From Congress to the States: Explaining the Emergence and Membership of Freedom Caucuses in State Legislatures

Abstract

Since 2017, Republican lawmakers in a growing number of US states have formed ideological intraparty organizations, modeled after the US House Freedom Caucus, that seek to move state policy further rightward. What explains the appearance of these state freedom caucuses, and what kinds of lawmakers are more likely to join them? We show that the creation of these caucuses was initially motivated by concerns that state-level legislative Republican parties are too ideologically heterogeneous but has since been driven by conservative entrepreneurs seeking to spread freedom caucuses nationally. We also provide evidence that conservative legislators are more likely to join a new state freedom caucus, as one would expect, but also that, in a few states, lawmakers who are more electorally vulnerable lawmakers or lack internal influence have also been more likely to join. These findings underscore how state-level ideological caucuses can appeal to members’ multiple goals and serve as instruments of vertical polarization in a federal system.

Introduction

Members of Congress have a long tradition of forming caucuses – “voluntary, organized associations” of lawmakers that are “without recognition in chamber rules” (Hammond 1998) – to help them overcome collective action problems and achieve shared policy and power goals. Though most caucuses are issue-based, some are explicitly ideological, with members drawn entirely from one party who share a particular set of beliefs and policy preferences.

The US Congress is not the only legislative entity where such “intraparty organizations” (Bloch Rubin 2017) may form. A growing number of state legislatures have become home to self-described “freedom caucus” ideological member organizations, modeled after the congressional House Freedom Caucus (HFC) which was founded in January 2015 by a handful of disgruntled conservative lawmakers in the US House of Representatives. These state-level freedom caucuses (SFCs), consisting entirely of Republicans who espouse conservative policy goals, emerged in a few states in 2017, but by mid-2023 they had spread to at least one legislative chamber in 17 states. Many have aggressively sought to push state policy further rightward, even if it means obstructing legislative business, dividing their own party in the legislature, and incurring the wrath of party leaders.

The emergence and rapid spread of SFCs raises two important questions. First, why have they appeared in some state legislatures but not others? Second, what compels individual lawmakers to join SFCs when they are created, especially if they require members to dedicate scarce resources to their maintenance and risk retaliation by party leadership? In this study, we offer answers to both questions. First, we show that the origins of these caucuses have varied over time. In an initial wave from 2017 to 2020, they were more likely to emerge in more ideologically heterogeneous Republican legislative parties, a finding consistent with the hypothesis that conservative legislators create SFCs to make their parties more unified around conservative policy principles. However, the second and ongoing wave that began in 2021 has been more top-down, driven by the entrepreneurial efforts of conservative political actors, illustrating how national interest groups and elected officials can contribute to state-level polarization. We then provide evidence that, unsurprisingly, conservatives have a greater probability of joining SFCs, but that these caucuses sometimes also appeal to lawmakers who lack internal influence in the legislature or who are more electorally vulnerable. This last finding underscores the multiple objectives that intraparty caucuses can potentially help legislators achieve.

Prior literature

Most caucuses in legislatures are nonideological, formed around common regional, constituency, or economic interests. The reasons that members of Congress join such caucuses, and their influence on legislative outcomes and lawmaker influence, have been well studied (e.g., Ainsworth and Akins 1997; Hammond 1998; Victor and Ringe 2009). Some congressional caucuses, however, are explicitly ideological, bringing together like-minded lawmakers who belong to the same political party. Scholars have increasingly sought to explain what these intraparty organizations in Congress do, why they form, and their influence on national politics. In her book-length study of the subject, Bloch Rubin (2017) argues that congressional intraparty caucuses overcome collective coordination problems by offering members selective benefits, like access to party leaders or beneficial signals to constituents, and by creating opportunities to convert public goods into excludable accomplishments (see also Clarke 2020). They also employ internal rules that smooth group decision-making to gain bargaining power (Bloch Rubin 2017).

