Bill Sponsorship and Congressional Support for Policy Proposals, from Introduction to Enactment or Disappearance

Article excerpt

Research on policy change tends to focus on legislative successes (bills that are enacted), policies that are especially important or controversial, and the final stages of the policy process. This article attempts to show how to improve our ability to trace support for policy proposals through the entire legislative process, for failures as well as successes and for less-visible proposals as well as more visible ones. We refine the concept of a "policy proposal"-a particular proposed solution to a public problem-as a set of identical or nearly identical bills introduced into one or more congresses; show how to find such bills, and examine a stratified random sample of 60 considered by the U.S. Congress; describe how much support the proposals receive; show that, in line with some views of legislative activity, proposals are generally on the agenda for only a short time; and suggest that trends in sponsorship provide a good way to measure support for particular proposals for policy change. It is argued that the approach developed in the article will aid subsequent studies of the determinants of policy change.

We have learned much in recent years about the causes of congressional action on policy proposals-about how Congress is affected by public opinion, the party balance, interest groups, and other factors. Our conclusions have been based, however, on a very partial picture of the policy process. Most research focuses on legislative successes (bills enacted, money appropriated, etc.), the final stages of the policy process, and important or controversial issues. We know relatively little about congressional action on bills that do not make it to the final stages of the legislative process, or are neither especially controversial or important-that is, the vast majority of bills (on these points, see, e.g., Baumgartner and Leech 1998: 38, 40; Talbert and Potoski 2002: 865; Edwards, Barrett, and Peake 1997: 547; Krutz 2000).

Why so partial a picture? We focus here on what we believe are two key reasons. First, we cannot explain the ebb and flow of congressional support for policy proposals from the time they are introduced because we cannot measure it. second, we can say little about congressional action on the vast majority of bills because the concept of random sampling has not been applied to the study of congressional action on policy-we have focused on biased samples of bills or policies without much concern for how this undermines our ability to generalize.

This article shows how to begin filling in our picture of the legislative process. We propose a way to measure congressional support for policy proposals, based on sponsorships and cosponsorships, and use the measure to gauge support for a (stratified) random sample of policy proposals from the 101st congress, 1989-1990.

Our measure of support is intended to be used as a dependent variable in subsequent studies of policy change, making it possible to test our theories of policy change over the whole policy process, studying the determinants of sponsorship as well as roll-call voting and enactment. And our approach to sampling is intended to improve our ability to generalize about congressional action.

LEGISLATIVE SUPPORT FOR POLICY PROPOSALS

Analyses of the determinants of legislative action most often try to explain enactment or some clear result of enactment, such as expenditures. Researchers know they should examine earlier phases of the legislative process, but seldom move back beyond roll-call outcomes (particularly in quantitative work).1 We thus know little about support for policy proposals never voted on, or about what distinguishes proposals that are voted on from those that are not.

There are two key reasons why we know little about support for such policy proposals. First, we have no satisfactory way to measure it. Second, when we try to develop such a measure, we realize that there is an even more fundamental problem: we have no good operational definition of the thing to be measured-of the concept of "policy proposal" itself.

Defining "Policy Proposal"

When we discuss congressional support for something that has not been voted on, what exactly is the "something"? It could be what Kingdon (1995: 150) calls a policy proposal: a particular proposed solution to a policy problem. Intuitively, it is easy to imagine tracing the rise and fall of support for such a proposal.

But what does this mean operationally? How should a "proposal" be defined? How are we to identify a particular proposal and distinguish it from others?

We could define a policy proposal as a bill, as Kingdon (1995: 150-51) does when writing of airline deregulation that "Senators Howard Cannon and Edward Kennedy had agreed on a proposal, and the administration simply adopted that bill wholesale." Kingdon provides this as an example, not a definition, but it makes sense to focus on bills: they are distinct, easily identified entities, central to the legislative process, proposing solutions to public problems.

Such a definition seems too narrow, though, because it doesn't conform to how we normally discuss politics. Those who write about politics often describe changes in support for what they probably think of as policy proposals, reporting, for example, that congressional support for repeal of the estate tax increased over the course of a few years, or that support for a balanced budget amendment to the Constitution rose a few years ago but then fell. Intuitively, such descriptions seem to convey important information, but they are not descriptions of support for single bills. For one thing, bills die at the end of every congress, so support for one can't extend beyond two years; for another, members of Congress often introduce multiple, identical bills, and it wouldn't make sense to gauge support for one and ignore the others.

