Judicial Decisions as Tax Legislation



Written by: Nancy C. Staudt[*] René Lindstaedt and Jason O’Connor

ABSTRACT

This article provides a new understanding of the Court-Congress dynamic. It responds to an important literature that for several decades now has misconstrued inter- branch relations as fraught with antagonism, hostility, and distrust. This unfriendly dynamic, it is argued, is evidenced by the repeated congressional overrides of Supreme Court cases. This claim, while true in some circumstances, ignores the friendly relations that exist between these two branches of government—relations that may be far more typical than scholars suspect.

In this article, Professors Staudt, Lindstaedt, and O’Connor undertake a comprehensive study of congressional responses to Supreme Court cases and make a surprising finding: Overrides, although the sole focus in the extant literature, account for just a small portion of the legislative activity in response to the Court. In fact, Congress is just as likely to support and affirm judicial decision-making through the codification of a case outcome as they are to undermine a decision through an override. To investigate fully the nature of congressional oversight of Supreme Court decision-making, the authors undertake both qualitative and quantitative analyses of all the different types of legislative review. In doing this the authors make a series of important and robust findings that challenge and build on the Court-Congress literature. They identify the legal, political, and economic factors that explain why legislators take notice of Supreme Court cases, and are able to predict the factors that are correlated with congressional activity.

This is the first comprehensive and all-encompassing study of Court-Congress relations and for this reason it has important normative and positive implications. The study highlights a complex and nuanced inter-branch dynamic and at the same time shows that the justices themselves are able to impact the legislative agenda to a far greater extent than heretofore understood. This study also provides an important challenge to scholars who have questioned whether the Supreme Court should have jurisdiction over complex issues, such as those in the economic context, given the justices’ lack of training as well as their apparent tendency to “bungle” difficult cases. Professors Staudt, Lindstaedt, and O’Connor argue that scholars have little or no need to worry about Court decision-making in these areas: Not only do legislators routinely review the Court’s decisions, but they also confirm the outcomes as valuable contributions to national policy-making via the codification process.

I.INTRODUCTION

If existing commentary is to be believed, Members of Congress bully Supreme Court justices on a routine bases. Although the framers anticipated two co-equal branches of government in our system of separated powers,[1]the legislature has come to demand a level of respect and deference that strongly suggests congressional preferences must be prioritized over those of the Court in the policy-making arena. When the justices ignore this mandate—Congress quickly strikes back with override legislation.[2]

Though it is true that legislators hold hearings, draft override legislation, and reverse Supreme Court decisions, this characterization is at best an incomplete account of Court-Congress relations. In reality, Congress responds to Supreme Court cases in a myriad of ways and most of these responses cannot be labeled “unfriendly,” “antagonistic,” or “hostile.” Legislators, in fact, are often quite supportive of Supreme Court opinions and frequently codify case outcomes, thereby cementing the justices’ views into new legislation and re-expressing them as their own. Congress, in short, not only defers to the Court—it often transforms judicial decisions into federal legislation.[3]

Moreover, much of the legislative activity that emerges in response to Court cases is neither critical nor particularly sympathetic, but rather content-neutral. Members of Congress, for example, frequently look to Court opinions to glean an understanding of current judicial approaches to statutory interpretation.[4] Legislators then rely on these interpretive norms to craft statutes that will withstand scrutiny down the road should the statute be challenged in federal court. In other contexts, legislators refer to Court cases in debates and hearings as a means to signal constituents, journalists, and other interested parties of important and emerging issues.[5 ] As an empirical matter, proposed overrides account for just thirty percent of the legislative responses to Supreme Court cases,[6]but a review of the literature would lead one to believe that when legislators act in response to the Court—they override cases.[7] As an empirical matter this is wrong.

We believe it is important to understand the full array of congressional responses to Supreme Court decisions; after all these activities are costly in terms of time, energy, and resources. When legislators respond to the Supreme Court, they must believe it to be a useful expenditure of resources—an effective means to achieve their ambitions whether associated with reelection, advancement in the political hierarchy, or simply good policy- making. Identifying all the cases and controversies that gain congressional attention and not just those that spark a negative response will improve our understanding of the role that federal courts play in legislative politics—a role that scholars have investigated but have not yet fully theorized or understood.[8]

In this article, we seek to build on the important and influential body of work that has explored the Court-Congress dynamic that has emerged over the course of the last century. Many noteworthy scholars—too many to name here—have contributed significantly to our understanding of how and why Congress responds to the Supreme Court and many have examined how this oversight impacts the judicial decision making process. But no scholar—not a single one—has ever attempted a comprehensive study of the all the forms of congressional oversight that takes place with respect to Supreme Court opinions. Scholars and commentators focus exclusively on legislative overrides, a notable but minor component of the legislative review process.

Our strategy for executing this comprehensive and all-inclusive study encompasses both quantitative and qualitative approach for understanding Court- Congress relations. In Section IIA, we describe our data collection procedures and in Section IIB we describe our empirical findings in some depth. While scholars interested in inter-branch dynamics have suggested that congressional oversight of the Supreme Court is relatively infrequent, we find that Congress monitors judicial decision-making with surprising regularity: We find that legislators discuss and analyze one out of every two cases decided in the Supreme Court.[9] This statistic is important and revealing—it suggests that legislators do not take an ad hoc approach to the Supreme Court, but systematically review judicial decisions and assess outcomes.

In Section III, we investigate the factors that lead legislators to spend time, energy, and resources on Supreme Court decisions. Our theory for explaining why the legislature responds to Supreme Court cases revolves around one key factor: Issue salience. Issues decided by the Supreme Court often become important to legislators not because they are inherently interesting, but because they have achieved policy significance through the work of agenda entrepreneurs (which include congressional experts, lobbyists, journalists, and the justices themselves).[10] Moreover, we investigate the factors associated with economics, politics, and legal reform that increase the likelihood that an issue will emerge at the center of debate and controversy in and around the Capitol. In Section IV, we test our theories and estimate three different statistical models to understand and explain how these factors impact the incidence of oversight, as well as its frequency and speed.

Section V turns to a qualitative analysis of legislative oversight. In this section, we first describe the cases subject to the different types of responses and then investigate the factors that explain why Congress overrules certain cases but codifies others. It is easy to understand a congressional override—legislators disagree with the outcome—but why would Congress codify a Court decision? After all, the Court decisions govern all jurisdictions and Congress need not get involved to assure the outcome has national implications. In this section, we discuss the factors associated with the different types of responses and find interesting and robust answers. Congress, we find, is quite a bit more likely to codify a Supreme Court case when the justices issue a unanimous decision.[11]

This suggests the judicial vote operates as a signal: When a body of nine diverse thinkers can agree on a single outcome—a similarly diverse body (albeit much larger) can expect to agree as well. We obtained this insight, along with many others, because we expanded our study of Congress and the Court well beyond override activity.

Finally, in Section VI we turn to the positive and normative implications of our study. We first note that our findings document and confirm a far more complex dynamic between Congress and the Court than offered heretofore in the literature. Moreover, our study demonstrates that the justices play an important and key role in setting the congressional agenda; through their opinion writing and vote patterns the justices are able to increase the probability of a legislative response to surprisingly high levels. Although many scholars have argued that the justices follow—or should follow— the preferences of the legislators when interpreting statutes,[12]we find influence also runs in the opposite direction. Legislators are just as apt to follow the lead of the justices when writing and rewriting their statutes: Judicial opinions can and do become legislative enactments. Finally, our study provides an important challenge to the scholars who have argued that the Supreme Court is particularly incompetent when it comes to certain areas of the law, such as those involving economic issues, and thus should cede jurisdiction in these cases to a specialty court created by Congress.[13] Our study shows that legislators routinely review the Court’s decision-making in these areas, thereby assuring the Court does not bungle the difficult issues that it addresses and—and more importantly—Congress frequency expresses approval of the Court’s decision-making via a codification of the outcome. This suggests the justices have far more skill and competence to render decisions than scholars have ever before acknowledged.

II.A DESCRIPTION OF THE COURT-CONGRESS DYNAMIC

A.The Extant Literature Inadvertently Conceals Valuable Insights

Scholars investigating Court-Congress relations generally seek to explain inter- branch dynamics by analyzing legislative responses to judicial decisions in a wide range of legal areas simultaneously.[14] By clustering numerous legal fields together, scholars are able to generalize empirical findings and for this reason aggregation of diverse topic areas can be useful. But drawbacks to this methodology exist. Congress’ inclination to respond to Supreme Court cases may vary—in fact is likely to vary—from issue to issue and this means that the tendency to use facts and circumstances in one context to draw inferences and conclusions in another is problematic.

The political motivation to react, for example, may differ in tax, criminal law, and employment law. Studies suggest that legislators’ ideology best explains votes on civil rights legislation, while interests groups, lobbyists, and macro-economic factors may better explain congressional decision-making in the economic context.[15] Moreover, because Congress allocates jurisdiction over the various legal issues to a range of different congressional committees, it is possible that Court-created issues get distinct treatment based solely on the parlementarian’s referral decision.[16] As Mark Miller notes, evidence indicates that the Judiciary Committee, largely comprised of lawyers, tends to view the Supreme Court as having the final word on legal controversies and thus views congressional overrides as appropriate only in extreme circumstances.[17] The Energy and Commerce Committee members, by contrast, view the Court as a friend or foe depending on the political context: “If it is good politics to treat the Court with respect, then this is how the Energy and Commerce Committee will respond; but if it is good politics to attack Court decisions, then the committee will not hesitate to impose its policy preferences by modifying or overturning court pronouncements.”[18]

To be sure, rigorous scholars, both individually and in teams, have amassed data that begins to address the problem of aggregation. Henshen and Sidlow examine congressional responses in the labor and anti-trust contexts and thus limit their analysis to two House and Senate committees.[19] While their study provides useful statistics on the frequency of legislative responses to Supreme Court decisions, they do not attempt to explain why the Committees respond differently beyond offering an explanation grounded largely in intuition.[20]  Ignagni, Meernik & King focus on a single committee, the Judiciary Committee, and thus appear to avoid the problems associated with grouping diverse issues together for purposes of empirical analysis.[21] In fact, however, the Judiciary Committee controls a wide range of legal and political issues including freedom of information, immigration, bankruptcy, anti-trust, criminal law, federal jurisdiction, patents, copyrights, and so forth.[22] In short, Ignagni, Meernik & King replicate the problem of grouping diverse issues into a single study without controlling for this diversity in any way (a fact the authors did not ignore).

Eskridge as well as Spiller and Tiller have sought to understand the circumstances that lead legislators to respond to federal court decisions, focusing on outcomes solely in the civil rights context and thus do not suffer from the aggregation problem.[23] These studies are grounded in positive political theory and both offer clever and important contributions to the literature on the Court-Congress dynamic. We admire the studies, but also note that neither seeks to offer an empirical test of the theoretical models presented with the aid of statistics. We do not criticize these authors for this failing as this was not their goal, but we do believe subjecting their theories to rigorous empirical testing is key for understanding fully the interplay between the Court and Congress and we seek to fill this gap here.[24]

Moreover, every single empirical study of the Court-Congress dynamic every single one focuses on congressional override activity and we view this as a serious impediment to a full comprehension of inter-branch relations. By focusing exclusively on override proposals (and their success rates), scholars interpret all other types of responses to Supreme Court opinions as equivalent to no response at all. Of course, in one way, this conclusion is accurate given that the different congressional responses all allow the Court decision to remain good law except an override. But we believe good reasons nevertheless exist for examining the range of congressional responses that show up in legislative histories, in floor debates, and in studies and analyses. In fact, by ignoring a substantial amount of legislative behavior vis-à-vis the Supreme Court, we believe scholars’ findings may be somewhat problematic and unrepresentative, problems we discuss further below.

While we emphasize gaps and drawbacks in the extant literature, our goal here is to build and expand upon it. We hope to highlight new and exciting features of the Court-Congress dynamic that previous work has concealed. In fact, by pulling back the

curtain entirely on the oversight process, we expect to offer a more complete description of inter-branch relations—a goal no scholar or team of scholars has attempted to achieve before.

B. A New Methodological Approach

In order to avoid the pitfalls of the existing literature and at the same time to build on its insights, we focus on a single substantive area of the law in control of a single committee in each chamber of Congress. To this end, we have decided to focus on economic issues, specifically those in the taxation context, an area of the law that garners quite a bit of attention in both the Supreme Court and in Congress and one that satisfies our jurisdiction requirement.[25] Our rational for focusing on economics, and taxation in particular, is linked to the fact that both Congress and the Court seem to privilege these areas in the decision-making process.