The HFC, one of the most prominent recent ideological caucuses to appear in Congress, exemplifies this perspective of intraparty organizations. Founded in 2015 by a dozen conservative members of the US House of Representatives, the HFC uses a binding rule to keep its members unified and limits its membership to Republicans who are willing to vote against party leadership, thereby maintaining internal cohesion and forming a pivotal voting bloc that gives the group political leverage. Indeed, research has found that the Freedom Caucus, along with other ideological caucuses in Congress, have successfully passed or defeated legislation, shaped the legislative agenda, determined party tactics and strategy, changed chamber rules, and selected party leaders (Green 2019; Green and Crouch 2022; McGee 2020; Bloch Rubin 2017), and that they develop different donor bases and new campaign funding networks with each other (Clarke 2020; McGee 2017).¹

This research has shed valuable new light on the politics of ideological caucuses in Congress. Yet with a handful of exceptions (e.g., Mahoney 2018; Rouse, Hunt, and Essel 2022), ideological caucuses in state legislatures have not been subject to scholarly analysis, despite possessing the same, if not greater, potential influence over legislative politics.² This lack of scholarship is especially striking insofar as state legislatures have become an increasingly important source of policymaking (e.g., Grumbach 2022) and provide an opportunity to conduct comparative analyses of the origins, membership, and activity of intraparty organizations. This includes state freedom caucuses, which have appeared in over a dozen state legislatures in recent years.

An overview of state freedom caucuses

Table 1 lists summary information about each state’s freedom caucus formed between 2017 and August 2023. Note first that, just as intraparty organizations in Congress are more common in the House than the Senate (Bloch Rubin 2017), most SFCs are only found in the state House. Just one state (Washington) has Senate freedom caucus members but no House caucus members, and of the five that have bicameral membership, three were initially House-only organizations.³ Second, the rate at which SFCs have been created accelerated after 2020: while only 5 states established freedom caucuses between 2017 and 2020, 12 more did so between January 2021 and April 2023, and two previously existing freedom caucuses, in Mississippi and North Carolina, “re-formed” during that period.

Table 1. State freedom caucuses: Summary data (2023)

Note: Data is current as of August 2023.

^a Bold and underline indicate that (a) Republicans are the majority party in the chamber and (b) the SFC is at least twice the size of the seat margin between Republicans and Democrats, making it a potential pivotal or “swing” bloc on the chamber floor for votes decided by a simple majority.

^b Initially established only in the state house; later expanded to the state senate (either because senators joined or because house caucus members were elected to the senate).

^c News accounts suggest that, as of March 2023, the caucus is larger (34 members in North Carolina; 40 members in New Hampshire; and 26 members in Wyoming, which would make it a pivotal bloc in that state).

Finally, some caucuses have affiliated with the State Freedom Caucus Network (SFCN), a national organization established under the auspices of former US HFC chair Mark Meadows (R-SC) that promotes the creation of freedom caucuses in state legislatures. The SFCN offers affiliated SFCs a number of benefits, such as a full-time paid staffer, the opportunity to coordinate with other freedom caucuses on policy, a mechanism for bringing national attention to state issues, and even advice on how to vote on particular bills and amendments (Brown and Metz 2023, Wolfson 2023a, 2023b).⁴ The SFCN, in turn, sees SFCs as a way to strengthen states vis-a-vis the national government, move legislative Republican parties further to the right, enact conservative policies at the state level, check the power of moderate governors and lobbyists, expose “corrupt” procedural practices that benefit incumbents, and – as SFCN President Andy Roth explained in a media interview – provide “full-time conservative oversight” of state government (Interview with SFCN staffer, January 31, 2024; Reynolds 2023a; State Freedom Caucus 2023).

A review of press accounts of state freedom caucuses reveals that many have been highly active, taking positions on legislation, sponsoring bills, pushing for changes to their chamber’s rules, endorsing candidates for elected office, posting on social media, issuing press releases, and holding news conferences. Some have used more unorthodox tactics, such as sponsoring public protests, lobbying state and federal officials, hosting constituent town hall meetings, filing lawsuits, openly criticizing their own party’s leaders, filibustering their party’s bills, and creating political action committees. Occasionally their activities have become a source of major intraparty conflict. In the South Carolina House, for instance, caucus members were effectively kicked out of the party for failing to sign a pledge in early 2023 that they would not challenge incumbents in GOP primaries, while the Speaker of the Wyoming House openly criticized the caucus’s “thirst for power” and “vicious discord” in a guest editorial published later that year (Herlihy 2023; Sommers 2023).

