Setting the Agenda: Responsible Party Government in the U.S. House of Representatives / Edition 1 available in Paperback
Setting the Agenda: Responsible Party Government in the U.S. House of Representatives / Edition 1
- ISBN-10:
- 0521619963
- ISBN-13:
- 9780521619967
- Pub. Date:
- 09/12/2005
- Publisher:
- Cambridge University Press
- ISBN-10:
- 0521619963
- ISBN-13:
- 9780521619967
- Pub. Date:
- 09/12/2005
- Publisher:
- Cambridge University Press
Setting the Agenda: Responsible Party Government in the U.S. House of Representatives / Edition 1
Buy New
$40.99Buy Used
$17.65
-
$17.65
-
SHIP THIS ITEM
Temporarily Out of Stock Online
Please check back later for updated availability.
-
Overview
Product Details
| ISBN-13: | 9780521619967 |
|---|---|
| Publisher: | Cambridge University Press |
| Publication date: | 09/12/2005 |
| Edition description: | New Edition |
| Pages: | 352 |
| Product dimensions: | 6.30(w) x 9.25(h) x 0.79(d) |
About the Author
Mathew D. McCubbins is a professor of political science at the University of California, San Diego. His authored works include Legislative Leviathan: Party Government in the House (1993); Under the Watchful Eye: Managing Presidential Campaigns in the Television Era (1992); Recent co-edited books include The Origins of Liberty: Political and Economic Liberalization in the Modern World (1997); and Elements of Reason: Cognition, Choice, and the Bounds of Rationality (2000). His most recent book is Stealing the Initiative: How State Government Responds to Direct Democracy (2001) with Elisabeth Gerber, Arthur Lupia, and D. Roderick Kiewiet. McCubbins is also the author of numerous articles in journals such as Legislative Studies Quarterly; Journal of Law, Economics, and Organization; Law and Contemporary Problems; and the American Journal of Political Science. He is the coordinator of the Law and the Behavioral Sciences Project and was a Fellow at the Center for Advanced Study in the Behavioral Sciences for 1994–95.
Read an Excerpt
Cambridge University Press
0521853796 - Setting the Agenda - Responsible Party Government in the U.S. House of Representatives - by Gary W. Cox and Mathew D. McCubbins
Excerpt
1 Modern democracy is unthinkable save in terms of the parties. - Schattschneider 1942 The party system that is needed must be democratic, responsible and effective - a system that is accountable to the public, respects and expresses differences of opinion, and is able to cope with the great problems of modern government. - American Political Science Association 1950 Thus, if the parties were in trouble, so too was democracy. - White 1992
Introduction
1.1. INTRODUCTION
For democracy in a large republic to succeed, many believe that responsible party government is needed, with each party offering voters a clear alternative vision regarding how the polity should be governed and then, if it wins the election, exerting sufficient discipline over its elected members to implement its vision (cf. Ranney 1951; American Political Science Association 1950). America was once thought to have disciplined and responsible parties. Indeed, students of nineteenth-century American politics saw parties as the principal means by which a continental nation had been brought together: "There is a sense in which our parties may be said to have been our real body politic. Not the authority of Congress, not the leadership of the President, but the discipline and zest of parties has held us together, has made it possible for us to form and to carry out national programs" (Wilson 1908: 218, 221; cf. Bryce 1921: 119).
Since early in the twentieth century, however, critics of American politics have often argued that congressional parties are largely moribund. Some contend that they have become nothing more than labels for like-minded politicians who act together when they agree but otherwise pursue their own agendas and careers (Mayhew 1974). A chorus of critics depict members of Congress as dedicated to the pursuit of graft, campaign contributions, and the emoluments of office and as captured by interest groups who seek to turn public policy into private favors (McConnell 1966; Fiorina 1977a; Weingast, Shepsle, and Johnson 1981; Buchanan 1968; Becker 1983; Stigler 1971; Lowi 1969; Schattschneider 1960; for surveys see Cox and McCubbins 1993, Chapter 1, and Munger and Mitchell 1991).
Even though Congress does suffer from many infirmities, we will argue that a hitherto unrecognized form of responsible party government has characterized U.S. politics since the late nineteenth century. As in the traditional view of responsible party government, our theory depicts congressional parties as electorally accountable and legislatively responsible, at least to an important degree. We differ from the traditional view, however, in at least two ways.
First, whereas traditional theories stress the majority party's ability to marshal a cohesive voting bloc as the source of its legislative power, our theory stresses the majority party's ability to set the agenda as the key to its success. The importance of this distinction can be suggested by recalling that the most prominent line of criticism of partisan theories focuses directly on the issue of voting cohesion.
