Authors: Banks Miller and Brett Curry
Publication: Social Science Quarterly 105: 1193-1204
Original source: ssq_judicial_wealth_decision_making_2024.pdf
Conversion note: this Markdown was generated by text extraction from the hosted file. Review equations, tables, notes, and citations before treating it as a final web copy.
Received: 4 December 2023 Revised: 11 June 2024 Accepted: 18 June 2024 DOI: 10.1111/ssqu.13412 ORIGINAL ARTICLE Exploring the influence of wealth on judicial decision making Banks Miller1 Brett Curry 2 1Political Science Program, School of Economic, Political and Policy Sciences, University of Texas at Dallas, Richardson, Richardson, Texas, USA 2Department of Political Science & International Studies, Georgia Southern University, Statesboro, Georgia, USA Correspondence Brett Curry, Department of Political Science & International Studies, Georgia Southern University, Statesboro, GA 30460, USA. Email: bcurry@georgiasouthern.edu
Objective: Attention to levels of socioeconomic inequality has given rise to work investigating its potential to influence political actors and, thus, policy outcomes. These studies have focused on elected actors in legislative contexts. Ours fills this gap by assessing the consequences of a judge’s wealth on decision making. Methods: We obtain data on federal appellate judges’ wealth and explore its influence in cases involving economic issues. We estimate logit models controlling for other factors and include fixed effects for circuit, specific issue area, and year of the decision. The models consider both the main effects of wealth and its potential interaction with judicial ideology to influence decision making. Results: We find no direct relationship between a judge’s wealth and her decision making in economic cases. However, wealth interacts with ideology to exert a significant influence on decision making; here, greater wealth amplifies the ideological predilections of more conservative judges. Conclusions: This is the first study to document a link between wealth and the choices judges make on the bench. It adds a new dimension to the ways economic inequality can influence political actors and raises new questions about themechanisms by which such inequality is consequential for elite political behavior. KEYWORDS courts of appeals, judicial decision making, judicial wealth An increasing focus on economic inequality in wealthy democracies has brought renewed attention to the potential biases in political decision making it can produce. One important aspect of this attention, often referred to as “vertical political inequality” (e.g., Abizadeh 2021) or “representational inequality” (Carnes and Lupu 2023, 258), considers the extent to which socioeconomic “difference[s] between decisionmakers and citizens” (Elsässer and Schäfer 2023, 470) are impactful. Not only do these differences exist Social Science Quarterly. 2024;1193–1204. wileyonlinelibrary.com/journal/ssqu © 2024 Southwestern Social Science Association. 1193
1194 MILLER AND CURRY between decisionmakers and citizens, but they also exist within the ranks of the political class. 1 As Stacy (2020, 2) observes, “wealth is relative and…[e]ven if they do not apply to most other individuals, the differences that distinguish millionaires from billionaires, or the wealthy from the less-wealthy, are potentially quite meaningful.”2 Yet, with important exceptions from the legislative context, knowledge about the relevance of individual affluence to the behavior of elite political actors has been scant. In this article, we rely on findings from the legislative context as well as social background theory— a paradigm that has been employed extensively in judicial scholarship (see George and Weaver 2017) and in making sense of political action more generally (e.g., Burden 2007; Carnes 2013)—to make a somewhat exploratory but novel contribution to the literature. 3 We assess what, if any, consequences individual-level variation in personal wealth has for decision making by political actors who are neither electorally accountable nor dependent upon campaign contributions nor subject to traditional lobbying: judges on the U.S. Courts of Appeals. We utilize originally collected data on noncriminal cases involving economic issues from 2013 to 2017 and treat the ideological direction of a judge’s decision in a case as the dependent variable. We restrict our focus to economic matters because prior work suggests that social background variables are likely relevant to judicial decision making only in situations where the legal issues at hand are salient to the characteristic under examination (e.g., Boyd, Epstein, and Martin 2010).4 In the next section, we briefly summarize key findings from existing studies of how the socioeconomic status of government officials has been found to influence policy making; we necessarily focus on the legislative environment, as this has been the key area for such investigations, and then consider the potential implications of these findings for judicial behavior. Though extant work has succeeded in establishing the existence of relationships between socioeconomic status and governmental decision making, it has been less successful at settling on the precise drivers of those relationships (e.g., Elkjaer and Klitgaard 2021). As Elsässer and Schäfer (2023, 481) put it, “[t]he exact mechanisms of how different kinds of social inequality translate into political inequality are not well-understood.” As such, our chief aim is to explore whether variations in judicial wealth have implications for decision making. Though we consider several possible drivers of the patterns we observe in the concluding section, more thorough attempts to adjudicate the specific mechanisms at play must await additional investigation. After discussing these studies, we describe our rationale for relying on wealth as our key explanatory variable in this study of judicial decision making. Next, we describe the process by which we gathered that data on judicial wealth and assembled and coded the economic cases we analyzed. Here we explicate our variables and set forth our hypotheses. We then present and assess our empirical results and conclude by touching on opportunities for future research. EXISTING STUDIES AND THEIR IMPLICATIONS FOR JUDICIAL BEHAVIOR There is little debate that American political institutions are and always have been populated to a significant degree by the “haves” of society (e.g., Levinthal, Beachum, and Levine 2017; Carnes 2012,6 – 7 ) . Nevertheless, it was just a decade ago that one scholar lamented that “political scientists simply don’t 1 For reasons we discuss, we focus here on personal wealth. Based on the data we have gathered on federal judges and describe herein, the median net worth for federal appellate judges is approximately $2 million. This is roughly twice the average net worth for members of Congress. https://www . opensecrets.org/news/2020/04/majority-of-lawmakers-millionaires/ 2 Scholars have uncovered important variations that relate to attitudinal differences even within the highest echelons of the wealth distribution (e .g., Klontz et al. 2015; Suhay, Klašnja, and Rivero 2021). 