(PDF) (DOC) (JPG)December 20, 2014

THOMAS BAUERSCHMIDT SWEENEY [*]

A defining feature of the United States Senate is senators’ division into electoral classes with staggered election cycles. Every two years one third of the Senate faces reelection to a new six-year term, while the other two classes continue their terms uninterrupted. The system of staggered elections causes distinct forms of political exchange between the classes of the Senate in a “cohort effect.” This study analyzes the effect of this political exchange in the context of “reversal legislation.” Reversal legislation occurs when Congress responds to a Supreme Court decision by passing legislation that effectively overturns the Court’s ruling. This paper examines the question of how Senate election cycles affect when senators cosponsor legislation reacting to high profile Supreme Court rulings, a question which falls at the center of several extensive fields of literature concerning elections and the dynamics between Congress and the Court. I create a dataset covering three instances of legislation reversing high profile Supreme Court decisions and test for the presence of a cohort effect using a multivariate regression. On the basis that senators are driven by the need for reelection, I predict that senators in the cohort facing reelection, next, will be more likely to cosponsor legislation that effectively reverses a high profile Supreme Court ruling. The results lead to a more nuanced understanding of when a cohort effect is likely to occur, as well as deeper understanding of reversal legislation.

Literature and Theory

It is easiest to begin with an examination of David Mayhew’s Congress: The Electoral Connection. In this piece, Mayhew establishes that senators are driven by the need for reelection, and “whether they are safe or marginal, cautious or audacious, congressmen must constantly engage in activities related to reelection” (Mayhew, 1974, p. 49). According to Mayhew, the focus on reelection permeates congressmen’s decision making and cannot be discounted.

Mayhew argues that representatives engage in three distinct forms of election seeking behavior: advertising, credit claiming, and position taking. The first of these forms, advertising, attempts to create a “favorable image” associated with a senator’s name. Essentially, “a successful congressman builds what amounts to a brand name” (Mayhew, 1974, p. 49). The image is generally empty of real content, primarily attempting to generate name recognition in voters. Generally speaking, “to be perceived at all is to be perceived favorably” (Mayhew, 1974, p. 50). However, congressmen attempt to associate themselves with qualities such as “experience, knowledge, responsiveness, concern, sincerity, independence, and the like” (Mayhew, 1974, p. 49). Incumbents generally have a significant advantage over challengers in advertising because of the resources at their disposal and the name recognition that is generated simply by being in office.

He defines credit claiming as the practice of generating a belief within constituents (or other relevant political actors) that the representative is responsible for desirable actions taken by government. “The emphasis here is on individual accomplishment (rather than, say, party or governmental accomplishment) and on the congressman as doer (rather than as, say, expounder of constituency views)” (Mayhew, 1974, p. 53). Credit claiming, according to Mayhew, most often occurs in the form of “particularized benefits.” These benefits are given to specific group, which is defined narrowly enough that a single congressman can claim credit for providing the benefits. Also, “each benefit is given out in apparently ad hoc fashion (unlike, say, social security checks) with a congressman apparently having a hand in the allocation” (Mayhew, 1974, p. 54). The most desirable of these credit claiming opportunities (from a congressmen’s point of view) are those which generate large, highly visible, concrete results. It is worth noting that a major limitation of credit claiming opportunities is a congressman’s ability to effectively claim credit for large, government-wide changes. At a certain scale, voters are unlikely to believe that a single congressmen is responsible for such massive changes, limiting the types of legislation/benefits congressmen can take credit for.

The third and final type of election seeking behavior Mayhew describes is “position taking,” defined as “the public enunciation of a judgmental statement on anything likely to be of interest to political actors” (Mayhew, 1974, p. 61). In this arena, “the electoral requirement is not that he make pleasing things happen but that he make pleasing judgmental statements” (Mayhew, 1974, p. 62). Mayhew argues that, while one of the most common forms of position taking is certainly role call votes, congressmen can be quite creative in generating opportunities for position taking. One option is cosponsoring legislation important to constituents. In terms of position taking, congressmen can cosponsor legislation and then go to their constituents, pointing to their support of a particular bill. This is distinct from credit claiming, since the emphasis is on the congressmen’s position on the bill, rather than attempting to generate the perception that the congressmen is responsible for the legislation, as would be the case in a credit claiming endeavor.

Richard Hall complicates this story with his thesis in Participation in Congress (Hall, 1996). Hall argues that “the limits of time and the press of legislative and nonlegislative demands are substantial and have become increasingly so in recent years” (Hall, 1996, p. 23). Essentially congressmen and women are playing a zero-sum game when it comes to participation in Congress, meaning that their limited time and resources restrict the number of ways they can engage the political process. “One of the first lessons that a neophyte legislator learns is that while the demands placed upon him seem to expand geometrically, the supply of time available to meet those demands is, by virtue of universal law, inelastic” (Hall, 1996, p. 28). The opportunities for participation far exceed any given congressman’s time and resources, which means that any effort spent pursuing one avenue of participation could have been spent on some other form of participation and congressmen are forced to choose one over the other.

Hall uses this theory of participation as a zero-sum game to frame the way political scientists ought to examine acts of participation. He argues that, while “the temptation is to distinguish between… those [activities] which are substantive and serious and those which are legislatively superfluous or symbolic… [this temptation] ought to be strongly resisted” (Hall, 1996, p. 25). Hall argues that because of the opportunity cost that accompanies any form of participation, there is no such thing as a superfluous or insignificant act of participation. Any form of participation must have been deemed more important than alternative forms, meaning that any action taken by congressmen and women holds a certain significance that should not be overlooked.