It would be more reasonable to think of a policy proposal as a particular alternative manifested in any number of highly similar bills, even if introduced in different congresses. Total support for the proposal would be the sum of support for all the bills; changes in support could be tracked from congress to congress for as long as the proposal was on the agenda.

Unfortunately, such a definition may be problematic in its own way. How similar in content must bills be, to be viewed as manifestations of the same policy proposal? When bills are identical, this is no problem. And when they are very different, it is no problem either.

But what about intermediate cases, when bills are similar but not identical? On one side would be bills that address the same problem but propose different solutions. This is what observers mean when they say that there are "three patients' rights bills" or "two welfare reform bills" on the agenda. It would not make sense to see these as representing the same policy proposal. On the other side would be bills that differ only slightly; these it would make sense to treat as expressions of the same proposal. The problem, then, is to distinguish between bills so similar they may be treated as identical, and those not similar enough.

If the proposals addressing an issue are few and readily distinguishable, each could be treated separately and support calculated for each. If, however, there were many proposals, not easily distinguished from each other, it would make little sense to think of there being congressional support for a clearly delineated proposal; attempts to describe congressional action would have to go in another direction.

As the research reported below was beginning, there was reason to hope that the proposals addressing particular issues would be few and distinct (Burstein 1998; Boli-Bennett and Meyer 1978). Thus, it made sense to describe congressional policy proposals and how support for them changes over time. But how should support be measured?

Measuring Support for Policy Proposals

There have been two approaches to measuring congressional support for policy proposals early in the policy process. Both conclude that sponsorship is a plausible measure of support.

The first approach is based on commonsense notions of what it means to say that congressional support for repeal of the estate tax is rising, or that support for a balanced budget amendment rose and then fell. For many historians, political scientists, and journalists, sponsorship is a good measure of such support (see the review in Burstein 1998: ch. 2).

A second approach was that of political scientists arguing on more theoretical grounds that sponsorship (including cosponsorship) may indicate support for legislative change. As part of the public record, sponsorship may be like rollcall voting, signifying support for particular policies (Krehbiel 1995: 906, 910; Talbert and Potoski 2002); it may used by legislators to win constituents' support or by their challengers to arouse opposition (Schiller 1995: 189); and it may be used by some legislators to signal others about commitment to a coalition favoring a bill.2

Though both approaches point toward using sponsorship as a measure of support, neither provides a satisfactory way to do so. The first gauges changes in support over time, but does not carefully define what is being supported, typically focusing on a vaguely defined set of bills that vary among themselves in ways not described. The second approach is clear about what is being supported-nearly always, single bills-but because bills die at the end of each congress, it cannot describe changes in support over time (see, e.g., Browne and Ringquist 1985; Campbell 1982; Kessler and Krehbiel 1996; Krehbiel 1995; Regens 1989; Wilson and Young 1997).

We propose to gauge support for policy proposals by combining the strengths of both approaches, taking into account support, over time, for all highly similar bills. If our attempt succeeds, our ability to study the earlier stages of the legislative process will be enhanced.

To evaluate the potential utility of the new measure, we ask four questions:

1. Does the new measure produce different results than previous ones? Specifically, does support as gauged by sponsorship of policy proposals differ substantially from support as gauged by sponsorship of single bills?

2. Does the new measure enable us to learn something new about the policy process? Here we ask specifically: Do members of Congress try to build support for favored proposals over fairly long periods of time, or do they give up and abandon proposals that do not quickly win enactment?

3. Do proposals get farthest in the legislative process when sponsorship is at a peak, as we would expect if sponsorship is a good measure of support?