How important are these SFC activities for shaping political or policy outcomes? One may argue that they constitute little more than symbolic actions of electoral position-taking, particularly media-focused acts like holding press conferences and posting on social media.⁵ In addition, unlike the HFC, a number of SFCs make their membership public, encouraging those lawmakers to tout their caucus affiliation to voters rather than press the caucus to shape policy outcomes behind the scenes (see Table 1). Furthermore, in many chambers, the SFC is not large enough to exercise the kind of influence that the US HFC often does. For instance, as of August 2023, eleven chambers had membership that constituted less than 15% of the GOP conference, constraining their ability to sway their party. In the 118th Congress, by contrast, nearly 20% of House Republicans are Freedom Caucus members. Relatedly, only four⁶ state freedom caucuses in 2023 were large enough to serve as a pivotal floor bloc, the traditional source of leverage for intraparty organizations (Bloch Rubin 2017), and five SFCs were in chambers where Republicans are the minority party, giving them little influence over the chamber agenda (but see Clarke, Volden, and Wiseman 2023).⁷

This does not mean that SFCs are mere position-taking vehicles that lack influence, however. Table 2 provides a sample of substantive policy victories won by several state freedom caucuses between 2017 and mid-2023. In some cases, the caucus was large enough to be pivotal on the chamber floor, and it exploited that advantage by joining with Democrats to form a cross-party floor majority, defeating the majority GOP in a “disappointment” vote (Jenkins and Monroe 2015). In other cases, the freedom caucus was still influential despite its limited size. Some caucuses successfully exploited minoritarian chamber rules that allow a smaller number of lawmakers to influence the legislative process. At least one, South Carolina’s HFC, used an extra-legislative tactic – litigation – to bring about change. In addition to these examples, it may be possible for an SFC to move its party’s agenda further rightward, especially if the caucus constitutes a sizable proportion of the party.⁸ Though this kind of influence is harder to observe, since it is often exercised out of the public eye, anecdotal evidence suggests that it has happened in some chambers, such as the Arizona House (Small 2023b). Chambers with freedom caucuses have also introduced conservative “culture war” legislation at a higher rate than chambers without SFCs. For example, between January and August 2023, state legislatures with SFCs introduced an average of 5.71 bills targeting the LGBT community, versus an average of 4.7 anti-LGBT bills in non-SFC chambers.⁹

Table 2. Examples of state freedom caucus influence

In short, the data indicate that state freedom caucuses can be quite active and influential, irrespective of their size, even when facing opposition from GOP leaders in their chambers. It also suggests a potential role for a DC-based organization in the formation and operation of these caucuses. In the following section, we test a number of hypotheses that may explain why these caucuses have appeared in some states but not others.

Which states are more likely to adopt freedom caucuses?

Though state freedom caucuses offer state Republican lawmakers several potential electoral and policy benefits, only some legislative chambers have adopted them. Understanding the pattern by which they have been created can provide further insight into the conditions that encourage the formation of intraparty organizations in general and SFCs in particular.

We test five hypotheses for why a state legislature would adopt a freedom caucus. The first two are derived from a commonly stated mission of state freedom caucuses: to unify the GOP around a conservative agenda. As SFCN President Andy Roth explained in an interview, SFCs in GOP states counter the influence of more moderate lawmakers who are elected as Republicans in such large numbers that “the chambers are effectively controlled by moderates and Democrats” (Reynolds 2023a). Similarly, Wyoming Freedom Caucus chairman John Bear explained that the purpose of having a state freedom caucus is about “providing that differentiation for the people of Wyoming to see who are the conservatives and who are not” (Wolfson 2023b). Put another way, legislative chambers that are less polarized – where the GOP is more heterogeneous and closer to the Democratic Party – are more likely to see the formation of SFCs.¹⁰

Hypothesis 1: An SFC is more likely to form in a state legislative chamber with a more ideologically heterogeneous Republican Party.

Hypothesis 2: An SFC is more likely to form in a state legislative chamber where the Republican and Democratic parties are less distant from each other.