Many prominent scholars, including Schattschneider (1942: 131-2), Mayhew (1974), and Kalt and Zupan (1990), view legislators' votes as driven primarily by their constituents' and their own opinions, with partisan considerations playing a distinctly secondary role. Building on such views, Krehbiel (1993, 1998) argues that the two parties' attempts to influence votes either are negligible or cancel each other out.
If, as these theories suggest, party pressures cancel out, however, then the majority party cannot marshal its troops effectively, as required by traditional theorists of responsible party governance. Instead of being driven toward the platform promises of the majority party by the force of its discipline, policies in Congress will be driven to the center of congressional opinion by the logic of the famous median voter theorem.
If one accepts the traditional view that parties are strong only to the extent that they can affect their members' behavior on substantive votes, and if one views congressional votes as positioning policy along a single left-right continuum, then Krehbiel's argument is persuasive. In particular, given these two assumptions, majority parties matter only if they can secure nonmedian policy outcomes, and, in order to do this, they must engage in the unenviable and unlikely-to-succeed task of regularly pressuring their centrist members to vote against their constituents' and/or their own opinions.
Our emphasis on agenda control deflects this canonical criticism of partisan theories in the following way. We do not model voting in Congress as if there were a single vote on a single dimension (per the standard unidimensional spatial model); rather, we envision a series of votes on different issues. This opens up the possibility that, even if the majority party were unable to secure a nonmedian outcome on any given issue considered in isolation - a debatable premise - it might nonetheless greatly affect the overall legislative outcome if it prevents some issues from being voted on at all.
Figure 1.1 Illustration of leftward and rightward policy moves
| Image not available in HTML version |
To see how agenda-setting power can affect legislative outcomes, imagine a newly elected Congress and the set of existing government policies - label them q1,..., qn - that it faces. Each of these policies could in principle be adjusted, sliding them further to the left or right (e.g., less stringent or more stringent regulation of abortion). The newly elected members of the House have opinions regarding how each of the n policies should ideally be positioned along their respective left-right dimensions. Denote the center of congressional opinion (the median ideal position) regarding each policy by F1,..., Fn for the n policies.
Note that one can divide the existing government policies into two main categories, depending on the relationship between the legislative median, Fj, and the status quo, qj. In one category are policies that lie to the left of the current center of congressional opinion, qj < Fj. If the House votes on a bill to change such a policy from the status quo (e.g., qjR in Figure 1.1) to the floor median, Fj, the result will be a rightward policy move. In the second main category are policies that lie to the right of the current center of congressional opinion, qj Fj. If the House votes on a bill to change such a policy from the status quo (e.g., qjR in Figure 1.1) to the floor median, Fj, the result will be a leftward policy move.
Now suppose in this simple example that Democratic majorities can block bills that propose rightward policy moves from reaching votes on the floor, thereby killing them without the necessity of a clear floor vote on the bill itself. The Democrats' blocking actions might take many forms, such as a chair refusing to schedule hearings, a committee voting not to report, the Rules Committee refusing to report a special rule, or the speaker delaying a particular bill. Each of these actions might in principle be appealed to the floor and reversed via a series of floor votes. It is a maintained assumption of our approach that the transaction costs involved in such appeals are typically so high (see Chapter 4) that the majority's delaying tactics are effective in killing (or forcing changes in) the bills they target.1 To the extent that they are successful, the Democrats will produce a legislative agenda on which every bill actually considered on the floor proposes to move policy leftward. As a natural consequence, a majority of Democrats will support every bill.
This example, we hasten to add, overstates what our theory actually predicts (e.g., there are rightward policy moves that even the Democrats would like to make and, similarly, leftward policy moves that even the Republicans would support, when the status quo is extreme enough). Nonetheless, the discussion so far suffices to illustrate the potential power of a minimal form of agenda control (just the power to block) and makes clear that our theory sidesteps critiques that focus on the debility of party influence over floor votes (such as Krehbiel's). We can deny both the notion that parties must secure nonmedian outcomes issue by issue in order to matter and the notion that parties must exert discipline over how their members vote on bills in order to matter.2 Agenda control alone suffices - if it can be attained - to exert a tremendous influence over policy outcomes.