This theory meshes nicely with Mayhew’s arguments, particularly those concerning credit claiming and position taking opportunities. Hall argues that seemingly “superfluous or symbolic” legislative participation is in fact significant. Mayhew’s theory helps to explain the potential benefits of engaging in legislative activity that seems unlikely to generate productive legislation. Even if a congressmen’s support for a piece of legislation is unlikely to produce real change, such participation may be at the very least an opportunity to engage in position taking. Ultimately, Hall argues that Congressmen are faced with two choices: Whether or not to participate, and how intense the level of participation should be. Mayhew’s theory helps to explain the value and importance of supporting legislation that may be unlikely to actually pass. Any legislative action taken was chosen over alternative uses of congressmen’s resources, and is therefore significant. Even if participation is not expected to result in the passage of a bill, it may very well be a genuine attempt to generate opportunities for credit claiming or position taking and should be considered a significant political act.

This story is further extended by the work of Kenneth Shepsle regarding the effect of overlapping generations in the Senate. Shepsle seeks to explain “the implications of staggered terms” in predicting which senators support legislation that is particularly popular or unpopular among voters (Shepsle, Dickson, & Van Houweling, 2002, p. 2). Shepsle et al.’s argument assumes that constituents “observe a retrospective voting rule (Fiorina 1981) that we call the WHYDFML Principle (“What have you done for me lately?”)” (Shepsle, Dickson, & Van Houweling, 2002, p. 13). This assumption argues that the more recent a representative’s actions are, the more weight constituents will place on those actions in evaluating their senators. Shepsle et al. argue that the WHYDFML theory explains why senators engage in the behaviors outlined by Mayhew in an “especially intense and animated [way] just before elections,” but not why senators would be willing to “[take] ‘hits’ to reputation, [make] sacrifices, [cast] embarrassing votes, and [pass] up opportunities for gain” (Shepsle, Dickson, & Van Houweling, 2002, p. 30, 31).

Shepsle et al. make the point that these seemingly counterintuitive behaviors have been empirically observed, citing the example of a 1991 bill which raised Senate pay in spite of strong voter opposition. The authors argue that which senators supported the bill was largely influenced by the Senate’s structuring into cohorts with staggered election cycles. Shepsle et al. describe a model where senators operate within an “institutional norm” that creates a series of “inter-temporal deals.” The authors claim that, given the opportunity, all senators would place all of the legislation which constituents view positively at the end of their term, immediately before reelection, to maximize the effects of WHYDFML. However, since this is not a feasible option, senators engage in an exchange of sorts, where those senators who are not in the cohort next facing election take the hits, make unpopular votes, etc. with the understanding that the same will be done for them when their cohort is facing reelection.

Shepsle et al. acknowledge that “even with the effects of WHYDFML, [reputation burnishing] is the sort of activity in which [we would expect] a senator [to] engage almost continuously throughout his or her term” (Shepsle, Dickson, & Van Houweling, 2002, p. 30). WHYDFML does not explain why senators would forgo these types of opportunities. Shepsle et al. suggest a punishment regime. In their formulation, senators who break from the exchange by failing to give preference to senators facing reelection are punished later on. When the senator who violates the exchange finds him or herself facing reelection, he or she will be excluded from the exchange. This regime ensures that senators who are not facing reelection will generally give those facing reelection preference on legislation viewed positively by voters.

This result of overlapping generations in the Senate has implications for the way we understand a broad range of topics, particularly how the Senate interacts with other government institutions. In a 2009 study, Shepsle et al. examine how this exchange affects interactions between the Senate and the House. However, the ways in which this exchange might affect interactions between the Senate and the Supreme Court seems to have gone uninvestigated.

There is a broad base of literature covering relationships between Congress and the Courts. A traditional, oversimplified description of separated powers often places the Court beyond Congress’ sphere of influence. However, the literature indicates that there is in fact a complex and nuanced relationship between the two, and that this interaction is an inherent part of the United States legislative structure. The system was designed to blur the edges of interpretive authority between the three branches and prevent any single branch from becoming a final arbiter (Agresto, 1984) (Choper, 1980). Furthermore, this interaction is a political reality (Macedo, 1990) (Choper, 1980) (Meernik & Ignagni, 1997) (Hettinger & Zorn, 2005).

The precise nature of the relationship between Congress and the Court is complex, nuanced, and somewhat contested. Some authors suggest that while the Court may act strategically when required, it is relatively insulated from Congressional backlash (Segal, Westerland, & Lindquist, 2011) (Segal J. A., 1997) (King, 2007). Other authors such as Anna Harvey and Barry Friedman argue that “the Court is bound by institutional constraints to take account of congressional preferences when making constitutional decisions” (Harvey & Friedman, 2006). Harvey and Friedman argue that these constraints are applied to the Court in constitutional cases as well as statutory rulings. In their 2006 study, Harvey and Friedman do not attempt to identify exactly “where congressional constraints bind [the Court],” or what precisely those mechanisms are.

Tom Clark explores one possible point of constraint in a piece from 2009. Clark argues that Court curbing is a crucial mechanism in Court-Congress relations. Clark’s argument brings together two bodies of literature, covering Court curbing and statutory reversals. While these two topics have often been treated distinctly, Clark brings them together, arguing that “Court curbing can affect judicial independence because it can be a credible signal about waning judicial legitimacy… In particular, the justices believe that legislative attacks on the Court are signals about a lack of public support for the Court” (Clark, 2009, p. 972, 973). Clark argues that Congressional responses to Court rulings can serve as a mechanism to signal the Court on public opinion, and that the Court adjusts its behavior accordingly.

