Issue Politics in Congress / Edition 1 available in Paperback
- ISBN-10:
- 0521671329
- ISBN-13:
- 9780521671323
- Pub. Date:
- 10/10/2005
- Publisher:
- Cambridge University Press
- ISBN-10:
- 0521671329
- ISBN-13:
- 9780521671323
- Pub. Date:
- 10/10/2005
- Publisher:
- Cambridge University Press
Buy New
$27.99Buy Used
$18.19
-
$18.19
-
SHIP THIS ITEM
Temporarily Out of Stock Online
Please check back later for updated availability.
-
Overview
Product Details
| ISBN-13: | 9780521671323 |
|---|---|
| Publisher: | Cambridge University Press |
| Publication date: | 10/10/2005 |
| Edition description: | New Edition |
| Pages: | 222 |
| Product dimensions: | 5.98(w) x 9.02(h) x 0.51(d) |
About the Author
Read an Excerpt
Cambridge University Press
0521855217 - Issue Politics in Congress - by Tracy Sulkin
Excerpt
1
Electoral Challenges and Legislative Responsiveness
Going into his first reelection campaign in 1992, Senator Bob Graham was about as secure as any incumbent facing a challenge could hope to be. Though new to the Senate, he had a long history in Florida politics, including many years of service in the state legislature and two terms as governor. Endorsed by a host of newspapers and interest groups, he was described by many as the state's most popular politician, largely as the result of his reputation for action on environmental and economic issues of interest to his constituents (Duncan 1993, 321). During his active and well-funded campaign, Graham further leveraged these strengths by highlighting his competence and interest in the economy, the environment, and the proper role of governmental regulation.
In contrast, his opponent, Republican representative Bill Grant, faced an even greater uphill battle than most challengers. Hurt by his record of overdrafts at the House bank and his recent switch from the Democratic to the Republican Party, Grant experienced considerable difficulty raising funds, eventually raising only $200,000 to Graham's $3 million. Characterized by St. Petersburg Times political reporter Bill Moss as having "drive, but no fuel," (1992, 1B) he nonetheless launched a spirited campaign against Graham, focusing on health care, economic issues, and the need for a balanced budget.
Not surprisingly, Graham won the race easily, netting nearly two-thirds of the vote. When he returned to Washington, he continued to pursue his interests in the environment and in general economic issues, introducing and cosponsoring legislation and serving on the Environment and Public Works and Banking, Housing, and Urban Affairs committees. More unexpectedly, perhaps, he also became much more active on Grant's priority issues from the campaign. In the term following the election, he introduced twenty-three bills, resolutions, and amendments relating to health issues (on topics ranging from public health and health education to ensuring emergency care and preventive health benefits to health maintenance organization enrollees to regulating the quality of hospitals), more than twice as many as he had introduced in the previous term. He also introduced thirteen measures on balancing the budget, more than four times as many as he had before.1
Graham's attentiveness to Grant's issues is not a coincidence, nor is it unusual. To the contrary, it reflects a widespread yet largely unrecognized mode of responsiveness in the U.S. Congress. Winning legislators regularly take up their challengers' priority issues from the last campaign and act on them in office, a phenomenon I call "issue uptake." Congressional campaigns thus have a clear legacy in the content of legislators' agendas, influencing the areas in which they choose to be active and the intensity with which they pursue these activities. Moreover, as I will show, the extent of this legacy varies in a predictable way across individual representatives and senators, across legislative activities, across time, and across chambers of Congress. Understanding the factors leading to variation in uptake therefore offers fresh insight into some of the most important and enduring normative and empirical questions in American politics regarding the electoral connection in legislative behavior, the role of campaigns and elections, and the nature and quality of congressional responsiveness.
