Redistricting and Party Loyalty in the State Supreme Courts

Abstract

In this research, we examine redistricting plan review by state supreme courts to determine what factors influence party-aligned voting in judicial decisions. We analyze whether judicial selection systems matter, as well as the ideological extremism of judges. Our judge-level data includes votes on redistricting cases heard across the American states from 1961 to 2022. We find that judges who are ideologically extreme are more likely to cast party-aligned votes, but only when selected by appointment with life tenure or by partisan elections.

In cases that offer measurable benefit to political parties, what factors impact whether judges will vote in alignment with their party? This line of inquiry has historically focused on the federal courts (Cox and Katz 2002; McKenzie 2012) on cases with electoral consequences, specifically redistricting decisions. Consequently, the extent to which state supreme court judges will vote to support their party, and what makes such a decision more or less likely is not yet known. It is undoubtedly an important question.

That judges will side with their own party in redistricting cases is a common assumption, at least for political parties that bet on loyal judges. It is an assumption that helps explain costly efforts to “flip” seats on state supreme courts to alter the partisan balance of judges on these courts, especially when a single seat can have direct and meaningful consequences for statewide politics. Consider the 2023 Wisconsin supreme court election – the most expensive judicial race in US history – in which challenger Janet Protasciewicz defeated conservative incumbent Dan Kelly to create the first liberal majority on the state’s high court in 15 years. Upon taking her position, Protasciewicz joined liberal colleagues to declare the state’s Republican-drawn legislative maps unconstitutional, altering the balance of political power in the state for the first time in a decade. Of course, this assumption also explains why a judge who votes contrary to this expectation may attract considerable attention. When Ohio Chief Justice Maureen O’Connor (a Republican) voted against Republican-favored maps in that state in 2022, there were immediate calls for impeachment by state Republican leaders. ¹

Research on state supreme court redistricting votes is both intuitive and timely, given that contemporary US Supreme Court action has re-specified the significance of state court jurisdiction in the redistricting process. Research in state judiciaries also offers the comparison of institutional variation. Judges on state high courts can be appointed like federal jurists, but they can also be elected in partisan and nonpartisan contests. While a few have the political insulation of life tenure, most keep their jobs through some electoral mechanism. Given this, state supreme courts offer a unique setting that can help us understand how institutional rules shape political outcomes.

For these reasons, we examine redistricting plan review in the state supreme courts. Specifically, we analyze how state constitutional protections, individual ideological proclivities, and institutional rules impact the degree to which a judge will demonstrate party loyalty in these decisions. Our judge-level data includes votes on redistricting cases heard across the American states from 1961 to 2022. Out results indicate that ideological extremism is associated with a greater likelihood of party-aligned voting, but only for judges that are appointed or elected through partisan elections. While we find no differences between Republican and Democratic judges, we do find that judges are more likely to support their party when they are a member of the majority party.

The partisan stakes of state supreme court redistricting

At the risk of sounding colloquial, motivation for this research comes directly from political headlines. ² While state supreme court involvement in redistricting is not a recent phenomenon, the jurisdictional importance of these courts has significantly increased over time. This heightened role for state supreme courts in political reapportionment is best explained by the confluence of several factors.

The first pertinent factor comes from action taken by the US Supreme Court in Rucho v. Common Cause 588 U.S. ___ (2019), which effectively removed federal court jurisdiction over partisan gerrymander claims, citing them as nonjusticiable political questions. Chief Justice Robert’s majority opinion suggested that state courts (or state legislative action) constituted a more appropriate redress for such grievances. The court’s decision was immediately perceived as consequential by partisan litigants. Eric Holder, former Attorney General and chair of the National Democratic Redistricting Committee (NDRC) gave remarks to the press soon after: “We’ll be fighting in the states to ensure that we have a fair redistricting process. We will use the state courts where we are no longer able to use the federal courts.” ³ Jason Torchinsky, general counsel for the National Republican Redistricting Trust (NRRT) also provided commentary: “That opens a Pandora’s box at the state level. State judiciaries are going to have to wrestle with the same questions” that the Supreme Court just did. ⁴

The reaction by Holder and Torchinsky was prescient, as over half of all redistricting challenges since the Rucho decision have been filed in state courts (16 of 29). ⁵ Further, of the 55 total map disputes in the 2020 redistricting cycle, 30 were processed in the state courts, and more than half of those (18) were partisan gerrymander claims. ⁶ Of course, litigants continue to challenge redistricting maps in federal courts, particularly when a claim involves potential Voting Rights Act violations. However, the significance of the Rucho decision is that redistricting resolution is more directly focused on state judiciaries.

