Interest Group Factions in the Expanded Democratic Party

Casey B. K. Dominguez

Department of Political Science and International Relations

University of San Diego

This paper begins with the premise that political party organizations should be treated as open, fluid networks of actors. The "Expanded Party" (Bernstein 1999) encompasses party committees and officials, but also activists, donors, consultants, candidates, officeholders, and interest groups. It defines interest groups that have an orientation to the Democratic party as those that, when they give money, give more than half of that money to Democrats. The paper then looks for factions among these interest groups in the Expanded Democratic party. The paper defines factions to be durable coalitions of different actors that try to influence the party's internal organization, and have some rational basis for doing so. Examining the 2002 congressional primary candidates that were supported by Democratic party-oriented Political Action Committees, it finds that there is a consensus on one primary candidate in about 70% of primary races. This finding undermines the conventional wisdom that Democrats are frequently deeply divided, especially in primaries. In the races where these PACs do not support the same primary candidate, there are no discernible factions organized along recognizable issue lines. When there is any disagreement about a primary candidate, it is usually the case that Leadership PACs and labor groups split their support among multiple primary candidates. The paper concludes that more work needs to be done defining the boundaries of the Expanded Party, and defining and measuring faction.

Paper prepared for presentation at the Western Political Science Association Annual Meeting, Las Vegas, NV, March 8-11, 2007
Interest Group Factions in the Expanded Democratic Party

Upon first inspection, many political observers would agree with the statement, “the Democratic party is factionalized,” or perhaps with the corollary, “the Democratic party is more factionalized than the Republican party.” Both of these statements imply that the Democratic party is so factionalized that it can’t agree on what it is, or what it stands for, or who its leaders are. Intuitively, one might say that “labor” or “environmentalists” are some of these supposed factions. These nuggets of conventional wisdom beg the question: are there factions in the Democratic party? Do interest groups, or groups of interest groups, make up some of those factions? The pages that follow apply at least one reasonable definition of faction to the interest groups that loosely affiliate with the Democratic party, and conclude that at least at a broad level, there are no durable, sensible, identifiable factions among them. Rather, interest groups do what they do best: pursue their own individual group’s agenda, not necessarily in concert with any other group(s). While this may be somewhat surprising, perhaps the more startling conclusion is that, at least in primary elections, interest groups in the Democratic party do not fight against each other all that often. Since this is a preliminary study, the paper concludes with the caveat that more work can certainly be done on these questions.

In order to fully understand the operation and potential influence of modern American political parties it is important to look beyond their official organizations and committees and examine how various actors operate in parties’ informal, network-like structures. Although empirical studies of party organization have focused on the official representatives of the party (Silbey 1990; Herrnson 1988; Cotter et al. 1984), theories of party structure support a broader definition that includes elite party actors.[1] An emerging literature is attempting to reconceptualize the roles that parties play in elections. These scholars base their approach on the assumption that the party structure is open and fluid, and at different times, different individuals, in different roles, can be influential.[2] For example, several recent studies (Doherty 2005, Masket 2004, Monroe 2001, Schwartz 1990) that focus on the interpersonal connections between party officials, legislators, contributors, and activists conclude that conceptualizing a party as a network rather than a hierarchy allows for a much fuller understanding of how it actually works. According to this perspective, a definition of the “Expanded Party” (Bernstein 1999) can encompass both the party’s formal hierarchical structures and the loyal big-money donors who support the organization, the activists who staff candidate campaigns, the highly partisan hired-gun strategists who help party candidates compete for office, and at least some of the interest groups that support the party’s efforts. This definition is totally compatible with more traditional definitions of party (for example, Aldrich 1995) that argue that a party is fundamentally composed of office-seekers and benefit-seekers. Its contribution is to broaden the operational definition of benefit seekers and ask more questions about how they relate to, and influence, the office seekers.

Recent work has shown that it can be fruitful to consider how elite party participants in the Expanded Party can affect the political process, and particularly the politics of party nominations. Recent research into the endorsements received by candidates for presidential nominations has shown that candidates who receive support from a broad swath of party elite endorsers are more likely to win the nomination (Cohen et al. 2001). In addition, several studies have shown that staff (both campaign professionals and those who work in political or members’ offices in government) are overwhelmingly party loyal and that when they select candidates or members to work for, that can be a signal of party insider support for that person (Monroe 2001; Bernstein 2000; Bernstein and Dominguez 2003).

