An Analysis of Policy-Based Congressional Responses to the U.S. Supreme Court's Constitutional Decisions

Abstract

While Congress can attempt to overrule constitutional decisions of the Supreme Court by initiating the constitutional amendment process, an amendment is rarely a practicable option. Instead, Congress regularly tries to modify the impact of constitutional decisions with ordinary legislation. I analyze policy-based responses to the Supreme Court's constitutional decisions that were initiated in Congress between 1995 and 2010. For each responsive proposal, I consider the relationship between the proposed legislation and the Court's legal holding and the relationship between the proposal and the public policy associated with the Court's decision. I find that Congress enjoys considerable success in reversing the policy impacts of the Court's decisions but is limited in its ability to overcome the Court's legal rules.

When members of Congress dislike an opinion announced by the United States Supreme Court, they can express their disapproval in several ways. Congressional responses range from the nearly costless issuance of public statements voicing criticism of the Court to the daunting task of shepherding proposed constitutional amendments through Congress in the hopes of sending amendments to the states that will overcome the Court's constitutional interpretations. Other legislative proposals may reverse statutory interpretations adopted by the Supreme Court or revise policies that have been declared unconstitutional in the hopes of satisfying judicial scrutiny. Instead of attempting to alter the policy announced by the Court, members of Congress may engage in institutional attacks¹ designed to weaken the Court. I consider the use of ordinary legislation to limit or modify the impact of constitutional decisions of the Supreme Court. I call these proposals “policy-based responses” to differentiate them from institutional attacks, which are often assumed to be Congress's preferred or only available vehicle for responding to the Court's constitutional decisions. I argue that policy-based responses are a regular and important part of the interaction between Congress and the Supreme Court.

I adopt the well-founded assumption of legislative scholarship that members of Congress are motivated primarily by preferences over policy, either their own or those induced by their constituents (Clausen 1973; Fenno 1973). I predict that, to the extent possible, members of Congress will use the legislative process to refashion policies announced by other branches of government to conform to those preferences (Martin 2001). Constitutional amendments provide one way for Congress to annul judicial interpretations of the Constitution, but Congress has employed this tool with success only four times (Devins & Fisher 2004:23). In order for Congress to send an amendment to the states, a proposal needs the support of supermajorities in both chambers. Ratification requires the assent of three-fourths of the states. Because of these onerous requirements, a constitutional amendment will rarely be a practicable option. I expect, instead, that legislators will attempt to use ordinary legislation to minimize the impact of constitutional Supreme Court decisions with which they disagree.

The analysis presented herein is based on an extensive search of congressional publications for references to legislative proposals that respond to Supreme Court cases in which judicial review was exercised. I find that members of Congress regularly introduce legislation with the goal of affecting the impact of the Court's constitutional decisions. I address three questions about the use of these proposals and their content:

1. 1. What are the characteristics of the constitutional Supreme Court decisions for which Congress considers responsive legislation?

2. 2. Do responsive proposals attempt to reverse the policy and legal impacts of the Court's constitutional decisions?

3. 3. How successful is Congress in its attempts to modify the impact of constitutional decisions of the Supreme Court through ordinary legislation?

I find that Democratic and Republican majorities use policy-based responses regularly and that these proposals target Supreme Court decisions in numerous issue areas. The consideration of the content of the proposals reveals that they vary in their treatment of the Court's legal holdings and associated policies. Most, however, attempt to partially or fully reverse the Court's policies while simultaneously working within the bounds of the legal rules announced by the Court. Inspection of the responsive proposals that are enacted reveals that Congress enjoys appreciable success in modifying Court-announced policies but that is limited in its ability to countermand the legal holdings announced by the Court.

The analysis makes several contributions to the separation of powers literature. First, it provides a systematic and comprehensive analysis of policy-based responses to the Court's constitutional decisions. Second, it provides descriptive information about numerous congressional responses to the Court's constitutional decisions. The most comprehensive analysis of policy-based responses to the Court's constitutional decisions is Meernik and Ignagni's evaluation of the presence and success of decision-reversal legislation (1997). This work, while providing a thoughtful analysis of the factors associated with the presence and success of legislative responses, does not provide any discussion of individual legislative proposals. In contrast, Fisher (1988) and Devins and Fisher (2004) provide numerous examples of congressional responses to Court decisions but do not provide an indication of the frequency with which different types of responses are proposed or passed. An overview of the substance of responsive proposals will allow scholars to begin to evaluate the scope and limits of this congressional tool.

In Section 1, I review key insights of the “governance as dialogue” approach to the study of Congress-Court interaction and argue that policy-based responses to the Court's constitutional decisions are an important part of the Congress-Court dialogue. In Section 2, I review the methods used to identify the sample of responsive proposals introduced in Congress between 1995 and 2010. In Section 3, I use these data to evaluate hypotheses about the use of policy-based responses.

Section 1: Existing Literature and Theory

Interaction between Congress and the Supreme Court has received considerable scholarly attention since the publication of Dahl's (1957) seminal inquiry that led him to conclude that the Supreme Court is a part of the dominant national alliance and that it will be reigned in by Congress when it steps outside the bounds of what the national majority finds acceptable. These insights have led scholars to consider the impact of actual and anticipated congressional action on Supreme Court decision-making and have generated a literature rich with arguments about the ways in which Congress may constrain the Supreme Court. Scholars have considered the existence of a congressional constraint in statutory (Eskridge 1991; Ferejohn & Shipan 1990; Hansford & Damore 2000; Segal 1997; Spiller & Gely 1992) and constitutional cases (Harvey & Friedman 2006; Martin 2001; Segal et al. 2011).