3 As we make clear shortly, the crux of social background theory is that a judge’s “attributes and experiences” can impact her decision making over and above attitudinal considerations; the construct on which we focus—personal wealth—is a key socioeconomic “attribute,” and such considerations ar e well-grounded within social background theory (see George and Weaver 2017, 286–288; Mau ˇcec and Dothan 2022). 4 Indeed, a recent review notes the importance of investigating judges in this way to ascertain how well the developing consensus that “the economic backgrounds of politicians can matter for economic policy … travels across institutional contexts” (Carnes and Lupu 2023, 264). 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
WEALTH AND JUDICIAL DECISION MAKING 1195 know much about how inequalities in the class composition of American political institutions affect policy making in the United States” (Carnes 2013, 9). Many studies in this area—all of which have focused on elected legislators—have considered just one measure of socioeconomic status at a time, which has resulted in the collective analysis of “a broad range of economic characteristics” (Carnes and Lupu 2023, 254). We summarize key findings from studies utilizing a variety of socioeconomic measures here; we consider their implications for this study and discuss our own decision to measure the socioeconomic status of federal judges by capturing their net worth in the next section. Legislative studies Studies examining the composition of legislatures have shown that the overrepresentation of the upper class in those institutions tends to bias outcomes toward elite interests: white-collar lawmakers are less supportive of the social safety net and progressive taxation, and they tend to see economic issues through a “more conservative lens” than do their working-class counterparts (Carnes 2013, 12–21). Investigations into the relationship between legislators’ wealth and certain policy perspectives have echoed this sentiment. Griffin and Anewalt-Rensburg ( 2013) conclude that, all else being equal, wealthier legislators were more supportive of repealing the estate tax than their less affluent colleagues. Kraus and Callaghan ( 2014)fi n d that wealthy members of the House of Representatives cast more conservative votes than do less wealthy members, and a study of Swiss politicians also notes a correlation between higher income and conservatism on economic issues (Wuest and Rosset 2019). Recent work by Stacy (2020) adds to the narrative on economic inequality’s relationship to policy outcomes; relying on legislative effectiveness scores, he finds the wealthiest members of the House are far more effective in advancing their policy agendas than their counterparts. More broadly, our study draws on what Burden ( 2007) calls the “personal roots” of legislative representation. In key respects, his framework coincides with what law and court scholarship refers to as social background theory, which we discuss shortly. Burden (2007, 15) characterizes this perspective as recognizing the “direct substantive consequences of descriptive traits.” Not only does it encompass what he refers to as “traditional categories of interest” such as race and sex; but it also extends to “[l]ess visible traits such as religion, sexual orientation, and wealth…” (Burden 2007, 140). Economic issues Notwithstanding some exceptions, existing work on the relationship between the socioeconomic status of governmental actors and policy outcomes has tended to examine economic policy issues. This is sensible for several reasons. First, economic matters represent the “set of issues for which social class seems to matter the most in opinion studies” (Carnes 2012, 11; see also Hout 2008). Further, economic issues are of heightened salience to the affluent, which should strengthen the relationship between their policy attitudes and attributions (Suhay, Klašnja, and Rivero 2021). One important study underscores this point by finding the socioeconomic differences between the positions legislators take are larger on security and economic issues as opposed to matters of social policy (Lax, Phillips, and Zelizer 2019). This is largely consistent with the thrust of social background theory as scholars of judicial behavior have understood it—individual, nonideological traits or characteristics are only likely to be consequential for behavior in areas that tap policy issues adjacent to those characteristics (e.g., Miller and Curry 2023; George and Weaver 2017; Boyd, Epstein, and Martin 2010). As noted, speculation about the mechanisms that may underlie the relationships legislative scholars have uncovered between socioeconomic strata and economic policy positions is more muddied. Accounts in the United States have pointed to economic self-interest (e.g., Griffin and Anewalt-Rensburg 2013) or the influence of campaign contributions and donor class preferences (e.g., Gilens 2015) as likely drivers of these patterns; however, important studies outside the United States have validated the robustness of this relationship between 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1196 MILLER AND CURRY representational inequality and policy responsiveness in contexts where such factors make less theoretical sense. For instance, Elsässer, Hense, and Schäfer (2021) examine Germany, a far more egalitarian society with a very different campaign financing system from the United States. Even so, they conclude socioeconomic