[Note: This post is a lot longer than I expected. It will probably be of interest to students taking an American politics course or, specifically, an interest groups course. As of 12/4/09 I don’t have the bibliographic references listed, but will add later on.]
Concerns about groups in American politics date back at least to Madison’s (1787) concerns at the founding of the republic, when he warned against the danger of factions in our national politics. Factions were, he lamented, inevitable in a free society, and posed one of its most serious risks. Tocqueville (1835) noted at the start of the Jacksonian era that Madison’s factions were one of the defining characteristics of the American society at all levels. It was not, however, until Bentley (1908) that attempts to observe, classify, and methodically examine groups in political life entered into the domain of political science. Since that time, the focus on and understanding of group politics has grown exponentially. Despite the empirical advances in understanding, the normative questions about whether the group system is good for democracy remain unsettled. Further, the biggest gaps in our understanding seem to persist in regard to measuring interest group influence and bias within the group system. Although the scholarship on interest groups is expansive, I will attempt to survey the field by organizing it into studies on group formation, maintenance and growth, mobilization, the shape of the interest group community, the relation of groups to different government actors, normative issues, and suggestions for further inquiry
Truman (1951, 1958) made a pluralistic argument for groups, contending that they form as a result of disturbances or threats to society. Truman perceived of disturbances as impacting segments of society. When individuals were impacted by a disturbance, Truman assumed that they would band together to respond. Thus, groups would emerge in waves, and would only exist for as long as was needed to restore an equilibrium regarding the specific disturbance. Under this conception, groups are short-lived and their members are all equally mobilized for collective action. Olson (1965) noted that this view of groups is at odds with the majority of people that do not participate. From this, Olson challenged Truman on the inevitability of individual participation. Olson argued that rational individuals would free ride on the benefits achieved by the collective. If an individual could gain the benefits of collective action, but could do so without acting, then they could rationally be expected to abstain from acting. Therefore, individuals could not be assumed to rationally act collectively given a disturbance (more on incentives below).
Walker (1983, 1991) goes even further, arguing that it is not necessarily a threatened collective that alone forms a group. Without the necessary resources and organizational efforts of a leader, groups could not easily form in the United States. Walker contends that without the resource mobilizing efforts of patrons, groups remain merely a possibility. Salisbury (1969) emphasizes the necessity of an entrepreneurial leader for group formation. While Truman emphasizes the goal-oriented aspects of collective action, it is the work of scholars who succeed him that break apart the group, and differentiate between patrons, leaders, and members. Each of these different individuals agrees to join groups for different reasons, as I explain below.
MAINTENANCE AND GROWTH
If collective action is not inevitable, then we must move beyond Truman’s explanation to account for it. At the heart of Truman’s view is the assumption that all group members are motivated by the same reasons for joining a group. Olson explains why this view will not result in action at all. In its place, Olson argues that selective incentives for individuals will motivate them to participate in action that achieves collective benefits for the group. This solves the free rider problem. For example, a welder knows that his contribution will not add much utility to the union’s ability to secure higher wages. He abstains from joining. The union knows that if everyone behaved this way, they would have no leverage to raise wages. So, the union offers a scholarship fund for welders’ children, and a union hall for socializing after a hard day’s work. The union has offered other incentives, which the welder could not obtain without joining, to encourage union membership, which allows it to secure higher wages.
Clark and Wilson (1961) call these inducements material incentives. They also suggest that different types of incentives will motivate different people, at different times, and for different types of action. Clark and Wilson argue that there are different types of goals that animate group action – material, solidary and purposive. Material goals involve direct material benefits to individuals, such as tax benefits, earmarks, and higher wages. Solidary goals are those related to being in association with other people. And, purposive goals are those that achieve a policy or ideological end, such as voting rights, partisan gains, or changes in foreign policy. Clark and Wilson argue that groups tend to align primarily with one type of goal and offer a range of incentives, depending on the potential membership, to encourage people to join and remain in a group.
Going back to Salisbury and Walker, differences in the motivations and incentives of members differ from those of organization leaders, or their patrons. Salisbury proposes an exchange theory, in which leaders gain materially in exchange for the specialization and organizational benefits they provide the group in order for it to survive. Patrons, on the other hand, provide the necessary resources to initiate and continue groups (Walker 1983, 1991). Patrons themselves have many motivations, such as solidarity. Often, however, patrons are government agencies that wish to legitimize certain groups over others, or to provide more means of interest intermediation to diffuse social conflict. Thus, groups require individuals associated with them to have differentiated motivations and goals.
As Rosenstone and Hansen (1993) argue, individuals generally do not spontaneously organize into politics. Rather participation is mobilized by strategic politicians who attempt to shape the calculus of politics by including more individuals. This perspective resonates with Schattschneider’s (1960) view of politics as conflict. If politics is essentially a series of conflicts, he warns, “Watch the crowd.” Schattschneider argues that the outcome of a conflict is determined by which side is the largest. Because conflict is contagious, and crowds are impossible to control, the scope of the conflict is expanded generally by the losing side. The more privatized a conflict is, the easier it is for one side to control the outcome. Therefore, interest groups often attempt to expand the scope of conflict to gain more leverage. Rosenstone and Hansen concur, stating that it is easier to add new partisans to your ranks than to convert old partisans from the other side.