4. Is sponsorship correlated with public opinion, as would be expected of a valid measure of support?

DATA

Because one of our goals is to generalize about congressional action, we must define a population and a sampling method. We began with the 5,977 public bills introduced into the 101st congress, 1989-1990 (U.S. Library of Congress 1998: Table 6-1). Because our concern was bills explicitly proposing new policies, appropriations bills were excluded from the sampling frame. A stratified random sample of House bills was chosen: 50 at random, and ten more that were reported out of committee (to ensure the inclusion of some bills that moved well into the legislative process; see appendix).3

Summaries of the 60 bills were found in the Thomas database of the U.S. Library of Congress (2001). To determine whether it made sense to think of the bills as manifestations of a policy proposal-of a particular set of elements, duplicated (or nearly duplicated) in a set of identical (or nearly identical) bills-the following procedure was devised: "Bill tracking reports" were found in the Lexis-Nexis Congressional Universe (2001) and the CRS index terms at the end of each report were used as key words to search for bills in the Thomas database. (The CRS index terms were more useful than those in Thomas, but the latter had more information on the bills themselves.) Where it seemed appropriate, variations in key words and in their order were used as well. If the original bill was an omnibus bill addressing multiple subjects, we selected the first substantive title and treated it as a bill for our purposes.

We then sought all bills with identical or virtually identical summaries in the 101st congress (both House and Senate), and previous and subsequent congresses. The search went back in time until the first such bill was found, and forward until the policy proposal was either enacted or disappeared from the congressional agenda. (None remain on the agenda.)

We found that it is indeed useful to think of legislative activity in terms of policy proposals. Sets of nearly identical summaries were not difficult to identify, or to distinguish from other summaries. Members of Congress regularly introduce bills summarized in virtually identical ways, both within congresses and over time, and rarely propose alternative ways of expressing the same policy ideas.4

We treated bills with virtually identical summaries as manifestations of the same policy proposal. It is these bills that comprise our data set.

Because the procedure for seeking nearly identical bills is new, we provide an example. One bill selected initially was H.R. 5598, the "Patent Competitiveness and Technological Innovation Act of 1990." This was an omnibus bill, so we selected Title I, called the "Patents in Space Act," as our bill. The summary of this bill, in Thomas, was

Title I: Patents in Space-Patents in Space Act-Amends Federal patent law to provide that any invention made, used, or sold in outer space on a space object or component under the jurisdiction or control of the United States shall be considered made, used, or sold within the United States unless: (1) it has been specifically identified and otherwise provided for in an international agreement which the United States has signed; or (2) it is carried on the registry of a foreign state in accordance with the Convention on Registration of Objects Launched into Outer Space. Declares that any invention carried on such registry shall be considered made, used, or sold within the United States if it has been specifically so agreed in an international agreement between the United States and the state of registry.

A search for bills in the 101st congress under "patents" produced 48 bills; "space policy" produced 38, some duplicating those related to patents. Three of the summaries were identical to that of H.R.5598: H.R. 352, 2946, and S. 459 (enacted as Public Law 101-580), and one was very similar, Title III of H.R. 5145. By way of comparison, H.R. 5145 says that it relates to inventions made in outer space "on a vehicle or payload" instead of H.R. 5598's "space object or component"; and that it applies "except for a vehicle or payload (1) under an international agreement" instead of H.R. 5598's "unless: (1) it has been specifically identified and otherwise provided for in an international agreement. . . ." Much of the rest was identical, and, overall, the similarities were so great that both were seen as manifesting the same policy proposal.

Other bills found in the search were very different; bills with the keyword "patents" dealt with transgenic farm animals (H.R. 1556), patent infringement by state governments (H.R. 3886), and other topics, while those under "space policy" dealt with weapons in space (H.R. 966), the export of satellites (H.R. 2624), etc. No bills on patents in space took another approach. Other keywords (e.g., "inventions") led only to bills that had already been found.

Identical searches for earlier congresses found relevant bills in the 99th and 100th congresses, but none earlier. Because S. 459 was enacted into law, there was no need to search subsequent congresses, but such a search was conducted for all bills not enacted during the 101st.

For each congress, we ascertained how many members of Congress sponsored any bill manifesting each proposal, and how far each proposal got in the legislative process, described in terms of the furthest advance of any relevant bill: (1) bills referred to committee only; (2) subject of a committee hearing; (3) reported out; (4) debated on floor but passed neither house; (5) passed one house; (6) passed both houses; (7) enacted into law (cf. Wilson and Young 1997).