bias “in the makeup of parliaments matters for substantive representation” (Elsässer, Hense, and Schäfer, 2021, 1890). Similar findings have emerged elsewhere beyond the United States (e.g., Hemingway 2022; Wuest and Rosset 2019; Courtney 2015). To summarize, “across different economic measures and different national contexts, politicians from different economic backgrounds consistently appear to think differently about economic issues, potentially contributing to unequal congruence” (Carnes and Lupu 2023, 259). Such results represent important replications of the basic empirical findings by studies of American politics; however, the similarity of those results in different democratic political contexts has complicated efforts to settle on some sort of universal mechanism(s) that could be driving them (e.g., Elkjaer and Klitgaard 2021, 5). We build on this work and explore a different group of political actors in a context that has gone unexamined by scholars investigating socioeconomic differences and their relevance to the behavior of political actors. Considering the influence of ideology Before turning to our measurement and data collection strategies, it is important to touch on studies that have examined the socioeconomic status of political actors in conjunction with their ideological predilections or partisan affiliation (e.g., Lax, Phillips, and Zelizer 2019; Maks-Solomon and Ribgy 2020). Simply put, even when income groups have opposing preferences, “representational disparities … are better explained by partisanship than affluence” (Elkjaer and Klitgaard 2021, 1; see also Branham, Soroka, and Wlezien 2017; Brunner, Ross, and Washington 2013; Enns 2015). Burden ( 2007, 16, 39) also makes this general point; he notes that, while ideology represents just one factor that comprises legislative preferences, personal experiences or characteristics should not necessarily be expected to trump ideology’s influence across the board. Lax, Phillips, and Zelizer’s ( 2019) work, which examines economic and partisan biases in the behavior of government officials, is especially instructive. These authors focus on instances where U.S. senators can either vote in a way that is consistent with partisanship or consistent with affluent interests, but not both simultaneously. They, in effect, test which impulse predominates when “party and purse” pull in opposite directions and find senators—both Republican and Democratic—are much more responsive to partisan positions than ones affiliated with affluence; as they summarize it, “partisanship induces, shapes, and constrains affluent influence” (Lax, Phillips, and Zelizer 2019, 917). Here again, such conclusions mesh well with the ways judicial studies have conceived of social background theory—the theory considers relevant attributes or background characteristics apart from ideology as explanations for judicial choice but simultaneously recognizes the additional influence of policy preferences (see George and Weaver 2017, 286; Miller and Curry 2023; Boyd, Epstein, and Martin 2010; Szmer, Christensen, and Kaheny 2015). Scholarship on representational inequality seems to bear out this possibility, even as it disagrees as to which side of the ideological or partisan spectrum is more heavily influenced by socioeconomic status. Most studies of elected politicians suggest Democrats and ideological liberals are susceptible to engaging in socioeconomic class-based legislative voting on economic issues while their more conservative Republican colleagues are not (Kraus and Callaghan 2014; Grumbach 2015; Carnes 2013). These studies find Republicans favor legislation that increases income inequality regardless of socioeconomic status, but that lower- and upper-status Democrats are divided in this area. On the other hand, Republican voters are divided along socioeconomic lines when it comes to economic issues whereas Democrats in the electorate have traditionally been more unified (Bartels 2018). Maks-Solomon and Rigby ( 2020) identify this cleavage of opinion in their assessment of how voters are represented in the U.S. Senate: “on economic issues, the Republican rich are more conservative than the Republican poor,” but Democratic voters of all socioeconomic stripes are generally in sync on economic issues. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
WEALTH AND JUDICIAL DECISION MAKING 1197 Hypotheses With all this in mind, we consider hypotheses pertaining to the influence of a judge’s wealth on her decision making in economic cases. In light of durable guidance from legislative work across different socioeconomic measures and national contexts (see Carnes and Lupu 2023, 259), Hypothesis 1 makes a directional prediction: Hypothesis 1. All else being equal, wealthier judges will vote in more ideologically conservative ways than their less wealthy counterparts. In our second set of hypotheses, we explore the potential interactive influence of wealth in tandem with ideology. Here, as summarized above, extant work sends different messages. The literature on elected political elites suggests an interactive effect for liberals and Democrats, while conclusions drawn from the broader electorate anticipate wealth exerting a disproportionate impact on the behavior of conservatives and Republicans. In other words, are federal judges akin to elected politicians or do they share more in common with ordinary voters? Like members of Congress, federal judges are certainly elite political actors. But there are also potentially important contextual differences between them. They do not stand for election or have constituent pressures, which eliminates other constraints including the need to prioritize the interests of key partisans who comprise the donor class. They are not susceptible to traditional lobbying attempts, and they engage in retrospective adjudication rather than formulate prospective, generally applicable rules (e.g., Posner 2008, 81–2). And, like other tenured workers, “effort aversion…play[s] an unusually large role in the utility functions of judges” (see Epstein, Landes, and Posner 2013, 31). In sum, because traditional notions of politics are of limited consequence to federal appellate judges who hold office in perpetuity, “its impact on their conceptions of conservative and liberal positions is more limited than that impact