SHAPE OF THE INTEREST GROUP COMMUNITY
Because Truman expected groups to die off when disturbances had been corrected, he predicted that the overall size of the interest group community would remain relatively small, fluctuating according to threats but never steadily growing. However, this has been far from the case. Groups have proliferated since the 1950s, often as a result of the expansion and increased complexity of government operations. Until the 1970s, interest groups were conceived of as existing in iron triangles with agencies and congressional committees. The view shaped the notion that relationships between the three entities were stable, mutually interested, and relatively impervious to outside intervention. Heclo (1978) challenged this conception, arguing that government is far too complex to be administered by rigid iron triangles. Instead, he contended that “issue networks” characterize the group system, with information specialists in congress, the bureaucracy and the public, shaping policy and government operations. The issue networks bind individual actors and groups in policy domains not by inflexible relationships, but by affinities toward particular types of policy. Browne (1990) went even further, claiming that government had grown so complex since the 1970s, and undergone so many changes (e.g. post-Watergate congressional reforms) that a conception of “issue niches” was more appropriate. In either case, the bird’s eye view of the interest group community sees it as intertwined with various government actors and the public around common policy interests. But, what kinds of groups are intertwined in these networks and niches?
Berry (1999) notes the dramatic rise of liberal citizens’ groups since the 1970s. He notes that such groups have displaced liberal material groups, like unions, in their pursuit of postmaterial policy agendas, such as reproductive rights and environmental policy. Still, Schlozman and Tierney (1986) note that occupational groups dominate the landscape. This comports with Schattschneider’s (1960) concern about the “upper-class accent” of the interest group system. Many other prominent scholars, such as Walker (1991) and Baumgartner and Leech (1998), also note this bias. However, the empirical fact of occupational group proportion (as opposed primarily to citizen groups) has often been equated with the over-representation of occupational interests. Whether occupational interests are overrepresented or not, this is not the correct evidence to support the argument. (I will below return to whether this bias in the proportion of occupational groups corresponds with disproportionality in the representation of occupational interests.) Nonetheless, the terrain of who’s who in the interest group universe has frequently been mapped.
GROUPS AND GOVERNMENT ACTORS
I will here briefly address the relationships between interest groups and various government actors. Much of the literature has focused on the relationship of groups to Congress. These studies often focus on the issues of access and influence to legislators, the role of groups in elections, and case studies of particular issues. Denzau and Munger (1986) consider the role of interest groups and constituents vis-à-vis legislators. They find that unorganized interests are indirectly represented by groups before their representatives, because the demands of the constituency set the boundaries for acceptable policy positions of the legislator. Therefore, groups will not solicit and influence a legislator on policy positions that would be unacceptable to her constituents. Instead, only groups with policy goals within the range of acceptability of constituents will lobby legislators and, thus, indirectly represent unorganized interests in the district. Other studies examine the nature of group behavior in relation to legislators. Hojnacki and Kimball (1998), for example, find that groups will lobby allies during policy formation and fence-sitters during final passage. Hall and Wayman (1988, 1990) found that PAC donations buy access to legislators, particularly in committee, but not votes. Hall and Deardorff (2006) expand on this by casting contributions as a “subsidy” for friendly legislators, where contributions in the aggregate might influence a policy agenda but not specific votes.
The literature considers group relations to other government actors, as well. Caldeira and Wright (1990) attempt to identify group influence in the Supreme Court. They find that the specific arguments or positions do not sway justices to grant cert to cases, but rather the overall volume of briefs is used by justices as a gauge of the case’s saliency. Research on the Court is, of course, more constrained than research on Congress because of data restrictions and general opaqueness of court operations. One area where data is widely available, and increasing, is the bureaucracy. Much of the research on groups and the bureaucracy is concerned with the ways in which groups influence the interpretation of congressional mandates in rulemaking. One view perceives of groups as providing an additional oversight role of the agencies for congress. This accords with McCubbins and Schwartz’s (1984) view of fire alarm oversight. Kerwin and Furlong (1995, 2005) also examine the role of interest groups in rule-making, finding that much of group influence is difficult to consider because it occurs at non-public stages of rulemaking.
I also briefly note the recent significant contribution of Nownes (2006), which considers the range of lobbyist behavior in relation to the executive, legislature, and bureaucracy, on all levels of government (local, state, and federal). Nownes, through considerable fieldwork across the country, finds that lobbyists basically work to three main ends: policy, procurement, and land-use. What is striking is that lobbyists essentially work the same way at all levels. The difference in who is lobbied lies in the goal. Those seeking policy goals lobby the legislature, those seeking procurement lobby the bureaucracy, and those working for land use lobby the executive. Land use is not a major issue at the federal level, but it is at the state and local levels. Additionally, procurement occupies a significant amount of attention of the lobbyists in Nownes’ sample, at all levels.