Surprisingly little attention has been devoted, however, to one of the key insights offered by Dahl—that Congress was able to reverse the policies announced by the Supreme Court when it invalidated federal laws through the passage of ordinary legislation. To the contrary, the dominant assumption in much research on Congress-Court interactions is that the primary means available to a Congress that wishes to reverse the dictates of the Supreme Court's constitutional decisions is the constitutional amendment process. Martin (2006:4) summarizes the view dominant in the literature: “Congress can overturn statutory decisions by amending or changing a statute but must pursue a more arduous process to overturn constitutional decisions.” Martin asserts that Congress is limited in its ability to overcome the Court's constitutional decisions because it must rely on either the constitutional amendment process or engage in institutional attacks to check the Court. This assumption underlies an extensive literature on judicial decision-making that regularly assumes that the Court has little reason to constrain itself in anticipation of congressional reaction (see, for example, Segal & Spaeth 2002).²

The assumption of judicial finality has also been a fixture of debates on constitutionalism (Agresto 1984; Black 1960; Devins & Fisher 2004). The notion that the Court is the sole and final interpreter of the Constitution is challenged by scholars that see a serious role for Congress and other non-judicial actors in constitutional deliberation. Miller (2009:7) characterizes the movement that reacts to the dominance of the assumption of judicial supremacy as the governance as dialogue movement. This approach conceives the processes of constitutional interpretation and policymaking as the result of inter-institutional conversations. Its supporters see an important role for Congress in responding to even the constitutional decisions of the Supreme Court.

Devins and Fisher (2004:230–33) offer a list of examples to counter the “last word doctrine” that holds the Supreme Court definitively resolves constitutional questions, thereby stymying attempts by Congress to enact alternative policies. They remind us that when the Supreme Court upholds the constitutionality of a measure, Congress may respond by discontinuing it. When the Supreme Court holds that a practice is not prohibited by the Constitution, Congress may often prohibit or restrict it by statute. When the Supreme Court holds that an action is not protected by the Constitution, Congress can offer protection above the minimum requirement established by the Court. These types of responses are in addition to the forms of legislative response to judicial invalidations which may include the revision of a policy to satisfy judicial scrutiny or the imposition of the same policy under a different constitutional authority (Meernik & Ignagni 1997).

Ignagni and Meernik offer empirical support for the governance as dialogue model (1994, Meernik & Ignagni 1995, 1997). They find that members of Congress regularly propose, consider, and pass legislation in response to the Supreme Court's constitutional decisions. Meernik and Ignagni (1997) found that Congress attempted to reverse 22 percent of the cases in which the Supreme Court declared a state or federal law invalid between 1954 and 1990, and that it was successful in 33 percent of its efforts. They demonstrate that while institutional attacks and constitutional amendments may be used by Congress to influence the Court and to modify judicially announced policies, so too are legislative policies that take on the Court's policies directly.

In his analysis of Congress-Court interactions in the area of federalism, Pickerill (2004) suggests that Meernik and Ignagni (1997) overstate the frequency with which Congress reverses the Court's constitutional decisions by failing to account for differences in the nature of decision-making in Congress and in the Court. Pickerill argues that Congress operates on a public policy dimension while the Court operates primarily on a constitutional policy dimension. When the Court reviews the constitutionality of a statute, justices are less concerned than members of Congress with how a statute affects public policy and more concerned with the application of the relevant constitutional value(s) to the statute in question. The implication is that dialogues between Congress and the Court ought not be conceived as battles over public policy in a unidimensional space. Members of Congress, recognizing the role of law in Supreme Court decision-making but being less constrained by it themselves, may draft legislative responses to the Court's decision that satisfy the Court's constitutional tests while retaining or resuscitating their preferred policies. Pickerill (2004) suggests that Congress can reverse the Court's policy without reversing the Court's ruling.

I build on the insight of the governance as dialogue movement that Congress can and does respond to the constitutional decisions of the Supreme Court to advance its preferred policies. Like Meernik and Ignagni, I argue that Congress is able to modify the impact of the Court's constitutional rulings through the passage of ordinary legislation. I also draw on Pickerill's arguments that coordinate construction scholars overestimate the importance of constitutional doctrine to members of Congress and that congressional responses typically fail to reverse the Court's constitutional decisions in the legal sense.

I contend that while Congress and the Court shape public and constitutional policy, the actors within each institution prioritize these types of policy differently. Opinions announced by the Supreme Court provide information to members of Congress about the types of policies the Court recognizes as legitimate under its preferred readings of the Constitution. The interest of members of Congress in constitutional policy is instrumental; they can use information about the Court's preferences over constitutional policy to draft responsive legislation that will survive judicial review.

I expect that members of Congress will propose and enact responsive proposals that are consistent with the Court's legal rules. These hypotheses are grounded in the assumption that members of Congress care about the ultimate state of public policy. While members may accrue some benefit for associating themselves with policies that are not adopted or that are subsequently invalidated by the Court, position-taking is likely to yield higher payoffs when it is coupled with tangible and enduring policy outcomes (Martin 2001; Mayhew 1974). Legal rules will vary in how much they constrain Congress's pursuit of policy, so responsive proposals will vary in the extent to which they reverse the policies associated with Court rulings.

My goals for the analysis are to discern how regularly Congress responds to the constitutional decisions of the Supreme Court with policy-based legislation, to evaluate the extent to which these legislative responses attempt to modify the policies and legal rules announced by the Supreme Court, to ascertain the frequency with which such proposals are enacted, and to explore the relationship between proposal content and the probability of passage. Towards these ends, I will evaluate the following hypotheses:

Hypothesis 1: Members of Congress will regularly introduce legislative proposals with the purpose of modifying the impact of the Supreme Court's constitutional decisions.

Hypothesis 2: Responsive proposals will vary in the extent to which they would actually reverse the public policies associated with the Court's constitutional decisions, but will usually be consistent with the Court's legal holdings.

Hypothesis 3: Responsive proposals that contradict the Court's legal holdings will be less likely to be enacted than proposals that are reconcilable with the Court's legal holdings.

Section 2: Overview of Data Collection Process

I identified responsive proposals based on statements made by members of Congress. Like Meernik and Ignagni (1997) I adopt Stumpf's (1965:382) definition of decision-reversal legislation which characterizes proposals intended “to modify the legal result or impact, or perceived legal result or impact of a specific Supreme Court decision, or decisions” as decision-reversal legislation. (Because I contend that modifying the impact of a decision is not the same thing as reversing it, I call the proposals policy-based legislative responses instead of decision-reversal legislation.) If a member of Congress made a statement in the Congressional Record or in a published committee report indicating a legislative proposal (a bill or an amendment)³ was intended to respond to or reverse the impacts of a Supreme Court decision, I included it in the data (subject to some exclusions detailed below).