in some other segments of political elites” (Baum 2017, 18). As such, Baum (2017) views affect toward sociopolitical groups rather than values as the key linkage between issues and ideology. We find arguments about the interactivity between socioeconomic status and ideology compelling. However, since existing scholarship leaves us conflicted about directionality—principally because it is unclear to us whether we should anticipate federal judges to be more similar to elected politicians or ordinary voters—we offer two distinct and oppositely directed predictions about the influence of wealth on decision making in combination with judicial ideology: Hypothesis 2a. Among ideologically conservative judges, those with higher levels of personal wealth will vote more conservatively compared to their ideologically conservative colleagues with lower levels of personal wealth. Hypothesis 2b. Among ideologically liberal judges, those with higher levels of personal wealth will vote more conservatively compared to their ideologically liberal colleagues with lower levels of personal wealth. DATA AND MEASUREMENT “There is no one right way to measure the economic backgrounds of politicians,” because the “right measure” depends on factors including the population being analyzed and what is being tested (Carnes and Lupu 2023, 255). Indeed, political studies have captured socioeconomic status with a variety of measures, including but not limited to personal wealth (e.g., Griffin and Anewalt-Remsburg 2013), education (e.g., Borthwick et al. 1991), or occupational status (e.g., Carnes 2012). We capture the socioeconomic status of federal judges on the U.S. Courts of Appeals with a measure of personal wealth for several reasons. First, traits such as occupation or education simply do not vary among federal judges. All have 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1198 MILLER AND CURRY
Judge 2013 Wealth 2017 Wealth Percent change Judge A $3,293,215.90 $5,261,986.78 59.8% Judge B $1,719,159.00 $2,892,710.00 68.3% Judge C $4,016,661.00 $3,258,778.00 −18.9% attended law school and are lawyers of some sort. 5 Second, there is no consistent, reliable, or comprehensive way to capture other potential proxies for an appellate judge’s socioeconomic circumstances in early life. We gathered data to create this key independent variable, judicial wealth, from the self-reports judges must submit annually to the Administrative Office of the U.S. Courts (AO). These forms require judges to report all assets—including equity investments, mineral rights, and property holdings with the exception of their primary residence—with their corresponding values. We requested, obtained, and coded these AO reports for all available federal appellate judges, including those on senior status, between 2013 and 2017. The form of this reporting varies slightly from judge to judge—as an example, some disclosures may signify a stock or mutual fund by name and others by ticker symbol. In all cases, though, the value of the holding is reported within a standard range specified on the form; exact values are rarely reported. The online Appendix contains an example of the AO’s annual disclosure form and our approach to coding its contents. For each judge, we aggregated the items to create a measure of total wealth, defined as the net value of reported property and assets at the end of each calendar year. Federal appellate judges are about 16 times as wealthy as the average American household, with a median net worth of about $2 million. 6 There is, however, significant variation among these judges: between 2013 and 2017 the standard deviation for wealth is $1.15 million. Our judicial wealth variable is logged and corrected for skew and inflation. Wealth can vary both across individuals and within them over time; for example, 2017 votes are analyzed with the measure of judicial wealth gleaned from a judge’s disclosure for the 2017 calendar year. 7 Consider three examples in our data, summarized in Table 1. Judge A’s wealth increased by nearly $2 million from 2013. In her case, the holdings were largely the same and the increase was primarily due to the value of several Vanguard mutual funds moving from the $100,001–$250,000 range into the $250,001–$500,000 band. In raw dollars, Judge B’s aggregate wealth grew from $1.7 million to $2.9 million during that time span; this is largely explained by some increases in the number of investments as well as their value. For instance, by 2017 Judge B had inherited an account from Schwab with a variety of holdings. There are also occasional instances where a judge’s net worth declines over the period we examine, and Judge C is a good example of this rare occurrence. His reported net wealth decreased by more than $750,000 over the 5 years we examine, primarily due to the disappearance of some investments altogether and lagging performance in several others. To capture our dependent variable, we collected data on economic cases by executing Westlaw searches of U.S. Courts of Appeals decisions, both published and unpublished, decided from 2013 through 2017. 5 Of course, we can consider more fine-grained distinctions in terms of the caliber of law school a judge attended, whether they attended a private undergraduate institution, the type of legal position they held prior to joining the federal judiciary (e.g., private practice, law school professo r, state judge) or clerked for a Supreme Court justice. We control for these possibilities (see Table 2A in the online Appendix), and doing so does not affect our relationships of interest: wealth, ideology, and decision making. 6 Our wealth measure averages between the high and low ranges contained in the reports. (See online Appendix for an example.) To ensure there are no potential distorting effects of relying on the mean, we reran our models substituting the low and high values for the mean wealth measure; the results were the same as those obtained utilizing the mean. Context is important when assessing perceptions of wealth inequality (e.g., Xu and Garand 2010) and part of that context involves one’s own circumstances and relevant comparison peer groups (see Stacy 2020). 7 As such, and in keeping with work on affluence as it relates to elected officials (e.g., Wuest and Rosset 2019;S t a c y2020), we refrain from making causal claims throughout the article and instead reference observed associations. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