We examined the Supreme Court docket over the course of the that last five decades and found that the number of economic disputes nearly always outnumbered civil rights-type cases on the judicial agenda. Figure 1 below depicts comparative data from 1953 through 2002 and highlights the Court’s apparent predisposition to grant certiorari to cases in the economic context; although the Court occasionally reviewed more civil rights than economics or financial cases, overall it is clear the Court has decided many more economic controversies over the course of the last half century than those involving civil rights.[26] We do not mean to suggest that civil rights controversies are not politically, economically, or culturally salient issues—of course they are rather we suggest that investigating and explaining institutional dynamics in the economics context will lead to an understanding of a larger portion of the work that preoccupies the Court and thus possibly Congress when engaged in the oversight process.

Figure 1 Civil rights and economic cases as proportion of statutory interpretation cases, 1953-2002 Terms N=2,905

Judicial Decisions as Tax Legislation_html_m713f9cb8

At a more specific level, we find that the tax statute is one of the most if not the most litigated statutes in the Supreme Court in modern times. If we disaggregate the data in figure 1, it becomes clear the Court has granted review to a disproportionate number of tax disputes. And figure 2 indicates this trend shows no sign of abating between 1986 and 2002, the Supreme Court decided more statutory controversies involving the tax code than any other federal statute or rule.[27] Importantly, the other most litigated laws and rules also involve issues outside the realm of civil rights, including the Federal Rules of Civil Procedure, the Bankruptcy Code, and the Employee Retirement Income Security Act. Accordingly, our focus on economic questions, and in particular taxation cases, enables us to illuminate a great deal of the Court’s decision- making activity.

Figure 2 Most litigated statutes and rules during the Rehnquist Court era, 1986-2002 terms

Judicial Decisions as Tax Legislation_html_721c9699Our goal in this paper, however, is not to explain Supreme Court decision-making but rather how justices’ decisions end up on the legislative agenda.[28] Many students of Congress have noted the importance of the tax laws in legislative politics, as well as the extraordinary power that members of the tax-writing committees—the House Ways and Means and Senate Finance Committees—wield in the legislative process.[29] But do the legislators pay any attention to the Supreme Court when wielding this power? Before investigating this question, we first describe our data collection procedures and or preliminary findings.

C.The Frequency and Type of Congressional Oversight: Descriptive Statistics

For purposes of our investigation, we collected all 284 Supreme Court tax cases decided between 1954-2005 (i.e between the 1953-2004 terms).[30] We then collected every legislative response to the cases decided in this time period.[31] The evidence in the taxation context confirms what scholars have found elsewhere: Supreme Court cases routinely gain the attention of the legislators. Since 1954, the Supreme Court has decided 284 tax cases and legislators have discussed and analyzed 55% of these cases, by name, in legislative hearings, floor debates, committee reports, studies, and so on.[32] These oversight activities took place over the course of fifty-one years and individual legislators and Committees from both the House and the Senate participated in this oversight process. There were a total of 744 references to the cases: 292 surfaced in Joint Tax Committee Reports; [33]232 from individual Senators or Senate Committees; 177 from individual House members or House Committees; and 41 in Conference Committee Reports. Once Congress noted a case, it responded differently in different contexts. Legislators proposed an override in 19% of the cases, sought to codify 14% of them, and held hearings on or cited to an additional 30% of the cases decided since 1954.[34]Table 1 depicts the form and frequency of responses generated by the cases.

Table 1 Congressional responses to Supreme Court tax cases, 1954-2004, N=284.

 

Congressional Response

Frequency

Cases that Did Not Trigger a

Response

127 (45%)

Cases that Did Trigger a Response

157 (55%)

Type of Response:

1. Override Proposal (47 succeeded)

55 (19%)

2. Codification Proposal (33 succeeded)

41 (14%)

3. Positive Citation

52 (18%)

4. Negative Citation

22 (8%)

5. Cited for Purposes of Understanding Court’s Approach to Statutory Interpretation

10 (3%)

6.Used in Nomination Proceeding

4 (1%)

 

This data along with that presented above are both more narrow and more expansive than that offered in the existing literature on Court-Congress interactions. Our study is narrow because, with the exception of Hettinger and Zorn, we are the only scholars to depict congressional responses in a single legal area. Our study is more expansive than that found in the extant literature, however, because it highlights all forms of congressional responses, including mere “position-taking activities,” as well as codification and override efforts. We think our narrow focus will enable us to describe and explain congressional activities with more precision than that seen in the current literature. And the expansive feature of our study highlights what must by now be obvious: The existing literature’s biased focus on overrides leads scholars to ignore a substantial amount of congressional activity. Indeed, given that only 55 of the references in Congress were associated with an override proposal, scholars studying Court-Congress relations in the economic context would disregard at least 70% of congressional activity in response to Supreme Court decision-making. This limited focus in the literature highlights what we perceive to be a serious limitation in current empirical studies and one we think is important to remedy.

III.SUPREME COURT CASES ON THE LEGISLATIVE AGENDA

Policy issues must gain traction before serious debate will take place in Congress on the topic. If an issue or problem fails to surface on the legislative agenda or if it does not gain salience in the policy deliberations, the existing state of affairs—the law as set out by the Supreme Court—remains the status quo. Once the matter or issue begins to attract attention, however, legislators will have increased opportunities to take positions and propose policy reforms that advance their political aims and goals.

In this section, we seek to explain why some but not all cases appear on the legislative agenda. As noted in Section II above, once a case attracts attention, legislators will respond in a number of different ways that may include simply referring to the case in floor debates and hearings or attempting to impact the law substantively through legislation that codifies or overrules the case. In this section, we limit our investigation to the factors that influence which issues emerge as salient; later we explore the substance of the legislative oversight and the type of legal reform that emerges in response to the case outcome. Here we limit our exploration to the factors that explain 1) why a case appears on the agenda as a general matter; 2) why some cases attract sustained attention over many years while others are subject to perfunctory notice and comment in the congressional record; and 3) why some cases surface quickly—within days of the Supreme Court’s publication of a case—while other cases do not attract attention for years. Because we are interested in three different features of congressional oversight— incidence, frequency, and speed—we will estimate three different statistical models.

Before we present our statistical findings, we briefly discuss our theory of oversight. Our theory for explaining why the legislature responds to Supreme Court cases revolves around the idea of issue salience. Issues decided by the Supreme Court often become important to legislators not because they are inherently interesting (this might be particularly so of tax cases!) but because they have achieved significance through the work of agenda entrepreneurs or because of unique economic, political, and legal circumstances.

A.Theory and Hypotheses: Why Do Legislators Review Supreme Court Cases?

It is no surprise that legislators spend their time addressing salient policy issues; the more difficult puzzle is how and why do U.S. Supreme Court cases become salient topics in legislative circles? Existing studies suggest a variety of forces, including individual legislator activities, media coverage, lobbying efforts, and social and economic crises, may impact the type of issue that becomes worthy of legislative notice and comment.[35] Although much of the literature focuses on the congressional agenda as a general matter and not specifically with regard to Supreme Court cases,[36] we hypothesize that many of the same factors—as well as others—explain how and why the justices’opinions grab the attention of Congress.

1. Agenda Entrepreneurs

a. Legislators: Members of the Tax and Budget Committees in Congress

Legislators recognize that discussion and debate of important policy problems in the halls of Congress can lead to increased media coverage and heightened constituent satisfaction. This reality means legislators will compete to place issues within their area of expertise on the legislative agenda, which in turn enables them to advance their policy goals but also reap political rewards associated with increased national attention.[37] With respect to Supreme Court tax decisions, we expect the agenda entrepreneurs in Congress will be members of the tax and budget committees. The House and Senate have a total of thirty-five standing committees, but only five have expertise, jurisdiction and control over tax and budget issues.[38]

The tax committees, which include the House Ways and Means Committee, the Senate Finance Committee, and the Joint Committee on Taxation, have responsibility for holding hearings, submitting legal reports and summaries to Congress on tax issues, and for drafting and amending the tax statute.[39] These three committees, like all the other committees in Congress, seek to place their issues at the center of national debate and at a substantive level—are completely aware that Supreme Court cases can impact legislative decisions and bargains codified in the statute.[40] For this reason, we expect that these three committees will be particularly attentive to Supreme Court tax cases and will be more likely to notice and comment on them in debates and hearings and will refer to them when proposing legal reform in the taxation context. We also hypothesize that the House and Senate Budget Committees, which together have authority to establish and enforce the national budget, will pay close attention to Supreme Court tax decisions. Budgetary goals are closely linked to the level of revenue raised through taxes and spent through tax expenditures, and the Budget Committee members also face the reality that Supreme Court decisions can have a sizeable effect on budgetary policies through pro government outcomes that may provide unexpected revenues or pro-taxpayer outcomes that lead to revenue losses.[41]

Given that tax lawmaking impacts taxpayers all across the nation, we recognize that nearly all legislators—regardless of committee assignment—will have incentive to comment on Supreme Court tax cases when the Court’s outcome has a notable impact on constituents back home. But we theorize that Members of the Tax and Budget Committees will be far more likely to notice and comment on Supreme Court tax decisions than other Members of Congress and will seek to make the issues salient even when they appear to be irrelevant to existing policy debates.

Our data demonstrate that the Tax and Budget Committee members indeed are quite a bit more likely to monitor Supreme Court activity. Members of these committees were responsible for 81% (540/664) of the comments made on tax cases in the legislative histories, documents, reports, and so forth. The House Tax and Budget Committees authored 20% of all the comments, the Senate Tax and Budget Committee authored 18%, and the Joint Tax Committee authored the remaining 42% of all the remarks and comments. A preliminary analysis of the correlation between these actors and oversight of Supreme Court tax outcomes confirms suggests that tax and budget committee members operate as agenda entrepreneurs when it comes to Supreme Court tax cases.

b. Interest Groups

Interest groups play an important role in the development of the legislative agenda. As many scholars have noted, when organized and cohesive, these groups can increase the salience of an issue by providing legislators with information, data, and credible arguments for reform.[42] In this way, interest groups act as agenda entrepreneurs regularly drawing attention to their issues and problems including the judicial decisions rendered by the Supreme Court. Interest groups harmed by the Court’s outcome will seek a legislative override, whereas those advantaged by the decision will seek a codification of the ruling to cement the outcome into law or perhaps to broaden its effect.

Although it is feasible to identify the individuals and groups that appear before Congress as witnesses in hearings or in some other formal role, it is fruitless to attempt to identify the range of interests that contact legislators through unofficial channels. A proxy for such lobbying efforts, however, may exist. The cases that reach the Supreme Court tend to impact a wide range of taxpayers beyond the named litigants and this leads interested parties to file amicus curiae briefs in order to advocate their position. Individuals and groups willing to spend time, energy, and resources filing these briefs are likely to lobby legislators after the justices render a decision in the controversy.[43]

Accordingly, when interest groups get involved in the issue, we hypothesize that these groups (or related groups with similar interests) will seek congressional oversight and that the likelihood, speed, and frequency of legislative responses will increase.

During the period of our study, individuals and groups filed amicus curiae briefs in 42% of the tax cases. Some controversies generated a single brief (generally on behalf of the taxpayer) while others involved more than twenty briefs with hundreds of signatories. In St. Martin Evangelical Lutheran Church v. South Dakota,[44]a case in which the Court held that religious schools are not subject to the federal unemployment tax rules, for example, twelve religious organizations filed briefs on behalf of the taxpayers;[45]in United States v. United States Shoe Corp,[46]a case in which the Court found the Harbor Maintenance Tax to be unconstitutional, thirteen corporate organizations filed on behalf of the taxpayer; [47]and in Newark Morning Ledger v. United States,[48]a case in which the Court permitted taxpayers to depreciate the value of newspaper customer lists (an item the newspaper had purchased in a merger), 21 amicus filed briefs with the Court.[49] Other cases such as United Dominion Industries v. United States[50]and United States v. Cleveland Indians Baseball,[51]initiated just one amicus brief filing. A slight majority of the cases (58%) had no amicus filings.