Legislators' motivation for engaging in uptake behavior is simple; they undertake it because they believe that doing so will help them to achieve their electoral goals. Because challengers focus their campaigns on the incumbent's weaknesses, their choices of campaign themes provide signals to winning legislators about important issues that they may have previously neglected. To the extent that legislators act on these signals, taking up salient issues and making them a part of their agendas, they can remedy any weaknesses, strengthening their records before the next campaign and inoculating themselves against possible attacks. Uptake thus has the potential to promote individual legislators' electoral goals as well as the health and legitimacy of the representative process, as legislators adjust their activity in office in response to electoral challenges. The central theme of this book is to explore these possibilities by investigating the dynamics of uptake, explaining why it occurs, how it varies, what it reveals about legislators' motivations and decision making, and how it impacts the policymaking process.
Linking the Electoral and Legislative Arenas
The claim that legislators' behavior is influenced by their past campaign experiences seems perhaps intuitive, and it is certainly not new. As far back as 1960, in his work on norms and behaviors in the U.S. Senate, Donald Matthews noted that it was "difficult, really, to understand the senators, how they act and why, without considering what happens to them when they are running for office" (1960, 68). However, despite the central role given to elections in democratic theory, systematic examinations of how they influence legislative politics and policymaking remain remarkably absent in the literature. We know very little about how legislators' experiences as candidates shape their subsequent activity in office, even though the "reelection-oriented representative" lies at the core of our conception of legislative behavior.
Why is there such a dearth of knowledge about how campaigns influence legislative behavior? It is certainly not due to a lack of scholarly interest in campaign effects or representation and responsiveness, two research agendas that enjoy considerable attention in the literature on American politics. Nor is it likely due to disagreement among scholars working in these fields that the situation described earlier demonstrates a campaign effect or an instance of responsiveness. Instead, the lack of attention to these linkages is the natural result of the traditional divisions of labor in political science research that have limited our ability to recognize and address political phenomena that lie at the intersection of these divisions.
Within legislative studies, the major substantive division mirrors the general institutions-behavior divide of the American field as a whole, with some scholars focusing on the structures and processes of Congress as an institution and others studying the elections that bring representatives and senators to Washington in the first place. There is little overlap, and even those scholars who do research in both areas seldom address both in a single work. As a result, the idea of the "Two Congresses," originally developed as an analytical device to distinguish legislators' careers in Washington, D.C., from their careers in their districts or states, actually accurately describes the current state of research in the field.2 Most research treats representatives' and senators' experiences as candidates as separate from their subsequent behavior as legislators even though the two are inextricably linked.
The most damaging result of the bifurcation between electoral and institutional studies is that we neglect many important questions about the representative process. Most studies of legislative behavior assume that representatives and senators are strategic and forward-looking, using their time in office to promote their future reelection prospects. Following Mayhew's (1974) lead, scholars have asked, "If legislators were proactive and concerned about reelection, what patterns of behavior would we expect to see?" Explorations of this question in a variety of settings have provided substantial insight into how electoral considerations influence incumbents' roll call voting decisions (Fiorina 1974; Kingdon 1989); their participation in committees (Fenno 1973; Hall 1996); their efforts at legislative entrepreneurship (Schiller 2000; Wawro 2000); their "home styles" in their districts or states (Fenno 1978); and their choices about casework and constituency service (Bond 1985; Johannes 1984; Parker 1980). What is missing in this literature is a retrospective approach investigating how past experiences influence legislators' current choices about their activities. If campaigns teach legislators which strategies work and which don't, and what their strengths and weaknesses are in the eyes of the constituency and of potential opponents, then they should play an important role in shaping responsive behavior in the next term. An approach to studying Congress that neglects this potentially important linkage between the electoral and legislative arenas yields a necessarily incomplete picture of legislative behavior.