Because I relied on statements of members of Congress to identify responsive proposals, the data exclude responsive proposals for which no member took to the floor or entered a statement into the Congressional Record asserting a relationship between their proposal and a Court decision unless the bill was the subject of a published committee report that makes clear that connection. While this method may, therefore, not identify the complete universe of responsive policies, it is the most comprehensive and systematic method available for the identification of such proposals. It is in statements made upon introduction and during debate that members are most likely to explain their reasons for supporting legislation. Relying on statements by members of Congress to identify responsive proposals has the additional benefit of ensuring that the responses I have identified are purposive, thereby excluding policies that only incidentally affect the impact of a decision.

Other studies of attempts to modify or reverse the impact of particular Supreme Court decisions have made entry into their samples conditional on varying thresholds of legislative success. Eskridge (1991) includes only cases subject to a committee hearing. Meernik and Ignagni (1995, 1997) exclude proposals that are not subject to at least one floor vote. Elsewhere, Ignagni and Meernik (1994) require bills to be reported by committee to earn inclusion in their sample. I depart from these studies by including bills that are introduced that are not the subject of further legislative activity. According to Clark (2011), extant research underestimates the potential relevance of bill introduction in the Congress-Court dialogue. He argues that the introduction of Court-curbing proposals signals public displeasure with the Court and may induce sophisticated behavior by the Court. Observation of the introduction of Court-curbing bills is the mechanism by which members of the Supreme Court become aware of threats to their legitimacy that would result from diminished public confidence. Whether or not policy-based responses perform this function is an open question, but in order to evaluate it, the complete set of introduced proposals is needed. I explore differences between proposals that are the subject of post-introduction activity and those that are not below.

The text of congressional committee reports and the Congressional Record is available online at the Library of Congress's website (THOMAS.loc.gov). I used the search feature of THOMAS to retrieve every entry in the House, Senate, and Extensions of Remarks sections of the Congressional Record and every committee report that included the words “Supreme Court” in the 104^th through the 111^th Congresses. Using the Find feature of a web browser, I searched within each document to find the reference(s) to the Supreme Court. I reviewed each document to determine whether or not it included a reference to a piece of legislation that, according to its supporters, would satisfy the Stumpf definition above.⁴ For the 16-year period under study, this required the review of over 1300 committee reports and 10,000 articles in the Congressional Record. I excluded responses to cases in which the Supreme Court did not exercise judicial review⁵ and, because of a lack of data about case characteristics, responses to decisions announced prior to the Supreme Court's 1946 Term.

Often proposals are clearly identified by their supporters as responsive proposals. For example, upon introduction of the Civil Rights Restoration Act of 2006, Senator Mike DeWine (R-OH) said, “In my view, Board of Trustees v. Garrett (2001) and Kimel v. Florida Board of Regents (2000) were wrongly decided. And, they should be overturned. My bill will do just that” (DeWine 2006, S8842). In other instances, members make clear their intent to modify the impact of a decision without saying their bill would reverse the Court. Congressman Jerrold Nadler (D-NY) argued that the Jessica Gonzales Victim Assistance Program would “restore some of the effectiveness of protective orders” in the wake of the Court's decision in Town of Castle Rock v. Gonzales (2005) (U.S. House 2005, 652). Because Nadler suggests that his proposal would modify the impact of the case (here, a decrease in the effectiveness of protective orders), I included the proposal in the sample.

When a supporter indicated that a proposal responds to multiple Supreme Court decisions but associates the cases with the same policy and legal rule, I coded the proposal as responding to the most recent Supreme Court decision. In contrast, when a member indicated that a proposal responds to multiple Supreme Court decisions and discusses separately the rule or impact of each decision, I coded a separate observation for each case. For example, in C & A Carbone, Inc. v. Town of Clarkstown, New York (1994), the Supreme Court invalidated a local flow-control ordinance that required all waste to pass through a certain waste processing plant for discriminating against interstate commerce in violation of the Dormant Commerce Clause (U.S. Const. art. I, sec. 8, cl. 3). In Fort Gratiot Sanitary Landfill, Inc. v. Michigan Department of Natural Resources (1992), another Commerce Clause case, the Court invalidated waste import restrictions that limited the ability of waste processing plants to accept waste from other counties, states, and countries. Because both cases deal with waste management, it is no surprise that legislative responses to the decisions in several congresses arose as a package. The responses, however, were distinct. In response to Fort Gratiot, the proposals would have allowed states to impose restrictions on the import of solid waste. The responses to C & A Carbone, Inc. were separate and more limited. The proposals would not have allowed the creation of new flow-control ordinances, but would have “grandfathered in” flow-control ordinances that had been established prior to the Court's decision (U.S. Senate 1995).

In some instances, multiple bills embodying the same policy were introduced in response to a single Supreme Court decision in the same Congress. Because it is improbable that Congress would pass multiple bills in the same Congress that impose the same policy, I included only one of the bills in the sample. When two bills had the same policy (as determined by their identification in THOMAS as related bills and through a comparison of their bill texts to ensure policy equivalence), I assumed that the bill that enjoyed the most legislative activity was the primary vehicle for policy change in an issue area and included only that bill in the sample. When multiple bills with the same objective were introduced and failed at the same stage of the legislative process, I included in the sample the bill that was introduced first.

I do not contend that the presence of multiple similar bills is without consequence. To the contrary, I expect that the introduction of multiple bills signals salience of the relevant Court decision to Congress and is likely associated with an increased probability of passage of one of the proposals. However, including multiple bills that embody the same policy would have artificially inflated the sample size. I identified 111 responsive proposals (contained in 105 bills) introduced between 1995 and 2010. These proposals respond to 43 unique Supreme Court decisions decided between 1954 and 2010.

Section 3a: What Are the Characteristics of the Supreme Court Decisions for Which Congress Considers Responsive Legislation?

A comprehensive test of the relationship between Supreme Court case characteristics and their influence on the probability that a legislative response is initiated in Congress would take as the unit of analysis the Supreme Court case and predict the presence of a response. The works of Joseph Ignagni and James Meernik are exemplars of that approach (Ignagni & Meernik 1994; Meernik & Ignagni 1995, 1997). Because I have adopted the responsive proposal as my unit of analysis, my ability to draw firm conclusions about relationships between case characteristics and the presence of a congressional response is limited. However, the data can be used to draw preliminary insights into the relationships between case characteristics and responsive legislation.