FINDINGS

What the Policy Proposals Address

The 60 proposals vary widely by subject and potential impact. Of great importance were those enacted as the Americans with Disabilities Act, and the Financial Institutions Reform, Recovery, and Enforcement Act (popularly known as the "savings and loan bailout"). Some were of moderate significance, such as those enacted as the Student Right-to-Know and Campus security Act and the Amtrak Reauthorization and Improvement Act; of significance only to small groups were proposals addressing tanker traffic in Puget Sound, and pecan, mushroom, and lime promotion and research programs. Others address foster care, taxes on capital gains, the admission of refugees on an emergency basis, the disposal of solid waste, and the boundaries of Rocky Mountain National Park. A complete list is available on the authors' websites.

Cosponsorship of Bills vs. Sponsorship of Policy Proposals

The first question is whether the new measure produces new results. Does support as gauged by sponsorship of policy proposals differ substantially from support as gauged by sponsorship of single bills, providing additional information about the policy process?

The answer is that support gauged the new way differs a great deal from support gauged by sponsorship of single bills. Focusing on the 101st congress alone, the 60 original buis had a total of 1,190 sponsors, the 60 policy proposals (that is, all the identical bills), 2,643. What's more, Congress enacted ten of the original bills, but 15 of the policy proposals.

If the entire history of the policy proposals, from initial introduction to enactment or disappearance, is considered, the disparity is even greater. Here the focus is on cumulative support. Within a congress, sponsors of a policy proposal are counted only once. Over time, though, sponsors are counted separately for each congress in which they sponsor a particular proposal; because it requires action to reintroduce a proposal at the start of every congress, it makes sense to think of each re-introduction as conveying useful information about support.

Viewed this way, the 60 proposals had 7,090 sponsors during the entire time they were on the agenda, six times as many as the original 60 bills', 20 proposals were enacted, double the number of the original bills. Focusing on single bills, as most studies do, leads to substantially underestimating support for policy proposals (Wawro 2000: 31-32).

How Long Are Policy Proposals on the Agenda?

A view of the policy process increasingly popular among political scientists suggests that policy change comes about through the conjunction of long- and short-term processes. Developing a proposal and winning attention to it often take a long time, as proponents refine the proposal and try to soften up the system, pushing for consideration in many ways and in many forums (Kingdon 1994: 210, 226).

For proposals in the system, though, ultimate success or failure may be determined very quickly. Often, according to Kingdon, enactment depends upon the opening, typically for a short time, of a "policy window," a conjunction of circumstances that moves some proposals onto the decision agenda, to be either enacted or decisively rejected (Kingdon 1994: 174-75). If ingredients necessary for policy change are present-a well-articulated definition of a public problem, a proposed solution, mobilized interest groups, engaged political entrepreneurs-then rapid, dramatic change will occur; if key ingredients are missing, the opportunity will be lost.

Some aspects of this scenario have been examined in detail (Baumgartner and Jones 1993), but not what happens to specific policy proposals. Are many proposals actually on the agenda for a long time, as sponsors introduce and reintroduce bills, try to build support, and wait for a policy window to open? Or might proposals first be introduced when a window appears about to open, only to be withdrawn shortly if they fail to pass, as members of Congress conclude that the window never opened, or opened and closed too quickly?

It turns out that Congress tends to act quickly on proposals or not at all. Proposal time on the agenda averaged 2.6 congresses; only 22 percent were considered for four or more. Half the proposals enacted were enacted in just one congress, and four in each of the next two-18 of 20 within three congresses. Proposals that fail are on the agenda longer than those that succeed-2.8 congresses versus 2.2-but the difference does not seem large, and is not statistically significant.

Seldom do members of Congress try patiently to build support for a policy proposal while waiting for a policy window that will make enactment possible.5 Rather, they act more as if they believe the possible opening of a policy window provides an opportunity for quick enactment of a new proposal. If they prove wrong, they usually abandon their effort, acting as if one try, or maybe two or three, suffices to show that further effort would be pointless.

This is not to say that issues are on the agenda for only a short time, only particular ways of addressing them. Those who fail to win support for a policy proposal may abandon it, but maintain their general goal; they may try to reframe the issue and win support for a new proposal very different from the old (Baumgartner and Jones 1993: especially ch. 10). But nobody has any evidence on this point.