Studies of congressional elections are similarly limited by their sole focus on a single stage in the representative process. In fact, perhaps the best illustration of the problems inherent in the separation of electoral and institutional behavior falls on the elections side in the debate about whether or not campaigns matter. The ongoing scholarly and popular concern about the quality of campaigns is predicated on the notion that they do, but empirical confirmation of these effects has proven more elusive. Fifty years of quantitative research on elections has confirmed that, while campaigns generally increase voters' information levels, they rarely change many vote decisions (Campbell et al. 1960; Holbrook 1996; Lazarsfeld, Berelson, and Gaudet 1944; Miller and Shanks 1996; Nie, Verba, and Petrocik 1976). More evidence exists for their indirect influence on efficacy and turnout, but there is still considerable disagreement about the magnitude and importance of these effects (see, for example, Ansolabehere and Iyengar 1995; Freedman and Goldstein 1999; Wattenberg and Brians 1999). As a consequence, debates about the potential impacts of campaigns remain among the most vigorously contested in the literature on American politics.
Unfortunately, the scope of these discussions has been quite limited, focusing almost exclusively on how campaigns may or may not influence citizens' attitudes and vote choices. Whatever the eventual conclusion drawn from this research, a solely voter-centered approach to campaign effects can provide only a partial answer to the question of whether or not campaigns matter. The limits of this approach become clear when we consider that the outcomes of congressional elections are generally quite predictable - incumbents win the vast majority of the time. For this reason, any impact of campaigns on voters will be important only at the margins. By expanding our focus to include their potential influence on winners' behavior in office, we can get more analytical leverage. Specifically, if campaigns inform the content and intensity of legislators' subsequent legislative activity, then they clearly do matter, apart from any effect they may have on voters. To assess the true impact of campaigns, we must therefore look for effects that extend beyond Election Day. However, as Fenno noted in his recent work on representation, "for most political scientists most of the time, the study of elections has meant only the study of voters and their voting behavior" (1996, 76), a perspective that limits our conception of the role of campaigns and results in a great gulf in our understanding of the linkages between electoral and legislative politics.
Defining Uptake
By explicitly specifying the process through which legislators' activity in office is related to their campaign experiences, the study of uptake makes a significant advance in closing this gap. Though I focus on uptake as a mechanism for connecting campaigns to legislative behavior, it has its scholarly roots in other subfields of political science, particularly political communication and political theory. Scholars in these fields studying deliberation and discursive processes have been interested in determining the conditions under which ideas expressed by certain participants are incorporated into the arguments of others.3 If, for example, in a discussion about the merits of a particular tax policy, one discussant raises the issue of its impact on the budget deficit, do other discussants also address this point? To the extent that they do, uptake has occurred, and the deliberative process itself can be judged to be more successful, at least from a procedural standpoint.
In the congressional setting, uptake is conceived of in a slightly different way, going beyond language to focus on other types of behavior, in particular, participation in the legislative process, including legislators' sponsorship and cosponsorship of bills, resolutions, and amendments and their statements on the floor about pending legislation. If a challenger focuses her campaign on agriculture, defense, and education, uptake is measured as the amount of attention the winning legislator devotes to these issues in his legislative activity when he returns to Congress. Graham's introduction of measures on health care and the balanced budget is thus evidence of his uptake of Grant's themes.
At its roots, then, uptake is fundamentally about legislative responsiveness, as representatives and senators use their time in office to respond to and address their previous challengers' critiques. Uptake levels provide an indicator of this responsiveness, both for individual legislators and for Congress as a whole. At the individual level, we can compare the rates of uptake across legislators to place an individual legislator's behavior in context. For example, we can determine whether Graham's level of attention to his challenger's themes is relatively high, relatively low, or somewhere in between. Such analyses also enable us to identify the factors that distinguish those legislators who are highly responsive to their challengers' critiques from those who are less so and to predict uptake levels based on characteristics of a legislator or his or her constituency. Similarly, aggregate uptake patterns across all legislators allow us to assess the strength of institutional responsiveness - the success of the institutions of representative democracy at transmitting issues from one stage of the process to another as issues are introduced in campaigns, incorporated into legislative activity, and translated into public policy. We can investigate whether the legacy of campaigns is stronger in certain types of activities than others, or at certain times within legislative terms, and use these findings to estimate the longevity of uptake effects. Although the opportunity to study individual and institutional agenda-based responsiveness is in some ways a side consequence of modeling the linkages between campaigns and legislative behavior, it is a critically important one. By quantifying and measuring uptake levels, we gain a new perspective on the dynamics of the relationship between the electoral and legislative arenas as well as a more nuanced understanding of the nature of legislative responsiveness.