WEALTH AND JUDICIAL DECISION MAKING 1199
Economic issues Percentages of data Securities 39 Bankruptcy 3 Tax 4 Insurance 48 Miscellaneous business 6 After excluding criminal cases, 8 we narrowed our results by focusing on categories involving economic matters identified by Westlaw , including securities, finance, banking, taxation, insurance, and healthcare. We then created a limited number of mutually exclusive categories to capture the main subject matter of each case, to account for the possibility that the influence of judicial wealth might not be uniform across these areas. Table 2 displays the distribution of our data across these categories. Securities and insurance-related cases appeared most frequently and, while we control for subissues using fixed effects in our models, these categories did not influence our results; a model focusing on each of the two major issue areas independently results in conclusions that are substantially similar to those presented here. 9 Since securities- and insurance-related cases appeared most frequently, we offer an example of each to illustrate our coding protocols. First, consider a 2014 securities case from the Third Circuit, Gold v. Ford Motor Company . Here, individual shareholder Gold alleged the Ford Motor Company had violated a provision of the Securities Exchange Act of 1934 by failing to comply with a notice requirement before announcing a distribution of preferred securities. Similarly, in a 2016 case against Progressive in the Sixth Circuit, the estate of a pedestrian killed in a traffic accident alleged the insurer had acted in bad faith and in violation of state law when it conditioned its offer for compensation on an agreement to release and indemnify the insured. Both appellate rulings favored Ford and Progressive, and those outcomes were coded as being ideologically conservative because each favored the party in the dispute with more financial resources.10 Our dependent variable is the individual judge’s vote in a case , which equals 1 when the judge favors what observers would characterize as the “have,” or the party with greater resources in these economically oriented cases—in the parlance of traditional ways of thinking about judicial voting, this is equivalent to an ideologically “conservative” vote (see, e.g., Pacelle, Curry, and Marshall 2011). To illustrate, we coded judges as voting for the “have” (coded 1) when siding with the securities entity being sued by investors; the creditor facing the debtor; the insurer over the insured; the bank or mortgage company over the customer; the health-care company instead of the patient; and so forth. Votes against these interests were coded zero (see also Spaeth et al. 2021, 52–3). Because most decisions in the Courts of Appeals are affirmances (e.g., Songer and Sheehan 1992), we code for the direction of the lower court decision being reviewed using this same scheme—decisions by the district court favoring what we term the “have” are coded 1, with all others coded zero. To measure judicial ideology, in the models that follow we rely on CFscores created by Bonica, which rely on revealed preferences rather than the identities of appointment actors (Bonica and Sen 2017; see also Bonica 2014).11 8 We focus on noncriminal economic cases to ensure social background variables that might be salient in criminal economic matters (e.g., one’s prior service as a prosecutor) do not obscure our focus on judicial wealth. 9 We also modeled the role of wealth in more socially oriented issues. Using previously collected data on Second Amendment cases and on cases involving issues around voting, we find that there is no relationship between wealth, ideology, and conservative voting. 10 Gold v. Ford Motor Co. (2014), 577 Fed. Appx. 120; Shaheen v. Progressive Casualty Insurance Company (2016), 673 Fed. Appx. 481. 11 Bonica’s measure relies on campaign contribution information to create a measure of judicial ideology (Bonica and Sen 2017). Importantly for our purposes, the dollar amounts involved do not factor into the CFscore; Bonica’s scores are generated on the proportion or distribution of giving rather than the amounts contributed. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1200 MILLER AND CURRY
Model 1 Model 2 Model 3 Model 4 Model 5 Coefficient (SE) Coefficient ( SE) Coefficient ( SE) Coefficient ( SE) Coefficient (Clustered SE) Wealth −0.00 (0.05) 0.04 (0.06) 0.08 (0.07) 0.16 (0.07) * 0.16 (0.07)* Ideology 0.23 (0.06) ** −2.37 (1.24) −4.03 (1.42)** −3.73 (1.48)* −3.73 (1.37)** Wealth * Ideo. — 0.17 (0.08) * 0.26 (0.09)** 0.24 (0.09)** 0.24 (0.09)** Controls Dist. Ct. Conserv . — — 2.56 (0.11) ** 2.52 (0.12)** 2.52 (0.18)** Circ. Ideo. — — 0.51 (0.09) ** 0.50 (0.11)** 0.50 (0.14)** Constant 0.89 (0.85) 0.24 (0.91) −2.23 (1.05) −3.96 (1.14) −3.97 (1.14) Circuit FE — — — Yes Yes Year FE — — — Yes Yes Issue FE — — — Yes Yes N 2942 2942 2942 2941 2941 Number of judges 229 229 229 229 229 Number of cases 1143 1143 1143 1143 1143 χ2 19.48 (0.00) 23.92 (0.00) 719.34 (0.00) 865.54 (0.00) 270.24 (0.00) PRE 0.00 0.00 0.32 0.35 0.35 * p < 0.05; ** p < 0.01 (all tests are two-tailed). Higher values on this variable represent greater conservatism and lower values signify more liberal ideological leanings. Using this measure, we construct a measure of the median ideology on the panel on which each judge serves. Again, higher values on this variable signify more conservative three-judge panels and are expected to increase the probability of an ideologically conservative decision. Finally, we include fixed effects for circuit, issue area,a n dyear in our models. We cluster simultaneously on judge and case to control for any sources of nonindependence among the errors. 12 RESULTS We estimate a series of logit models with standard errors simultaneously clustered by case and by judge. The models fit the data reasonably well, and the difference between them is the set of controls and interaction effects that we include. Results are displayed in Table 3. To begin, in Model 1 we include only the two key independent variables: judicial wealth and judicial ideology, omitting all control variables. In this simple model, only ideology is a statistically significant predictor of vote choice. 