Our finding may still seem implausible to some readers, who may recall policy proposals on the agenda for a long time. Indeed, the sole previous analysis of a policy proposal (Burstein 1998) found some sponsors persisting for 30 years before winning most of what they wanted. It is here that the importance of random sampling becomes apparent. Burstein analyzed an equal employment opportunity proposal, part of an extraordinarily long and intense struggle for civil rights-the sort of struggle likely to stay in the minds of those who lived through it, and mislead them about how long most proposals are on the agenda.

Sponsorship of Proposals as a Measure of Support or Policy Change

If sponsorship is a good measure of support for policy proposals, it should be related to plausible consequences and causes of such support. Here we consider whether sponsorship is related to movement through the political process and to public opinion.

There are three ways in which sponsorship may gauge support. How far a policy proposal advances in the legislative process may be related to (1) the absolute number of sponsors it gets; (2) the number relative to other proposals; and (3) changes in the number-that is, to trends in sponsorship.6 If at least one of these is fairly strongly related to how far policy proposals progress, then sponsorship would seem a good measure of support.7

Past work suggests that advance through the legislative process is related to neither the number of sponsors (Krehbiel 1995) nor how many a proposal gets relative to others (Browne 1985; Wilson and Young 1997). What seems to matter are trends in support for particular proposals (Burstein 1998); proposals may be heterogeneous in ways not fully understood, so that sponsorships have the same weight for particular proposals over time, but vary among proposals.

Findings for the 60 policy proposals follow this pattern. Absolute number of sponsors is not a good measure of support; of the nine proposals that got the greatest number of sponsors in a single congress (100 or more), only a third were enacted, the same proportion as for all 60 proposals. Proposals that are enacted have more sponsors than those not enacted, but none of the correlations between total sponsorships and measures of legislative success-how far they got on the seven-category legislative action scale, and enactment"was statistically significant.

Trends in sponsorship, in contrast, do seem related to how far proposals get in the legislative process. Here it is necessary to reconceptualize the unit of analysis. Up to this point, the unit of analysis has been the policy proposal (N = 60). For the following analyses, however, the unit of analysis is each congress a proposal is on the agenda (as in Burstein 1998); call this-one proposal on the agenda for one congress-a congress-proposal. For example, the Americans with Disabilities Act was on the agenda for eleven congresses, so it contributes eleven observations-eleven congress-proposals-to our data. What we have may be thought of as very roughly similar to an event history analysis, in which we see whether a proposal moves from "on the agenda" to "enacted" when sponsorship is at a peak, and not otherwise.

First we consider whether enactment occurs when sponsorship is at its peak, as we would expect if sponsorship is a good measure of support. Ten of the 20 enacted proposals were on the agenda for more than one congress. Our N, the number of congress-proposals, is 35. When sponsorship peaked, enactment was likely; of the ten congresses when sponsorship was at a peak, proposals were enacted in seven (Table 1a, upper left cell). When sponsorship was lower, in contrast, proposals were not enacted; of 25 congresses when sponsorship was not at a peak (across the ten proposals), proposals were enacted in only three (second row). Altogether, enactment (or non-enactment) was consistent with peak (or non-peak) sponsorship in 29 of the 35 cases, or 83 percent.

Next we extend the analysis to all 31 proposals on the agenda for two or more congresses, whether ultimately enacted or not (Table 1b). The results are very similar to those for enacted proposals. Among the 89 congress-proposals, when sponsorship was at a peak, proposals got as far as they ever did (on the seven-category legislative action scale) in 14 of 22 instances (first row), and when sponsorship was not at its peak, proposals did not get as far in 59 of 67 instances (second row). Altogether, advance through the legislative process was consistent with peak and nonpeak sponsorship-in line with whal would be expected of a good measure of support-in 73 of the 89 cases, or 82 percent. Both sets of relationships-sponsorship with enactment and with advance through the legislative process-are statistically significant.

Public Opinion and the Sponsorship of Proposals

If sponsorship is a good measure of support for policy proposals, it should be related to public opinion, as enactment often is. And if trends in sponsorship seem to provide the best measure of support, it makes sense to relate trends in sponsorship to trends in public opinion. Thus, it makes sense to adopt Page and Shapiro's (1983) approach, gauging the impact of opinion on congressional action by showing how often changes in public opinion were congruent with (moved in the same direction as) changes in sponsorship.