For purposes of testing our prediction that amicus filings in the Supreme Court will impact legislative activity subsequent to the decision, we created three measures: 1) a continuous variable equal to the number of amicus briefs filed on behalf of the taxpayer in each case, 2) a dichotomous variable indicating whether amicus briefs were filed or not, and 3) a dichotomous variable indicating whether briefs were filed on behalf of the taxpayer of the government. Together these variables enable us to assess the impact of brief filing on legislative oversight as well as the impact of increasing numbers of interest groups on the process. As we note below, the number of briefs filed is less important than the fact a brief is filed. Moreover, our data indicates that taxpayer brief-writers are more active in getting legislative oversight to Supreme Court cases than those supporting the government.

c. Journalists

Although legislators and interest groups often seek to place issues on the congressional agenda for their own political and economic purposes, factors outside this context can contribute to issues salience. Media coverage, in particular, operates as an agenda-setting function in Congress—numerous studies have demonstrated that legislators respond to issues made germane through articles in newspapers and magazines.[52] It is debatable whether the public opinion attached to salient issues can actually influence policy outcomes in the long run, few question politicians’ strong inclination to respond in some fashion to the leading issues of the day reported in the national media.[53]

While many commentators consider Supreme Court tax cases “boring”[54]and therefore seem unlikely to receive the level of media attention devoted to, say, civil rights controversies, in fact, at least 52% of the cases in our database were subject to coverage in a prominent newspaper.[55] When the Court reported its decision in South Carolina v. Baker,[56]holding that Congress had the power to tax state and local bonds, for example, the New York Times reported the development the day after the Court rendered the decision. While the statutory provision at issue in the case had the potential to impact very few bond holders, the New York Times suggested that the decision was of “major importance to state and local governments” and noted that “In the credit markets, prices of municipal bonds fell immediately after word of the decision . . . [although] they recovered as market experts pointed out that the ruling did nothing to alter statutes already on the books.”[57] The case received extensive media coverage across the nation and this led many state and local governments to worry that the justices’ decision would ‘cast a pall’ over their ability to raise capital.[58]

Although the decision in South Carolina v. Baker expanded congressional power, members of Congress critiqued the decision for “permitting unwarranted intrusion into state and local affairs.”[59] Soon after the Court issued the decision, the House proposed a Constitutional Amendment that would eliminate the power to tax; when the constitutional amendment failed to gain support, both the House and the Senate adopted resolutions guaranteeing that they would not support a new tax on state and local bonds.[60] Since 1988, when the Supreme Court issued the decision, Congress has discussed the case nearly every year in committee or on the floor and has adopted twenty-one different resolutions promising not to tax state and local bonds. Whether these resolutions have any significance for the viability of capital markets is not clear, but given their frequency, it is clear legislators believe the resolutions are important for “position-taking” purposes.

Not all cases receiving media coverage spark the level of congressional response received by South Carolina v. Baker—although nearly all the cases discussed in the media are subject to at least some level of commentary in Congress down the road. In United States v. Bisceglia, for example, the Court held that the Internal Revenue Service could compel banks to make private individuals’ records available for inspection through “John Doe” summons.[61] This case also received prominent news coverage in the New York Times the day after the Court issued its opinion and members of Congress did comment on and study the decision, but only for a brief period of time.[62] Notably South Carolina v. Baker, which has been subject to repeated monitoring, did not lead legislators to amend the tax statute, while Congress did modify the law in response to Bisceglia, suggesting the intensity of the oversight is not necessarily correlated to actual policy changes.[63]

For purposes of this article, we measure salience by the coverage the case received in the four major newspapers across the nation within one week of the decision including the New York Times, Wall Street Journal,L.A. Times, and Chicago Tribune.[64]

In hypothesizing a correlation between media coverage and legislative responses to Supreme Court taxes, we are not suggesting that legislators (or staff members) necessarily read the newspapers and then proceed to discuss the issue on the chamber floor or respond with a bill intended to override, codify, or amend the Supreme Court’s holding. Instead, we expect that media coverage contributes to the salience of the case in public debates and among constituents, and thus will be correlated to legislator’s inclination to take some sort of action with regard to the judicial decision.

d. Supreme Court Justices

In addition to legislators, interest groups, and journalists, the justices themselves act as agenda entrepreneurs. In a surprising number of cases, the justices have attempted to attract Congress’

attention through language in their opinions that explicitly invites a legislative response. The justices do this often by pointing to the drawbacks of the majority opinion and then inviting Members of Congress to remedy the perceived problem. In United States v. Byrum,[65]for example, the Court considered whether an owner of stock who transferred his shares to his children but retained substantial voting rights should be forced to include the property in his estate upon death.[66] The Court ruled against the government in the controversy by accepting the so-called “general view of the tax statute” but noted the rule did not impose the appropriate tax consequences given the retention of a substantial property interest.[67] Justice Powell, writing for the majority, argued the Court should follow established rules but the opinion implied Congress should reconsider the outcome. After noting the drawbacks of the Court’s approach, Powell noted:

When a principle of taxation requires re-examination, Congress is better equipped than a court to define precisely the type of conduct, which results in tax consequences. When courts readily undertake such tasks, taxpayers may not rely with assurance on what appear to be established rules lest they be subsequently overturned.”[68]

Congress responded to Byrum with the “anti-Byrum rule” reversing the outcome and allowing the government to obtain the desired rule through legislation.[69]

In Dickman v. United States,[70]the Court ruled for the government in the context of a gift tax controversy imposing a tax on imputed income from an intra-family loan agreement that did not call for interest payments.[71] Justices Powell and Rehnquist, in a dissenting opinion, called for a legislative response to the majority opinion, which they argued had not only overstepped the bounds of judicial authority, but had created both an ill-advised and inequitable rule of taxation.[72] Congress also responded to this case but codified and expanded the rule giving the government even a bigger win than it obtained in Court.[73]

Scholars such as William Eskridge have noted that the justices are likely to request congressional action for institutional reasons—they believe Congress is the appropriate body to make major policy decisions in certain legal contexts.[74] This is especially true in circumstances in which the justices believe a long-standing interpretation of existing statutes should be modified or changed completely. In fact, the language and context of both Byrum and Dickman suggest this is precisely what led the justices to draw legislative attention to the cases. This interpretation of judicial motive is not universally accepted, however. Pablo Spiller and Emerson Tiller argue that the Court’s invitation may not reflect deference to Congress but rather a strategic maneuver that allows the justices to achieve simultaneously their preferred position on two different policy dimensions.[75] The justices may prefer, for example, to adhere to a particular interpretive approach in the statutory context (e.g., plain meaning) but adopting this approach could lead to an undesirable policy outcome (e.g. the government prevails in tax case in which the justices prefer a pro-taxpayer outcome). To achieve both preferences, Spiller and Tiller suggest the Court will decide the case using the preferred canons of interpretation and at the same time invite a congressional override.

We do not have an opinion as to why the Court invites legislative responses,[76]but we do hypothesize that when the justices publicly call for congressional action in response to a Court opinion, the request will increase the likelihood that the case would show up on the legislative agenda. Of the 284 decisions the Court rendered between 1954-2004, the justices invited legislative oversight 31 times. We created a dichotomous variable to test our hypothesis that invitations increased the probability of congressional oversight: a case with a judicial invitation (in any opinion) is coded 1 and all other cases as 0.

A second way that the justices may heighten the salience of a case is to render an opinion with a divided vote. When justices dissent, they signal that a genuine legal controversy continues to exist and thus some form of legislative response is necessary to assure a clear and final outcome on an important legal topic. This controversy could relate to any number of aspects of court decisions making, including a difference of opinion as to the outcome itself or disagreement as to which branch of government is best positioned to create national policy. A unanimous decision, by contrast, suggests little or no dispute and this too may be an important signal to Congress. Unanimity in decision- making by a heterogeneous body such as the Supreme Court suggests that regardless of one’s political or policy viewpoint, one single rational interpretation of the statute exists. We pick up on this point further below when we explore the substance of the congressional responses, but for now we note reasons exist to suspect that both divided and unanimous decisions grab the attention of Congress.[77]

In the period of our study, the Court reached a consensus in 109 (38%) of the tax cases it decided. Because we expect that both unanimous and divided votes will get the attention of Congress at the preliminary stages of congressional activity (and thus neither will show up as highly correlated with oversight), we created a dichotomous variable equal 1 if the decision was unanimous and 0 if divided. This allows us to investigate our hypothesis and to examine whether the vote has any impact on legislators’ inclination to comment on a case as many scholars have argued in the past.[78] In Section V, we examine the impact of divided/unanimous votes on the substance of the legislative oversight.

2.Economic Crises

Political economists have argued that national economic problems tend to increase the salience of fiscal and monetary matters in and around Congress.[79] Although unemployment, inflation, budget deficits, and GDP growth may not impact the content of legislative action, experts agree these forces move tax issues onto the legislative agenda—they act as “legislative vehicles” that motivate reforms and initiatives.[80] Indeed, the tax-writing committees often choose titles for their tax bills that reference the economic problem to be addressed, such as the “Economic Recovery Act of 1981”[81]or “Balanced Budget and Emergency Deficit Control Act of 1985.”[82]

Why would economic factors that make tax issues salient as a general matter lead Members of Congress to focus specifically on Supreme Court tax cases? Our theory revolves around data that suggest the Court grants certiorari to tax cases that raise issues of national importance and empirical studies show that the justices believe important issues are those that have major revenue implications for the federal fisc.[83] We hypothesize that decisions impacting federal revenue intake during periods when the economy has heightened salience will take on increased importance in legislative debates. Of course, this theory would also suggest that policy changes emanating the Treasury or federal district and appellate courts will also get heightened attention but in this article we focus on just one vehicle of change: Supreme Court tax cases.

Although we expect that the tax cases are more likely to show up on the legislative agenda during times of perceived (or claimed) economic crises, we do not expect legislators will pay particular attention to pro-government or pro-taxpayer outcomes. As James Verdier has noted, legislators often have differing views on how best to respond to an economic problem and even when a consensus emerges in one era, it could easily change in the next. In the 1950s, for example, Congress preferred a balanced budget to the tax cuts that would stimulate a weak economy, whereas legislators in the 1960s and 1970s used the recession as a justification for a range of investment and business tax incentives that produced a deficit.[84] By the early 1980s, however, Congress began to worry about the increasing size of the federal budget deficit and proposed bills aimed at reducing its size.[85] Tax issues were on the agenda at every turn but differences in policy perspectives led to different responses.

There are also strategic reasons why legislators may be inclined to comment on tax cases generally, but not necessarily those favoring either the government or the taxpayer even if a consensus exists as to the appropriate tax remedy for the economic problem at hand. Legislators may prefer pro-government decisions in periods when the deficit is high, for example, but they may also fear reprisal by the interest groups who will suffer under spending cuts (i.e. the loss of tax preferences). If it is the Court that makes the difficult decision—i.e. the pro-government decision—then the legislators are able to gain the policy advantages of decreasing government spending while avoiding the responsibility of the tax increase. Indeed, legislators may critique the pro-government decision simply to convey to interest groups and constituents they are paying attention to their concerns, but in fact have no intention of proposing an override. This may have been Senator Dole’s strategy when he spoke out against Arkansas Best Corp. v. Commissioner[86]—a case that closed a loophole long criticized by tax experts and policy analysts.[87] Arkansas Best arguably helped to reduce the level of the budget deficit but it had the potential to seriously harm interest groups, especially manufacturing interests in farm states such as Kansas.[88] By speaking out against the pro-government decision, Dole—who represented Kansas and chaired the Senate Finance Committee—signaled to his constituents he cared about their interests even if he had no plans to propose substantive legislation that would reverse the Court’s outcome. In short, we expect economic factors to impact the level of attention that Supreme Court cases draw in Congress, but not the direction of the commentary.

To test whether a correlation exists between economics and case salience, we collected data on the level of unemployment inflation, the federal budget deficit as a percentage of GDP, and GDP growth. The measures are all denominated in real terms (fiscal year 2000 values).

3.Political and Ideological Differences

Much of the literature on Court-Congress dynamics considers the ideological position of each branch of government for purposes of understanding and predicting congressional responses to Supreme Court cases. Theorists hypothesize that as the political distance between Congress and the Court expands, the probability of oversight will also increase. This theory is grounded in empirical data that shows 1) justices decide cases in a manner that furthers their own political and ideological preferences,[89]and 2) legislators respond to salient events in a manner that promotes their own political agenda.[90] These findings imply that a conservative court will be subject to increased oversight when the legislature is ideologically left and vice versa.[91] In fact, a preliminary investigation of our data indicates that as the ideological distance between Congress and the Court increases, the frequency and speed of legislative oversight also increases.