13 Moving from the 10th percentile of conservatism to the 90th increases the likelihood of a conservative vote by 10 [5, 14] percentage points (95 percent confidence intervals are in brackets throughout). Models 2 through 5 add interactions and control variables. Most importantly, from our perspective, is that the interaction between ideology and wealth is a significant predictor in all these models. 14 This provides strong support for Hypothesis 2a, 12 In the online Appendix, we include a model with additional control variables for gender and minority status. The inclusion of these variables does not alter our conclusions. We also include a model that controls for the change in wealth over the period of our study. The inclusion of this variable does not alter the conclusions we present here. 13 Although not displayed in Table 3, we estimated a model similar to Model 5 without any interaction between wealth and ideology and our results do not differ significantly from those presented in Model 1. Wealth remains an insignificant predictor of voting, while ideology is strongly significant. 14 In Model 2, a χ2 test of the interaction yields a significant result ( χ2 = 23.54 (p < 0.00)). Results of χ2 tests in the other models are similarly significant. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
WEALTH AND JUDICIAL DECISION MAKING 1201
which suggests the influence of wealth on decision making would be limited to more conservative judges. In what follows, we interpret the size of effects using Model 5, which includes a full suite of controls, including circuit-, issue-, and year-level fixed effects and standard errors clustered by case and by judge. 15 Note that including the three fixed effects causes us to lose one observation, hence the differing N from Model 3 to Models 4 and 5. It is easiest to illustrate the conditional effect of wealth by graphing the results of the ideology and wealth interaction, as displayed in Figure 1. Note that in the figure a conservative judge is one at the 90th percentile of our conservatism ideology measure and a liberal judge is one at the 10th percentile. The trend among conservative judges is clear: as wealth increases so too does the likelihood of a conservative vote. Across the range of wealth in our data, the increase in conservative voting among conservative judges is 16 [8, 25] percentage points. The effect becomes distinct from liberal judges (with no overlap among the 95 percent confidence intervals) at $9.3 million, approximately the 80th percentile in our data. There is a slight decrease in the likelihood of a conservative vote among liberal judges as wealth increases, but substantively the decline is slight—from about a 70 percent chance of a conservative vote to about a 66 percent chance. Therefore, any effect of wealth on voting is cabined among the conservative judges in our data. In Model 5, two control variables are statistically significant, both as expected. First, a conservative decision at the district court level implies a high likelihood of a conservative decision at the circuit court, reflecting the tendency of appellate courts to confirm the holding of lower courts. Second, as the other members of a panel at the circuit court level become more conservative so too does the likelihood of a conservative vote. Shifting from a median panel member in the 10th percentile of conservatism to one in the 90th percentile increases the likelihood of an individual judge’s conservative vote by 13 [6, 20] percentage points. In Table 2A in the online Appendix, we display results for two additional models that help to con- firm the robustness of our results. In Model 6, we substitute a simple measure of judicial partisanship (based on the party of the appointing president) for our more sophisticated measure of ideology based on campaign contribution data. In Model 6, there is essentially no difference in our results, with judges appointed by Republicans evincing increasing conservativeness with increasing wealth and essentially no effect of wealth for Democratic judges. In Model 7, we add a series of additional controls to the model 15 We estimated a model similar to Model 5, substituting raw values for the logged values of wealth. As wealth increases, we find that conservative judges are much more likely to vote conservatively as compared to their liberal colleagues. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1202 MILLER AND CURRY based on background characteristics of the judges. These include the rank of law school attended, whether an individual attended a private undergraduate university, service as a Supreme Court clerk, prior service as a judge in another position, and previous work in private practice. None of these variables are statistically significant predictors of judicial behavior nor does their inclusion alter our conclusion about the conditional effect of wealth on conservative judges. DISCUSSION AND CONCLUSION Our aim was to explore whether variations in wealth can influence the decisions of federal appellate judges. Departing from previous studies, we considered wealth’s role in shaping the behavior of nonelected elites and discerned an interactive effect with ideology: as wealth increases among conservative judges, so too does their likelihood of casting conservative votes in economic cases. To recall, we offered no a priori expectation for the direction of this relationship due to conflicting guidance from two strands of existing literature. The result that ultimately emerged parallels findings about the divide on economic issues within the ranks of ordinary voters (e.g., Bartels 2018); it is not consistent with findings about personal wealth’s influence on legislative behavior, where socioeconomic distinctions are evident among liberals and Democrats. At least for present purposes, then, federal judges appear to be more similar to ordinary voters than to elected political elites. Why might this be the case? Though we cannot speak definitively, we can speculate about some possibilities. The most general point, which we raised in framing our hypotheses is probably also the most important—the legislative and judicial contexts are different. The most obvious of those differences center on things like elections, the reality of lifetime terms for federal judges