Issue Areas and Salience

Table 1 reports the distribution of cases and responses by issue area. The third column reports the number of constitutional cases decided by the Supreme Court within each issue area for the 1995 to 2010 terms. If legislative responses to the Supreme Court's decisions are distributed across issues randomly, the percentages in Columns 1 and 2 should approximate the percentages in Column 3. In fact, there are more responses to the Court's decisions that involve the First Amendment, federalism, privacy, and economic activity than would be expected by chance and Congress responds to fewer cases related to criminal procedure than chance would predict.

Table 1. Distribution of Cases in Sample, Legislative Responses, and Cases Heard by the Supreme Court by Issue Area

A partial explanation for the variance in attention to different issue areas is that the Supreme Court's treatment of cases varies across the issue areas in ways that are meaningful to Congress. In its 1995 through 2010 terms, the Supreme Court invalidated federal policies in 4 percent of the cases in which it exercised judicial review. However, the rate of invalidation of federal laws was significantly higher for First Amendment (10 percent) and federalism cases (19 percent). The attention of Congress to these issues may reflect a desire to replace policies invalidated by the Court. In its cases regarding economic activity, the Court never invalidated an act of Congress, but in more than 20 percent of cases invalidated state laws or regulations. In several instances, members of Congress introduced legislation to allow states to reinstitute policies invalidated by the Court. Higher numbers of judicial invalidations of policies in an area are likely to increase the number of responses in that issue area.

Case salience is also an important factor that appears to affect the likelihood of a congressional response. Most (65 percent) of the cases that engender legislative responses during the period under study satisfy Epstein and Segal's (2000) measure of case salience, by which cases receiving front-page coverage in the New York Times the day after their announcement are considered salient. The overrepresentation of salient cases in the sample suggests that members of Congress are more likely to respond to salient cases than non-salient cases. This comports with the findings of earlier work that demonstrate congressional responses are more likely when there is public awareness of and opposition to Supreme Court decisions (Ignagni & Meernik 1994; Meernik & Ignagni 1995, 1997).⁶ Salient cases appear to be more likely to engender legislative responses, generally, and more likely to engender multiple responses both over time and within the same Congress. For example, the high number of responses to the Court's privacy cases is driven by repeated efforts of Republican legislators to counter the Court's abortion decisions. In this sixteen-year period, there are eight responses to Roe v. Wade (1973) and four to Stenberg v. Carhart (2000).

Timing of Responses

Figure 1 reports the number of responsive proposals introduced by Congress. An average Congress in the period witnessed 13.88 unique response attempts. Clark (2011:276–97) identifies an average of 16.57 Court-curbing measures introduced per Congress between 1995 and 2010. While Court-curbing may be the more frequent mechanism used by members of Congress to signal their displeasure with the Court, policy-based legislative responses are also a regular part of the Congress-Court dialogue.

Figure 1. Number of responsive bills introduced by Congress.

In this period, an average of 9.34 years elapsed between the announcement of a decision and the initiation of a congressional response. This statistic is misleading because of the skew of the variable. (The distribution is reported in Figure 2.) Twenty-seven percent of responses are initiated within one year of the relevant Supreme Court decision and more than half of responses are initiated within five years. At the other extreme, however, 15 percent of responses target decisions that are more than 25 years old. While some Supreme Court decisions are targeted in their advanced age, the general pattern suggests that Court decisions are most susceptible to legislative responses in the few years after their announcement.

Figure 2. Distribution of “years to response” variable.

Party Effects

I do not observe significant differences in the tendencies of Republican and Democratic congressional majorities to introduce responsive legislation. During the 104^th, 105^th, 106^th, 108^th, and 109^th Congresses, the Republican Party held a majority of seats in both the House and the Senate. On average, these Republican majorities introduced 14.6 responsive proposals. During the congresses with Democratic majorities in both chambers (the 110^th and 111^th), an average of 13 responsive proposals were introduced. This finding stands in sharp contrast to the use of institutional attacks, which, in recent years, have been primarily associated with the Republican Party (Clark 2011; Miller 2009).⁷

As would be expected, most bills sponsored only by Democrats (83 percent) responded to conservative decisions of the Supreme Court while most Republican-backed bills (65 percent) were targeted at liberal outcomes.⁸ Partisan differences also emerge when I consider the primary issues in the Supreme Court cases to which responsive proposals respond. Table 2 reports the number of responses by issue area, grouped by the partisan makeup of the members of Congress that served as original cosponsors for the proposal. (I code a proposal as bipartisan if it had at least one Republican and one Democrat as an original cosponsor.) Responses to the Court's First Amendment cases provide the most interesting result—70 percent of such proposals were supported by a combination of Democrats and Republicans. Responses to First Amendment cases may be the easiest upon which to build broad coalitions. The data include responses to decisions invalidating bans on indecent and pornographic material on the internet and flag-burning—restrictions on speech that may be broadly popular with constituents across parties. Other issues seem to belong primarily to one party or the other. There are no examples of Democratic bills introduced to respond to the Court's privacy or due process cases, but 70 percent of the responses to the Court's civil rights cases were sponsored only by Democrats. Republicans were more active in responding to the Court's economic activity decisions.

Table 2. Responsive Proposals by Party Membership of Original Cosponsors

Section 3b: Do Responsive Bill Attempt to Reverse the Policy and Legal Impacts of the Court's Constitutional Decisions?

The data above reveal that members of Congress regularly respond to the Supreme Court's exercise of judicial review in various issue areas and that their attention is divided between new and old cases that advance liberal and conservative policies. This offers support for Hypothesis 1. Next, I consider the potential impacts of responsive proposals. I characterize both the policy and legal implications of responsive proposals to evaluate the extent to which they would effect reversals of Court decisions, allowing for the possibility that “reversal” means different things to different institutional actors.