For purposes of testing the hypothesis further, we rely on the legislative “Common Space Scores” generated by Poole and Rosenthal[92]as well as on the judicial “Common Space Scores” generated by Epstein, Martin, Segal, and Westerland.[93]

Because the scores for both the Court and Congress are on the same scale we are able to compare directly the ideologies of the two branches of government. To investigate our hypothesis, we compared the preferences of the Court and 1) the median member of each chamber, 2) the median member of the party in control of each chamber, and 3) the median member of the gate-keeping committee in each chamber. We use these three different congressional medians to account for the divergent theories of congressional power found in the political science literature.[94] We believe that the chamber majorities, the political party in control of each chamber, as well as the tax-writing committees all wield significant power at different times and thus investigate each group separately.[95]

The “common space scores,” for both Congress and the Court, are bounded below by –1 and above by 1, with lower scores indicating a more liberal ideology. We calculated the distance between the median member of each body and used this measure to determine whether the legislators and justices were politically aligned or divided at the time the Court rendered the decision as well as when Congress acted in response to the Court. Although theoretically our political distance measure could be as great as 2 (Congress and the Court would have political preferences located at the opposite ends of the continuum) or as small as 0 (preferences are perfectly aligned), we do not observe either extreme: the maximum distance of the divide in our dataset is equal to .7160[96]and the minimum is equal .000095.[97] The average distance for each group (i.e. each chamber of Congress, the majority parties, and the tax committees vis-à-vis the Court) lies between .13 and .29. For purposes of comparison and for understanding these scores, consider that the ideological distance that existed between the House and the Supreme Court in 2004 was .254 and the distance between the Senate and Supreme Court median was .095. At this time, all three institutions were fairly conservative; the voting records of the median member of the House (Jim Walsh (R-New York)) and the median member of the Senate (Alan Specter (R-Penn)) indicate both were slightly more conservative than the median of the Supreme Court (Justice O’Connor).[98]

4.Periods of Legal Reform

We have suggested that salient Supreme Court decisions operate as a catalyst to legislative action: legislators who dislike an outcome are apt to critique it and propose an override, those who approve will work to cement the decision into statutory law via codification, and others will comment purely for position-taking purposes. But it is also possible that the causal arrow works in exactly the opposite direction: when legislative initiatives gain prominence, Members of Congress are likely to investigate existing law for purposes of drafting new statutes or for debating its merits. Put differently, the move for reform within Congress may provoke increased interest in Supreme Court decisions.

Our database suggests as much. Consider a 1984 initiative that called for a tax on individuals who relinquished their American citizenship as a means to avoid taxation on income. In debates and hearings addressing the proposal, Members of Congress relied on the Supreme Court’s definition of income to justify the new statute. Taxing expatriates was legally permissible, it was argued, because the justices had adopted an expansive definition of “taxable income” in cases such as Commissioner v. Glenshaw Glass and General American Investors v. Commissioner.[99] Similarly, when considering various presidential budget proposals, the Joint Committee on Taxation issued reports noting that the proposal would have no impact on existing Supreme Court law such as Commissioner v. P.G. Lake and Commissioner v. Gillette.[100]And when Congress was in the midst of enacting the major reforms in 1986, committees issued numerous reports and studies of existing law, which included references to relevant Supreme Court cases.[101]

To investigate the impact of preexisting reform proposals on legislators’ inclination to notice and comment on Supreme Court tax cases, we identified seven historically important and notable pieces of legislation in the tax and budget context adopted between the years 1954 and 2004. We expect that when Congress undertakes fundamental reform, it will assess the existing state of the law more rigorously and thus review of Supreme Court cases will increase. The major tax statutes we include in our study are the Revenue Act of 1964,[102] the Tax Reform Act of 1969,[103] the Economic Recovery Tax Act of 1981,[104] and Tax Reform Act of 1986.[105] In the budget context, we include the Congressional Budget and Impoundment Control Act of 1974,[106] the Balanced Budget and Emergency Deficit Control Act of 1985 (commonly known as the Gramm-Rudman-Hollings Act),[107],and the Budget Enforcement Act of 1990 (also know as PAYGO).108 We hypothesize that for two years prior to the adoption of these reforms, the probability that the Supreme Court cases gain notice and attention will increase.[109]

Once the legislation is adopted, interest in the cases—and citation to them in committee reports, legislative, and hearings—will decrease in number.

The one exception to this decline in attention involves the Budget Enforcement Act of 1990. This law, extended through 2002, mandated revenue neutral legislation, which allowed legislators to pursue new tax expenditures but only if they were offset by revenue increases.[110] We believe Supreme Court cases will take on increased importance during this era because of their potential to impact the revenue implications of existing statutory enactments. Newark Morning Ledger Co. v. U.S.[,111]for example, is a case that involved a newspaper’s attempt to amortize the cost of customer lists. The government argued against the amortization noting that if the justices allowed the taxpayer to prevail, the fisc would millions in revenue due to unexpected tax deductions.[112] The Court held for the taxpayer, which meant that an existing provision in the tax code immediately cost the government quite a bit more than anticipated by the budget analysts in Congress. The opinion is noteworthy not only because of the fiscal implications but to override the outcome would effectively implement a tax increase—making it feasible to adopt a whole new collection of tax expenditures while still abiding by the revenue neutral mandate found in the Budget Enforcement Act. Not surprisingly various legislators argued for a reversal of the opinion with accompanying legislation that would increase tax preferences in other areas.[113] Other legislators argued that the outcome should be codified—this would be a revenue neutral move given that the legislators themselves would not add a new expense to the fisc but would cement an existing rule (albeit created by the Supreme Court) into the statute.[114] Accordingly, we hypothesize that the Budget Enforcement Act of 1990 will lead to increased attention to Supreme Court decisions between 1988 and 2002, the time period when Congress was considering the legislation for adoption and the point at which it expired.

For purposes of testing our hypothesis, we created two variables. The first is a dichotomous variable that account for major tax bills under consideration—specifically years 1962-64, 1967-69, 1979-81, and 1984-1986. The second variable accounts for the major budget bills and includes the years: 1972-74, 1982-84, 1988-2002. Luckily, there is no overlap, so we are able to assess the impact—if any—of tax versus budgetary lawmaking on the citation of Supreme Court cases. We expect, however, general increased activity during all these years but not in the 23 years when no major legislation was under consideration.

B.Summary of Hypotheses

Our goal in this section IIIA of the article is to understand and explain legislators’ inclination to notice and comment on Supreme Court cases in the tax context. We expect tax cases to show up on the legislative agenda through the work of agenda entrepreneurs: Tax and Budget Committee Members in Congress, interest groups, journalists, and the justices themselves. We also expect that economic crises and major legal reform will increase the probability that legislators will notice Supreme Court tax cases. With regard to political and ideological differences between the Court and Congress—we expect that as the ideological distance between Congress and the Court increases, oversight will also increase. We conducted preliminary analyses of the data and found correlations exist between many of these factors and congressional oversight of the tax cases. Below, we explore the factors in fully specified statistical models. Specifically, we examine three different features of oversight: its presence, frequency, and speed in the Court-Congress dynamic and find that the variables have different roles to play in different contexts.

IV. THE OCCURRENCE, FREQUENCY, AND TIMING OF CONGRESSIONAL OVERSIGHT

In this section, we move from hypothesizing to testing whether the variables described above impact the congressional decision to begin the oversight process of a Supreme Court tax case. In Section IVA we investigate the variables that trigger congressional oversight as a general matter, as well as the frequency of this oversight; in Section IVB we explore the speed with which oversight takes place.

A. The General and Count Models

As noted above, Congress commented on 55% percent of all the tax cases decided between 1954-2004. We first seek to understand why legislators comment on certain cases but leave others undisturbed in the legislative process.[115] Next we seek to explain why some cases generate a single comment in Congress, while others lead to extensive debate and controversy over the course of several decades.[116] We use two different models to investigate the variables that are correlated to the occurrence and frequency of legislative monitoring. Our explanatory variables are the same in both models and all focus on the role of agenda entrepreneurs, specifically lobbyists, journalists, and the justices themselves. For now, we must ignore the activities of the tax-writing committees, as well as the variables associated with economic crises, political ideology, and legal reform—all of which we pick up in the next section using a model that allows us to investigate these time-varying factors.[117]

Our findings for both the general oversight and the frequency of oversight are presented in table 2 below and highlight the role of lobbyists and the Supreme Court justices as agenda setters in the legislative process. Our findings indicate that lobbyists and justices signal legislators on emerging and important legal issues—and most importantly—legislators are attentive to both groups. Specifically, our general model, which examines the likelihood of oversight as a in the main, demonstrates that when the justices invite oversight of their own decision-making process, legislators are quite a bit more likely to comment on a case then ignore it. Our count model, which allows us to investigate why certain cases are subject to repeated and extensive oversight, also highlight the important role of the justices, but also shows that when lobbyists get involved the level of oversight increases. Our models show a statistically significant correlation between judicial invitations and congressional oversight both as to oversight generally but also as to the number of times legislators mentions a case in hearings, debates, and so forth. Lobbying activities are also correlated with oversight at statistically significant levels, but only in the count model measuring the number of responses to individual cases.

Table 2: Congressional review of Supreme Court cases. Statistical significance at the .01 level marked * and robust standard errors are found in parentheses.

 

EXPLANATORY VARIABLES

GENERAL (LOGIT) MODEL FINDINGS118

COUNTMODEL

FINDINGS119

Lobbyists

Journalists

Justices: Invitation for Review Justices: Unanimous Decision Constant

alpha

.404 (.267)

.313 (.265)

1.740* (.557)

.106 (.259)

-.403 (.223)

.665* (.211)

.353 (.207)

.994* (.308)

.181 (.217)

.305 (.116)

2.445 (.298)

 

Table 2 presents raw coefficients and robust standard errors and show important correlations, but these statistics do not by themselves convey the substantive impact of each variable on the oversight process. Accordingly, we translate this information into probabilities to provide greater detail into the role of the agenda setters on congressional activities. We find that on average, each case has a 54% chance of being subject to some kind of congressional oversight.[120] If the justices write an opinion that invites legislative review of the case, however, the probability of oversight increases to 83%–the justices are able to increase the probability of legislative activity by nearly 30%! In short they are able to make it a near certainty that legislators will react. We interpret this statistic as showing the level of inter-branch dialogue that is both possible and predictable in the context of national economic issues. If all the agenda entrepreneurs get involved (i.e. lobbyist, journalists, and the justices), the probability of oversight increases to 88% suggesting that it is the justices who have the greatest impact on the legislators’ decision to respond.

With regard to the frequency of oversight, our data indicates—as we would expect—that legislators are far more likely to comment on a case just once than they are to comment 50 or more times over the course of the oversight process. If the legislators do comment, the probability of a single comment is 15%; the probability of ten comments is 1%; the probability of thirty comments is .03%; and the likelihood of a case receiving fifty comments in Congress is equal to .001%. Although theoretically, the number of responses to a specific case is infinite, the maximum in our dataset is 57 citations and the average is 3.1.

The count model presented in table 2, column 2 explores the frequency of oversight and indicates that the rate of congressional commentary increases when private actors get involved and when the justices issue invitations for review of the Court opinion; their impact, however, is not constant. When we examine the substantive impact of these players, we find that lobbyists and justices do not explain why one case receives a single mention while another case receives two references in the congressional oversight process. But as the frequency of oversight increases, the role of each of these actors also increases. Put differently, the probability that legislators will comment one time is 15% and this likelihood does not increase when private groups lobby or the justices request oversight. However, when the level of oversight is already intense—then the lobbyists and the justices appear to have more of an impact. For example probability that legislators will comment 10 times on a case is 1%, but if lobbyists and justices get involved—the likelihood of oversight increases to 2%. We think the substantive impact of these two agenda entrepreneurs—although statistically significant—is quite small and thus we expect that quite a bit more is going on in the process than we can identify in statistical analyses of the data when it comes to the number of times that legislators refer to a case.

Accordingly, we conduct a qualitative analysis of the oversight process and find that many of the cases subject to more than perfunctory oversight were also the subject of codification and override proposals and not merely commentary for purposes of position- taking or for purposes of understanding the Court’s approach to statutory interpretation. This suggests that the frequency of commentary is linked to the specific outcome of the case and to actual statutory reform down the road.[121] We offer additional qualitative analysis of the substance of the oversight and comment further on legislative enactments in Section V below.