Page and Shapiro (1983) analyzed the relationship between public opinion and public policy in the U.S. between 1935 and 1979 for all the issues on which they could find both public opinion and policy changing-231 issues. Opinion and policy were described as congruent when both moved in the same direction, and as noncongruent otherwise. They found congruence in opinion and policy in 66 percent of the 231 instances, noncongruence in the rest.

Our findings with regard to sponsorship are at least as strong. Data on trends in public opinion were available for 17 of the 60 proposals. Of these, trends in opinion and sponsorship were congruent in f6 (94 percent; for more details, see Burstein, Bauldry and Froese 2001). These results are what we would expect if opinion affected policy. And sponsorship is related to public opinion just the way we would expeci if il were a good measure of support for policy proposals.

Does this mean that trends in sponsorship are a good measure of support? There is no by-the-book way to answer this question. But the strength of the connection of sponsorship with enactment, movement through the legislative process, and public opinion seems consistent with that conclusion. What's more-and this is crucial-up to now there has never been a measure of support for particular policy changes from the time they are first introduced until they are enacted or disappear from the agenda. Using sponsorship as such a measure will make it possible to analyze the pre-policy phase of the legislative process in a way not possible before.

CONCLUSIONS

This article began by contending that our conclusions about policy change in the United States are not as convincing as they might be, because they are based disproportionately on studies of the final stages of the legislative process and on a very atypical set of issues.

Our goal has been to improve our ability to examine the earlier stages of the legislative process and to generalize from our results. To do this, we proposed three linked innovations: (1) operationalizing the concept of a "policy proposal;" (2) measuring support for such proposals in terms of sponsorships; and (3) analyzing a (stratified) random sample of policy proposals. We hoped to show that our approach can both produce new and interesting descriptions of congressional action, and make possible new work on its determinants.

We believe we have shown it possible to operationalize the concept of policy proposals in a manner that is both rigorous and consistent with everyday discourse about politics; that trends in sponsorship represent a good way to gaugesupport for such proposals; and that it is both practical and enlightening to sample such proposals randomly. We have found that traditional approaches greatly underestimate the amount of support for policy proposals; that proposals are generally on the congressional agenda for only a short time, and that trends in sponsorship appear to be a good indicator of support. Our findings about the importance of sponsorship parallel those in some recent work (Talbert and Potoski 2002; Wawro 2000), and should encourage further research on that aspect of the policy process.

This article has at least two implications for tuture research. First, the methods used provide a promising approach to the study of legislative failures as well as successes, less-visible issues as well as more visible ones, and the earlier stages of the legislative process. Measures of support for policy proposals may be used as dependent variables in studies of policy change, enabling us to test theories about the impact of public opinion, interest groups, the party balance, and other forces on support for policy change over the entire time proposals are on the agenda.

Second, our approach may improve our ability to describe the evolution of policy and the processes through which change proceeds. We have learned in recent years, for example, how proposals rejected in one venue may be reframed and introduced by their supporters into another (e.g., Baumgartner and Jones 1993). So far, however, there has been no way systematically to track the ebb and flow of support for particular proposals or to follow them through whatever transformations they undergo. The methods developed here may provide a way to do so.

Our effort must necessarily be seen as preliminary. We have not, in this article, used our approach to test hypotheses or help adjudicate between competing theories. We believe, however, that our approach is practical enough, and our results promising enough, to justify the efforts that would lead to hypothesis testing and theoretical conclusions.

[Sidebar]

NOTE: We would like to thank Frank Baumgartner, Alexander Hicks, Bryan Jones, Wendy Schiller, and John Wilkerson for helpful advice and comments. Work on this article was supported by NSF grant SES-0001509.

1 There are many studies of committee support for proposals, but this is not the same as support in the legislature as a whole.

2 Talbert and Potoski 2002; Wawro 2000: 30-32; Wilson and Young 1997: 28. Although it is sometimes important to distinguish between sponsorship and cosponsorship, often both together are seen as measures of support (Wawro 2000: ch. 2). Here "sponsor" includes both.

3 Sixty was the largest sample that could be managed with the resources available. Bills were sampled, not policy proposals; it was completely impractical to analyze so many bills and sort them into policy proposals before sampling.

4 Rarely were summaries in the gray area between extremely similar and clearly different. Then the entire bills were read and compared; a few times, we looked at interpretations of the bills in the Congressional Quarterly Almanac, academic journals, and the New York Times.