The third hypothesis is that SFCs are more likely to appear in chambers where rank-and-file legislators lack influence over legislative outcomes. As noted previously, research has found that intraparty caucuses can provide leverage for individual lawmakers with limited opportunities to shape the legislative agenda. Just as members of the US HFC often bristled against the arbitrary exercise of power by Speaker John Boehner (R-OH), so too have founders of some state freedom caucuses complained about a top-down, Speaker-driven legislative process that marginalizes conservative lawmakers (e.g., Montgomery 2019; Ulrich 2022). Chambers with leaders that possess more formal tools of influence are better positioned to constrain the opportunities of rank-and-file lawmakers. We therefore expect that state legislatures with more powerful leaders are more likely to see the formation of freedom caucuses.¹¹

Hypothesis 3: An SFC is more likely to form in a state legislative chamber with more powerful leadership.

Our last two hypotheses are suggested by prior research showing that entrepreneurs play an important role in the formation of legislative caucuses (Mahoney 2018; Schickler 2001). These entrepreneurs can be members of the legislature themselves, but they may also be external to the legislature. For instance, the Conservative Opportunity Society, an ideological caucus headed by then-backbencher Newt Gingrich (R-GA), had first been suggested to Gingrich by former president Richard Nixon (Green and Crouch 2022). This seems especially true for state caucuses, since states are often (and increasingly) subject to influence from national parties and special interest groups, and there are examples of state caucuses aided if not managed by a national organization (Berry and Berry 1990; Grumbach 2022; Hertel-Fernandez 2019; Shipan and Volden 2008; Sullivan and Winburn 2011).

Two types of external entrepreneurs are relevant for the formation of SFCs.¹² The first is the SFCN, which was created with the explicit purpose of encouraging freedom caucus formation in state legislatures. Though SFCN President Roth has claimed that the group does not instigate the establishment of SFCs, he has also suggested that it does play an important role in their formation.¹³ In addition, lawmakers in some states have credited the SFCN with recruiting them to create freedom caucuses, and others have explicitly identified their freedom caucus as part of the SFCN.¹⁴

Hypothesis 4: An SFC is more likely to form in state legislative chambers targeted by the SFCN.

The other type of outside entrepreneur likely to encourage the formation of a caucus is a member of the US HFC.¹⁵ Lawmakers from a state that is represented by a member of the HFC could be spurred to follow suit in their own chamber, but anecdotal evidence also suggests that some caucus members have actively lobbied Republicans from their state’s legislature to create freedom caucuses (e.g. Juhlin 2023; Wolfson 2023b).¹⁶ Further suggestive of this hypothesis is that, in a few states, the state freedom caucus’ inaugural press conference featured an HFC member from their House delegation (e.g. Alexander 2022; Juhlin 2023; Ulrich 2022). One SFCN staffer described HFC lawmakers as “absolutely essential” to SFCs, serving as “mentors” to state lawmakers in a caucus (Interview with SFCN staffer, January 31, 2024). In Illinois, caucus founder Chris Miller – whose wife, Rep. Mary Miller, is a member of the US Freedom Caucus – went so far as to call the state caucus an “umbilical to the [House] Freedom Caucus” (Adams 2022).¹⁷

Hypothesis 5: An SFC is more likely to form in a state represented by one or more members of Congress who are in the HFC.

We use logit regressions to test these hypotheses. The dependent variable is a dichotomous measure of whether or not a state has an SFC, with the population being all US states except for Nebraska’s unicameral and nonpartisan legislature. To ensure the results are consistent across states, and because some of the predictors apply only to the lower chamber, we exclude Washington’s freedom caucus, since it is the only SFC that was formed exclusively in the state senate. That leaves 16 lower chamber caucuses established between 2017 and August 2023 in a population of 49 states.

To test the first two hypotheses, we use Shor and McCarty’s NPAT scores, which estimate individual lawmakers’ ideology from their roll-call votes (Shor 2023).¹⁸ We measure ideological heterogeneity with the standard deviation of NPAT scores for all Republicans in each state House; larger numbers should have a positive effect if the first hypothesis is correct. To measure party distance, we use the absolute difference between the median NPAT scores of Republicans and Democrats; if the second hypothesis is correct, this variable should have a negative coefficient.