With this expanded view of the party organization, it is important to consider the roles that outside groups like organized labor and conservative Christians play in performing essential party functions like recruiting, funding, and getting out the vote for the party’s candidates. Cohen et al, for example, include in their presidential candidate endorsement lists such important interest groups as the UAW in the Democratic party and the NRA in the Republican party. But asking how “interest groups” fit into the Expanded Party begs the question of how we should decide which groups qualify as “partisan”. We might say that some interest groups are more integrated in to the political parties (Skinner 2004), but others may act more as competitors to the party establishment. Once the boundaries of the Expanded Party have been established, a second, and more important set of questions arises: to what degree do these party-oriented groups divide into factions? And if there are identifiable factions in the Expanded Party, how often and under what circumstances do they try to remake the party in their own image, regardless of the wishes of other important party constituents?

These are the questions that will be addressed in the following pages. I argue that interest groups themselves are not equivalent to factions. Factions should be considered to be groups of actors who regularly engage in similar behavior in competition with other similar groups. I identify some factional groups and behavior in the Expanded Democratic party, but conclude that the party is more fractured than factionalized.

Faction. As Bernstein (2004) notes, the literature on party factions in the American context is practically nonexistent. V.O. Key explained the degree of factionalism in southern Democratic one party rule, but as noted by Benedict (1985), his definition of faction, “a combination, clique or grouping of voters and political leaders who united at a particular time in support of a candidate” is so general that it would imply that modern candidate-centered politics is totally factionalized (Benedict 1985, 365). There is a literature on the formation and maintenance of durable factions in multiparty contexts, particularly in Italy and Japan (see, for example, Cox, Rosenbluth and Thies 2000), but because the electoral context is so different, it is difficult to make analogies from this body of work to the American context. And even in a comparative context, there does not seem to be a scholarly consensus about how to identify and measure factions within party organizations. Belloni and Beller (1976) organize some of the literature on party factions in a helpful review. They found, albeit thirty years ago, that most studies of factions were specific to particular countries and circumstances. They also found, perhaps because of each study’s narrow focus, that each defined faction somewhat differently. Some authors argue that factions only exist prior to the formation of parties, others that factions only exist within fully functioning parties. Some characterize factions as issue-oriented, while others say that they arise around personalities. All of the studies discussed by Belloni and Beller imply that a faction must be organized, but there is no apparent agreement about the criteria for that organization. Benedict (1985) seems to argue that clientelism is an important component of factions, but it is not clear how essential that criterion is to the definition.

Contemporary journalists also use the concept of faction to refer to different phenomena. A recent blurb in the Tampa Tribune asserts that “incoming Governor Charlie Christ and U.S. Senator Mel Martinez represent different factions of Florida’s Republican party…a ‘moderate populist’ in the future governor and a ‘religious conservative’ in the new party chairman.” This definition of faction implies a broad, ideological agreement and an existence that goes beyond any one leader or candidate. Another recent Washington Post article describes the post election blame-game between moderate and conservative Republicans. It lists several potential groupings of Republicans:

Moderate Republicans quickly concluded that the party needs to be more moderate. Conservative Republicans declared that it should be more conservative. Main Street is angry at Wall Street, theo-cons are angry at neo-cons, and almost everyone is angry at President Bush and the GOP congressional leadership.(Grunwald 2006).

Here, faction implies both loose groupings of ideologically similar Republicans, but also seems to imply that the President and his allies, and separately, the Congressional Republicans, might comprise distinct factions of the party. Another article, this time about the Democrats, refers explicitly to the factions that Speaker Nancy Pelosi will encounter in the House: Blue Dogs, New Democrats, and Progressives (Epstein 2006). This definition of faction refers not to loose ideological tendencies in the parties, but rather to self-conscious organizations of legislators.

It could be argued that a good deal of research on the US Congress is inherently interested in the formation and behavior of factions within the parties in government (particularly studies of caucuses, Leadership PACs, or most famously the “Conservative Coalition” of the middle-twentieth century). But these studies, while relating ideology to voting behavior in Congress, do not explore potential relationships between those factions-in-government, and the broader organizational party. One could ask, for example, whether factions-in-government relate to regular, “factional” connections between coalitions of legislators, the campaign staff they hire, the interest groups with whom they affiliate, and the fellow candidates to whom they lend their support. Or, more broadly, one could ask to what extent the parties are“controlled” by a given faction. How does a faction achieve that power within the party? What is the structure of the competition within a party? Which actors or institutions mediate conflict between these factions? To what extent are the parties really serving to broker among competing groups, to forge compromises between their constituents’ interests? Questions like this are intriguing, but they beg the question of how to identify a faction in the first place.