Coding Response Types

Congress may replace or revise policies invalidated by the Supreme Court and it may discontinue, prohibit or modify policies upheld by the Supreme Court (Devins & Fisher 2004; Meernik & Ignagni 1997). I classified each bill based on the extent to which it would reverse the policy implications of the Supreme Court decision to which it responds. There are four categories—(1) complete reversals, (2) partial reversals, (3) non-reversals, and a residual category for policies that are not clearly classifiable (4). To classify proposals, I again relied first on statements of members of Congress (from either the Congressional Record or from congressional committee reports) that characterize the intended policy implications of responsive legislation that they support. In order to be characterized as a complete policy reversal, a proposal must meet a modified version of Eskridge's (1991, 3320) definition of an “override:” complete policy reversals would modify the impact of a decision such that subsequent cases presenting the same facts will be decided differently. Bill supporters must make explicit that their bill would impose a substantively identical policy. In addition to looking to members' statements, I also compare the proposed legislation to the syllabus of the relevant Supreme Court decision. If the bill supporters' characterization of the proposal is plausible, I include the proposal in the complete reversal category.⁹ If bill supporters indicate that a proposal would impose a substantively identical policy to one previously invalidated or would prohibit a policy allowed by the Court, the policy is coded as a complete reversal unless the legislative text makes clear that the bill's impact would be more narrow.

If bill supporters indicate that significant modifications to a proposal are required or that a policy is narrower in scope than one considered by the Court, it is characterized as a partial reversal. If proposed legislation would impose a different type of policy than the one at issue in the relevant Supreme Court case but supporters state that it furthers the same policy goal, the proposal is characterized as a partial reversal. Bills that would limit the application of the Supreme Court's policy or exclude its application in some cases are also characterized as partial reversals.

If a proposal would impose a different type of policy than the one at issue in the relevant Supreme Court case and bill supporters do not explain the connection between the responsive policy and the policy at issue in the case to which it responds, it is coded as a non-reversal. If no substantive change to policy would result if a policy was adopted, it is a non-reversal. Finally, if a proposal responds to a case in which the Court exercised judicial review but responds to an issue unrelated to the constitutional question, the policy is coded as a non-reversal.

I next characterize each congressional response based on its relationship to the Court's constitutional holding. I define four categories of legal response type. Proposals that include language seemingly at odds with the relevant Court decision that are characterized by their supporters as reversals of the Court's rules are characterized as being in conflict with the Supreme Court's holding. At the other extreme are responses that afford protection of some right above the floor established by the Supreme Court in the decision that triggers the response. Responses that are characterized as attempts to correct legal defects identified by the Supreme Court are included in a third category. Finally, a fourth category includes responses that do not challenge the Court's legal rule.¹⁰ Table 3 reports the frequency of response types across categories.

Table 3. Relationship Between Policy Response Type and Legal Holding

A survey of the bills introduced in response to the Supreme Court's constitutional decisions reveals variation in the breadth of their potential policy impacts. Some would reinstate policies previously invalidated by the Court. For others, passage would not significantly alter the policy associated with the relevant Supreme Court decision. I identified 37 proposals that would completely reverse a policy announced by the Court. Sixty-three proposals would have the effect of partially reversing a Court-announced policy. Ten bills would clearly not have the effect of reversing the policy impact of the relevant Court decision and the policy impact of one proposal is ambiguous.

There is also variation in the relationship between Congress's responsive legislation and the Court's legal holdings. While 11 responsive proposals are openly in conflict with the Supreme Court holdings to which they respond, for more than half of the proposals (58), there is no challenge to the Court's legal holdings. Between these extremes, a number of proposals purport to correct legal defects identified by the Supreme Court. To illustrate the variation in the intended substantive and legal impacts of congressional responses to the Court's constitutional decisions and to illustrate the classification scheme, I review examples of bills across different categories of response type.

Complete Policy Reversals That Do Not Challenge Supreme Court Holdings (12 Percent of Responses)

I categorized 13 responses as complete policy reversals that do not affect the Court's legal holdings. If enacted, these 13 proposals would countermand the policy implications of a Supreme Court case without challenging the associated legal holding. Five bills in this category respond to decisions in which the Court upheld a policy in the face of a constitutional challenge while the remainder respond to invalidations of challenged policies. An example from the former category is a response to the Court's decision in Morton v. Mancari (1974). In Morton, the Supreme Court upheld federal Indian preference laws against a Fifth Amendment challenge. Representative Curt Weldon (R-PA) introduced the Native American Equal Rights Act in 2000 that would have repealed all federal Indian preference laws. When the United States Supreme Court issues a decision saying a government can do something, it is often within Congress's power to make a determination that the government should not exercise that right and to codify that preference.

Some responses to judicial invalidations attempt to reinstate invalidated policies without upsetting Court-announced legal rules. The Religious Liberty Protection Act (RLPA) (1999), which would have reinstated the policy invalidated in City of Boerne v. Flores (1997) on different constitutional grounds falls into this category. The RLPA was proposed after the Supreme Court invalidated the Religious Freedom Restoration Act (RFRA) (1993) as it had been applied to the states. RFRA required that strict scrutiny be used to evaluate free exercise claims even when the policy at issue was a neutral law of general applicability. In Boerne, the Supreme Court held that Congress had exceeded its enforcement powers under the Fourteenth Amendment when it made RFRA applicable to the states (U.S. Const. amend. XIV, sec. 5). The RLPA would have imposed the same policy under Congress's authority to regulate commerce (U.S. Const. art. I sec. 8 cl. 3). When bill sponsors reestablish an invalidated policy on different constitutional grounds, they can render irrelevant the constitutional question that has been decided against their policy preferences. With the RLPA, bill supporters attempted to shift the relevant constitutional question without sacrificing the substance of the invalidated policy. Understandably, members of Congress may view this policy as reversing the Court's decision, but judicial scholars should recognize the limited nature of this type of reversal. It is limited to the Court's policy and does not reach its holding.

Complete Policy Reversals That Are Openly in Conflict with Supreme Court Holdings (8 Percent of Responses)

In nine instances, members of Congress introduced bills that offered interpretations of the Constitution clearly at odds with those announced by the Supreme Court.