For the third hypothesis, because state freedom caucuses often point to the Speaker specifically as the source of marginalization of rank-and-file conservatives, we use an index of Speaker power developed by Mooney (2013), updated to 2018, that takes into account the formal powers of the speakership in each chamber. Larger values represent more powerful speakers, so a positive coefficient would be consistent with this hypothesis.¹⁹ To test whether the SFCN encouraged state-level caucuses to form, we use a proxy measure of whether the state sent lawmakers to the group’s “inaugural gala” in Atlanta, Georgia on December 14, 2021 (Conservative Partnership Institute 2023). The event was scheduled to coincide with the announcement of the first new SFCN-affiliated caucus (in Atlanta), possibly to encourage other gala attendees to create caucuses in their own state chambers.²⁰ Since some SFCs had formed before the SFCN was established, we divide our analysis into two separate periods (2017–2020 and 2021–2023) and test the variable only for SFCs formed in the second period. States that had formed a freedom caucus in the first period were excluded from the second period.²¹ Finally, as a test of the fifth hypothesis, we include a dummy variable coded 1 if the state has members of the HFC in its state delegation and 0 otherwise.

We also add several control variables. Since legislative professionalism is associated with a number of important facets of state legislative politics, could conceivably affect the capacity or incentive of lawmakers to form SFCs, and has been tested in other studies of caucus formation in state chambers (e.g. Clark 2019), we include the Squire Index measure of the professionalism of state legislatures (Squire 2017). We also include a dichotomous variable measuring whether the state legislature has a Republican majority in case party control is associated with the creation of freedom caucuses.²² Since this variable predicts the formation of SFCs in the first period perfectly, we are only able to include this control in our models for the second period. Using alternative control variables such as state-level ideology, partisan lean (Cook’s PVI), and Republican seat share (e.g. Clark 2019) does not substantively change the results. Finally, given the importance of race in conservative politics (Blum 2020; Gervais and Morris 2018; Mason 2018; Noel 2013), we included the percentage of each state’s white non-Hispanic population as an additional control variable.²³

Table 3 presents the results of the logit analysis. The baseline model for the states that formed in 2020 or before provides support for the first hypothesis: chambers with more ideologically heterogeneous Republican parties were more likely to see the formation of freedom caucuses. This suggests that lawmakers who created SFCs were motivated by a desire to increase ideological agreement within their party. None of the other hypotheses explain the appearance of these caucuses before 2021, however: while the signs of other predictors in the model are in the expected direction, they do not reach statistical significance. Adding controls for state legislative professionalism and the racial composition of the state population does not change these results.

Table 3. Likelihood of state Freedom Caucus formation

Note:

* p < 0.1

** p < 0.05

*** p < 0.01.

The results differ when looking at caucuses formed starting in 2021. Ideological heterogeneity is no longer statistically significant.²⁴ However, there is evidence for Hypothesis 5: though the inaugural gala variable is not statistically significant, having a member in the state delegation from the US House Freedom Caucus increases the likelihood that the state legislature forms its own freedom caucus, and the variable is weakly statistically significant (p < 0.1).²⁵ In other words, after 2020, Republican state lawmakers who created SFCs were not necessarily driven by concerns about their party’s lack of homogeneity but rather were likely inspired or lobbied by key members of Congress to form their own caucuses. Adding controls for professionalism, GOP majority status, and race does not meaningfully change the results.

To illustrate the substantive effect of these variables, Figure 1 shows the change in predicted probability in the first, “base” model (2017–2020) of forming an SFC as party heterogeneity increases, comparing states that have one or more HFC members in its delegation with states that have none. The effect is substantively significant. For states with one or more US House members who are in the HFC, a one standard deviation increase in ideological heterogeneity from the mean (i.e., from 0.28 to 0.39) increases the probability that an SFC will form from 15% to 33%.

Figure 1. Marginal effect of GOP heterogeneity on the probability of SFC formation (base model 2017–2020). The two lines show the predicted probability for states that have one or more US House members in the House Freedom Caucus (solid) or have no members in the Caucus (dashed).

Who joins state freedom caucuses?