versus the more uncertain tenure of legislators, lobbying, and campaign donations. Nor can we eliminate the possibility that different priorities by recent Republican and Democratic presidents when it comes to judicial nominations, such as Federalist Society membership (Scherer and Miller 2009) or attention to descriptive representation (e.g., Scherer 2023; Hurwitz and Lanier 2012) may also be part of the story. However, our hunch is that the apparent dissimilarity when it comes to the influence of wealth between judges and these other political elites might stem less from these factors than the different types of decisions the two sets of actors make. Legislators make prospective judgments resulting in generally applicable rules; lower court judges make retrospective decisions that, though they can have broader implications, are intended to produce a discrete outcome for particular persons—the immediate parties to the case. As such, judges engage in uniquely “evaluative” behavior while legislators have a different “utility function” (see Posner 1993, 10, 22). Beyond that, and in contrast to legislators, a common legal training teaches judges to approach problems in a particular way (e.g., Schauer 2012). Apart from ways legislators and judges differ, how might judges be similar to voters? Posner (1993)h a s famously analogized the behavior of federal appellate judges to everyday voters who respond rationally to ordinary incentives. He contends that, as is true of voters, “a single decision rarely (not never) has a big impact” (Posner 1993, 18) and the voting calculus of appellate judges and ordinary people share additional similarities (Posner 1993, 18–20). Further still, judges are susceptible to many of the same cognitive forces that characterize ordinary citizens (e.g., Guthrie, Rachlinski, and Wistrich 2000). This last idea, in particular, is consistent with what scholars have concluded about how socioeconomic differences map onto how ordinary voters think. Socioeconomically elite voters tend to be dispassionate and acontextual in their judgments, while those of lower socioeconomic status are more egalitarian, empathetic, and engaged with the needs of others (Piff et al. 2010). Likewise, wealth—for judges and people generally—exposes individuals to different social networks and life experiences than the less affluent. This reality may still contribute to the role we see for wealth in judicial decision making, even if it does not derive from expressly self-interested concerns. This represents one plausible explanation of our result but, in the end, our study is unable to isolate the exact mechanisms at play. Nevertheless, this is the first study to document a link between the wealth of federal judges and the choices they make on the bench. It adds a new dimension to the ways economic inequality can influence political actors—here, by interacting with ideological predilections—and simultaneously raises new questions about the particular mechanisms by which such inequality is consequential for elite political behavior. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
WEALTH AND JUDICIAL DECISION MAKING 1203 Just as it is critical for scholars to assess the avenues by which differences in the socioeconomic status of governmental officials matters for their decision making, future research is also needed to flesh out the extent to which they can influence judicial decision making in other circumstances. One potential route of investigation could be to examine the role of judicial wealth in the decisions of U.S. District Court judges—another group that has no electoral constituency but operates in an environment, that is, unique from judges on the Courts of Appeals. The district court context could provide an opportunity to investigate whether wealth and ideology interact differently with demographic factors like race or gender in shaping judicial decision making. Another avenue for future research is extending lines of inquiry into the state courts, where variations in methods of selection and state-level differences in income inequality could be leveraged. Finally, the availability of financial disclosure information from U.S. Bankruptcy Judges offers the opportunity to probe the implications of such differences in a specialized judicial context. ORCID Brett Curry https://orcid.org/0000-0002-6366-0714
Abizadeh, Arash. 2021. “Representation, Bicameralism, Political Equality, and Sortition: Reconstituting the Second Chamber as a Randomly Selected Assembly.” Perspectives on Politics 19(3): 791–806. Bartels, Larry M. 2018. “Partisanship in the Trump Era.” Journal of Politics 80(4): 1483–94. Baum, Lawrence. 2017. Ideology in the Supreme Court . Princeton, NJ: Princeton University Press. Bonica, Adam. 2014. “Mapping the Ideological Marketplace.” American Journal of Political Science 58(2): 367–87. Bonica, Adam and Maya Sen. 2017. “A Common Space Scaling of the American Judiciary and Legal Profession.” Political Analysis 25: 114–21. Borthwick, George, Daniel Ellingworth, Colin Bell, and Donald MacKenzie. 1991. “The Social Background of British MPs.” Sociology 25(4): 713–17. Boyd, Christina L., Lee Epstein, and Andrew D. Martin. 2010. “Untangling the Causal Effects of Sex on Judging.” American Journal of Political Science 54(2): 389–411. Branham, Alexander, Stuart N. Soroka, and Christopher Wlezien. 2017. “When Do the Rich Win?” Political Science Quarterly 132 (1): 43–62. Brunner, Eric, Stephen L. Ross, and Ebonya Washington. 2013. “Does Less Income Mean Less Representation?” American Economic Journal: Economic Policy 5(2) 53–76. Burden, Barry C. 2007. Personal Roots of Representation . Princeton, NJ: Princeton University Press. Carnes, Nicholas. 2012. “Does the Numerical Underrepresentation of the Working Class in Congress Matter?” Legislative Studies Quarterly 37(1): 5–34. Carnes, Nicholas. 2013. White-Collar Government: The Hidden Role of Class in Economic Policy Making . Chicago: University of Chicago Press. Carnes, Nicholas and Noam Lupu. 2023. “The Economic Backgrounds of Politicians.” Annual Review of Political Science 26: 253–70. Courtney, Michael. 2015. “Social Background and Intra-party Attitudes in Ireland.” Irish Political Studies 30(2): 178–98. Elkjaer, Mads Andreas and Michael Baggesen Klitgaard. 