Research on citizen awareness of Supreme Court decisions further informs this theory. James Gibson and Gregory Caldeira argue that the average citizen pays far more attention to the Supreme Court than is generally thought. In a 2009 piece, Gibson and Caldeira examine two data sets generated from national surveys, finding that “people are vastly more knowledgeable about politics than is typically portrayed” (Gibson & Caldeira, 2009, p. 435). Citing work by Johnston, Hagan, and Jamieson (2004) and Valerie Hoekstra (2000,2003), Gibson and Caldeira argue that, while citizens are generally well informed regarding the Court, they are particularly so “as an election approaches,” when the controversies involved are well covered by the media, and when a decision directly affects them (Gibson & Caldeira, 2009). This grants further legitimacy to theories that public opinion may affect the Court via Congressional action. Additionally, it opens the possibility for congressmen and women to leverage Court decisions for election-seeking purposes.

If voters are aware of Court rulings, and particularly so prior to elections, Congressional response to unfavorable Court decisions may serve as powerful position taking and credit claiming opportunities for representatives, particularly in response to high profile decisions. This concept is further supported by the work of James Meernik and Joseph Ignagni (1997). Meernik and Ignagni examine Congressional attempts to reverse constitutional rulings by the Supreme Court. Meernik and Ignagni argue that “because its members are substantially motivated by the drive for reelection, they give voice to the views of the general public, special interest groups, and the interests of the states they represent” (Meernik & Ignagni, 1997, p. 451). Their data supports this argument, as they find that the most influential factor in predicting attempts to reverse a constitutional ruling by the Court is public awareness.

This paper seeks to examine this complex relationship between the Court, Congress, and the public. Specifically, if the cohort effect observed by Shepsle affects senators’ decisions to cosponsor legislation that attempts to reverse a salient Supreme Court ruling. My broad hypothesis is that senators in the cohort next facing reelection will be more likely to cosponsor legislation reversing Supreme Court rulings. The opportunity to chastise the court may provide an opportunity for Mayhew’s credit claiming, and certainly creates opportunities for position taking. The significant impact of public opinion on reversal legislation, found by Meernik and Ignagni (1997), suggests that Senators will be conscious of publicly salient cases and have an incentive to capitalize on such an opportunity. Shepsle’s research indicates that the members of the other two cohorts have an incentive to back off and give these opportunities to members of the vulnerable cohort. I examine co sponsorship with the idea that it is a more substantial and meaningful (according to Hall’s theory) form of participation than simple roll call votes, as well as the fact that it extends the range of possible pieces of legislation to bills that were never voted on. I analyze reversals of statutory and constitutional rulings to cover a broader variety of high profile Supreme Court rulings.

Methods

For this study I constructed a dataset covering three separate instances in which Senate legislation was introduced which would effectively reverse a ruling by the Supreme Court. Each piece of legislation was introduced in a different Congress, creating three hundred distinct observations. For each subject, a variety of data was collected, discussed below in greater detail. The data was then run through a multivariate probit regression, controlling for several important variables. To begin, I selected three pieces of legislation responding to salient Supreme Court rulings. The study is restricted to legislation responding to salient cases for several reasons. One is that “congressional overrides typically occur only in particularly salient or egregious cases of statutory misinterpretation” (Hettinger & Zorn, 2005, p. 8). Additionally, the theory of electoral incentives previously discussed indicates that Senators will have the greatest incentive to cosponsor reversal legislation when public awareness is high. High public awareness makes position taking and credit claiming more effective, making participation in high profile situations more beneficial for senators seeking reelection. While it may be possible and worthwhile to examine differences between highly publicized cases and less salient decisions, that is beyond the scope of this paper.

The cases in question all appeared in both the New York Times and Wall Street Journal, the two most widely read newspapers in the United States (Paperboy, 2013). The Court rulings were controversial and highly publicized, making them ideal selections for this study. The first decision drew both strong criticism and staunch support from opposing public policy groups on the issue (Greenberger, 2002). The second and third decisions were each opposed by a majority of the American public (Eggen 2010; Castle Coalition 2014). The high levels of publicity and controversy surrounding these bills should provide ideal opportunities for credit claiming and position taking, making these cases ideal selections for this study. Each bill selected specifically references the corresponding Court ruling, and makes clear the intent of the legislation to reverse the effects of the Court’s decision. Each bill was introduced in a different congress, with a different class facing reelection for each bill. This selection process is designed to create variety within the dataset while still focusing on cases that are likely to matter to the general public, allowing us to test if a cohort effect makes senators more likely to cosponsor reversal legislation of high profile Supreme Court rulings.

 

Bills

S.2520: PROTECT Act

The first bill included in the dataset is the Prosecutorial Remedies and Tools Against the Exploitation of Children Today Act (PROTECT ACT). In April of 2002, the Supreme Court delivered a ruling striking down the Child Pornography Prevention Act (CPPA) of 1996. The Court ruled that the bill was overbroad in proscribing certain types of images, particularly the banning of “any … [virtual,] computer-generated image or picture” that “is, or appears to be, of a minor engaging in sexually explicit conduct,” and “any sexually explicit image that is ‘advertised, promoted, presented, described, or distributed in such a manner that conveys the impression’ it depicts ‘a minor engaging in sexually explicit conduct’” (The Oyez Project, 2014). The ruling dealt a blow to prosecutors nationwide who “had come to rely on the law in bringing prosecutions” (Greenhouse, ‘Virtual’ Child Pornography Ban Overturned, 2002).