A second question raised by the emergence of state freedom caucuses is which individual lawmakers are more likely to join them. The answer to this question matters because, as with any legislative caucus, the motivation for legislators to become members of an SFC reflects the caucus’ overall mission and, in turn, is likely to drive its strategy and tactics. Lawmakers who are members of state freedom caucuses may also have advantages over colleagues who are not, as is the case for members of organized factions in the US House.

Drawing from the goal-oriented theoretical approach of Richard Fenno (1973) and prior research on legislative caucuses, we test three claims for why a lawmaker might be more likely to join their chamber’s state freedom caucus. First, the caucus may satisfy a lawmaker’s policy goals by acting as a voting bloc or bargaining unit in the development of legislation. Prior research has found that legislators at the spatial extremes of their respective parties have the greatest incentive to join ideological intraparty organizations in Congress (Bloch Rubin 2017, 22), and this seems especially likely for state freedom caucuses since they are explicitly ideological and tout conservative-leaning policy agendas. We should expect more conservative lawmakers to join an SFC if this claim is true.

Second, a state freedom caucus may achieve a legislator’s electoral goals by providing them with electoral resources and serving as a brand to help them get votes from like-minded constituents, just as ideological caucuses do in Congress (Clarke 2020). They may also believe that they benefit electorally, at least in primaries, by being labeled as a “true” conservative fighting against “establishment” Republicans. If lawmakers who are more electorally vulnerable in primaries are more likely to join, that would be consistent with this hypothesis. Finally, given that intraparty organizations in Congress are a valuable tool for rank-and-file legislators to leverage their numbers in negotiations with party leaders (Bloch Rubin 2017), a state freedom caucus may appeal to a lawmaker’s influence goals. Incumbents who have less internal influence in their chamber, such as those with less seniority or who do not serve in a leadership position, should be more likely to join an SFC if this hypothesis is correct (Thomsen 2017; see also Rouse, Hunt, and Essel 2022).

To test these claims, we employ a logit model, with the dependent variable equal to 1 if a lawmaker joined her state’s freedom caucus when it was first established, for all SFCs formed between 2017 and 2021 for which caucus membership could be discerned.²⁶ To estimate the influence of policy goals, we use a variable measuring the aforementioned Shor/McCarthy vote-derived NPAT estimates of ideology for each lawmaker. This variable is larger for more conservative state lawmakers, so it should be statistically significant and positive if the hypothesis is correct. (Since NPAT scores are available through 2020, using this measure for chambers after that year excludes lawmakers elected for the first time in 2020.) We use American Conservative Union (ACU) lifetime scores instead of NPAT scores for two state chambers: in the Texas House, because NPAT scores predicted membership in the state freedom caucus perfectly (itself strong evidence for the policy goal hypothesis); and in the Idaho House, because a full model could not be estimated using NPAT scores.

To test whether electoral concerns motivate membership in a new SFC, we include a variable measuring the (logged) percent of the two-party vote won by lawmakers in their most recent primary election. Legislators who win their primaries by more narrow margins will presumably be more concerned about future electoral challenges from the right, and so have a greater incentive to join an SFC. We also include a variable measuring the (logged) percent of the two-party vote garnered by lawmakers in their most recent general election, since it is possible that incumbents who win their seats by more narrow margins are from more competitive districts and, as a result, have a disincentive to join a caucus that would likely be perceived by the median voter as too ideologically extreme. Finally, to test the importance of internal influence, we use two dichotomous variables: the first is equal to 1 if the legislator is the chair or ranking member of a committee, and the second is equal to 1 if the lawmaker is a party leader. We also include a variable measuring the number of terms served by GOP incumbents, under the assumption that those who have served longer have more influence than newly elected lawmakers and are thus less likely to join an SFC.

The results of the regressions for seven freedom caucuses that were formed between 2017 and 2021 are shown in Table 4. We find the most support for the claim that conservative policy preferences drive membership in a new SFC. Measures of legislator ideology are statistically significant (p < 0.1 or better) in six of the seven states – strong evidence that more conservative lawmakers are generally more likely to join their chamber’s newly formed freedom caucus regardless of state or chamber. In terms of electoral goals, the evidence is less strong. In the Washington state senate, the variable measuring election margins in primaries is negative and statistically significant, so more lopsided primaries are associated with a lower probability of joining a state freedom caucus, as hypothesized. However, while the variable is also negative in five other chambers, it is not statistically significant. Furthermore, and counterintuitively, the variable measuring election margin in a general election is also negative, and at least modestly statistically significant (p < 0.1) in the North Carolina House and the Texas House. It may be that incumbent Republicans in these chambers who won by more narrow margins believe that joining the freedom caucus will help increase Republican voter turnout enough in subsequent general elections to compensate for the loss of support from swing voters.