5 Ideas for policy change may have been developed in other venues after a policy idea has been turned into a bill, but our focus is on Congress.

6 We are not considering who the sponsors are-for example, if bills sponsored by committee chairs are more likely to pass. That will be the subject of future work.

7 Not, of course, as a perfect measure; see Krehbiel 1995: 921-22 on the difficulty of gauging legislators' support for particular policies.

[Reference]

REFERENCES

Baumgartner, Frank R., and Bryan D. Jones. 1993. Agendas and Instability in American Politics. Chicago: University of Chicago Press.

Baumgartner, Frank R., and Beth L. Leech. 1998. Basic Interests. Princeton, NJ: Princeton University Press.

Boli-Bennett, John and John W Meyer. 1978. "The Ideology of Childhood and the State." American Sociological Review 43: 797-812.

Browne, William p. 1985. "Multiple Sponsorship and Bill Success in U.S. State Legislatures." Legislative Studies Quarterly 10: 483-88.

Browne, William P., and Delbert J. Ringquist. 1985. "Sponsorship and Enactment: State Lawmakers and Aging Legislation, 1956-1978." American Politics Quarterly 13: 447-66.

Burstein, Paul. 1998. Discrimination, Jobs, and Politics. Chicago: University of Chicago Press

Burstein, Paul, Shawn Bauldry, and Paul Froese. 2001. "Public Opinion and Congressional Support for Policy Change." Paper presented at the annual meeting of the American Political Science Association, San Francisco.

Campbell, James E. 1982. "Cosponsoring Legislation in the U.S. Congress." Legislative Studies Quarterly 7: 415-22.

Edwards, George C., III, Andrew Barrett, and Jeffrey Peake. 1997. "The Legislative Impact of Divided Government." American Journal of Political Science 41: 545-63.

Kessler, Daniel, and Keith Krehbiel. 1996. "Dynamics of Cosponsorship." American Political Science Review 90: 555-66.

Kingdon, John W. 1994. Agendas, Alternatives, and Public Policy. New York: HarperCollins.

Krehbiel, Keith. 1995. "Cosponsors and Wafflers from A 10 Z." American Journal of Political Science 39: 906-23.

Krutz, Glen S. 2000. "Congressional Leaders and the Winnowing of Legislation." Paper presented at the annual meeting of the American Political Science Association, Washington, D.C., August 31-September 3.

Mayhew, David. 1974. Congress: the Electoral Connection. New Haven, CT: Yale University Press.

Page, Benjamin I., and Robert Y. Shapiro. 1983. "Public Opinion and Public Policy." American Political Science Review 77: 175-90.

Regens, James L. 1989. "Congressional Cosponsorship of Acid Rain Controls." Social Science Quarterly 70: 505-12.

Schiller, Wendy J. 1995. "Senators as Political Entrepreneurs: Using Bill Sponsorship to Shape Legislative Agendas." American Journal of Political Science 39: 186-203.

Smith, Mark A. 2000. American Business and Political Power. Chicago: University of Chicago Press.

Talbert, Jeffrey C., and Matthew Potoski. 2002. "Setting the Legislative Agenda: The Dimensional Structure of Bill Cosponsoring and Floor Voting." Journal of Politics 64: 864-91.

Thomas, Scott, and Bernard Grofman. 1993. "The Effects of Congressional Rules about Bill Cosponsorship on Duplicate Bills." Public Choice 75: 93-98.

United States Library of Congress. 2001. Thomas Legislative Information on the Internet. Available online: http://thomas.loc.gov.

_____. Congressional Research Service. 1998. Vital Statistics on Congress.

Wawro, Gregory. 2000. Legislative Entrepreneurship in the U.S. House of Representatives. Ann Arbor: University of Michigan Press.

Wilson, Rick K., and Cheryl D. Young. 1997. "Cosponsorship in the U.S. Congress." Legislative Studies Quarterly 22: 25-43.

[Author Affiliation]

PAUL BURSTEIN AND SHAWN BAULDRY, UNIVERSITY OF WASHINGTON

PAUL FROESE, BAYLOR UNIVERSITY

[Author Affiliation]

Received: January 30, 2004

Accepted for Publication: April 28, 2004

burstein@u.washington.edu

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.