In sum, traditional theories of responsible party government see a Democratic (or Republican) majority as mattering because the majority can marshal its troops on a given issue and thereby attain policy outcomes that differ from those preferred by the median legislator on that issue. Aldrich and Rohde's theory of conditional party government shares this perspective: "most partisan theories would yield the expectation that the majority party would have sufficient influence...to skew outcomes away from the center of the whole floor and toward the policy center of [majority] party members" (Aldrich and Rohde 1995: 7). Such theories are vulnerable to Krehbiel's critique and its predecessors. In contrast, our theory sees a Democratic (respectively, Republican) majority as mattering because the majority can prevent reconsideration of status quo policies lying to the left (respectively, to the right) of the current median legislator on a given policy dimension - thereby filling the agenda mostly with bills proposing leftward (respectively, rightward) policy moves.
We should add that we do not view American parties as incapable of disciplining their troops. Indeed, we believe they regularly seek additional support on close votes, employing both carrots and sticks in the process. Such efforts can even lead to nonmedian outcomes on particular issues (typically via procedural maneuvers, such as closed rules, rather than by outvoting the opposition on the merits). However, the majority party's efforts on the floor are designed to complement whatever degree of agenda manipulation has already occurred by corralling a few votes on the margin, not to coerce moderate members to cast risky votes in order to maximize party cohesion. Picking which bills will be voted on at all - that is, which status quo policies will be at risk of change - is the primary technique; garnering enough votes to eke out a victory is important but secondary.
A second way in which our theory differs from traditional notions of responsible party government is that the latter stress the enactment of new policies - as promised in the party platforms - as the main normative criterion by which one should judge whether party government is operating successfully. In contrast, our theory stresses the avoidance of party-splitting issues, hence the preservation of some existing policies, as the key to the political survival of majority parties (whatever its normative merits).
We do not claim that parties cannot or do not compile positive records of accomplishment and are restricted merely to the preservation of portions of the status quo. Even the most heterogeneous majorities in congressional history, such as the Democrats of the 1950s, were able to agree on a number of new legislative goals and accomplish them. Thus, we have argued previously (Cox and McCubbins 1993), and will argue again in Chapter 10, that control of the legislative agenda can also be translated into the enactment of some or all of the majority party's platform. However, as we explain in Chapter 10, the majority's success in changing policies, unlike its success in preserving policies, depends on its internal homogeneity.
Another way to frame this second difference is to say that we envision two stories in the edifice of party government, not just one. The first, or bedrock, story involves securing a super-proportional share of offices for the party's senior members, imposing a minimal (primarily negative) fiduciary standard on those senior officeholders, and thereby ensuring that the party collectively is able to prevent items from appearing on the floor agenda. The second, or superstructural, story consists of enhancing the ability of the party's officeholders to push (as opposed to preventing) bills, imposing a more demanding fiduciary standard upon them (one requiring that they use their new powers for the collective benefit) and thereby enhancing the party's collective ability to push items onto the floor agenda.
By shifting the terms of debate from the majority party's ability to marshal its troops on a given issue to the majority party's ability to decide which issues are voted on to begin with, and from the majority party's ability to change policies to its ability to preserve policies, we seek to provide a new theoretical grounding for partisan theories of congressional organization - and to defend it empirically. In what follows, we will show that our theory explains important features of the postbellum history of the U.S. House of Representatives extremely well. To set the stage for that demonstration, in this chapter we provide a précis of our theory and outline the remaining chapters.
1.2. A PRÉCIS OF PROCEDURAL CARTEL THEORY
There are two main approaches in the literature on congressional organization. One view stresses how well congressional organization serves members' nonpartisan goals. For example, the House is declared well organized to (1) promote the reelection of its members (Mayhew 1974, 1991; Shepsle and Weingast 1984c), (2) make gains from legislative trade possible (Weingast and Marshall 1988), (3) make specialization and the efficient generation of information possible (Gilligan and Krehbiel 1989, 1990), and (4) aid in bargaining with the other chamber or other branches of government (Diermeier and Myerson 1999; Epstein and O'Halloran 1999). Political parties are explicitly denied a consequential role in these theories.