Table 4. Predicting the likelihood of joining a new state freedom caucus (2017–2021)

Note: Tested for incumbent Republicans only. Committee leaders in the Nevada House could not be identified. ACU scores are used for the Texas House because NPAT scores predict caucus membership perfectly, and for the Idaho House because coefficients for the full model could not be estimated using NPAT scores. A full model for Nevada could not be estimated without excluding the term in office variable.

* p < 0.1

** p < 0.05

*** p < 0.01.

There is more evidence for the claim that lawmakers are encouraged to join an SFC because they seek greater internal influence. The variable measuring whether a lawmaker held a top party leadership post had to be excluded from all the models because in no state did a leader join a freedom caucus; in other words, the caucus was unattractive to those who already had positions of power in the party. Seniority is statistically significant in two states (Idaho and North Carolina), and the coefficient has the expected (negative) sign in every state where it could be tested. Interestingly, serving as a committee leader (chair or ranking member) is only statistically significant in Idaho and Washington, and the coefficient in those two states is positive, meaning those in committee leadership positions were more likely to join a freedom caucus, not less so.

Discussion and conclusion

SFCs present a unique opportunity for scholars of American politics to examine the causes and consequences of organized party factionalism beyond the US Congress. We find evidence that early caucuses were more likely to emerge in more ideologically heterogeneous parties. This is not surprising, given that the stated purpose of SFCs is to make state legislative parties more uniformly conservative. However, caucuses formed in 2021 and after are more often the result of a top-down entrepreneurial effort involving national political actors affiliated with the HFC. In terms of individual membership, we provide strong evidence for the claim that conservatives are more likely to join new SFCs. In addition, in some state chambers, members who lack influence in the legislature or who are more electorally vulnerable are also more likely to join freedom caucuses. Taken together, this suggests that lawmakers join these organizations first and foremost to achieve their policy objectives, but also sometimes to help get reelected or achieve greater influence in their chamber.

More broadly, the formation of these caucuses suggests the continuation of the populist, anti-establishment, anti-government sentiment that reflects the spirit of the Tea Party movement in its quest to remake the GOP, harnessed by Donald Trump in his successful bid for the 2016 presidency (Blum 2020; Gervais and Morris 2018). Their activity and occasional conflict with GOP leaders in their chambers may also represent an escalating struggle within Republican state parties between leadership-aligned forces emphasizing ideological moderation in the pursuit of electoral and policy success and far-right members expressing a desire to turn the party into an agent of conservative-populist change, regardless of short-term political costs.

Further research is needed to better understand these developments, as well as to test our hypotheses for other state caucuses that are associated with congressional equivalents. The Progressive Caucus, for example, has a nonprofit affiliate that has brought state lawmakers to Washington to develop political strategy with members of Congress (Congressional Progressive Caucus Center 2020). Another important area of future research is how and why state freedom caucuses may change in size, influence, and strategy over periods of shifting political contexts, such as a flip in party control of the chamber.

Such research faces nontrivial challenges, not least the difficulty of obtaining caucus membership at the state level. Nonetheless, using the data we have been able to gather on state freedom caucuses, we have provided evidence that speaks to the perceived value of intraparty organizations to lawmakers in a time when most state governments face one-party control. The findings presented here also show the importance of considering the influence that members of Congress and national political actors can have on the decision of state lawmakers to create and join intraparty caucuses. Furthermore, even in an era of unified party state government, the proliferation of these caucuses is a reminder that internal divisions can emerge within parties – putting to the test whether a state legislative party can function in spite of those divisions or if, as one set of political scientists put it, it is “so paralyzed by factionalism that it ceases to cooperate as a party at all” (Koger, Masket, and Noel 2010).