As may be obvious, my approach is greatly influenced by the agenda-setting tradition in public policy, political communication, and public opinion research. At the core of the agenda-setting paradigm is a focus on issues: how they arise, how they change, and how they are communicated and acted on by different actors in the political process. Within the public policy field, this tradition is exemplified by the work of scholars who have studied agenda formation and change at the national level (Baumgartner and Jones 1993; Cobb and Elder 1983; Downs 1972; Kingdon 1984). In political communication research, the focus has been on the microfoundations of agenda-setting, examining in particular the relationship between the amount of media coverage given to issues and the extent to which they are perceived as important problems by the public (see, for example, Iyengar and Kinder 1987; McCombs and Shaw 1972). Public opinion scholars have, in turn, attempted to unite the insights of these two fields to focus on the relationships between the governmental, public, and media agendas (Erikson, MacKuen, and Stimson 2002; Page and Shapiro 1992).
The agenda-setting perspective is much less familiar to scholars of legislative behavior, where the focus has traditionally been not on legislators' relative attention to different issues, but on the positions they take on roll call votes, which are generally interpreted as manifestations of their underlying ideological preferences. This focus on preferences has informed nearly all of the literature on legislative representation and responsiveness (but see Hall 1996). An agenda-based approach like uptake thus constitutes a fundamentally different way of thinking about responsiveness. The leverage provided by this approach becomes most apparent when we compare it to the traditional way of conceptualizing representational linkages. As such, I begin with a review of the literature on representation and responsiveness in Congress.4
Research on Legislative Responsiveness
Concerns about representation are as old as the history of democratic theory. As Pitkin (1967) notes, modern theorizing about the proper role of representatives is still largely centered on the work of classical theorists like Burke, Rousseau, and Mill, who all concerned themselves with the standards for representative legitimacy. For instance, should representatives act as delegates, letting their constituents' directives dictate their behavior? Or should they act as trustees, gathering information and making their own judgments as to which actions would be in the best interests of their districts or states? With the growth of quantitative political science, scholars became interested in operationalizing these concepts and determining whether or not legislators behaved as theorists prescribed. Research on representation has therefore enjoyed a central place in the study of American politics over the past fifty years.
Different studies propose slightly different conceptualizations of legislative representation, but most focus on a particular type, called "policy responsiveness."5 The basic premise underlying policy responsiveness is that, in accordance with the delegate standard of representation, legislators' behavior in office should reflect the interests of their districts or states. If we ask what sorts of behavior might indicate such responsiveness on the part of legislators, a number of possibilities come to mind. For example, legislators' roll call voting decisions should reflect the preferences of their constituents on these matters, so their issue positions should serve as one important indicator of their responsiveness. Similarly, responsive legislators should devote attention to important issues, adjusting their agendas in response to outside events, the desires of the constituency, critiques from challengers, and the like. For the most part, however, the literature on policy responsiveness has focused solely on issue positions, an approach that can be traced back to one of the earliest and still most influential empirical studies of representation, Miller and Stokes's (1963) article, "Constituency Influence in Congress."
Miller and Stokes's primary goal in their now-classic piece was to assess the claim that constituents have considerable influence on or control over legislators' behavior, particularly their voting decisions. Though this is an assertion that would be strongly questioned today, it was, at the time of their writing, "commonly thought to be both a normative principle and a factual truth about American government" (1963, 45). Coining the term "policy congruence" to refer to the relationship between legislators' positions on issues and their constituents' opinions on those same issues, Miller and Stokes compared the results of district opinion polls to representatives' roll call voting records to assess the strength of the correlation between the two across three broad issue areas - social welfare, foreign affairs, and civil rights. Contrary to the expectations of the textbook models of representation, which predicted high levels of congruence, they found that the relationship between legislators' and constituents' positions was positive, but relatively weak, and varied markedly by issue area.