Beyond Rucho, factors related to the states themselves have also been significant. Specifically, some state constitutional provisions offer protection for claimants not found in the US Constitution. Over half of all state constitutions (29) have a requirement that elections be “free,” and a subset of these (18) further specifies that elections be both free and “open” or “equal.” Beyond this, some states provide redistricting commissions with specific criteria using “free elections” language. ⁷ If litigants in states with constitutional “free” election language argue that maps which advantage one party over another result in inherently unequal elections, these provisions create a viable legal pathway for challenges in state judiciaries that does not exist in the federal courts.

A final factor is the success of a state-court strategy for litigants, particularly for challengers from the Democratic Party. The tactic has proven so successful that Republican strategists describe redistricting litigation as the cornerstone of “sue until blue” politics. ⁸ While Democratic redistricting legal victories occurred prior to the Rucho decision, they were more headline-worthy in the 2020 redistricting cycle. ⁹ By early 2023, high-profile judicial actions in Ohio, Pennsylvania, North Carolina, Wisconsin, and Virginia supported electoral maps more favorable to Democrats, with pending cases in other states. ¹⁰

The high-stakes nature of state supreme court redistricting is underscored by journalistic emphasis. In 2022, CNN’s coverage of the North Carolina Supreme Court decision in Moore v. Harper declared the courtroom a place “where redistricting has become a blood sport as intense as the Duke vs. UNC basketball rivalry.” ¹¹ Clever journalism aside, redistricting influences “who wins elections, who is at the table when laws are considered, and what laws are passed.” ¹² Decisions in the state judiciaries can ultimately dictate the balance of political power across the United States, and thus the partisan stakes of state supreme court redistricting decisions are sizeable.

State court involvement in the redistricting process

Redistricting authority in each state is designated by state law. Most states assign this role primarily or fully to the state legislature. In roughly half of all states, “commissions” may also participate in the redistricting process with varying degrees of authority. Even in states with commission structures in place, state legislatures continue to have input and influence on the redistricting process. Redistricting plans are like other state laws in that once maps are created, they can generally be subject to gubernatorial veto. Similarly, redistricting plans or processes can also be subject to legal challenge in a judicial forum. ¹³

There are several ways that state courts may be involved in the redistricting process. Some states have a prescribed role for judges to either select members of commissions or serve on commissions (Levitt and Wood 2010). Courts may also be asked directly to draw district lines. District line-drawing by state courts can be prompted when legislatures reach an impasse over maps, or when timely revisions are needed on commission plans. Courts may also draw district lines when their own decisions have upheld challenges to redistricting maps, and upcoming elections necessitate immediate action. In the 2000 and 2010 redistricting cycles, state legislative district lines were drawn by courts in 11 states, and congressional lines in nine states (Levitt and Wood 2010).

While such direct involvement by courts in redistricting is impactful, it is uncommon. Thus, the primary role that state courts have in the redistricting process occurs once litigants challenge the legality of the map-drawing process, or the district maps themselves. Such litigation is common, and state supreme court review of redistricting plans is either automatic or triggered by citizen request in nearly half of all states (Levitt and Wood 2010). In these cases, judges review redistricting plans that are crafted or influenced by state political majorities, and their decisions reflect a judge’s support (or not) of those political majorities.