The Senate responded quickly. On May fifteenth, Senator Hatch introduced the PROTECT Act. In his opening statement, Senator Hatch stated, “While I firmly respect the Supreme Court’s role in interpreting the Constitution, the decision left some gaping holes in our nation’s ability to prosecute child pornography effectively”

(Senator Hatch (UT), 2002). Senator Hatch went on to highlight the most significant provisions of the act, including provisions to empower “prosecution to proceed when the child pornography includes persons who appear virtually indistinguishable from actual minors,” and prohibiting “the pandering or solicitation of anything represented to be obscene child pornography” (Senator Hatch (UT), 2002). Nine other senators joined with Senator Hatch, cosponsoring the legislation, and the bill was passed in the Senate with unanimous consent (Library of Congress, 2002).

 

S.1313: Protection of Homes, Small Businesses, and Private Property Act

The second bill included in the data set comes out of the 109^th Congress. In 2005, the court ruled on the case of Kelo v. City of New London. The case arose out of the Connecticut town of New London, where the city seized citizen’s private property with the intent of selling the land to private developers to stimulate the economy. The city claimed that such a seizure was viable under the Fifth Amendment, but “the property owners argued taking private property to sell to private developers was not public use” (The Oyez Project, 2014). The court handed down a 5-4 decision in favor of New London, ruling “that public use was properly defined more broadly as ‘public purpose’” (Greenhouse, Justices Rule Cities Can Take Property for Private Development, 2005). Many viewed the ruling as an assault on property rights, including Justice O’Connor, who warned “ The government now has license to transfer property from those with fewer resources to those with more” (Greenhouse, Justices Rule Cities Can Take Property for Private Development, 2005).

The Senate responded with the Protection of Homes, Small Businesses, and Private Property Act of 2005. Introduced by Senator John Cornyn, the bill specifically addressed the Kelo ruling, describing the case as “alarming,” and arguing that “the Court has `effectively . . . delete[d] the words `for public use’ from the Takings Clause of the Fifth Amendment’ and thereby `refus[ed] to enforce properly the Federal Constitution’”

(Library of Congress, 2005). The bill’s provisions were incredibly straight forward, stating simply that “the power of eminent domain shall be available only for public use,” “the term `public use’ shall not be construed to include economic development,” and that the act applies to federal and state uses of eminent domain (Library of Congress, 2005). The bill was cosponsored by thirty-two senators in addition to Senator Cornyn, and ultimately died after being referred to the committee on the judiciary.

 

S.3295: DISCLOSE Act

The third bill included in the data set was introduced in reaction to the decision by the Supreme Court in the case of Citizens United v. Federal Election Commission. The case was brought by Citizens United, against the FEC over Citizens United’s movie titled Hillary: The Movie. Citizens United argued that portions of the Bipartisan Campaign Reform Act of 2002 were unconstitutional violations of free speech (The Oyez Project, 2014). In a “sharp doctrinal shift” the Court ruled in favor of Citizens United, overruling precedents established in Austin v. Michigan Chamber of Commerce (1990) and McConnell v. Federal Election Commission (2003) (Liptak, 2010). The ruling was quite broad, and liberated corporations to spend unrestricted and undisclosed quantities in independent expenditures supporting or attacking political candidates. The case received a huge amount of publicity and was hotly contested among the Justices, with the minority accusing the majority of committing “a grave error in treating corporate speech the same as that of human beings” (Liptak, 2010).

The Senate responded within months, when Senator Charles Schumer introduced the Democracy Is Strengthened by Casting Light On Spending in Elections Act (DISCLOSE Act). In the bill’s General Findings, it is stated that the American public is right to attempt to regulate corporate spending in elections, and that “The Supreme Court’s decision in Citizens United v. Federal Election Commission on January 21, 2010, reverses established jurisprudence and sound policy to greatly increase the dangers of undue special interest influence over the democratic process. That decision has opened the floodgates for corporations and labor unions to spend unlimited sums from their general treasury accounts to influence the outcome of elections” (Library of Congress, 2010). The bill was designed to “[close] certain loopholes and [rely] on enhanced disclosure… to level the political playing field so that special interests do not drown out the voice of the average voter” (Senator Schumer (NY), 2010). Forty-nine senators joined Senator Schumer in cosponsoring the legislation, and the bill eventually died after being referred to the Committee on Rules and Administration.

Variables

Below is a discussion of the variables included in the regression analysis.

 

Cosponsorship

The main dependent variable for this study is whether or not a senator cosponsors legislation. Sponsorship is lumped in under this category. While it is probable that there are differences between motivations for sponsorship and cosponsorship of legislation, there are not sufficient observations in this dataset to establish a meaningful relationship, and the logic of electoral incentives and cohort effect still apply. Cosponsorship is used as the dependent variable because it represents a measurable effort of participation, which is inherently significant (Hall, 1996). Additionally, focusing on cosponsorship allows for the inclusion of bills which never go to a vote. This is important because even if a bill is never voted on, it may still represent a legitimate attempt at position taking by a senator, and possibly credit claiming as well. To exclude such legislation would be to ignore potentially significant participation in the Senate. Data on cosponsorship was obtained from the Library of Congress’ online THOMAS database, and is coded as a binary variable with 0 indicating that a given senator did not cosponsor the legislation while 1 indicates that the senator did cosponsor the bill.