2021. “Economic Inequality and Political Responsiveness: A Systematic Review.” Perspectives on Politics 26: 1–26. Elsässer, Lea, Svenja Hense, and Armin Schäfer. 2021. “Not Just Money: Unequal Responsiveness in Egalitarian Democracies.” Journal of European Public Policy 28(12): 1890–908. Elsässer, Lea and Armin Schäfer. 2023. “Political Inequality in Rich Democracies.” Annual Review of Political Science 26: 469–87. Enns, Peter K. 2015. “Relative Policy Support and Coincidental Representation.” Perspectives on Politics 13(4): 1053–64. Epstein, Lee, William M. Landes, and Richard A. Posner. 2013. The Behavior of Federal Judges: A Theoretical and Empirical Study of Rational Choice. Cambridge, MA: Harvard University Press. George, Tracey E. and Taylor G. Weaver. 2017. “The Role of Personal Attributes and Social Backgrounds on Judging.” In The Oxford Handbook of U.S. Judicial Behavior , edited by Lee Epstein and Stefanie A. Lindquist, 286–302. New York: Oxford University Press. Gilens, Martin. 2015. “Descriptive Representation, Money, and Political Inequality in the United States.” Swiss Political Science Review 21(2): 222–28 Griffin, John D. and Claudia Anewalt-Rensburg. 2013. “Legislator Wealth and the Effort to Repeal the Estate Tax.” American Politics Research 41(4): 599–622. Grumbach, Jacob M. 2015. “Does the American Dream Matter for Members of Congress? Social-Class Backgrounds and Roll-Call Votes.” Political Research Quarterly 68(2): 306–23. Guthrie, Chris, Jeffrey J. Rachlinski, and Andrew J. Wistrich. 2000. “Inside the Judicial Mind.” Cornell Law Review 86: 777–830. Hemingway, Alexander. 2022. “Does Class Shape Legislators’ Approach to Inequality and Economic Policy? A Comparative View.” Government and Opposition 57(1): 84–107. Hout, Michael. 2008. “How Class Works: Objective and Subjective Aspects of Class Since the 1970s.” In Social Class: How Does it Work? , edited by Annette Lareau and Dalton Conley, 25–64. New York: Russell Sage Foundation. 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
1204 MILLER AND CURRY Hurwitz, Mark S, and Drew N. Lanier. 2012. “Judicial Diversity in Federal Courts: An Historical and Empirical Exploration.” Judicature 96: 76–83. Klontz, Bradley T., Paul Sullivan, Martin C. Seay, and Anthony Canale. 2015. “The Wealthy: A Financial Psychological Profile.” Consulting Psychology Journal: Practice and Research 67(2): 127–43. Kraus, Michael W . and Bennett Callaghan. 2014. “Noblesse Oblige? Social Status and Economic Inequality Maintenance Among Politicians.” PLoS ONE 9(1): e85293. Lax, Jeffrey R., Justin H. Phillips, and Adam Zelizer. 2019. “The Party or the Purse? Unequal Representation in the U.S. Senate.” American Political Science Review 113(4): 917–40. Levinthal, Dave, Lateshia Beachum, and Carrie Levine. 2017. “Supreme Court a Millionaire’s Club.” The Center for Public Integrity. https://theworld.org/stories/2017-06-09/supreme-court-millionaires-club Maks-Solomon, Cory and Elizabeth Rigby. 2020. “Are Democrats Really the Party of the Poor? Partisanship, Class, and Representation in the U.S. Senate.” Political Research Quarterly 73(2): 848–65. Mauˇcec, Gregor and Shai Dothan. 2022. “The Effects of International Judges’ Personal Characteristics on their Judging.” Leiden Journal of International Law 35: 887–95. Miller, Banks and Brett Curry. 2023. “When Advocates Become Adjudicators: Tracing the Effects of Prosecutorial and Public Defense Experience on Judicial Decision Making.” American Politics Research 51(6): 796–804. Pacelle, Jr., Richard L., Brett W . Curry, and Bryan W . Marshall. 2011.Decision Making by the Modern Supreme Court. New York: Cambridge University Press. Piff, Paul K., Michael W . Kraus, Stephane Cote, Bonnie Hayden Chang, and Dacher Keltner. 2010. “Having Less, Giving More: The Influence of Social Class on Prosocial Behavior.” Journal of Personality and Social Psychology 99: 771–84. Posner, Richard A. 1993. “What Do Judges and Justices Maximize? (The Same Thing Everybody Else Does).” Supreme Court Economic Review 3: 1–41. Posner, Richard A. 2008. How Judges Think. Boston: Harvard University Press. Schauer, Frederick. 2012. Thinking Like a Lawyer: A New Introduction to Legal Reasoning . Cambridge, MA: Harvard University Press. Scherer, Nancy. 2023. Diversifying the Courts: Race, Gender, and Judicial Legitimacy . New York: NYU Press. Scherer, Nancy and Banks Miller. 2009. “The Federalist Society’s Influence on the Federal Judiciary.” Political Research Quarterly 62: 366–78. Songer, Donald R. and Reginald S. Sheehan. 1992. “Who Wins on Appeal? Upperdogs and Underdogs in the United States Courts of Appeals.” American Journal of Political Science 36(1): 235–58. Spaeth, Harold J., Lee Epstein, Ted Ruger, Sarah C. Benesh, Jeffrey Segal, and Andrew D. Martin. 2021. “Supreme Court Database Codebook.” http://scdb.wustl.edu/_brickFiles/2021_01/SCDB_2021_01_codebook.pdf Stacy, Darrian. 2020. “Wealth and Policymaking in the U.S. House of Representatives.” Paper Prepared for Presentation at the American Political Science Association’s Annual Meeting. Suhay, Elizabeth, Marko Klašnja, and Gonzalo Rivero. 2021. “Ideology of Affluence: Explanations for Inequality and Economic Policy Preferences Among Rich Americans.” Journal of Politics 83(1): 367–80. Szmer, John, Robert K. Christensen, and Erin B. Kaheny. 2015. “Gender, Race, and Dissensus on State Supreme Courts.” Social Science Quarterly 96: 553–75. Wuest, Reto and Jan Rosset. 2019. “Legislator Income, Policy Attitudes, and Voting Behavior.” https://retowuest.net/papers/ wueest-rosset-leg-income-2019.pdf . Xu, Ping and James C. Garand. 2010. “Economic Context and Americans’ Perceptions of Income Inequality.” Social Science Quarterly 91: 1220–41. Cases cited Gold v. Ford Motor Co. 577 Fed. Appx. 120 (2014). Shaheen v. Progressive Casualty Insurance Company 673 Fed. Appx. 481 (2016). SUPPORTING INFORMATION Additional supporting information can be found online in the Supporting Information section at the end of this article. How to cite this article: Miller, Banks, and Brett Curry. 2024. “Exploring the influence of wealth on judicial decision making.” Social Science Quarterly : 1193–204. https://doi.org/10.1111/ssqu.13412 15406237, 2024, 4, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/ssqu.13412 by University Of Texas - Dallas, Wiley Online Library on [03/06/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License