Party loyalty and judicial behavior: Theory and evidence

Given that the primary role of state supreme courts in redistricting is voting in cases that involve challenges to redistricting plans or processes, this research seeks to understand how party loyalty may be expressed in these votes. Political scientists understand party loyalty most directly from observations of American voting behavior (Campbell et al. 1960; Miller 1991; Miller and Shanks 1996), but also from observations of the legislative and executive branches of government (i.e., Erikson, Mackuen, and Stimson 2002). For “single-minded seekers of reelection,” party loyalty can translate into electoral advantage (Cox and McCubbins 2005, 2007), although excessive partisanship can also create electoral liability, particularly on divisive votes (see Carson et al. 2010). Party loyalty can also determine institutional advancement for members of Congress, and consequently their legislative success (Coker and Crain 1994). Party loyalty is easily understood as motivation for those that serve in the “political” branches of government, but less so for judges who are historically perceived as neutral arbiters of the law. In the traditional adage, judges are legal-minded rather than electoral- or policy-minded.

Political scientists who study courts counter this traditional framework, claiming that it ignores both theoretical and real-world analysis of judicial decision-making. While judicial actors are influenced by legal considerations, scholars find that judges are also motivated by political preferences (or attitudes) and long-term strategy. Attitudinal motivation refers to judicial choice that incorporates sincere policy preferences with less regard for anything else, including the law (Segal and Spaeth 1993, 2002). Strategic motivation prioritizes long-term over immediate goals and incorporates a judge’s policy preferences alongside the actions of other relevant actors (e.g., Epstein and Knight 1998). Such long-term goals include policy preferences, but also include features like prestige, job satisfaction, and career advancement (Epstein and Knight 2013).

How do considerations of party factor into conventional models of judicial choice? A judge’s partisanship is most obviously not a legal consideration. Instead, partisan loyalty can be explained from both an attitudinal frame and a strategic frame. From an attitudinal perspective, Peterson (2019) explains that party loyalty may result from a judge’s efforts to advance their policy preferences, given that a judge’s policy preferences are more likely to be represented by their co-partisans in the legislature. In contrast, a strategically motivated judge would support their party in order to further their own political ambitions or long-term career goals. A “win” for a judge’s party may increase the likelihood of some future career milestone.

In terms of evidence, research mostly supports the idea that party loyalty impacts judicial voting. Most analyses that specifically focus on redistricting litigation examine the federal courts. In a study on US District Court redistricting decisions, Lloyd (1995) found that judges of both parties were more likely to strike down redistricting plans crafted by a legislature that they did not share partisanship. Cox and Katz (2002) similarly found that “friendly” courts were more likely to uphold redistricting plans than “hostile” ones. Peterson (2019) also found “sophisticated partisan calculations” in redistricting decisions, which tend to favor the electoral interests of a judge’s own party. Alternatively, McKenzie (2012) did not find a general effect of partisanship on redistricting decisions, but did find a “constrained” partisanship effect in that judges are more likely to strike down plans crafted by an opposition legislature when the plan is challenged under the Voting Rights Act. ¹⁴

While there is no analysis of redistricting decisions in state judiciaries, other election-law analyses yield mixed results. ¹⁵ Whereas Graves (2003) found evidence of partisan voting in ballot access cases, Kopko (2008) did not. Kang and Shepherd’s (2011, 2016) analysis of election dispute cases showed that partisanship influences judicial voting in significant and measurable ways, but their more recent work (2017) shows that legal issues matter as much if not more than party loyalty.

Factors that influence party-aligned voting

In this research, we are interested in factors that influence the likelihood that a judge will cast a party-aligned vote on redistricting litigation in state supreme courts. How much legal protection is built into the redistricting process varies among the states. States also vary in the institutional design of state judiciaries. Finally, judges also vary as individuals. In this section, we describe how variations in laws, individuals, and systems may impact the likelihood of party-aligned voting in redistricting cases.

Laws

As stated earlier, legal factors impact judicial decisions. Prior research suggests that redistricting law constrains redistricting decisions, at least when precedent is clear. In other words, if legal guidelines provide clear directives, judges should be less inclined to vote based on party loyalty. Meanwhile, ambiguous precedent creates an alternative voting “opportunity.” McKenzie offers that “motivated by either strategic or psychological forces, judges are more likely to favor their own party in redistricting when a lack of clarity… provides them with that window of opportunity” (2012, 802).