 

Senate Class

The primary independent variable used is a senator’s electoral class. This variable is based on Shepsle’s theory of staggered election cycles. Shepsle’s theory indicates that senators that are not facing reelection in the next election cycle will give preferential legislation to the class that is facing reelection. As such, this variable is also coded as a binary, with 1 indicating that the senator is in the class next facing reelection, and 0 indicating that he or she is a member of either of the other two classes. According to Shepsle’s theory, we expect to see a positive relationship between the Class variable and Cosponsorship. A positive relationship would indicate that the cohort effect is at work, and senators not facing reelection are giving preference to the class facing reelection in the coming election cycle. The legislation included in the dataset comes from the 107^th, 109^th, and 111^th Congresses. In each of these congresses a different class was facing reelection, giving a consistent variation within the data.

 

 

Electoral Vulnerability

This variable is added to accommodate variation in electoral pressures facing senators. Senators who are likely to face tough electoral competition in coming elections are likely to be more motivated to engage in the election-seeking behaviors described by Mayhew than senators in safer seats. More specifically in this case, senators are expected to engage in position taking and possibly credit claiming by cosponsoring reversal legislation. The electoral vulnerability variable is included to control for this dynamic. The electoral competition facing a senator is operationalized in the form of a Holbrook and Van Dunk index (HVD index). The index is formed using “district-level state legislative election results that account for the average margin of victory along with the presence of uncontested and ‘safe’ seats” (Shufeldt & Flavin, 2012). A higher HVD index indicates more intense competition in elections.

The HVD index is used rather than a Ranney index because the HVD index was designed to measure the level of competition between candidates, while the Ranney index was intended to measure the level of party competition within a state. Accordingly, “researchers should use the HVD measure when the mechanism of interest is the electoral pressure a legislator (and future candidate for reelection) faces from his or her constituents” (Shufeldt & Flavin, 2012). The HVD measure for each year was obtained from data publicly available from Carl Klarner through Indiana State University. For each observation, a four-year HVD index for that particular year was recorded, with a predicted positive correlation between the HVD index and cosponsorship.

 

Distance From Median ADA Score

This variable is included in order to account for ideological factors in the legislation. Cosponsoring legislation will only electorally benefit senators whose constituents support the bill’s position. A conservative senator, with conservative constituents, is not likely to benefit from supporting a piece of liberal legislation. The expectation of the cohort effect is dependent upon senators seeking reelection, and therefore it is crucial to control for the ideology of the bills included in the study. In order to do this, I use rankings from Americans for Democratic Action. ADA scores are based on a senator’s votes on twenty pieces of critical legislation from a given year. Scores range from 0 to 100, with 0 being totally conservative and 100 being completely liberal. For each piece of legislation I calculate the median ADA score of the senators who cosponsor the bill. Using the median ADA score as a measure of how liberal or conservative the bill is, I then calculate the difference between each senator’s ADA score and the median score for the bill. This number is converted to an absolute value to create a measure of distance from the median score, with a higher number indicating a greater ideological distance from the bill. The expectation for this control is an inverse relationship with cosponsorship, presuming that senators who are ideologically distant from a bill will be less likely to support, let alone cosponsor the legislation.

Tenure in office

This control is relatively straightforward. The measure is the number of years a senator had been in office at the time the legislation was introduced. The expectation is that senators with longer tenures will have a greater understanding of the opportunities presented by such legislation. In addition to appreciating the opportunity, veteran senators are more comfortable and experienced in introducing legislation, and are therefore more likely to engage in this type of behavior (Schiller, 1995). Therefore, tenure in office is expected to have a positive relationship with cosponsorship, operating on the theory that more experienced senators will have a greater understanding of the opportunity as well as being more comfortable cosponsoring legislation challenging Court rulings.

 

Leadership Position

This is a variable recording whether or not a senator holds a leadership position within the Senate. For this dataset the definition of a leadership position was obtained from the Senate Historical Office and includes the President pro tempore, Majority and Minority Leaders and Whips, the Republican Conference Chairman, Policy Committee Chairmen for both parties, and Senate Campaign Committee Chairmen from both parties. The information was obtained from the Congressional Directory for the 107^th, 109^th, and 111^th Congresses. The prediction for this variable is that there will be a positive correlation with cosponsorship, on the basis that senators in prominent positions attract more pressure from constituents and interest groups to propose legislation than less prominent senators (Schiller, 1995).

Variables and Specific Hypotheses

Dependent Variable: Cosponsorship of legislation, defined to include both bill introduction and cosponsorship, coded as a binary.

0 = Did not cosponsor bill

1 = Cosponsored bill

Figure 2.1: Independent Variable & Controls

* Primary Independent Variable

Hypotheses

• Hypothesis 1: Membership in the class facing reelection in the next election cycle will increase the likelihood of a senator cosponsoring reversal legislation.
• Hypothesis 2: Senators who are more electorally vulnerable will be more likely to cosponsor reversal legislation.
• Hypothesis 3: Senators who are ideologically distant from the bill will be less likely to cosponsor reversal legislation.
• Hypothesis 4: A longer tenure will make a senator more likely to cosponsor reversal legislation.
• Hypothesis 5: Holding a leadership position will make a senator more likely to cosponsor reversal legislation.

Data Analysis

Initial results of the regression show the P-value for the likelihood ratio chi squared value to be 0, indicating that the variables are related in a significant way. The coefficients for each variable are shown below.

Figure 3.1: Regression Results

* P-Value < 0.10    ** P-Value < 0.05  *** P-Value < 0.01

The most influential control is clearly the ADA Score, which controls for a senator’s ideological position relative to the bill’s other sponsors. This variable also passes the most robust measure of statistical significance, with a P-Value below 0.01. The other statistically significant variables were a senator’s tenure at the time the bill was introduced and whether or not the senator held a leadership position. The measures for senate class and electoral vulnerability were not statistically significant.