On the other hand, a new generation of partisan theories argues that the House is well organized to serve the collective interests of the majority party (Cooper 1970; Cooper and Brady 1981; Sinclair 1983, 1995; Stewart 1989; Kiewiet and McCubbins 1991; Cox and McCubbins 1993; Maltzman and Smith 1994; Binder 1997; Evans and Oleszek 2002; Gamm and Smith 2002). One variant of partisan theory, known as the conditional party government model (Rohde 1991; Aldrich 1995; Aldrich and Rohde 2001), focuses on how the majority party leadership's powers expand as the members they lead become more alike in political preference (and more different from the opposition) - leading ultimately to greater voting discipline and thus to greater success in legislating for the majority. Another variant, while accepting a version of the conditional party government thesis, focuses on an array of procedural advantages enjoyed by the majority party that are not conditional on its internal homogeneity. We call this variant "procedural cartel theory" (Cox and McCubbins 1993, 1994, 2002), the key aspect of which is the majority party's use of agenda control to achieve its desired outcomes.3
While we develop procedural cartel theory at length in Chapter 2 and in Legislative Leviathan (1993), here we can briefly note four key claims that distinguish our approach. First, legislative parties arise, we believe, primarily to manage electoral externalities involved in running campaigns in mass electorates. Second, legislative parties are best analogized, we believe, to legal or accountancy partnerships, with various gradations of junior and senior partners. Third, legislative parties - especially in systems where floor voting discipline is costly to secure, as in the United States - specialize in controlling the agenda, rather than in controlling votes. That is, they seek to determine what is voted on to begin with, rather than to dictate their members' votes issue by issue (although they do regularly seek votes on the margin). Fourth, a legislative majority party allocates both negative (delay or veto) rights and positive (accelerating or proposal) rights among its senior partners (and groups thereof), but the mix of such rights changes with the degree of preference homogeneity among the party's members.
To explain the last point, note that there is a trade-off between increasing veto power (and suffering higher negotiation costs in order to do anything) and increasing proposal power (and suffering higher externalities from the decisions made by those with such power).4 The more heterogeneous the preferences within a given coalition, the more that coalition's partners will wish to limit the proposal rights of other partners, which necessarily entails strengthening their own and others' veto rights. The value of the coalition then comes more and more in keeping certain issues off the agenda and stabilizing the associated status quo policies. The more homogeneous the preferences within a given coalition, the more that coalition's partners will agree to expand each other's proposal rights, which necessarily entails weakening their own and others' veto rights. The value of the coalition then comes more and more in pushing certain issues onto the agenda with the hope of changing the associated status quo policies. Regardless of the coalition's homogeneity or lack thereof, regardless of whether its value stems more from stabilizing status quo policies or more from changing status quo policies, it will continue to seize the vast bulk of offices endowed with special agenda-setting rights and thus to cartelize agenda power. In this sense, party government is not conditional on the level of agreement within the party; rather, the nature of party government simply changes, from a more progressive vision (implicitly taken to be the only party government worth having in most of the previous literature) to a more conservative vision.5
Having stated our inclinations on four important distinctions within the family of partisan theories, we can now diagram the elements of our theory (see Figure 1.2). Reading up from the bottom of the figure, we start with "majority party's control of delegated agenda powers," that is, with its control of the powers inherent in the various offices of the House endowed with such powers (e.g., the speakership and committee chairs). The better the majority party's control of such powers is, the more able will it be to fashion a favorable record of legislative accomplishment, although certainly other factors enter into this as well (such as the party of the president).6 The more favorable is the majority party's record of legislative accomplishment, the better its reputation or brand name will be, although again there are other factors that affect this, too (such as the president's actions). The better the majority party's brand name, the better will be the prospects for (re)election of its various candidates and the better will be the prospects for (re)attainment of majority status. The senior partners of the majority party care in particular about the latter because their ability to retain their chairs, speakerships, and other offices depends crucially on their party retaining its majority.
Figure 1.2 Majority-party agenda control and legislative success
| Image not available in HTML version |
Reading the diagram top-down, instead of bottom-up, one starts with individual members of Congress assumed to care both about (re)election to the House and (re)election to offices within the House. They recognize that (re)election (especially to internal posts) depends crucially on majority status, which in turn depends on maintaining a favorable brand name for the party. Maintaining a favorable brand name, in turn, depends on the party's record of legislative accomplishment, hence on its ability to solve the various cooperation and coordination problems that arise within the legislative process. The party solves these problems primarily by delegating agenda power to its senior partners.
Because the element in this theory that we most wish to stress concerns agenda power, we turn now to a more extended consideration - albeit still abbreviated - of the cartel thesis. The next chapter provides a fuller elaboration of our theory.
1.3. THE PROCEDURAL CARTEL THESIS
The broadest thesis of this book, which resonates in many legislatures worldwide, is that agenda power in busy legislatures is almost always cartelized. To put it another way, even though voting power in democratic legislatures is everywhere equal, proposal and veto power are everywhere unequal.