McKenzie’s (2012) analysis in the federal court system considers legal factors such as the type of legal challenge involved in a redistricting case (partisan gerrymandering, racial gerrymandering, or Voting Rights Act violation) as well as the type of redistricting plan under review (Congress, legislative, or both). These categories provide some legal direction in federal review, but arguably provide little evidence of clear legal precedent in the state court context, at least when considered alone. For this reason, it is instructive to consider state laws that prescriptively addresses these issues.

As discussed, some state constitutions include “free elections” language under which litigants challenge redistricting plans. These provisions provide additional legal criteria for judges in these states to consider on litigation involving elections and, in some states, the district map itself. Thus, there is additional legal guidance for judges in these states. Alternatively, state constitutions that do not include such language leave more room for ambiguity, as judges without constitutional directives have no additional legal criteria to consider in assessing redistricting plan conditions or outcomes. Given this, we control for state constitutions that contain “free elections” provisions in our analysis. We consider these provisions to be a legal constraint on judicial behavior, which leads to a general expectation:

H1: In redistricting cases, a judge will be more constrained when casting a party-aligned vote in states with constitutional “free elections” provisions than in those without.

Individuals

While partisanship and ideology are not directly interchangeable concepts, there is some obvious overlap in modern American politics. In research on the American judiciary, party identification was used as proxy for judicial ideology for many years. Further, evidence of partisan voting was interpreted as evidence to support the attitudinal model of judicial choice. As scholarly measures of judicial ideology have become more refined over time, scholars have recognized that ideology can factor into partisan behavior. Lloyd (1995) found that Republicans judges were generally more deferential to the creators of redistricting plans than Democratic judges, regardless of the partisan origin of the plan. Similarly, McKenzie (2012) found that conservative judges were generally more likely to uphold redistricting plans in Voting Rights Act challenges. ¹⁶

Judicial ideology can provide useful information for the likelihood that a judge will vote to support party, but not simply as a proxy for partisanship. Rather, the intensity that a judge holds their ideological preferences may be more useful information. In terms of party support, a party win almost certainly guarantees some advancement of policy goals for an ideological extremist. However, the same may not always be true for a moderate. A party win may well inhibit ideological goals for a moderate if government policy is already near a judge’s ideal point. At minimum, the propensity for party-aligned voting between ideological extremists and moderates may not be equivalent. Given this, we have the following general expectation:

H2: In redistricting cases, an ideologically extreme judge is more likely to cast a party-aligned vote than a moderate judge.

Institutions

While the individual proclivities of judges will certainly influence the likelihood of party-aligned voting, it is important to consider the institutional arrangements within which state supreme court justices operate. The institutional design that structures how state supreme court justices are selected includes systems of elite appointment with lifetime tenure, merit selection with retention elections, partisan election, and nonpartisan election. While these institutional rules for selection may not directly motivate judicial behavior, they may condition it. In other words, judges in some systems may feel more constrained in choosing to vote with their party than in others. For this reason, we posit that institutional rules interact with other factors that motivate judicial behavior. Here we explore the interaction of judicial selection rules and ideological extremism.

On the federal courts, appointment with life tenure insulates judges from the political repercussions for their decisions that could result in job loss. The institutional design of these courts empowers judges to pursue their policy preferences through their decisions. More importantly, the protection offered by life tenure exacerbates the tendency for ideologically extreme judges to cast party-aligned votes. Of course, lifetime appointments are a rare institutional design for state supreme courts that exists in only a handful of states. Instead, most state court judges are accountable to an electorate for job retention.

If research suggests that lifetime appointment relaxes institutional constraints on judicial behavior, it suggests that accountability mechanisms constrain it. Specifically, elected judges may alter their behavior to ensure reelection. Hall’s (1987, 1992) research showed that elected judges were more likely to alter voting behavior near the end of their terms. Similarly, Huber and Gordon (2004) found that elected judges imposed longer sentences as their reelection neared.