Results of Hypotheses

Hypothesis 1: Membership in the class facing reelection in the next election cycle will increase the likelihood of a senator cosponsoring reversal legislation.

The hypothesis for Senate class returned statistically insignificant results, indicating that the cohort effect described by Shepsle et al. was not apparent.

Hypothesis 2: Senators who are more electorally vulnerable will be more likely to cosponsor reversal legislation.

The hypothesis for electoral vulnerability also returned statistically insignificant results, indicating that electoral vulnerability was not an influential factor in a senator’s decision to cosponsor legislation.

Hypothesis 3: Senators who are ideologically distant from the bill will be less likely to cosponsor reversal legislation.

The hypothesis for ideological position relative to the median cosponsor returned statistically significant results. The coefficient is far and away the largest, and in the expected direction with an inverse relationship. This indicates that ideological positioning does play a major role in a senator’s decision to cosponsor legislation, and that the more ideologically distant a senator is from those sponsoring the bill the less likely he or she is to cosponsor the legislation.

Hypothesis 4: A longer tenure will make a senator more likely to cosponsor reversal legislation.

The hypothesis for a senator’s tenure returned statistically significant results, but in the opposite direction than was expected. This indicates that, rather than being more likely to cosponsor reversal legislation, the longer a senator held a seat, the less likely he or she was to cosponsor the bill. The coefficient is clearly the smallest of the statistically significant controls, however it is important to note the difference in the variable being measured. The coefficient for holding a leadership position (discussed below) is substantially larger than the tenure coefficient, but the leadership variable is a binary, meaning that at most it can by multiplied by a factor of one. Tenure on the other hand can be multiplied by notably larger factors, e.g. a senator with a tenure of twenty years would multiply the coefficient by a factor of twenty. Therefore, while the tenure coefficient is notably smaller than the coefficient for holding a leadership position, one variable does not necessarily have a measurably greater impact on the overall outcome.

Hypothesis 5: Holding a leadership position will make a senator more likely to cosponsor reversal legislation.

This hypothesis returned statistically significant results, but also in the opposite direction than was predicted. This indicates that holding a leadership position makes a senator significantly less likely to propose reversal legislation. The coefficient is notably smaller than that for the ideology measure, indicating that holding a leadership position is less influential than ideology, but as discussed above its overall impact may be comparable to that of the tenure variable.

Discussion and Conclusions

The returns on most of the hypotheses were unexpected, but consistent with one another. The primary hypothesis, that there would be a measurable cohort effect in sponsoring reversal legislation of high profile Supreme Court rulings, was disproven. This could indicate that senators do not view reversal legislation as a worthwhile exercise in position taking or credit claiming. It may very well be that even in high publicity cases constituents are not sufficiently aware of the Court’s decision or of the responding legislation to make reversal legislation a practical position taking mechanism. If such legislation would influence only a small fraction of constituents, Hall’s theory would indicate that senators are likely to pursue other, more efficient methods of position taking. It is important to note that this explanation does not necessarily contradict authors such as Gibson and Caldeira, who argue that citizens are well informed regarding Supreme Court action. Citizens may be conscious of Court decisions, but less so when it comes to legislation responding to those decisions. It may also be the case that citizens are aware and informed on rulings and reversal legislation, but senators still perceive a lack of constituent awareness and act accordingly.

The lack of an observable cohort effect raises interesting questions for Shepsle’s work. It seems likely that the cohort effect is more pronounced in different types of legislation. These results indicate that reversal legislation is not an area in which a cohort effect is measurable. However, this does not necessarily run counter to Shepsle’s theory. Given the electoral incentives driving the cohort effect, legislation which is well suited for position taking and credit claiming will likely demonstrate a strong cohort effect. Measuring the intensity of a cohort effect across various types of legislation could potentially allow scholars to identify the types of legislation which senators view as most beneficial in pursuing reelection. In this way, the cohort effect may be used to better understand senators’ motives in passing particular types of legislation.

The second hypothesis, that senators facing highly contested elections would be more likely to cosponsor reversal legislation, was also disproven. The conclusion that the cohort effect was not in play would lead us to expect this result. Additionally, the fact that electoral pressure did not play a significant role reinforces the theory that senators do not perceive this type of legislation to be significant to voters. In fact, if senators feel there are more efficient ways to use their time and resources, those facing contentious reelections would have an even greater incentive not to cosponsor such legislation, as it would detract from opportunities to engage in more efficient forms of position taking and credit claiming.

The third hypothesis, that senators ideologically distant from a bill would be less likely to cosponsor the legislation, was the only hypothesis which returned the expected results. The results are by no means revolutionary, but they may also tie into ideas of position taking and reelection. If constituents are not naturally aware of reversal legislation, there would be little to be gained from cosponsoring a bill. In such a situation, senators may default to party allegiance, reinforcing their position as a member of their respective party. Even if voters are not aware of a bill’s specific points, cosponsoring a bill with similarly aligned senators would allow senators to add one more piece of legislation to their résumé that reinforces their ideological position. In this situation, cosponsorship would be more of an opportunity for senators to simply maintain their ideological alignment.

The fourth hypothesis, that a longer tenure would make senators more likely to cosponsor reversal legislation, returned statistically significant results, but in the opposite direction than was predicted. This factor is difficult to explain, but may be due to longer-serving senators’ greater experience and specialization. It is possible that well established senators better understand how to focus their energies in pursuit of reelection, and appreciate that reversal legislation appears to hold low salience for voters, while newer senators are less informed, and therefore more willing to spend time and energy cosponsoring legislation which is of little significance to voters.