B.The Duration Model

We now turn from the occurrence and frequency of oversight to the period of time that a case lasts before congressional oversight takes place. This duration period is often very short—if legislators respond to a case they tend to do so fairly quickly after the justices render their decision. In fact, our data indicate that legislators are likely to respond to the Supreme Court within five years of the Court’s decision,[122]although some cases are left undisturbed for thirty years and then, for reasons never fully explored in the extant literature, surface onto the legislative agenda as a salient issue.[123] In this section, we seek to understand and explain why legislators take notice and comment upon certain Supreme Court tax cases soon after the justices render the decision while others are ignored for decades; put differently, why do some cases have a lengthy duration period before any mention in Congress, while others are addressed within months—even days— of the decision?

For purposes of understanding this question, we use a duration model, which enables us to investigate and explain the period of time that a case lasts without being commented upon by the legislators.[124] The variable we seek to explain is the speed of oversight and our explanatory variables include the agenda setters (lobbyists, journalists, and the justices) as well as the economic, political, and legal factors discussed in Section III above. We specified our model three different ways to account for our three distinct political measures—the ideological distance between each congressional chamber and the Court, the distance between each chamber majority party and the Court, and the distance between the tax-writing committees in each chamber and the Court.[125] These three models enable us to assess the role, if any, that politics plays in congressional oversight by the three different but important decision-making bodies in Congress. Finally, we include 51 “period-dummies” which allow us to assess the probability of a congressional response in each year after the Court decides the case.[126] We have 51 periods because the longest a case can last without comment is 51 years given that we study Court-Congress relations between the years 1954-2004.

Table 3 below presents our findings with regard to the sixteen independent variables discussed above; we exclude the period dummies for purposes of presentation.[127] Our models produce interesting and robust results. In all contexts, we find that Congress speeds up its response to the Supreme Court tax cases when lobbyists are involved and when the justices invite a response. These findings are consistent with the general and count models we explored above and confirm our theory that lobbyists and justices play an important and powerful role in setting the congressional agenda. On average, a case has a 3% chance of being noticed and commented upon by Congress in the year the decision is rendered, but if lobbyists and judicial invitations are present this probability increases to 9%.[128 ]Our findings do not depict a correlation between media coverage and congressional activity, which we find interesting because it suggests that journalists are not setting the congressional agenda, but are already aligned with it or perhaps are simply following it. Moreover, the lack of correlation between media coverage and the speed of oversight is surprising given that in other contexts scholars have hypothesized and found that journalists are capable of impacting the national agenda when civil rights are at issue.[129] Our findings also highlight the importance of disaggregating issues and examining each separately before making out-of- sample generalizations about the real world.[130]

With regard to the economic factors, we also find strong and interesting empirical results. We assessed the impact of unemployment, inflation, deficit as percent of the GDP, and the rate of GDP growth (all standardized to fiscal year 2000 values). Two indicators of national economic well-being, the level of the federal government deficit and the rate of GDP growth, are highly both correlated with legislative oversight. Our models indicate that high deficits and low GDP growth rates increase the likelihood that legislators will oversee Supreme Court decision-making in the tax context. In fact, when the deficit is at the highest and GDP growth is at the lowest, the probability that Congress will comment on a case within a year of the publication of decision is 50%–a notable increase from the average of a 3%. Although high deficits and a low GDP growth rate (both possible indicators of a problematic economy) speed up oversight, we found that high inflation does not correlate with congressional activity and that low unemployment rates actually increase legislative attention to the case. Given that high levels of deficit spending tend to decrease unemployment rates, however, we are not surprised that our data indicates that the two variables move in the same direction. Our findings are robust across all the models and are the virtually identical whether we include one-, three-, or five-year time lags, and irrespective of whether the taxpayer wins or loses in the Supreme Court.

The political and ideological differences between Congress and the Court did not have the relationship we expected. We hypothesized that as the ideological difference between the Court and Congress grew wider, oversight would increase. We examined the House and Senate chamber medians, the median of the majority party in each chamber, and the median of the tax-writing committees in each chamber and did not find that political differences explained much, if anything, about the process. In fact, we found no correlation with the House of Representatives and the Court and the correlations we found in the Senate suggested that as Senators were more closely aligned to the court—they were more likely to exercise oversight. Although the relationship between the Senate and the Court is potentially very interesting and may be suggestive of friendly legislators responding to friendly justices—as some scholars have argued[131]–we hesitate to give too much weight to this finding. First, our results are not robust across all three models, and second, our political measures may not successfully capture the views of the individuals who make the comments about the cases.[132]

Finally, we find that legal reform has no an impact on legislative activity. We expected that legislators would comment more readily on the tax cases when tax reform was on the agenda or when constrained by strict budgetary rules, but found no such correlation. This suggest that legal reform and legal rules are not motivating congressional review of the Court decisions, rather the economy, lobbyists, and the justices themselves are sparking legislative interest in the outcomes. Moreover, our findings indicate that the Court decisions, as a general matter, do no play a key role in budgetary decision or in major tax reform.

Table 3: Speed of congressional review of Supreme Court cases. Statistical significance at the .01 level indicated with * and robust standard errors found in parentheses.

 

EXPLANATORY VARIABLES

DURATION MODEL FINDINGS: CHAMBER POLITICS

DURATION MODEL FINDINGS: MAJORITY PARTY POLITICS

DURATION MODEL

FINDINGS: TAX COMMITTEE POLITICS

AGENDA ENTREPRENEURS

Lobbyists

Journalists

Justices: Invitation for Review

Justices: Unanimous Decision

ECONOMIC ENVIRONMENT

Unemployment

Inflation

Deficit as Percent of GDP GDP Growth

INTER-BRANCH POLITICS

House-Court Politics

Senate-Court Politics

House Majority-Court Politics

Senate Majority-Court Politics

House Ways &Means

Committee-Court Politics

Senate Finance Committee- Court Politics

LEGALREFORM Tax Reform on Agenda Budget Constraints Exist

Time-Period Controls

.513* (.205)

.182 (.191)

.837* (.281)

.023 (.198)

-.582* (.066)

-.487 (4.925)

-.373* (.062)

-18.811* (3.457)

.193 (1.368)

-4.267* (1611)

.428 (.340)

-.214 (.271)

——-

.494* (.206)

.230 (.194)

.798* (.295)

.003 (.198)

-.569* (.069)

-1.182 (5.261)

-.350 (.065)

-18.587** (3.538)

-1.439 (1.123)

.176 (1.181)

.384 (.342)

-.249 (.273)

——-

.516* (.204)

.146 (.190)

.797* (.272)

.025 (.199)

-.603* (.071)

-2.121 (5.315)

-.397* (.067)

-17.343* (3.699)

.464 (1.204)

-4.410* (1.279)

.393 (.356)

-.280 (.293)

——-

How do these factors impact the speed of legislative oversight when they are present together? We find, as noted, above that the likelihood of a congressional response to a case in the first year of decision is, on average, 3%. But if lobbyists are involved, the justices issue an invitation for oversight, the deficit is at its highest, and GDP growth is at its lowest—there is an 85% probability that the legislators will act on the case immediately. In the last section, we found that oversight as a general matter (i.e. ignoring the frequency and speed of oversight) was tied to judicial invitations and we find this factor increases the probability that legislators will review the cases sooner rather than later.

As to how our findings are similar to and different from the existing studies in the literature we find the following. Although no study focuses on all forms of oversight as we do and no study examines all the factors that indicate why certain cases become salient—there appears to be an emerging consensus that lobbyists and judicial invitations play an important role in the oversight process.[133] As we show below, however, the extant literature focuses on legislative overrides and our study confirms that lobbying and judicial invitations are highly correlated with this negative activity—but not to the positive responses that emerge in Congress that are associated with codification efforts.

Existing studies also suggest that unanimous decisions will decrease the likelihood that Congress will react but our study shows no such correlation.[134] We believe there may be an important reason for the difference in empirical finding: Unanimous decisions are positively correlated with legislative codifications of Supreme Court decisions but negatively associated with overrides—a finding that is impossible to uncover if scholars focus exclusively on overrides. With regard to the economic and legal reform variables that we investigate, no study has explored the importance of the economy for congressional policymaking vis-à-vis the Court decisions so comparisons are impossible. With regard to the political variables, the empirical literature is mixed on the importance of ideology and legislative oversight, but our study confirms the findings that suggest no such correlation exists.

We now turn to the substance of the oversight that occurs as well as the type of reform that legislators instigate in response to the cases and offer further analysis and comparisons.

V.THE SUBSTANCE AND DIRECTION OF CONGRESSIONAL OVERSIGHT

In this section, we turn from the factors that lead Congress to notice and comment on Supreme Court cases to the actual substance of the oversight when it takes place. As noted above, Congress responded to 152 cases decided during the years 1954-2004. The majority of the responses (66%) were positive or negative (or both) in nature, and the remaining 33% were impartial, in the sense that the comments were neither particularly critical nor favorable towards the Court’s outcome. In the qualitative analysis we provide below, we note that in circumstances in which legislators reveal preferences with regard to a specific case, they tend to include proposals to codify, override, or modify the outcome and roughly one-half of these proposals lead to substantive legal reform down the road. Impartial responses, by contrast, generally do not include legislative initiatives but rather summarize and report on the Court’s decision and its impact on the tax code. After exploring the cases subject to different legislative responses, we then investigate the factors that are correlated with each response category and who in Congress isresponsible for them. We find that the each different category is correlated with different factors discussed above.

A.A Qualitative Analysis of Four Categories of Oversight

1.Negative Responses and Overrides

The extant literature investigating legislative oversight of Supreme Court cases focuses exclusively on legislative activity that is unfavorable of Supreme Court decision- making and we find that quite a bit occurs in the context of taxation.[135] In our dataset, Congress responded in a negative manner to 28% (42/152) of the cases; 95% of these negative responses included a proposal to override or substantially modify the outcome; and 50% of the proposals actually led to new legislation. The earliest case to attract legislators’ negative attention is a 1955 case, U.S. v. Acri,[136]which involved the priority of government liens during bankruptcy proceedings and the most recent case is U.S. v. Fior D’Italia, which the Court decided in 2002 and involved the calculation of tip income for federal taxation purposes.[137]

Sixteen cases led to an actual override or modification, including well-known cases such as U.S. v. Grosso,[138]U.S. v. Davis,[139]Malat v. Riddel,[140]Commissioner v. Solimon,[141]and Gitliz v. Commissioner.[142]Eighteen cases led to override proposals that died in committee or were excised from the legislation during conference proceedings.[143]

The four cases that did not involve an override proposal but involved criticisms about the law addressed broad legal issues and problems. For example, the Senator Finance Committee cited the case of Berra v. United States to argue that the Supreme Court had contributed to the confusion in criminal tax law by failing to create clear and useful distinctions between felonies and misdemeanors;[144]Senator Laxalt pointed to Colonnade Catering Corp. v. U.S.[145]in raising concerns about the Internal Revenue Service and warrantless IRS inspections of documents;[146]and Senators Gordon and Domenici cited U.S. v. Hemme[147]and U.S. v. Carlton[148]when discussing the legality but unfairness of retroactive taxation.[149] These are cases that prompted exclusively negative responses— legislators who thought it worthwhile to address the Court’s outcome were apparently united in the view that it was problematic.

To our surprise, our qualitative analysis did not produce case-related factors that stand out as unique to Court opinions that received uniformly critical responses in Congress. The cases involved a range of different taxpayers (such as corporations, individuals, estates, and so forth),[150]both pro-government and pro-taxpayer outcomes,[151] both unanimous and divided opinions,[152]amicus curiae participated in some but not all of the cases,[153]the justices invited a congressional response in some but not most the cases,[154]the media reported on some but not all the cases.[155] The cases subject to override seem somewhat average when analyzed separately, but when these statistics are compared to the other categories—unique patterns emerge. We discuss this further below when we present our in statistical findings.

2.Positive Responses and Codifications

One of the least discussed features of congressional oversight involves legislators’ inclination to respond positively to judicial decision-making. Twenty-two percent of the responses (33/152) were wholly favorable; 94% of these responses involved a legislative initiative to codify the case; and 51% of these proposals succeeded in Congress.[156] The first case in the database to garner a positive response is a 1954 case involving the priority of government liens associated with overdue taxes in U.S. v. City of New Britain[157]and the most recent case is O’Gilvie v. U.S.[158]decided in 1996 involving taxation of punitive damages.