Retention mechanisms for judges vary across the states, and consequently reelection rates substantially differ across different systems. Judges subject to nonpartisan elections are reelected at much higher rates than those subject to partisan elections (Bonneau and Cann 2015). The information difference between the two helps to explain this, as voters tend to reelect incumbents in nonpartisan, low information environments, but cast votes based on partisan identities in higher information environments (Squire and Smith 1988). The party-vote share of judicial candidates in partisan elections closely mirrors the party-vote share in presidential elections (Lim and Snyder 2015).

While reelection rates generally differ between nonpartisan and partisan judicial elections, an exception can be found when races receive substantial media coverage. In these races, partisan voting patterns appear even in nonpartisan judicial races (Rock and Baum 2010), driven by partisan information that is communicated to voters (Bonneau and Cann 2015). Given this, judges subject to nonpartisan reelection have an incentive to minimize the amount of partisan information communicated through coverage of their decisions, particularly in high-profile cases. Because the salience of redistricting decisions is high, media coverage of redistricting cases is inevitable.

Based on what we know about retention politics and the high-profile nature of redistricting cases, it is reasonable to anticipate that judges in partisan election systems will behave like judges who have lifetime tenure: the more ideologically extreme they are, the more likely they are to cast a party-aligned ballot. For judges subject to nonpartisan and merit retention elections, revealing ideology or partisanship in high-profile cases might be problematic for reelection. Therefore, these judges may have less incentive to use ideology when casting a party-aligned vote. Given this, our expectation is:

H3: The influence of ideological extremism on voting in redistricting cases is conditioned by a state’s judicial selection method. The impact of ideological extremism should be more evident in appointment systems and partisan election systems, and less evident in nonpartisan election and retention election systems.

Control factors

The degree to which a judge votes with their party on redistricting cases may be influenced in other ways, which we consider as controls. One factor is whether the judge belongs to the majority in power. Judges from the majority party might be more concerned with the expectations of party leadership. They may also give greater weight to how their vote will be perceived by the broader public and may believe that ruling in line with their party can ensure support for future career advancement. It stands to reason that minority party judges, with similar logic, may exercise greater caution in ruling against the majority.

Another consideration is a judge’s partisan identity. While Republicans and Democrats likely have equal desire to see their party’s interests advanced in redistricting cases, there have been observable differences in partisan attitudes about redistricting over time. Republicans have placed more emphasis on state legislative influence, while Democrats have focused more on federal elections. Republicans have generally been more resistant to redistricting reform, while Democrats have increasingly supported independent commissions and other reforms. Finally, Republicans have generally been more effective (overall) at using redistricting to gain a political advantage than Democrats. While these differences may not directly influence party-aligned voting, it is nonetheless true that the redistricting issue itself may be conditioned through a partisan lens.

A final factor is the nature of the redistricting challenge. Specifically, it is important to consider whether redistricting issue being litigated is a partisan gerrymander challenge. ¹⁷ One of the main challenges in partisan gerrymandering cases is the absence of a clear, objective standard for determining when district boundaries are drawn unfairly for partisan advantage. While statistical tools exist to identify such gerrymandering, courts and legislatures have not universally agreed on how to use these metrics to identify gerrymandering. In the absence of a clear standard, such cases might open the door for an increased influence of partisanship on judicial votes.

Party-aligned voting in redistricting litigation

To understand party loyalty in redistricting decisions, we construct a model that predicts the likelihood of party-aligned voting. Our data include 1,412 state supreme court redistricting votes. These votes were cast by 757 judges in 220 redistricting cases heard from 1961 to 2022, roughly 30 cases each decade. Our dependent variable measures whether a judge voted with party in a redistricting decision. To determine whether a vote is party aligned, we obtained each judge’s party identification from data collected by Brace, Langer, and Hall (2000) supplemented with data from state judicial websites, judicial biographies, media searches, and Ballotpedia. ¹⁸ We identified 406 Democrats, 337 Republicans, and 14 judges of non-identifiable partisanship. The party of the legislature that proposed a redistricting plan was obtained from historical partisan composition timeline records published by the National Conference of State Legislatures (NCSL). We identified 93 plans created by Democratic majorities, 94 plans by Republican majorities, and 33 plans created under divided government. ¹⁹

Data from these two variables are summarized in Table 1. We define a party-aligned vote as one in which a judge votes to uphold a plan proposed by a unified legislature under the control of copartisans or votes to strike a plan proposed by a unified legislature under the control of non-copartisans. Our dependent variable is a binary variable that equals a 1 when a judge’s vote meets either of those two criteria, and 0 otherwise. ²⁰ In votes cast in which all necessary information is available, about 55% of votes demonstrate partisan alignment. This statistic is similar for both Democrat and Republican judges, varying by a single percentage point at most.

