Senators, interest groups, and lower court confirmations. Charlottesville, VA: University of Virginia Press.

  54. Szmer, J., & Songer, D. R. (2005). The effects of information on the accuracy of presidential assessments of Supreme Court nominee preferences. Political Research Quarterly,58(1), 151–160.

  55. Szmer, J., Songer, D. R., & Bowie, J. B. (2016). Party capability and the US Courts of Appeals: Understanding why the haves win. Journal of Law and Courts,4(1), 65–102.

  56. Voigt, S., & El Bialy, N. (2016). Identifying the determinants of judicial performance: Taxpayers’ money well spent? European Journal of Law and Economics,41(2), 282–319.

  57. Webber, S. S., & Donahue, L. M. (2001). Impact of highly and less job-related diversity on work group cohesion and performance: A meta-analysis. Journal of Management,27(2), 141–162.

  58. Zorn, C., & Bowie, J. (2010). Ideological influences on decision making in the federal judicial hierarchy: An empirical assessment. Journal of Politics,72(4), 1212–1221.

Download references


Christensen’s role in this research was supported by a National Research Foundation of Korea Grant from the Korean Government [NRF-2017S1A3A2065838].

Author information

Correspondence to John Szmer.


Appendix 1: U.S. Court of Appeals primer


Like most former British colonies, the U.S. has a primarily common law judicial system. Within that framework, judicial power is divided between the national and state governments. The U.S. Constitution created a single national court of last resort, the Supreme Court of the United States. It also empowered the U.S. Congress to create lower courts. While there are some limited subject matter courts designed to handle complicated specialties like tax, trade, and patent litigation, most federal cases are processed through a three-tiered hierarchy with general jurisdiction over “federal questions”—disputes that arise under color of U.S. national law. Additionally, the general jurisdiction courts may hear so-called “diversity jurisdiction” cases: importantFootnote 21 issues of state civil law involving opposing parties that reside in different states.Footnote 22

The U.S. District Courts are the courts of first instance. Each state (and some territories, like Puerto Rico) has at least one District Court, and no District Court crosses state boundaries. Appeals from District Court cases are heard by the U.S. Courts of Appeals.Footnote 23 The Courts of Appeals are divided into 12 geographic circuits.Footnote 24 The circuits are not divided equally. Tremendous variation exists with respect to the geographic scope of the circuit, caseload, number of sitting judges, and backgrounds of the judges (see Bowie et al. 2014; Haire and Moyer 2008).

Each circuit has jurisdiction over appeals from the District Courts within its boundaries. The First through Eleventh Circuits include the 50 states and territories, while the District of Columbia, home of the U.S. Capital, Washington, has its own Circuit. In general, the losing party at the District Court has a statutory right to appeal to the appropriate Court of Appeals. Similarly, after an adverse decision by the intermediate appellate court, litigants may appeal to the U.S. Supreme Court. In most instances, however, the litigant must petition the Court to hear the case. While the Court processes almost 10,000 petitions per term (October–June), fewer than 100 petitions are granted.


Judges serving on all three levels are first nominated by the President, and then ideally confirmed by a majority vote in the Senate. In practice, a single Senator from the state in which the vacancy occurred can block a lower court nomination. As such, following a norm often labelled “senatorial courtesy”, the President typically consults with the home-state Senators during the selection process (Steigerwalt 2010). While the confirmation of Supreme Court nominees has been contentious at different points in time, nominations at all levels are now routinely politicized. Once confirmed, however, federal judges serve for life. The U.S. does not have a mandatory retirement age, and only a handful of judges, almost all from the U.S. District Courts, have been removed from office for misconduct.

The politicization of the selection process routinely leaves more than a dozen of the 167 seats open on the 12 regional circuit courts of appeals (Bowie et al. 2014). Combined with rising caseloads, the appeals courts have turned to creative mechanisms for staffing the benches. Some circuits have to rely heavily on judges selected from pools of retired judges from the circuit or the Supreme Court, visiting judges from other circuits, and even district court judges.


Federal Rules of Appellate Procedure guarantee that the Courts of Appeals process cases fairly uniformly. However, within that broad framework, differences in formal local circuit rules, standard operating procedures, norms, leadership, and composition ensure significant variations in procedure and behavior (Bowie et al. 2014). With that in mind, we can make some generalizations about circuit court procedures. The lower court loser, or appellant, initially files a notice of appeal and then a written brief typically arguing that the District Court judge made one or more legally incorrect decisions at trial. The lower court victor, or appellee, then files its own legal brief.

At this point, most circuits have staff attorneys process the cases. In some instances, to help manage the caseload, the staff will recommend an expedited process. The cases are then assigned randomly to three judge panels. Each panel preprocesses the cases to determine which will require oral arguments. The judges then meet for a fixed period of time (maybe a week or two) to hear oral arguments and confer. During the post-oral argument conferences, they vote on the outcome of the case. Between strong norms of collegiality and consensus, combined with a docket often full of easy cases, the panel vote is usually unanimous. The most senior active judge on the panel in the majority then assigns one judge to write the opinion of the court. After all the cases assigned to the panel have been decided, the judges will return to their individual offices which are often dispersed across the circuit.

While a single judge is typically assigned to write the opinion of the court, the final product is influenced by multiple actors. First, while it varies for different judges, many rely on law clerks (typically recent law school graduates serving a 1-year term) to write the first draft (Bowie et al. 2014). Second, the panel continues to interact. The opinion author circulates the drafts to the other panelists and frequently incorporates their edits into the final product.

The final decision of the court is then published in an official reporter or the appendix to the reporter. This is a recent phenomenon. In the 1960s, faced with rising caseloads, the circuits began to distinguish between opinions with precedential value that had to be published and those without precedential value. The latter were distributed to the parties and maintained by the circuit, but they were not published and distributed to law libraries. Now cases deemed to have precedential value are published in the Federal Reporter, while the remaining cases, often decided by a short per curiam (unsigned) opinion, are published in the Federal Appendix. The latter are often called “unpublished” opinions, the correct nomenclature is “unreported”. Until 2007 most circuits prohibited citations to unreported cases.

After the case is decided, the losing party has at least three alternative routes to appeal the decision, though none of the appeals are guaranteed. They may petition the panel to rehear the case, petition the circuit to hear the case en banc (all the judges on the circuit meet to decide the case), or they can petition the Supreme Court. All three types of petitions are rarely granted, ensuring that the panel decision usually stands.

Appendix 2

See Table 4.

Table 4 Variance inflation factors greater than 3

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Szmer, J., Christensen, R.K. & Grubbs, S. What influences the influence of U.S. Courts of Appeals decisions?. Eur J Law Econ 49, 55–81 (2020).

Download citation


  • Judicial influence
  • Appellate courts
  • Citation rate
  • Case precedent

JEL Classification

  • K40
  • K41