At first cut it might seem strange that Congress would spend time and energy implementing code provisions that mirror Supreme Court outcomes; after all, the Court’s precedent applies nationwide and legislators need not cement the case outcome into statutory law to assure constituents benefit under the ruling. In fact, however, we found that Congress rarely “rubber stamps” Supreme Courts decisions into statutory law. Often codification occurs in the context of a larger reform effort that encompasses the issue decided by the Court in an earlier term. When Congress enacted legislation governing lease transactions in 1983, for example, it referred to and ultimately sanctioned the outcome of Frank Lyon v. U.S., which addressed leasing transaction and was decided in 1978;[159] when Members of Congress debated possible legislative remedies to thetax shelter problem they referred again to Frank Lyon as well as to Knetsch v. U.S.,[160]

Hillsboro v. Commissioner,[161] and various other cases that were helpful in closing loopholes in manner that replicated the outcomes found in the justices’ opinions.[162] In other circumstances, Congress sought to expand the Supreme Court outcome beyond the facts of the case and in doing so sanctioned the Court’s approach to the issues considered.[163] Legislators, for example, sought to extend the results of Harris v. Commissioner, U.S. v. Midland Ross Corp., U.S. v. Correll, and U.S. v. Goodyear Tire & Rubber Co. to a number of other related fact scenarios that the Court had not addressed or that were not raised by the litigants in the judicial proceedings.[164] Finally, in situations in which Congress does codify the case in a “rubber stamp” fashion, the legislative histories suggest that the Court opinion did not eliminate confusion on the issue.[165]

Understanding the nature of the codification efforts, however, does not provide insight into the types of cases that Congress is likely to spend time and energy commending and supporting. Like our analysis of the cases that Congress disfavors, we found that the judicial decisions that receive positive attention involve a range of different taxpayers,[166 ]both pro-government and pro-taxpayer outcomes,[167]amicus curiae are present in some but not all of the cases,[168]judicial invitations for a congressional response are not notably few or numerous,[169]and journalists reported on some but not all the cases.[170] There is, however, one factors that stands out as unique in our qualitative analysis: the number of unanimous decisions subject to codification efforts. When we examine the data in the aggregate, it is clear that Congress gives far more attention to Supreme Court decisions with divided votes; when we disaggregate the data, we again find that the legislators spend quite a bit more time focusing on divided decisions with one exception—the cluster of cases subject to positive reinforcement. Legislators are slightly more likely (17/33) to commend case outcomes with unanimous decisions but quite a bit less likely in all the other circumstances. This suggests that unanimous decisions operate as a signal to Congress—when a heterogeneous group of decision- makers share a view on an important legal issue—legislators (also with diverse viewpoints) are likely to agree with that outcome. Irrespective of whether legislators are driven by ideology or substantive policy-making, the unanimous decision suggests that individuals all along the spectrum can agree. If the issue is important and salient, then we predict that unanimous decisions will garner increased and positive attention in Congress.

3. Mixed Responses and a Combination of Overrides/Codifications

Legislators, as might be expected, do not always express a unified view on Supreme Court cases. Eighteen percent (28/152) of the cases subject to oversight sparked both negative and positive responses in Congress; 93% of these comments included a legislative proposal to override or codify (or both) the Court’s outcome; and 50% of the proposals led to actual legal reform. The earliest case in our database to spark debate and controversy is a 1955 case, U.S. v. California Eastern Line Co.,[171]which involved taxpayer rights to appeal lower court outcomes, and the most recent case is Newark Morning News Ledger v. U.S. decided in 1993 and involved depreciation of the cost of customer lists by publishers.[172]

Often the conflicting responses arose in a predictable fashion. Minority members of the Ways and Means Committee or the Senate Finance Committee, for example, would register formal opposition to a proposal to override or codify a case.[173] At times the sequence would be reversed: favorable comments on a case would be followed by a proposal to override the outcome (or vice versa). The House Ways and Means Committee, for example, commented approvingly on Arkansas Best v. Commissioner[174] and these comments were shortly followed by critical comments with an implicit suggestion that Congress should overturn the outcome by Senator Dole.[175] More interesting are the cases that engender competing proposals in the House and the Senate. The case, Dickman v. Commissioner, to name just one example, involved a debate concerning the taxation of imputed income associated with intra-family loans and sparked a House proposal to codify the outcome and a rival proposal in the Senate to override it.[176]

A qualitative analysis of the cases that fueled a mixed response suggests two notable differences from the cases that attract unidirectional reaction that is either positive or negative in nature. First, the majority of the Court cases subject to both types of responses were not unanimous,[177]suggesting that dissension on the Court is useful for predicting dissent and discord in Congress—just as an issue divides the justices it will divide Members of Congress. Second, the majority of the cases subject to conflicting views in Congress involved lobbyists—a factor that is also not present in the other category of responses.[178] This indicates that the interested parties (whether they win or lose) go directly to Congress to obtain a codification or override and given the salience of the issue as well as the conflicting views that exist—numerous and competing comments and proposals emerge in Congress. As to the other variables, including the identity of the taxpayer, the identity of the winning party, the number of judicial invitations for a congressional response, and media coverage: our qualitative analysis does not suggest these factors are unique with respect to highly controversial cases.[179]

4.Impartial Responses and Neutral Legislative Enactments

The final category of responses is the collection of neutral or impartial reactions to the Supreme Court tax cases. This is the largest grouping (49/152) and generally includes summaries of the law as presented in hearing and reports; only 18% of these responses included a legislative proposal (although these were not override or codification initiatives but were intended to address statutory provisions implicated by the outcome); and 33% of the proposals lead to a statutory reform. The earliest case in the database is Holland v. U.S.,[180]decided in 1954 and cited for its definition of the term “willingness” during impeachment proceedings of Judge Harry Clairborne,[181]and the most recent is U.S. v. Cleveland Indians Baseball Co.,[182]decided in 2001 and cited for purposes of understanding the Court’s approach to statutory interpretation.[183]

The fact that so few of the impartial responses involved legislative reform proposals immediately sets this category of congressional reactions apart for all the others discussed above and suggests strongly that legislators undertake a constant review of the tax law as it emerges from the Court. Not for purposes of legislative reform or for signaling constituents that legislators are aware of issues impacting local interests, but to understand the nature of the law itself. We expect this type of constant and apparently objective oversight to be linked to the existence of the Joint Committee on Taxation, a Committee comprised of both House and Senate members. Congress created the Joint Tax Committee in 1926 pursuant to Section 8022 of the tax code and authorized it to “investigate” and “report” on the existing state of the tax law but did not allocate legislative drafting powers to its members.[184] The majority of the neutral comments on the Supreme Court cases emerged in reports issued by the Joint Tax Committee; moreover, 40% (320/826) of all the legislative responses in the entire database originated in the Joint Committee Reports, suggesting that the Joint Committee frequently reports to Congress on the state of the tax law as mandated by the statute and as we would expect, includes Supreme Court case law as part of the record.[185]

That the Joint Tax Committee is actively reporting on the state of the law does not mean that all the neutral responses come from this Committee. As noted above, Senator Strom Thurmond referred to Holland v. U.S.,as well as Sansone v. U.S., and U.S. v. Bishop for purposes of defining “willfulness” in impeachment proceedings.[186]

Legislators referred to General American Investors v. Commissioner, Commissioner v. Acker, James v. U.S., and many other cases for purposes of understanding the Court’s approach to statutory interpretation when drafting legislation outside the tax context.[187]

And a small portion of the impartial responses emerged from legislators proposing a modification of the law in a manner that did not suggest Congress was either particularly critical or favorably inclined to the outcome reached by the Court. In these reform efforts, for example, Congress cited cases such as Haynes v. U.S. and U.S. v. Bisceglia when advocating a change in the law to avoid confusion on a different but related issue decided by the justices;[188]they cited U.S. v. Dalm when advocating a change in the law to address an issues the justices left undecided.[189] In response to United States Shoe Corp v. U. S., Congress reformed the law in a manner that literally followed the justices’ instructions found in the opinion for assuring that a customs tax satisfied constitutional mandates.[190]

With regard to the factors associated with judicial invitations to override, divided opinions, media attention, the date the Court published the decision, and so forth—the impartial responses do not stand out as notable.[191] The main distinction that we observe in our qualitative analysis of the cases and responses is that very few legislative proposals are linked to the impartial comments leading us to believe that the comments are driven by Section 8022, which requires the Joint Committee to issue periodic reports and summaries on the state of the law, more than a desire to initiate policy change or as a political move motivated by constituent interests.

B.A Quantitative Analysis of the Substantive Responses

Our qualitative analysis suggested a number of possible interesting correlations, but to test whether this relationship is statistically significant, we now turn to quantitative analysis of the our dataset. In this section, we use two models to investigate the factors we highlighted above and examine their impact on the probability of a negative, positive, mixed, and impartial response to a Supreme Court tax case. Our models grow out of our qualitative analysis of the cases and for this reason, we can use them to describe the data as well as predict the substance of future legislative oversight in this area of the law.

Additional research and exploration, however, is necessary before we can confirm the factors that lead legislators to act in substantively different ways to Supreme Court cases.

In order to investigate the factors that impact the different categories of oversight and legislative reform initiatives, we use the same explanatory variables discussed above, including the agenda entrepreneurs (tax and budget committee members, lobbyists, journalists, and the justices), the economic variables, and the political factors. Because we are trying to explain the substance the oversight, our dependent variable includes four different categories (i.e oversight and legal reform that is 1) positive, 2) negative, 3) mixed, or 4) neutral). This type of statistical analysis requires use of a model that is quite a bit more complicated to present in a table than the three models we used in our statistical analysis in Section IV.[192] Accordingly, we forego presentation of the raw coefficients and robust standard errors as above, and set forth only the probabilities that we generated from our statistical models.[193]

In section VB1 immediately below, we investigate the factors that explain the four different congressional responses generally. In section VB2, we turn to the variables that correlate with Congress’ decision to enact new law either codifying or overriding the Supreme Court decision. To aid our discussion, we highlight features of our findings in the tables below that are notable and have a strong impact on congressional activity.

1.Explaining Four Different Types of Congressional Responses Generally

Table 4 presents our findings with respect to the four categories of congressional responses. We present our results in the form of probabilities calculated from the raw coefficients;[194]this means that the tables below depict how each variable impacts the likelihood of congressional oversight that emerges in the negative, positive, mixed, or neutral form. In the table, a negative (positive) probability indicates a negative (positive) correlation between the particular variable and the type of legislative response under consideration. The table rows show the type of response generated by the Supreme Court opinion, and the columns indicate the particular variable of interest—each individual cell indicates a negative or positive correlation. To establish a baseline for comparison purposes, we first present the average level of oversight in each category. These numbers are depicted in column two and show that that negative comments surface in response to 32% of the cases examined in Congress; positive comment 25% of the time; a mixed collection of comments surface 10% of the time; and neutral comments emerged with respect to 32% of the cases.

We next examine the impact of the various agenda entrepreneurs on congressional oversight, including the Joint Tax Committee Members, lobbyists, journalists, and the justices themselves. We find that each of these groups has a unique impact on legislative oversight. As column 3 indicates, neutral responses are highly correlated with the activities of the Joint Tax Committee—increasing the level of neutral commentary by 43%! This finding confirms that the Committee faithfully undertakes its statutory obligation to report regularly to Congress on the state of the tax law. Column 4 indicates that lobbyists decrease the likelihood of a unidirectional response (solely negative, positive or neutral), but increase the probability that a case will receive mixed and conflicting comments in congressional debates and hearings. This finding suggests that interested actors—both those that support and oppose the Court’s outcome—go to Congress in an effort either to convince legislators to codify the case or override it. The extant literature has long suggested that lobbying activities are correlated with negative congressional activity, but our study indicates that lobbyist can also spark a positive response to the Supreme Court case. And this finding fits with a more general theory of lobbying found in the social science literature: When one party goes to Congress to advocate a particular outcome—an opposing party is likely to show up in order to advocate the opposite position.[195]

With respect to the role of journalists, column 5 indicates that media coverage is negatively correlated with neutral commentary, but has very little impact on the other categories of oversight. This suggest that the Joint Tax Committee is not at all driven by media concerns, but primarily focuses on its substantive goal of offering a comprehensive reports on the various cases to Congress. This is, perhaps, a compliment to the work of the lawyers, economics, and analysts on this committee.