Table 1. Number of Votes Cast in Redistricting Cases in State Supreme Courts, 1960–2022

In terms of independent variables, our theoretical discussion motivates legal, individual, and institutional factors that may influence the likelihood that a judge’s vote will reflect party-alignment. To operationalize our legal factor of interest, we include a dichotomous variable to indicate whether a state constitution contains a provision that guarantees “free elections.” For our individual factor, we include a state supreme court judge’s ideological extremism relative to colleagues. To calculate this measure, we utilize PAJID data from Brace, Langer, and Hall (2000), updated by Hughes, Wilhelm, and Xuan (2023). ²¹ A judge’s ideological extremity is calculated by mean-centering PAJID scores and then taking the absolute value. This results in a continuous variable in which higher values indicate a judge is more ideologically extreme. Roughly 10% of judges in our sample have a score that measures at least one standard deviation above the mean, and thus might be considered “ideologues”. ²²

Finally, our model includes an institutional indicator for a state judicial selection method. This categorical variable (transformed into a series of dummy variables for our analysis) indicates whether judges were selected by partisan election, nonpartisan election, merit selection, or appointment (the omitted category). ²³ Given our conditional expectation for this variable, we include interactions between ideological extremism and judicial selection method.

Several control variables are included in our model. These include indicators for majority party judges (approximately 60% of judges in our data), Republican judges (approximately 44% of judges in our data), and partisan gerrymandering claims (around 13% of cases). Summary statistics for all variables are presented in Table 2.

Table 2. Means of Model Variables

We control for the possibility of period effects in redistricting decisions by including fixed effects for each redistricting cycle. We also include random effects for each state and cluster our standard errors by state to control for autocorrelation within each case. To identify the causal relationships at interest, we use a selection-on-observables design that features logistic regression to control for confounders. ²⁴

Results

The results of our model that estimates the likelihood that a judge will vote with party are displayed in Table 3. The expected proportion of correct predictions is 0.57; if the model estimates were used to predict the outcome of future redistricting votes, we would expect the model to correctly classify 57% of the new observations (Herron 2000). This represents a 14% reduction in the classification error that would result if we relied on a naïve probability model to predict outcomes instead. Thus, the model explains a notable amount of variation in the data, though there is certainly noise remaining in the dataset.

Table 3. Predicting a Party-Aligned Vote, 1961–2022

Note: Omits random effects estimates by state.

* p < 0.05, ** p < 0.01, *** p < 0.001, two-tailed test

The coefficient for the “Free and Fair” elections variable is not statistically significant. Therefore, we find no evidence supporting Hypothesis 1. The coefficient for the judicial extremism variable, however, is statistically significant. The interaction term between judicial extremism and merit selection systems is also statistically significant; in addition, the interaction term between judicial extremism and nonpartisan election systems just misses statistical significance with a p-value of 0.069. With these results, we find support for Hypothesis 3.

Due to the difficulty of interpreting both logit models and interaction terms, we use the estimates in Table 3 to calculate the average discrete effect of ideological extremity for judges that are appointed (omitted), selected via merit, elected through partisan elections, and elected through nonpartisan elections. ²⁵ The results are displayed in Table 4. For judges that are appointed and elected through partisan elections, there is a positive and statistically significant relationship between ideological extremity and party-aligned voting. A one standard deviation increase in the ideological extremity of an appointed judge – equivalent to a 10-unit increase in that variable – is associated with an eight percentage point increase in the expected probability of a partisan vote. For judges selected via partisan elections, the likelihood of party-aligned voting increases by 12 percentage points.