Another possible explanation for newer senators’ greater inclination to cosponsor reversal legislation comes from Hall’s Participation in Congress. Hall endorses a theory of “rational apprenticeship,” in which freshman legislators tend to be less active than their more senior colleagues because they “consistently suffer from informational disadvantages… [and] a relative shortage of opportunities, networks, and resources to pay the marginal information and political transaction costs” (Hall, 1996, p. 102). It may be the case that less experienced senators who lack the social networks and informal influence carried by veteran legislators are forced to resort to more formal tools such as cosponsorship to pursue their political goals. More senior senators, on the other hand, may be less likely to cosponsor reversal legislation if they have more efficient and effective means than are available to less experienced senators.[†]

The fifth hypothesis, that holding a leadership position would make a senator more likely to cosponsor reversal legislation returned statistically significant results in the opposite direction than expected as well. While the hypothesis was disproven, it is in keeping with the other results. The expectation that leadership would be more likely to cosponsor the legislation was based on the premise that Senate leaders receive additional pressure from constituents and interest groups to propose and support legislation. However, the results from the other hypotheses indicate that reversal legislation is relatively insignificant in the minds of voters. If this is the case, then constituents and special interests would be pressuring Senate leadership to focus their energies on other forms of legislation, which would lead us to expect a leadership position to decrease the likelihood of cosponsoring reversal legislation.

While the results of this study largely disprove the proposed hypotheses, they nevertheless reveal important dynamics at work within the Senate. The results show that the cohort effect does not occur in the context of reversal legislation. This would appear to be in large part due to public indifference towards such legislation, or at least the senatorial perception of public indifference. The lack of a measurable cohort effect opens the door for a more nuanced understanding of Shepsle’s theory. These results indicate that the cohort effect may be restricted to certain types of high salience legislation, where voters are likely to be more responsive. If this is the case, testing for a cohort effect could allow scholars to better evaluate senators’ motives in supporting legislation. In instances with a measurable cohort effect, electoral concerns would most likely be at the forefront of a senator’s decision calculus. If there were not an observable cohort effect, as in this study, it would be likely that senators were being driven by motives other than pleasing constituents.

Additionally, the inverse relationships shown between tenure in office and cosponsorship, as well as the inverse relationship between holding a leadership position and cosponsorship, indicate that tenure and holding leadership positions affect the type of legislation a senator will sponsor. It may be the case that senior senators and Senate leaders are under greater pressure from constituents to sponsor certain types of legislation, and are therefore less able to spend resources on reversal legislation. It may also be the case that senior senators and leadership have knowledge that newer senators lack regarding the benefits and costs of reversal legislation, and that they are self-motivated to avoid such legislation. Further research into these motives would lend to a greater understanding of cosponsorship in the Senate.

This study creates several starting points for further research. The first is continued statistical analysis of the cohort effect in response to reversal legislation of high profile Court rulings. The scope of this dataset is relatively narrow, restricted to three instances of reversal legislation within the Senate. A more extensive dataset, covering a larger variety of reversal legislation would allow for important corroboration of these results. In addition to further examining cosponsorship of reversal legislation, future research might compare the presence of a cohort effect in various types of legislation. If legislation proven to be of high salience to voters demonstrates a more pronounced cohort effect, scholars could potentially test for a cohort effect as a means of determining senators’ motivations for supporting various types of legislation. In addition to the cohort effect, the results from the tenure and leadership position variables indicate that senior and high ranking senators are either pressured by constituents to support bills other than reversal legislation, or they are aware of some other factor at play which motivates them to avoid such legislation. Further examination of this dynamic would help develop the understanding of reversal legislation and why senators do or do not support such legislation.

The broad conclusion of this study is that the cohort effect is not observed in reversal legislation of high profile Supreme Court decisions. The results indicate that electoral factors do not play a significant role in a senator’s decision to cosponsor reversal legislation. In turn, this indicates that reversal legislation is not an efficient arena for the position taking or credit claiming activities described by Mayhew. Given the insignificance of electoral factors, there would appear to be forces other than reelection motivating senators to check the Court through reversal legislation. The results also indicate that there is a relationship between a senator’s seniority/rank and his or her motivations for sponsoring this type of legislation. These dynamics give us a more nuanced understanding of the cohort effect described by Shepsle, indicating that the exchange between Senate classes may be restricted to legislation that is highly salient for constituents. This more nuanced understanding of the cohort effect creates the possibility for future use of Shepsle’s theory in evaluating senators’ motivations as well as the salience of particular forms of legislation for constituents.

References:

Agresto, John. (1984). The Supreme Court and Constitutional Democracy. Ithaca, New York: Cornell University Press.

Castle Coalition. (2014). The Polls Are In. http://www.castlecoalition.org/index.php?option=com_content&task=view&id=43 (accessed 2014, 7-October).

Choper, Jesse H. (1980). Judicial Review and the National Political Process. Chicago: University of Chicago Press.

Clark, Tom S. (2009). “The Separation of Powers, Court Curbing, and Judicial Legitimacy.” American Journal of Political Science (Midwest Political Science Associattion) 53 (4). Pp. 971-989.

Eggen, Dan. (2010). Poll: Large majority opposes Supreme Court’s decision on campaign financing. 2010, 17-February. http://www.washingtonpost.com/wp-dyn/content/article/2010/02/17/AR2010021701151.html (accessed 2014, 7-October).

Gibson, James, and Gregory Caldeira. (2009). “Knowing the Supreme Court? A Reconsideration of Public Ignorance of the High Court.” The Journal of Politics (Cambridge University Press) 71(2). Pp. 429-411.