In 1996, Elton Gallegly (R-CA) proposed an amendment to the Immigration Control and Financial Responsibility Act that would have given states the option to deny public education to undocumented children. This proposal would have negated the policy impacts of Plyler v. Doe (1982) in which the Court held that a Texas law allowing the state to withhold state funds from local school districts for educating undocumented children violated the Equal Protection Clause of the Fourteenth Amendment (U.S. Const. amend. XIV, sec. 1). Supporters of the Gallegly Amendment agreed that the practical effect of the proposal was to reverse the impact of the Plyler decision and there is no evidence that the bill's supporters attempted to reconcile the legislation's seeming inconsistency with the Supreme Court precedent. Accordingly, I characterized it is an attempt to challenge the Court's holding. One may argue that this type of response should not be characterized as a reversal attempt because it is clearly in conflict with Supreme Court precedent and would be unconstitutional under Plyler. Such a characterization assumes away important empirical questions. Fisher (1988) and Meernik and Ignagni (1997) make clear that one possible form of legislative reversals of Court decisions is Supreme Court acquiescence to statutes at odds with the Court's precedents. Whether and when these sorts of bills are passed by Congress and survive scrutiny in the courts are questions worthy of further inquiry.

Complete Policy Reversals That Attempt to Correct Legal Defects Identified by the Supreme Court (14 Percent of Responses)

In 15 cases, legislative proposals were introduced to correct constitutional defects identified by the Supreme Court, thereby allowing Congress to reinstate a policy previously displaced by the Court. The congressional response to the Court's decision in United States v. Lopez (1995), in which the Supreme Court invalidated the Gun-Free School Zones Act of 1990, is an illustrative case. In Lopez, the Supreme Court held that Congress exceeded its power under the Commerce Clause (U.S. Const. art. I sec. 8, cl. 3) when it imposed a ban on firearm possession within 1,000 feet of a public, parochial, or private school. According to the majority opinion, possession of a gun in a school zone is in no sense an economic activity that might have a substantial effect on interstate commerce. The majority also indicated that the lack of a jurisdictional element “which would ensure, through case-by-case inquiry that the firearm possession in question affects interstate commerce” (Lopez, 514 U.S. at 561) was a problem. Congress responded promptly by considering the Gun-Free School Zones Amendments Act of 1995. The GFSZ Act as amended required that the government prove that a firearm has “moved in or the possession of such firearm otherwise affects interstate or foreign commerce.” Statements in the Congressional Record make clear that supporters saw the jurisdictional element as a means to save their policy while satisfying the rule announced in Lopez (Kohl 1995).

Partial Policy Reversals That Do Not Challenge Supreme Court Holdings (31 Percent of Responses)

While the discussion above reveals that Congress regularly attempts to reverse completely the policy implications of the Court's constitutional decisions, responsive proposals often include substantial revisions to the relevant policy. More than half of the proposals in the sample would institute partial reversals. The majority of partial reversals do not challenge the legal rules of the decisions to which they respond.

Some of the cases that generated complete policy reversal attempts also engendered responses that would have had less far-reaching policy impacts. Like the Religious Liberty Protection Act, the Religious Land Use and Institutionalized Persons Act (RLUIPA) was introduced in response to City of Boerne v. Flores (1997). The scope of the RLUIPA was considerably narrower than that of the RFRA or the RLPA. The RLUIPA re-instated the compelling government interest-least restrictive means test in cases that involved land use regulations and institutionalized persons. Supporters of the legislation indicated that, while narrower in scope than the RFRA, they believed the revised bill would protect free exercise in a way that would not be subject to the same challenge that succeeded in Boerne (Canady 2000).

I identified 33 bills in addition to the RLUIPA that would partially reverse or modify the policy impact of a Supreme Court decision in which judicial review is exercised without undermining the Court's legal holding.

Partial Policy Reversals That Are Openly in Conflict with Supreme Court Holdings (<2 Percent of Responses)

The Supreme Court's decision in Citizens United v. Federal Election Commission (2010) was met with disapproval by many in Congress. A string of legislative proposals soon emerged to limit the impact of the decision. Despite the Supreme Court's ruling that political speech may not be banned on the basis of a speaker's corporate identity, the Democracy is Strengthened by Casting Light on Spending in Elections (DISCLOSE) Act included a provision that would have prohibited domestic corporations from financing independent political broadcasts if more than 20 percent of their voting shares were held by foreign nationals. The proposal is a partial policy reversal because it targeted only one part of the Court's decision in Citizens United.

Partial Policy Reversals That Attempt to Correct Legal Defects Identified by the Supreme Court (15 Percent of Responses)

Seventeen proposals attempted to correct legal defects identified by the Supreme Court and in so doing, made significant changes to the policies invalidated by the Supreme Court in the related cases. The Supreme Court struck the federal flag desecration statute in United States v. Eichman (1990). In 2004, Senator Byron Dorgan (D-ND) proposed a flag-protection statute that would outlaw flag desecration intended to invite violence. The Flag Protection Act of 2004 is a response to the Court's Eichman decision but it does not completely reinstate the policy invalidated in that case. The modifications made to satisfy the Court's legal holding result in the policy prohibiting a narrower range of flag desecrations than the original federal flag desecration statute.

The Supreme Court's determination that the use of the line-item veto by the United States president violates the Presentment Clause (U.S. Const. art. I, sec. 7, cl. 2) in Clinton v. City of New York (1998) also engendered responses from Congress. Among the proposals was an amendment offered by Representative Paul Ryan (R-WI) to the Spending Control Act of 2004. The amendment would have established an enhanced recisions process that supporters asserted could satisfy the Court's constitutional requirements. Bill supporters made clear that the proposed language would further the same goal as the invalidated Line Item Veto Act (the elimination of wasteful spending) but that it did so differently in an attempt to satisfy the Court.

Partial Policy Reversals That Offer Protection Above Floor Set by Supreme Court (9 Percent of Responses)

When Congress believes the Court has not adequately protected rights it may attempt to secure those rights through legislation (Devins & Fisher 2004). In 10 instances, Congress responded to Supreme Court decisions in which the Court upheld a challenged policy by considering a law that would have created partial obstacles to similar policies in the future. For example, proposals to eliminate the death penalty for federal offenses despite the Court's determination that the punishment does not violate the Eighth Amendment's prohibition on cruel and unusual punishment (U.S. Const. amend. VIII) were introduced in six of the eight congresses in the sample. (These are partial policy reversals because they would not affect the imposition of the death penalty for state offenses.) Multiple responses to the Court's decision in Kelo v. City of New London (2005) also took this form. In Kelo, the Supreme Court held that a city did not violate the Takings Clause (U.S. Const. amend. V) when it took private property and sold it for private development. The Court determined that private development was an appropriate public use under the Fifth Amendment. Proposals considered in the aftermath of Kelo varied in their scope, but several would have prohibited the use of the power of eminent domain by the federal government or, in varying degrees, by states receiving federal economic development funds. (See, for example, the Private Property Rights Protection Act of 2005.)