Table 4. Average Discrete Effect of Ideological Extremity on the Probability of a Party-Aligned Vote by Selection System

Note: Results are calculated using the estimates in Table 3 and observed data values.

* p < 0.05, ** p < 0.01, *** p < 0.001

These results can be graphically displayed in a predicted probability plot, shown in Figure 1. When ideological extremity is equal to 10 – which is, incidentally, the 10^th percentile of that variable – the predicted probability that an appointed judge casts a party-aligned vote is 38%. When ideological extremity is bumped to 40 – the 90^th percentile of the data – the predicted probability jumps to 60%. A similar story can be told for partisan-elected judges: the probability a moderate judge casts a party-aligned vote is 32%, while for ideologically extreme judges it is 69%. For judges selected through merit systems and for judges elected through nonpartisan elections, the relationship is negative but not statistically significant; as such, they are left out of Figure 1.

Figure 1. Probability of a Party–Aligned Vote, by Ideology and Selection System

These results largely confirm the expectation that ideological extremism motives party-aligned voting, as conditioned by institutional rules. As expected, ideologically extreme judges in appointment systems are more likely to cast party-aligned votes than moderate judges in those systems. Perhaps not as expected is the same significant propensity for judges selected in partisan elections. While electoral accountability offers a theoretical constraint on the expression of judicial preferences in voting behavior, it may be that redistricting offers a specific legal scenario whereby party-aligned voting is encouraged (perhaps even expected) for judges who are more ideologically inclined to support party.

In terms of other results, the coefficient for a judge whose party matches the legislature that created the redistricting plan (majority judge) is positive and statistically significant. On average, a judge whose party controls the state legislature is 30% more likely to vote in a partisan matter relative to judges whose party does not control the legislature. The coefficients for other variables are not significant, and fixed effects for redistricting cycles are also not significant. ²⁶

Conclusion

In this research, we find that party-aligned voting in redistricting decisions is more likely to occur when judges are ideologically extreme than when they are moderate. This finding is consistent with attitudinal models of judicial decision-making, and probably not very surprising. A novel finding of this research is that this relationship is moderated by the institutional design of the state court system. Specifically, ideological extremism is associated with party-aligned voting in systems where judges are selected in partisan elections or appointed with life tenure. The relationship disappears when judges are elected via nonpartisan elections or selected by merit. A possible conclusion from this finding is that judicial selection mechanisms can encourage or discourage partisan behavior by judges. Moreover, changes to judicial selection may influence the degree of party-aligned voting in redistricting decisions.

Our research also suggests that majority party judges are more likely to cast partisan votes than judges in the minority. Obviously, judges who are in the majority want to stay in the majority. Meanwhile, it does not appear that judges are constrained by provisions in state constitutions that mandate “free” elections. While states that have such constitutional provisions provide a legal pathway for litigants to challenge redistricting plans, they do not provide a consistent legal answer for judges who review these claims.

The findings of this research give empirical credence to an assumption about judicial behavior that shapes contemporary redistricting strategy for both Democrat and Republican parties. This strategy has entered the next phase: judicial recruitment. Adam Kincaid, executive director of the NRRT, observed in a 2019 interview after the Rucho decision, “The next phase of redistricting is going to be about groups doubling down on their attempts to flip state courts.” ²⁷ Two years later, he reflected on events that included a successful “court flip” by Democrats in Pennsylvania: “I think a lot of people started to view the state supreme courts as partisan actors in redistricting after that point.” ²⁸ Of course, elected state supreme courts are not the only strategic targets, as appointed courts may be just as susceptible to court flipping attempts (albeit with potentially fewer constraints).

Unequivocally, the significance of state supreme courts in redistricting is not lost on either the right or left. Further, it is logical for parties to include state judiciaries in their strategies, as our research demonstrates that party organizations can benefit from courts staffed by judges that are more “friendly” to their redistricting efforts. In the future, we look forward to inquiry that investigates the success of partisan attempts to influence redistricting outcomes via judicial recruitment efforts on these courts.