A major, unresolved dimension to interest group studies is the normative impact on democracy and representation. Mills (1956) examined the power elite, and was convinced that most decision making was done by those at the top without the involvement of the citizenry. Truman’s (1951, 1958) and Dahl’s (1956, 1961) pluralistic system conceived of groups as essentially healthy and necessary for democracy. Groups emerged to respond to particular needs before government could, and everyone had the opportunity to participate in groups. Further, the pluralist view holds that the answer to “who governs?” is really dependent on which issue and at what time the question is asked. No one person has control in a pluralistic system and therefore power is balanced in the competition for the definitive aggregation of preferences. Schattschneider (1960) famously took exception to this view, noting that what emerged in the pluralist competition was a bias toward the powerful and wealthy. Olson (1965) also noted that it was far easier for certain types of groups to mobilize, such as business associations or unions, because they had a captive potential membership, ready resources, and easily defined goals. Thus, those individuals in already marginal positions, such as the homeless and the poor, have a more substantial collective action problem to overcome.
Following the initial reaction against pluralism, Lowi (1969) raised concerns about what he called “interest group liberalism.” The problem that he saw with the group system, besides being too large, was that it created fleeting and extra-legal ways around policy formation. The proliferation of groups, Lowi argued, prevented the strengthening of the laws and public institutions that should be the arbiters of public policy. Of course, the pluralists might refer to this characteristic as responsiveness and flexibility. Riker (1982, 1988) follows this concern and refers to the rise of groups as consistent with the populist tendency in American democracy. Riker agrees that the populist trend weakens institutions. Rauch (1994) is concerned that the entrenchment of groups causes a sclerosis in the American system, where entitlements become outdated and choke the ability of government to respond to new concerns. However, other scholars look to the group system and see it as an augmentation to democracy. For example, Denzau and Munger (1986) see groups as responsive to the preferences of a legislator’s constituency (see above). McCubbins and Schwartz (1984) see a role for groups in the oversight congressional oversight process.
One of the ongoing themes in the interest group literature, which is related to the concerns of Lowi, Riker, and Rauch, is the matter of influence and bias. Schattschneider’s classic concern about the bias in the system is echoed by Schlozman and Tierney (1986) and Walker (1991). However, as mentioned previously, the numerical advantage of occupational groups over citizen groups does not necessarily equal an advantage in influence. Group formation is often spurred by government activity, particularly regulation. Nownes (2006) notes that more business groups lobby because of the simple fact that they are more heavily regulated, and increasingly so. In order to rigorously determine whether certain groups, such as occupational groups, are disproportionately influential, one would first need to determine the entire range of interests. Furthermore, there is a problem in conceptualizing what influence is. Is influence the same as power? Is it achieving a particularly policy goal, or simply having it included on the agenda?
For example, Berry (1999) notes that liberal citizen groups have grown in number and influence since the 1970s. These are groups with ideologically liberal motives, but are often at odds with material groups. A notable case would be the conflict between unionized loggers and environmentalists. However, to which actors the influence accrues is crucial. Strolovitch (2007) notes that, even when previously marginalized groups gain prominence and influence, there is always a subset of that group that is still marginalized. An example would be that when homosexual rights groups gain prominence, individuals such as African-American lesbian women are still marginalized, and not necessarily represented by the larger group. Someone is always left out, influence is always redirected.
GOING FORWARD: VENUE STUDIES
The range of interest group scholarship is truly impressive. It considers the behavior of individuals and collectivities, across and between various institutions, and in different policy contexts. However, one vein of research that has yet to exhaust its merit is the consideration of venues. Baumgartner and Jones (1993) perhaps instigated this concept by discussing the venue shopping of groups on policy issues. Recently, Holyoke (2003) has taken up the idea again by asking scholars whether we can accept the venue, or locus of policymaking, as a given. While scholars have not intentionally circumscribed venues in their research, most group studies examine the interaction between groups and congress, groups and the bureaucracy, groups and the courts, and so on. However, Holyoke claims that we should treat venues as an empirical matter and follow issues and policies across the venues that they travel. Martha Derthick does this brilliantly in Up in Smoke, her study on the evolution of federal tobacco policy. Her primary object of interest is the policy, which leads her through different congressional committees, the federal courts, the presidency, and the mass media. This is the approach that Baumgartner (2007) and Mahoney and Baumgartner (2008) urge. Drawing on the emergent work in European political studies on “multi-level governance,” Baumgartner sees parallels between the difference in level of governance, and its varying patterns of lobbying, and the federalism of the US. This vertical conception of venues can be matched by the horizontal venues that exist in our separated system of checks and balances. Baumgartner, in fact, lays out a research agenda, at the top of which he urges his colleagues to follow issues longitudinally and across decision-making stages. Having attempted a research project using a venue-oriented approach, I can attest that it yields interesting results and a more holistic view of policymaking. It also takes seriously the concerns of Baumgartner and Leech (1998) that the interest group field has exploded with empirical studies, but remains without a correspondent explosion in understanding.