Turning now to the activities of the justices themselves. As depicted in column 6, judicial invitations have a surprising impact on the nature of the oversight. The invitations decrease the likelihood of unidirectional oversight, but increase the likelihood of mixed responses, as well as neutral responses. Like lobbying activities, the judicial invitations appear to spark a response by legislators who both support and oppose the Court’s decision. And just as interesting, the Court appears to garner the attention of the Joint Committee on Taxation and its work on summarizing the law. Apparently when the justices ask for review—Members of Congress respond with a wide range of commentary. This confirms our findings above indicating that when the justices speak directly to Congress—Congress listen.

One of the most intriguing findings of this part of our study is the role of unanimous decisions in sparking solely positive feedback. Unanimous decisions increase the likelihood of a positive response from 25% to 36%, but decrease the likelihood of all other types of responses. This finding suggests that when the Court is able to reach a consensus, the legislators take notice and tend to agree with the Court’s outcome. Scholars have debated the impact, if any, of unanimous decisions on override activity— and for the most part have rejected the idea that unanimity plays an important role in congressional decision-making. Our expanded study, however, confirms that the vote count on cases matters quite a bit to Congress—but only in the context completely ignored in the literature—the cases that spark a positive response.

Table 4: Probability of Different Types of Responses to Supreme Court Tax Cases; highlighted probabilities denote strong correlations between type of response and explanatory variable.

Judicial Decisions as Tax Legislation_html_449418cb

We also examined the impact of the economic, legal, and political variables discussed above in Section IV. In that section we found that when all the different responses were aggregated into one group, economics was strongly correlated with congressional oversight of Supreme Court cases. But in this analysis, where we disaggregated the different forms of oversight—we find to statistically significant correlation.

2.Explaining Different Types of Actual Statutory Enactments

We now turn from the type of congressional response that Supreme Court cases generated to the type of legislative reform that emerged after the publication of the judicial opinion. Overall, Congress adopted 47 legislative reforms in response to the Supreme Court cases since 1954. Table 5 below again presents our statistical findings in the form of probabilities and these numbers were calculated from the raw coefficients that we obtained from our statistical models. As a baseline and for comparison purposes, we found that 47% of these new statutory provisions overruled a Supreme Court case, 38% codified as case, and 10% were neutral in the sense that they related to the case but neither overruled nor codified the outcome. As these numbers suggest, on average there is a greater probability that Congress will adopt legislation modifying or overriding a Court decision than legislation that codifies or impacts the outcome in a neutral fashion.

When it comes to actual legislative enactments, we find the various agenda entrepreneurs have a different impact altogether than that observed with respect to the general commentary discussed above. Lobbying activities have little or no substantive impact on statutory enactments. But media coverage and judicial invitations are both strongly correlated with override activity and at the same time decrease the likelihood of a congressional codification of a Supreme Court case. These findings confirm the results in the existing literature and also lend confirmation to the work of Spiller and Tiller who suggest that the justices invite review in order to obtain a different outcome than that reached in the opinion itself.[196] A unanimous court decision decreases the probability of an override, but increases the likelihood of a codification: an exciting and important result that we discuss further below. Finally, when the economy is doing poorly—neutral legislation increases by 54%. Because we have only five neutral reforms in our database, however, we do not give much weight to this empirical finding.

Table 5: Probability of Different Types of Legislative Reform in response to Supreme Court Tax Cases; highlighted probabilities denote strong correlations between type of response and explanatory variable.

Judicial Decisions as Tax Legislation_html_1c3f7480

3. Summary of Findings on Substantive Activity

Our qualitative and quantitative analyses of congressional responses to Supreme Court decisions highlight new correlations never before considered in the literature. We find that when the justices invite Congress to review a decision and when lobbyists participate in the process—Congress is likely to produce both positive and negative responses (not just negative responses that lead to overrides as the existing literature suggests). Moreover, scholars have long debated whether a unanimous decision has any impact on the congressional decision to react. We find that when the Court unanimously issues a decision, Congress is quite a bit more likely to respond optimistically and positively both as a general matter and with regard to codification of an outcome particularly. This suggests that when the Court is united on an issue—Congress will be similarly united. With respect to override activity, we find that judicial invitations for congressional review are closely associated with this form of statutory enactment. This finding comports with similar findings in the literature.

VI.POSITIVE AND NORMATIVE IMPLICATIONS

Scholars have never attempted a comprehensive and all-inclusive study of the Court-Congress dynamic. Although the extant literature is filled with important contributions to the study of inter-branch relations, every one of the existing studies has focused on a single limited aspect of the interaction—congressional overrides of Supreme Court cases.[197] When Congress chooses to overturn a decision of the Court, it expresses a strong opinion on the merits of the judicial decision, but legislators are not limited to sending negative signals to the judiciary—they can also send positive signals. For purposes of this study, we collected and analyzed the full array of congressional responses to the Supreme Court and made surprising findings that advance our understanding of both Congress and the Court.

Most importantly, we demonstrate that the Court-Congress dynamic is not uni- dimensional but nuanced and varied. The existing literature implies when Congress responds to the Court, it does so in a hostile manner. To be sure, judicial decisions often spark a negative response in Congress, but just as often the cases lead to supportive and positive responses that ultimately produce codifying legislation.[198] We also find that legislators use the Court cases for position-taking purposes in an effort to signal to constituents and other interested parties that they are attending to important and salient matters. These position-taking activities surface in hearings and debates where legislators make forceful comments with respect to a particular case, but never actually submit a legislative initiative that would override, codify, or impact the case in any meaningful manner.[199]

Not only do our findings present a rich and varied picture of inter-branch dynamics, they also indicate that the justices play a key role in setting the legislative agenda. It is easy to see that when Congress adopts a controversial statute, it effectively sets the judicial agenda by virtue of the fact that controversies will make their way into federal courts; indeed, many scholars noted that statutory controversies seem to have monopolized the federal docket.[200] Our study demonstrates that agenda setting can also work in the opposite direction. Of course, when the justices grant certiorari they privilege an issue in national debates and increase the likelihood that legislators will attend to the issue, but the justices have found a device that virtually assures legislators will focus on a specific case in a speedy fashion—they issue a request for legislative review of the opinion itself. This request, as noted above, nearly guarantees the legislators will review the opinion in legislative debates and hearings soon after the opinion is rendered. In short, while many commentators view the Court as a body that follows the lead of the legislature when in comes to statutory interpretation—often Congress takes its direction from the justices. This is particularly true not only when legislators take up the issue found in the Court opinion, but also when they decide to codify the outcome in a manner that re-expresses the Court’s view as their own view of the law.

Finally, our study provides an important challenge to the scholars who have argued that the Supreme Court is particularly incompetent when it comes to certain areas of the law, such as those involving economic issues.[201] Various commentators, for example, have strenuously argued that the justices have no training in complex areas such as taxation—and thus they are not well positioned to decide issues of national importance that involve these areas of the law. Indeed, many have argued that Congress should “rescue federal taxation” from the Supreme Court by a creating specialty court comprised of experts in the field to review lower court opinions and render the final decisions.[202]

Scholars have made similar arguments in various other areas of the law with equal force.[203] Our study suggests that these arguments go too far, and may be altogether unwarranted. We find that legislators routinely review the Court’s decision-making, thereby assuring the justices do not bungle the difficult cases that they decide. More importantly, congressional responses to the Supreme Court tax cases suggest the tax bar has seriously underestimated the competency and skill of the Supreme Court. Congress frequently expresses approval of the Court’s decision-making via a codification of the outcome, demonstrating the justices are far more talented and gifted than ever before acknowledged in this area of the law.

APPENDIX

The table below offers summary statistics and coding information regarding the independent variables in our models.

Table A1 Summary of variables

 

Variable

Mean

(Std. Dev)

Minimum

Maximum

Media

Coverage

.051

(.220)

0

1

Budget Surplus

(in thousands of 1996 dollars)

-87,831

(112,995)

-432,509

219,500

Amicus Briefs

Filed

.425

0

1

Invitation for Congressional

Response

.109 (.495)

0

1

Corporate

Taxpayer

.355

0

1

Individual

Taxpayer

.397

0

1

Taxpayer

Prevailed in Case

.296

0

1

Unanimous Court Decision

.383

0

1

Ideological Distance

Between President and Court

.524

.040

.908

Ideological Distance

Between House and Court

.138

(.091)

.002

.338

Ideological Distance

Between Senate and Court

.113 (.082)

.008

.319

Ideological Distance Between Majority Party in House and Court

.349 (.155)

.019

.716

Ideological Distance

Between Majority Party in

Senate and Court

.358 (.154)

.022

.716

Ideological Distance

Between House Ways and

Means Committee and Court

.357

(.156)

.000

.854

Ideological Distance Between Senate Finance Committee and Court

.277 (.159)

.000

.653

 

 

[*]Nancy Staudt is the Class of 1940 Research Professor of Law at Northwestern University School of Law; René Lindstaedt is an Assistant Professor of Political Economy at Stony Brook University; and Jason O’Oconnor is an Economics Ph.D. student at Northwestern University. We thank many scholars who have commented on work, including Kathie Barnes, Randy Calvert, Charlotte Crane, Lee Epstein, Jason Johnston, Pauline Kim, Andrew Martin, Bob Pollak, Margo Schlanger, Emerson Tiller, Peter Wiedenbeck, Bob Wooten, Kathy Zeiler and participants in faculty workshops at Florida State, Georgetown, Northwestern, University of North Carolina, University of Pennsylvania, and Washington University Law Schools for helpful thoughts and comments. Please send comments to n-staudt@northwestern.edu.

[*]Nancy Staudt is the Class of 1940 Research Professor of Law at Northwestern University School of Law; René Lindstaedt is an Assistant Professor of Political Economy at Stony Brook University; and Jason O’Oconnor isan Economics Ph.D. student at Northwestern University. We thank Kathie Barnes, Randy Calvert, Charlotte Crane, Lee Epstein, Jason Johnston, Pauline Kim, Andrew Martin, Bob Pollak, Margo Schlanger, Emerson Tiller,Peter Wiedenbeck, Bob Wooten, Kathy Zeiler and the participants in the faculty workshops at Florida State, Georgetown, Northwestern, University of North Carolina, University of Pennsylvania, and Washington University Law Schools for helpful thoughts and comments. Please send comments to n-staudt@northwestern.edu.

[1]See, e.g., Peter L. Strauss, The President and the Choices Not to Enforce, 63 LAW AND CONTEMP. PROBS. 107, 119 (2000) (noting subordination of one branch by another is difficult to justify in a government of co-equal branches”); Charles Gardner Gehy & Emily Field Van Tassel, The Independence of the Judicial Branch in the New Republic, 74 CHI.-KENT L. REV. 31, 47 (1998) (noting that members of Constitutional Convention considered and assumed the three branches of government were co-equal actors in our system of separated powers); see also U.S. CONST. ART. I, II, & III (framers created three different but equally important, and apparently co-equal, branches of government).

[2]A vast literature exists that examines the congressional inclination to overrule Supreme Court cases. See generally, JEB BARNES, OVERRULED?: LEGISLATIVE OVERRIDES, PLURALISM, AND CONTEMPORARY COURT-CONGRESS RELATIONS (2004); COLTON C. CAMPBELL & JOHN F. STACK, JR., CONGRESS CONFRONTS THE COURT (2001); ROBERT KATZMANN, COURTS & CONGRESS (1997); JOHN R. SCHMIDHAUSER & LARRY L. BERG, THE SUPREME COURT AND CONGRESS (1972); WALTER F. MURPHY, CONGRESS AND THE COURT (1962); C. HERMAN PRITCHETT, CONGRESS VERSUS THE SUPREME COURT (1961); Pablo Spiller & Emerson Tiller, Invitations toOverride: Congressional Reversals of Supreme Court Decisions, 16 INT’L REV . OF LAW & ECON. 503 (1996); Joseph Ignagni & James Meernik, Explaining Congressional Attempts to Reverse Supreme Court Decisions, 47 POL. RES. Q. 353 (1994); William N. Eskridge, Overriding Supreme Court Statutory Interpretation Decisions, 101 YALE L.J. 331 (1991); Abner Mikva & Jeff Bleich, When Congress Overrules the Court, 79 CAL. L. REV. 729, 729-30 (1991); but see GEORGE I. LOVELL, LEGISLATIVE DEFERRALS: STATUTORY AMBIGUITY, JUDICIAL POWER, AND AMERICAN DEMOCRACY1-42 (2003) (arguing Congress intentionally adopts ambiguous statutes in an effort to defer to the Court on difficult issues).