Greenberger, Robert. (2002). Supreme Court Strikes Down Ban On ‘Virtual’ Child Pornography . 17-April. http://online.wsj.com/articles/SB1018955258589085120 (accessed 2014, 7-October).

Greenhouse, Linda. (2002). “’Virtual’ Child Pornography Ban Overturned.” New York Times. 17-April. http://www.nytimes.com/2002/04/17/us/virtual-child-pornography-ban-overturned.html (accessed 2014, 9-April).

____(2005). “Justices Rule Cities Can Take Property for Private Development.” New York Times. 23-June. http://www.nytimes.com/2005/06/23/politics/23wire-scotus.html?pagewanted=all&_r=0 (accessed 2014, 9-April).

Hall, Richard L. (1996). Participation in Congress. New Haven: Yale University Press.

Harvey, Anna, and Barry Friedman. (2006). “Pulling Punches: Congressional Constraints on the Supreme Court’s Constitutional Rulings, 1987-2000.” Legislative Studies Quarterly (Comparative Legislative Research Center) 31(4). pp. 533-562.

Hettinger, Virgina A., and Christopher Zorn.(2005). “Explaining the Incidence and Timing of COngressional Responses to the U.S. Supreme Court.” Legislative Studies Quarterly (Comparative Legislative Research Center) 30(1). pp. 5-28.

King, Chad M. (2007). “Strategic Selection of Legal Instruments on the U.S. Supreme Court.” American Politics Research 35, no. 621.

Library of Congress. Bill Summary & Status, 107th Congress, S.2520: All Congressional Actions. 2002, 15-November. http://thomas.loc.gov/cgi-bin/bdquery/z?d107:SN02520:@@@X (accessed 2014, 9-April).

—. “Bill Text 109th Congress S.1313.IS.” Thomas. 2005, 27-June. http://thomas.loc.gov/cgi-bin/query/z?c109:S.1313: (accessed 2014, 9-April).

—. “Bill Text, 111th Congress: S.3295.IS.” Thomas. 2010, 3-May. http://thomas.loc.gov/cgi-bin/query/F?c111:1:./temp/~c111zZHvJp:e1719: (accessed 2014, 9-April).

Liptak, Adam. (2010). “Justices, 5-4, Reject Corporate Spending Limit.” New York Times. 21-January.
http://www.nytimes.com/2010/01/22/us/politics/22scotus.html?pagewanted=all (accessed 2014, 9-April).

Macedo, Stephen. (1990). Liberal Virtues. Oxford: Oxford University Press.

Mayhew, David R. (1974). Congress: The Electoral Connection. New Haven: Yale University Press.

Meernik, James, and Joseph Ignagni. (1997). “Judicial Review and Coordinate Construction of the Constitution.” American Journal of Political Science 41 (2). pp. 447-467.

Paperboy. (2013). List of the Most Popular and Influential Newspapers in the USA – See more at: http://www.thepaperboy.com/usa-top-100-newspapers.cfm#sthash.BlsjGgOH.dpuf. 2013. http://www.thepaperboy.com/usa-top-100-newspapers.cfm (accessed 2014, 9-April).

Schiller, Wendy. (1995).“Senators as Political Entrepreneurs: Using Bill Sponsorship to Shape Legislative Agendas .” Midwest Political Science Association 39(1). pp.186-203.

Segal, Jefferey A., Chad Westerland, and Stefanie A. Lindquist. (2011). “Congress, the Supreme Court, and Judicial Review: Testing a Constitutional Separation of Powers Model.” American Journal of Political Science 55 (1). pp.89-104.

Segal, Jeffrey A. (1997). “Separation-of-Powers Games in the Positive Theory of Congress and Courts.” The American Political Science Review (American Political Science Association) 91(1). pp. 28-44.

Senator Hatch (UT). (2002). “Statements On Introduced Bills And Joint Resolutions-May 15, 2002.” Thomas: Congressional Record. 2002, 15-May. http://thomas.loc.gov/cgi-bin/query/F?r107:1:./temp/~r107xvvB4h:e0: (accessed 2014, 9-April).

Senator Schumer (NY). (2010). “Congressional Record, 111th Congress, S.3295 Introductory Remarks.” Thomas: Congressional Record. 03-May. http://thomas.loc.gov/cgi-bin/bdquery/z?d111:s.03295: (accessed 2014, 9-April).

Shepsle, Kenneth A., Eric S. Dickson, and Van Houweling. (2002). Bargaining in Legislatures with Overlapping Generations of Politicians. http://politics.as.nyu.edu/docs/IO/4751/shepsle1.pdf

Shufeldt, Gregory, and Patrick Flavin. (2012) “Two Distinct Concepts: Party Competition in Government and Electoral Competition in the American States.” State Politics & Policy Quarterly 12 (3). pp. 330-342.

The Oyez Project. Ashcroft v. Free Speech Coalition. 2014, 6-April. http://www.oyez.org/cases/2000-2009/2001/2001_00_795 (accessed 2014, 9-April).

—. “Citizens United v. Federal Election Commission.” Oyez. 2014, 9-April. http://www.oyez.org/cases/2000-2009/2008/2008_08_205 (accessed 2014, 9-April).

—. Kelo v. City of New London. 2014, 9-April. http://www.oyez.org/cases/2000-2009/2004/2004_04_108 (accessed 2014, 9-April).

[†] This explanation was initially proposed by Dr. Brian Harward while discussing these results.

Thesis Archives

| 2020 | 2018 | 2017 | 2016 | 2014 |