Given the presumption that the party involves multiple types of actors (candidates, donors, activists, staffers, strategists, groups) one could start looking for factions almost anywhere. My focus here is on interest groups. I ask whether there are identifiable factions among the interest groups that are loosely aligned with the Democratic party. To do this it is necessary to distill some of the essential qualities of a faction from the many definitions offered by scholars and political observers. I choose three of these qualities. The first is that a faction must be durable to at least a modest degree. It would be useless to observe, and really impossible to catalogue, the infinite pairings of groups that come together to support candidates or policies. So a faction must be observed to act in concert in multiple instances. Second, a faction should demonstrate some coordination, or give rise through its behavior to a belief that communication might exist between its constituent parts. That is to say, a faction must have some rational basis for acting in concert. It could happen that the Service Employees International Union and the gay-rights organization Human Rights Campaign support the same candidates in several different election cycles. But is there any rational basis for believing that they might be doing so on purpose? Do they have any policy or ideological goals in common? Are there any spoils of office that would benefit both? Third, a faction is inherently in competition with other similar groupings or potential groupings for control of its umbrella party’s agenda, message, or personnel. Coming together on a common softball team does not mean that two groups have formed a faction (though such informal connections could certainly form the basis for one). A party faction must try to influence the internal workings of the party in a meaningful way.

It is assumed in these criteria that a faction is not equivalent to an interest group itself. Faction implies that there is a loose, temporary amalgamation of interests who share at least one common goal. Because an interest group is a permanent organization, it cannot meet that definition.

Identifying Groups in the Expanded Party

Taking a step back from the problem of faction, we must start with an equally troubling problem, that of defining the groups that should count as part of the party’s coalition. Some groups may pursue a fully non-partisan, issue-oriented strategy, supporting candidates of both parties in order to gain access to them, and rewarding members of whichever party embrace their agenda more fully at any given point in time. These groups probably should not be considered to be part of either party. Other groups may, over time, or even at their inception, believe that their issue agenda will be best served by electing more members of a particular political party, and work uniformly to advance that party’s interests as a means to further their own. Some may, by their behavior, clearly prefer one party, but still act in clearly independent ways, while others may subordinate their primary issue positions to the good of the party. An important question for scholars of the Expanded Party is: which groups pursue which strategies, and how can you tell the difference? Certainly it is easier to tell with some groups than others. MoveOn.org, The Sierra Club, EMILY’s List, and the AFL-CIO all appear on lists of the biggest donors to Democratic Candidates. But what about groups that don’t give money, that give less money, or that give some money to Republicans?

The analysis that follows lays the groundwork for addressing these questions by drawing some initial boundaries around the Expanded Democratic Party. For reasons discussed below, I adopt a broad definition of which groups belong in the Expanded Party, though there is undoubtedly a continuum of “party-ness” not captured here. The universe of groups that will be examined here is defined by the groups’ behavior. In particular, I will examine the “party-ness” of Political Action Committees (PACs) that gave money to at least one candidate for a Democratic nomination for Congress in an open seat in 2002, and that have a general orientation toward the Democratic party. Why define the universe this way?

Nominations. If, as Aldrich (1995) and others tell us, the party is very critically made up of those who hold offices under its label, the party’s choice about who holds those offices is a fundamental one. It makes sense to examine the Expanded Party through a focus on nomination politics because in choosing a nominee, the party names an individual, with all of that person’s experience, policy positions, and symbolic characteristics, to represent the party as a whole. By supporting one candidate over another in a nomination contest, a group is therefore trying to affect the very composition of the party and what policies it fights for. By looking at nominations, any factions that are identified will be sure to meet the third criteria above, that of clearly trying to influence the internal operations of the party.

Open seats. In races to fill seats with no incumbents, the party actually has an opportunity to nominate a new candidate and stage an internal debate about the future of the party in that location. One certainly could examine interest group support (or rare opposition) to the re-nominations of sitting congressional incumbents. But because even contested primaries for congressional seats are so rare, and because interest groups have a strong tendency to support incumbents rather than give to challengers, it might be less fruitful to look at incumbent re-nomination contests for evidence of broad party cleavages. I leave that task for follow-up studies.