What do we mean by agenda power being cartelized in the specific case of the U.S. House? We provide a fuller explanation in the next chapter but for now suffice it to say that the agenda is cartelized when (1) special agenda-setting powers are formally delegated to various offices, such as committee chairs, the speakership, and the Rules Committee; (2) the majority party's members secure most of these offices, so that "agenda-setting services" can be procured only from members of the procedural cartel, just as certain kinds of economic services or goods can be procured only from the relevant economic cartel; and (3) the majority party's "senior partners," who hold these agenda-setting offices, act according to a minimal fiduciary standard - namely, that they do not use their official powers to push legislation that would pass on the floor against the wishes of most in their party.
Note that the features we have just listed also characterize most parliamentary governments (cf. Döring 1995): (1) Special agenda-setting powers are formally delegated to cabinet ministers, presiding officers, and directing boards (the analogs of the U.S. chairs, speaker, majority leader, and Rules Committee, respectively); (2) the governing coalition's members secure most, if not all of these offices (cf. Carroll, Cox, and Pachón 2004); and (3) the governing coalition's "senior partners," who hold these agenda-setting offices, respect a norm according to which no bills are pushed that would split the governing coalition (Tsebelis 2002). It is often true that many parliamentary procedural cartels expect an even greater level of cooperation between their senior partners than would have been expected of U.S. committee chairs during the uneasy alliance of Northern and Southern Democrats in 1937-60. Nonetheless, the structural design of the most basic form of party government is similar across a wide range of systems: break the theoretical equality of legislators by creating a class of agenda-setting offices, ensure that the governing coalition's senior partners secure these offices, and deal with the consequent problems of agency loss and floor discipline, as best the local conditions permit.
The basic design of party government within legislatures admits a trade-off between two costly methods of maintaining the power and advantages of these agenda-setting offices: procedural agenda control and voting discipline. Designing and maintaining rules that establish agenda control is costly; ensuring that members of the majority party vote with the party is also costly. Different legislatures, depending on their circumstances, choose different mixtures of these two costly mechanisms.
The question remains, why should agenda power be cartelized according to this basic recipe, in so many legislatures? Let us sketch out an answer to this question, based on our continuing research in comparative legislative studies (Amorim Neto, Cox, and McCubbins 2003; Cox, Masuyama, and McCubbins 2000; Carroll et al. 2004; Campbell, Cox, and McCubbins 2002; Den Hartog 2004). We plan to articulate this argument more fully in future work.
Although the details of legislative procedure differ widely across the world's democratic legislatures, one generalization holds universally: Important bills can only pass pursuant to motions formally stated and voted upon in the plenary session.7 The necessity of acting pursuant to formally stated motions means that every bill must consume at least some plenary time, if it is to have a chance at enactment. Simply put, plenary time is the sine qua non of legislation.
If all legislators have equal access to plenary time, then plenary time is a common pool resource, and rising demand for such time leads to various problems in the legislative process (cf. Cox 1987; Den Hartog 2004; Weingast and Marshall 1988). Can a coalition restrict access to plenary time, enhancing its own members' abilities to propose and block, while diminishing those of the opposition? At a very general level, models such as those of Duggan and Banks (2000), Calvert and Fox (2000), and Diermeier and Feddersen (1998) illustrate how stable restrictions on access to plenary time should be possible, as part of a repeat-game equilibrium. In these models, a majority of members benefit from better access, at the expense of the minority suffering worse access, to plenary time. The majority is held together by the threat that failure to abide by certain norms of behavior (stipulated in equilibrium play) will bring down the coalition, and with it each majority member's superior access to plenary time.
We suggest a somewhat more detailed and concrete recipe by which access to plenary time is restricted. A procedural cartel endows (or inherits) offices with agenda-setting powers, secures those offices for its senior partners, and ensures minimally fiduciary behavior by those senior partners.
How is it that such cartels stick together? In addition to the threat that the whole arrangement can come crashing down, depriving senior members of their offices or stripping those offices of their powers, we would add two additional reasons why access-hogging majorities, once formed, are stable. First, individual nonpivotal legislators in the majority have reason to fear punishment - such as lack of promotion, lack of aid for pet bills, demotion and, ultimately, expulsion from the majority - should they violate crucial norms of behavior (we elaborate such an argument in Cox and McCubbins 1993, 1994). Second, building up a mechanism by which to regulate access to plenary time (creating offices endowed with various special veto and proposal powers and ensuring that one's members secure those offices) entails large fixed costs and very low marginal costs on any particular policy or decision. The large fixed costs arise in creating and maintaining (1) the party's brand name and (2) the rules, procedure, precedent, and interpretation that establish and clarify the powers of agenda-setting offices (on the latter, cf. Chapter 4). To the extent that the parties succeed in establishing themselves as the only viable route to the top offices, they can become very stable indeed.
© Cambridge University Press