At least in part because of this surprising and somewhat disturbing conclusion, Miller and Stokes's work spawned a large number of studies examining the extent of constituency influence on legislators' vote decisions (for a review, see Bernstein 1989 or Jewell 1983). Some of these studies continued in their footprints, examining congruence within specific issue areas (see, for example, Page et al. 1984; Stone 1982) and further explicating the structure of the congruence relationship (Cnudde and McCrone 1966; McCrone and Stone 1986). Others took a different tack, attempting to tie variation in congruence to other factors like electoral vulnerability (Fiorina 1974; Kuklinski 1977; Sullivan and Uslaner 1978), legislators' role orientations (Jones 1973; Kuklinski and Elling 1977; McCrone and Kuklinski 1979), and proximity to the next election (Elling 1982; Kuklinski 1978). Still others offered methodological critiques of the measurement strategies used in assessing congruence and offered alternatives (Achen 1977, 1978; Erikson 1978; Stone 1979; Weissberg 1979). However, despite providing considerable refinement in how we think about policy congruence, these studies all generally reached the same conclusion: that responsiveness did exist, particularly on salient issues, but at lower levels and in more subtle and contingent ways than the framers might have wanted and the conventional wisdom pre-Miller and Stokes might have predicted.6 Perhaps because of the consistency of these findings, scholarly attention to policy responsiveness waned throughout much of the 1980s.
More recently, there has been a revival of interest in the subject, driven, at least in part, by the increased availability of data about public opinion and legislative voting and by widespread access to powerful software that enables scholars to explore more complicated models of policy congruence. Although these new studies have diverged from the approaches taken by the earlier work, they remain clearly influenced by Miller and Stokes's insights. For example, one important innovation in the recent literature on representation has been to take the policy congruence standard, but to apply it to Congress as an institution rather than to individual representatives. Most of the early empirical work on policy representation focused on dyadic representation, that is, the relationship between one representative and his or her constituency. However, as Weissberg (1978) noted, a different sort of policy representation was possible that focused on how well Congress represented the preferences of the public as a whole. Thus, even if dyadic representational linkages were relatively weak, strong collective representation could exist, a claim that was supported by Weissberg's preliminary analysis of the issue (but see Hurley 1982).
In the best-known examination of this sort of representation, Stimson, MacKuen, and Erikson's (1995) work on what they refer to as "dynamic" representation, the authors take up the issue of collective responsiveness, arguing that the cross-sectional nature of most representation research limits investigation into the process that produces the linkage between representatives and constituents (see also Erikson et al. 2002; Jacobs and Shapiro 2000; Page and Shapiro 1992). Process is important because the direction of the relationship is crucial for evaluating responsiveness - governmental action must follow public opinion. Stimson et al. (1995) advocate a time-series design that is capable of disentangling the reciprocal causal links between public opinion, policy activity, and policy outputs. This approach yields a more positive assessment of the linkage between Congress and the public, demonstrating that the ideological tenor of public policy does respond to changes in public opinion.7
These insights about the reciprocal nature of collective responsiveness have been taken up by another group of scholars, most notably Hill and Hurley (1999; Hurley and Hill 2003), who have sought to explore whether such reciprocal policy linkages exist at the dyadic level. While previous congruence studies conceived of constituency opinion as an independent variable, Hill and Hurley revisited Miller and Stokes's original data to demonstrate that the linkages between constituents and legislators go in both directions. They conclude that legislators are influenced by public opinion but also mold it, which they suggest yields a more optimistic conclusion about the strength of representational linkages.
© Cambridge University Press