Responses That Would Not Reverse Policy (9 Percent of Responses)

I identified 10 proposals that would not reverse the policy announced by the Supreme Court. These proposals, unsurprisingly, do not challenge the Court's legal holdings either. Despite the entreaties from Senator Dick Durbin (D-IL) that Congress needed to support the Violence Against Women Act of 2000 to overcome the Supreme Court's decision in Morrison v. United States (2000), the VAWA reauthorization would have had no such effect (Durbin 2000, S10221). Morrison invalidated the portion of the Violence Against Women Act of 1994 that provided a federal civil remedy for victims of gender-motivated violence. The reauthorization considered by the 106^th Congress would not have reinstated that policy. Also included in the non-reversal category are amendments offered by Phil Gingrey (R-GA) to appropriations bills in the 110^th and 111^th congresses, purportedly to limit the impact of the Court's decision in Kelo v. City of New London (2005). One such amendment would have added language stating that, “None of the funds appropriated or otherwise made available in this Act may be used to take private property for public use without just compensation” (H. Amdt. 1166 to 106 H.R. 6599). Neither the amendment nor the underlying legislation include definitions of public use or just compensation, so the amendment simply restates the language of the Takings Clause (U.S. Const. amend. V). Here, the amendment would have no policy impact. These and similar proposals make clear that caution is necessary in relying only on statements of members of Congress to evaluate the purpose and impact of proposed policies. With that caveat in mind, I note that proposals that are characterized as effective reversals that are, in fact, purely symbolic are the exception and not the rule. In most cases in the sample, members of Congress accurately reported the impact of their legislative proposals.

Responses Whose Policy Impacts Are Ambiguous (<1 Percent)

Finally, I identified one proposal for which I cannot determine the policy impact. In response to Hudson v. Michigan (2006), Representative Maurice Hinchey (D-NY) proposed a funding limitation on the Departments of Commerce and Justice, Science, and Related Agencies Appropriations Act, 2006. In Hudson, the Supreme Court held that the exclusionary rule does not apply to evidence obtained in violation of the “knock and announce rule” that requires police officers to wait 20–30 seconds after knocking and announcing their presence before they enter a home. The Hinchey Amendment would have prohibited funds in the CJS appropriations bill from being used to obtain evidence in contravention of the knock and announce rule. The impact of the policy is ambiguous because it is not clear that any funds in the bill would be used to obtain evidence in violation of the knock and announce rule in the absence of the amendment.

The above survey of proposals introduced in response to constitutional decisions of the Supreme Court reveals that members of Congress regularly introduce legislation that would reverse the policy implications of Court decisions, either in whole or in part. Notably, however, most of these reversals could occur without undermining the legal rules announced by the Supreme Court. If there is a typical legislative response to the Court's constitutional decision, it is a policy that proposes to modify the impact of a decision without disturbing the Court's holding. That is not to say that members of Congress never confront the Court's holdings head on. Contrary to my expectations, policies are occasionally introduced that are irreconcilable with the Court's positions. The variation in the potential impact of responsive proposals on the policy dimension offers support for Hypothesis 2 with the important caveat that there is also variation in the potential legal impacts of responsive proposals.

These findings challenge the view dominant in the empirical separation of powers literature that congressional responses to the Court's constitutional decisions are fundamentally different than responses to the Court's statutory cases. Alongside their proposals to strip the Supreme Court's jurisdiction and to enact constitutional amendments to reverse the Supreme Court's decisions, members of Congress regularly propose and consider legislative proposals that exhibit at least tacit acceptance of the Court's rulings. The bills identified here indicate that Congress has the potential to reverse the impacts of the Court's constitutional decisions without invoking the constitutional amendment process or institutional attacks. It is not the case that responses to the Court's constitutional decisions are uniformly (or primarily) institutional in nature when the decision is constitutional and policy-based when a decision is statutory. As Meernik and Ignagni (1995, 1997) suggest, Congress regularly responds to the Court's constitutional decisions by attempting to revise policy announced through ordinary legislation. They frequently do so without threatening the legal policy announced by the Court, suggesting that Congress may be able to reverse the Court's policies without undermining the Court's preferred constitutional interpretations.

Section 3c: How Successful Is Congress in Its Attempts to Modify the Impact of Constitutional Decisions of the Supreme Court Through Ordinary Legislation?

Members of Congress attempt to alter policies announced by the Supreme Court in its constitutional decisions through the passage of ordinary legislation. How often are they successful? Thousands of bills are introduced in every Congress and for most there is no real expectation that they will become law. An analysis of the stage of failure for responsive proposals makes clear, however, that this sample is not dominated by non-serious proposals. A significant percentage (39 percent) fail at the introduction stage but hearings and/or markups are held for 68 (61 percent) and 18 of the 111 responsive proposals identified became law, for an overall success rate of 16.22 percent. This is significantly higher than the average success rate for all bills in Congress. For the period of this study, the average number of bills introduced per Congress (excluding amendments) is 9090 and the average success rate for all bills is 4.82 percent.¹¹ While bills that attempt to modify the policies announced in the Court's constitutional decisions comprise a small portion of the congressional agenda, Congress responds to a substantial number of the Court's decisions. Figure 3 presents data on the stage of the legislative process to which all responsive proposals survive.

Figure 3. Survival data for responsive proposals.

To evaluate the extent to which bills that are introduced and experience no further legislative activity are different than the proposals that are the subject of some committee or floor action, I consider the frequency of response types across the two categories. (Table 4 reports these data.) The most notable distinction is that eight of the nine proposals classified as complete policy reversals that openly conflict with the relevant precedent experience no post-introduction activity. Proposals in this category are significantly less likely to experience post-introduction activity than others.¹² However, the remaining proposals that fail after introduction are scattered across the remaining response types. With the exception of the complete policy reversals in direct conflict with Court precedent, it is not possible to assert a priori which legislative proposals will be taken seriously by Congress.

Table 4. Legislative Activity and Success by Response Type

The successful responsive policies include seven bills previously identified as complete policy reversals, eight partial policy reversals and three non-reversals. This finding suggests that Congress can effectively reverse the policy implications of constitutional decisions of the Supreme Court, at least some of the time. This finding echoes the conclusion offered by Meernik and Ignagni (1997) that persistent and unified congressional majorities can undo the Court's constitutional decisions. But this is only half of the story. While a significant number of responsive policies were openly in conflict with the Court-triggering precedent, none of those bills were enacted. Twelve of the 18 successful responsive proposals offered no legal challenge to the Court's decision, five purported to correct legal defects identified by the Supreme Court and one extended the protection of a constitutional right above the minimum required by the Court. This finding, in conjunction with the finding that no proposals in the “Openly In Conflict with Relevant Precedent” were enacted, offers support for Hypothesis 3, that proposals consistent with the Court's legal holdings are more likely to be enacted than proposals that conflict with the Court's rules. Congress may be able to reverse the Court's policies, but it is not clear that it can reverse the Court in the legal sense. Table 4 reports the frequency of passage by response type.

I suggested above that the introduction of multiple proposals in the same Congress that promote the same policy goal would increase the probability of bill passage. I find this hypothesis supported. Thirty percent of proposals that were offered in multiple pieces of legislation were adopted. In contrast, only 6 percent of policies offered in isolation became law. The impact of introduction in successive congresses is less clear. In some cases, successive introduction suggests that a proposal is a position-taking vehicle rather than a policy-changing one. Some bills that are repeatedly introduced show little chance of becoming law like the “Right to Life” Acts that respond to Roe v. Wade (1973), and proposals repeatedly introduced to abolish the use of the death penalty by the federal government and flag protection proposals. However, a general claim that successive introduction is a sign of non-serious proposals is unwarranted. Some proposals that are introduced in successive congresses are eventually adopted. The Partial-Birth Abortion Ban Act that responds to Stenberg v. Carhart (2000) is an example. It also took Congress multiple attempts to pass a ban on virtual child pornography after the Supreme Court's decision in Ashcroft v. Free Speech Coalition (2002). Depending on the particular research question, it may make sense for other scholars using these or similar data to exclude bills that are successively introduced or to only include successively introduced bills if they respond to recent Supreme Court decisions.

Whether one considers the policy or legal impacts of a responsive proposal can dramatically impact the conclusions drawn about the ability of Congress to undo the work of the Supreme Court. In this sample, there are no examples of Congress passing a responsive proposal that is directly in conflict with the relevant Supreme Court precedent. In fact, of the 11 proposals identified that were clearly in conflict with the Supreme Court decision to which they respond, 9 (82 percent) were the subject of no post-introduction legislative activity. The remaining two proposals passed the House. This paints a dim picture of Congress's ability to overcome the Court in a strictly legal sense. In contrast, I find that 19 percent of complete policy reversal attempts are successful. The Court does not have the last word on the state of policy in its constitutional cases.

Discussion and Conclusion

Members of Congress can respond to Supreme Court decisions in a variety of ways. While recent scholarship has focused much attention on the initiation of institutional attacks in response to unfavorable constitutional decisions, members of Congress frequently attempt to counter decisions on policy grounds. By systematically analyzing responsive proposals, I have demonstrated that most responsive proposals attempt to reverse, at least partially, the impact of Supreme Court decisions but that most responses work within the constraint of the legal rules announced by the Court. The orthodoxy of judicial supremacy is partially true. To fully reverse a legal rule announced by the Supreme Court, Congress may need to rely on the constitutional amendment process. However, the assumption of judicial supremacy overlooks the fact that Congress can dramatically alter the impact of a Supreme Court ruling by engaging in statutory revision.

The analysis provides additional support for the governance as dialogue conceptualization of the policymaking process. As Fisher (1988:245) persuasively argues, “Courts are the ultimate interpreter of a particular case, not the larger issue of which that case is a part.” This restriction on the Court's pursuit of policy frees Congress to shape the impact of Supreme Court decisions through ordinary legislation. Analyses of congressional responses to the Supreme Court's constitutional decisions should take seriously the options available to Congress when it dislikes the Court's constitutional decisions. The pervasive assumption in the empirical separation of powers scholarship that Congress is powerless to do anything about the Court's constitutional decisions is contradicted by the data here.

This is an area of research ripe for growth and accumulation. The regularity with which Congress considers and passes legislative responses to the Court's constitutional decisions may provide a means to bridge subsets of the separation of powers literature that have developed separately. For example, responses to constitutional and statutory cases are often considered separately. However, most of the proposals considered here are not fundamentally different than the responses to statutory cases. Certainly, in constitutional cases, Congress behaves as though it is constrained by the legal confines announced by the Supreme Court, but it is likely that the bigger obstacles to a policy-based response are the vagaries of the legislative process than the matter of whether a case was decided on statutory or constitutional grounds. It is not readily apparent that modifying the policy impact of a constitutional decision is substantially harder than reversing a statutory interpretation. Future research should attempt to empirically evaluate that Congress treats statutory cases and constitutional cases differently, instead of assuming that the two sets of cases are incomparable. There are also obvious connections between this work on policy-based responses to the Court's constitutional decisions and the proposal of institutional attacks. The power of institutional attacks may be in the information that they convey to the Court, not in their passage or implementation (Clark 2009; 2011). Responsive proposals may work similarly, but they have the added impact of changing the Court's policy. Scholars should consider the relative costs and benefits to individual members of Congress of confronting specific Court decisions as an alternative to pursuing attacks that are targeted more broadly at the Court. Finally, the literature on institutional legitimacy suggests that the public views skeptically institutional attacks on the Court (Caldeira & Gibson 1992). As Caldeira (1987) demonstrates, there are no guarantees that the public will line up in support when a co-equal branch challenges the Supreme Court. Future research might consider the extent to which the public is aware of attempts to respond to Court decisions and whether or not the public views them as legitimate attempts to engage in coordinate construction or as attempts to usurp judicial authority.