[3]Seeinfranotes153-76 andaccompanyingtext.

[4]Seeinfranotes177-88andaccompanyingtext.

[5]Seeinfranotes177-88andaccompanyingtext.

[6]Seeinfranoted27-31andaccompanyingtext.

[7]Seesupranote2andcitationstherein.

[8]To take just one example: Theorists hypothesize that when Congress is ideologically distant from the Court, legislators are more likely to overturn a judicial decision. See, e.g, Virginia Hettinger & Christopher Zorn,Explaining the Incidence and Timing of Congressional Responses to the U.S. Supreme Court, 30 LEGIS. STUD . Q. 5 (2005). Although this claim (and others) may turn out to be true, it takes on a different meaning if Congress also codifies Court decisions in identical political circumstances. The meaning and the implication of the scholarly findings, in short, may need reconsideration in light of the full range of congressional activity in response to Court decisions.

[9]Seeinfranotes27-31andaccompanyingtext.

[10]Seeinfranotes34-75andaccompanyingtext.

[11]Seeinfranotes189-94andaccompanyingtext.

[12] Scholars have a range of views on the judicial role when it comes to interpreting statutes, but none argue the justices should completely ignore the preferences or intent of Congress. See e.g., Cass Sunstein, JusticeScalia’s Democratic Formalism, 107 YALE LAW J. 529, 532 (1998) (the goal of any system of interpretation is to constrain judicial discretion); William N. Eskridge, Textiualism, the Unkown Ideal?, 96 MICH. LAW REV. 1509 (1998) (judges must exercise humility in interpreting statutes and while they should be part critic, they must also be part agent to Congress); Martin Redish & Theodore T. Chung, Democratic Theory and the LegislativeProcess: Mourning the Death of Originalism in Statutory Interpretation, 68

TULANE L. REV. 803, 805 (1994) (originalist interpretive models view “judge’s role as limited to deciphering the these commands and applying them to particular cases”).

[13] See, e.g., Gary W. Carter, The Commissioner’s Nonaquiescence: A Case for a National Court of Appeals, 59 TEMPLE L.Q. 879 (1986) (arguing problems unique to taxation should lead Congress to pursue legislation authorizing National Court of Tax Appeals); Charles L.B. Lowndes, Federal Taxation and the Supreme Court, 1960 SUP. CT. REV. 222, 222 (“The thesis of this paper is simple: It is time to rescue the Supreme Court from federal taxation; it is time to rescue federal taxation from the Supreme Court); Oscar Bland, Federal Tax Appeals, 25 COLUM. L. REV. 1013 (1925) (criticizing the federal tax litigation process as “clumsy and time-consuming” and arguing for a single court with national jurisdiction over all tax disputes).

[14] See, e.g., James Meernik & Joseph Ignagni, Congressional Attempts on Supreme Court Rulings Involving Unconstitutional State Laws, 48 POL. RES. Q. 43, 45 (1995) (examining all cases in which the Supreme Court struck down legislation during the Warren, Burger, and early Rehnquist Court years); Harry P. Stump, Congressional Response to Supreme court Rulings: The Interaction of Law and Politics, 14 J. PUB. L. 377, 385-391 (1965) (identifying individual issue areas but aggregating cases for purposes of analysis).

[15] Scholars, of course, acknowledge the role of many factors in congressional policy making, but tend to focus on ideology in the civil rights context and interest groups in the economic contexts. See, e.g., Eskridge,supra note 2 at 302-390 (investigating political dynamic that exists in civil rights context between the Supreme Court, Congress, and the President); DOUGLAS ARNOLD, TH E LOGIC OF CONGRESSIONAL ACTION 149 (exploring roles of economic factors and interest groups in the making of economic and tax policy); E. SCOTT ADLER, WHY CONGRESSIONAL REFORMS FAIL: REELECTION AND THE HOUSE COMMITTEE SYSTEM, 149, tbl. 101 (2002) (tax-writing committees subject to interest group capture); Daniel Shaviro, Beyond Public Choice and Public Interest: A Study of the Legislative Process as Illustrated by Tax Legislation in the 1980s, 139 U. PA. L. REV. 1, 86-87 (1990) (tax law has been dominated “by interested groups that seek favors for themselves and that through a norm of logrolling, almost never oppose favors from each other”).

[16] See Roger H. Davidson, Walter J. Oleszek & Thomas Kephart, One Bill, Many Committees:

Multiple Referrals in the U.S. House of Representatives, 13 LEG. STUD. Q. 3, 21-22 (1988) (investigation of bill referral process); DAVID KING, TURF WARS: HOW CONGRESSIONAL COMMITTEES CLAIM JURISDICTION 2, 102, 113 (1997) (investigation of committee jurisdiction system).

[17] Mark Miller, Courts, Agencies, and Congressional Committees A Neo-Institutional Perspective, 55 REV. POL. 471, 484 (1993).

[18] Id.

[19] See Beth M. Henschen & Edward I. Sidlow, The Supreme Court and the Congressional Agenda Setting Process, 5 J. LA W & POL. 685 (1989); see also Beth Henshen, Statutory Interpretation of the Supreme Court:Congressional Response, 11 AM. POL. Q. 441 (1983).

[20] Henshen & Sidlow, supra note 19, at 719-24 (exploring role that Supreme Court plays in setting legislative agenda and noting roles of lobbyist the media); Henshen, supra note 19, at 441 (offering a variety of hypotheses on what attracts legislative attention to labor and anti-trust cases but offering no empirical investigation).

[21] Joseph Ignagni, James Meernick & Kim Lynn King, Statutory Constructions and Congressional Response, 26 AM. POL. Q. 459 (1998).

[22] U.S.CONGRESS, THE SENATE OF THE UNITED STATES COMM.AND SUBCOMM. ASSIGNMENTS, 108TH CONG. (2003) (describing jurisdiction of the Judiciary Committee).

[23] Eskridge, supra note 2, at 353-404 (exploring political factors that leads Congress to respond to Supreme Court); William N. Eskridge, Jr., Reneging on History?: Playing the Court/Congress/President Civil RightsGame, 79 CAL. L. REV. 613 (1991); Pablo Spiller & Emerson Tiller, Invitations to Override: Congressional Reversals of Supreme Court Decisions, 16 INT’L REV . LA W & ECON. 503 (1996) (theorizing on the role of judicial invitations in legislative decisions to override Supreme Court cases).

[24] Eskridge, supra note 2, at 390-95 (theory tested through qualitative analysis of cases in three different eras); Eskridge, supra note 23, at 617-42 (same); Spiller & Tiller, supra note 23, at 511-19 (theory tested by examining two Supreme Court cases). Hettinger and Zorn investigate the public choice theories set forth in Eskridge’s and Spiller and Tiller’s work and thus avoid the problem of aggregation and at the same time offer a quantitative analysis of the data. But they ignore a substantial amount of oversight activity by focusing exclusively on overrides of Supreme Court cases. See Virginia Hettinger & Christopher Zorn, Explaining the Incidence and Timing of Congressional Responses to the U.S. Supreme Court, 30 LEGIS. STUD . Q. 5 (2005) (limiting study to civil rights cases decided between 1967-1989).

[25] The Ways and Means Committee controls tax issues in the House and the Finance Committee controls taxation issues in the Senate. See Nancy Staudt, Redundant Tax and Spending Programs, 100 NW. L. REV. 1197 (2006); see also data depicted, infra notes 26-28 text accompanying notes 30-37; Note, 71

HARV. L. REV . 1324 & n 3 (1957) (noting “rapid interplay” between Court and Congress in taxation context).

[26] This number was computed from HAROLD J. SPAETH, U.S. SUPREME COURT DATA BASE

(December 9, 2004 release) with the following code (in Stata): generate civrights=1 if (analu==0) & (dec type==1 — dec type==6 — dec type==7) & (authdec1== 4 — authdec2==4) & (value==2) replace civrights=0 if civrights==.generate economic=1 if (analu==0) & (dec type==1 — dec type==6 — dec type==7) & (authdec1==4 — authdec2==4) & (value==8) replace economic=0 if economic==.

[27] Computed from HAROLD J. SPAETH , U.S. SUPREME COURT DATA BASE (October 14, 2004 release) using the law variable if (in Stata): (analu==0 — analu==3 — analu==5) & (dec type==1 — dec type==6 — dec type==7) & (term ¿1 985 & term != 2003).

[28] We plan to study Supreme Court decision-making in future articles. For a preliminary study of Supreme Court decision-making in the tax context, see Nancy Staudt, Lee Epstein & Peter Wiedenbeck, The IdeologicalComponent of Taxation (forthcoming publication in WASH. L. REV. (2007)); Nancy Staudt, Lee Epstein, Peter Wiedenbeck, René Lindstädt & Ryan J. Vander Wielen, Judging Tax Statutes and Interpretive Regimes, 38 LOY. L.A. L. REV. 1909 (2005); Nancy Staudt, Agenda Setting in Supreme Court Tax Cases: Lessons from the Blackmun Papers, 52 BUFF. L. REV. 889-922 (2005).

[29] See STEVEN S. SMITH & CHRISTOPHER DEERING, COMMITTEES IN CONGRESS 10 (1990)

(identifying tax-writing committees as “power committees); GARY COX & MATHEW D. MCCUBBINS, LEGISLATIVE LEVIATHAN: PARTY GOVERNMENT IN THE HOUSE 199-200 (1993) (noting the importance— and thus power—of the tax-writing committees in Congress).

[30] There are two reasons that we focused on this time period. First, Congress implemented major changes to the tax laws in 1954 but since that time the structure of the tax code has remained relatively stable. Seegenerally, W. ELLIOT BROWNLEE, FEDERAL TA XATION IN AMERICA: A SHORT HISTORY (1996). Second, the electronic databases made available through Westlaw and Lexis include in-depth documentation of congressional activities in the tax context going back to 1954, but prior to that time coverage is erratic.

For our data collection process, we first identified every taxation case in the U.S. Supreme Court and then reviewed each case produced by the search, retaining only those cases that involved an interpretation of a federal tax statute. Our unit of analysis is the case citation and we found 284 cases. If we use docket number as our unit of analysis, the figure increases to 554 (many cases are consolidated). We included only orally argued cases that resulted in a per curiam judgment or an opinion of the Court.

To reproduce our search, go to the Supreme Court library on Lexis and use this search: (federal w/s tax!) or (excise w/s tax!) or (estate w/s tax!) or (user w/5 fee) or (user w/s tax!) or (tax! w/s fraud) or (irc) or (i.r.c.) or (stamp w/s tax!) or (income w/s tax!) or (internal w/s revenue) or (tax! w/s lien) or (tax! w/s code) or (tax! w/s evad!) or (tax! w/s evasion) or (corporate w/s tax!) or (payroll w/s tax!) or (employment w/s tax!) or (social w/s security) or (26 usc) or (26 u.s.c.) or (tax! w/s refund) or (tax! w/s deficiency) or (unemployment w/s tax!) or (gift w/s tax!) or (fica w/s tax!) or (f.i.c.a. w/s tax!).

[31] This entailed a Lexis search of House and Senate documents using these Lexis libraries: House and Senate Documents, Congressional Full Text Bills, Congressional Rules and Parliamentary Procedures Committee Prints, Congressional Record, All Congresses Combined, Legislative Histories; CIS/Historical Index; Tax Legislative Histories. We searched for any reference to the Supreme Court tax case (either by name or citation—and we checked for misspelled entries) and also searched by subject matter but only included a congressional response in our database if the case was explicitly referenced by name or cite. We collected only the hearings, debates, reports, speeches, legislative proposals, and so forth that referred to the specific tax case and excluded any documents that discussed the issue in the case but did not refer to a particular Court decision. Although this approach is under-inclusive in the sense that it nearly guarantees that we overlooked legislative responses to some Court decisions, it also assures that our database contains only the instances in which legislators intended to discuss and respond to the Supreme Court.

[32] These Court-Congress interactions have emerge when taxpayers believe Congress has imposed an unconstitutional economic burden, such as customs tax litigated in United States v. United States Shoe Corp., 530 U.S. 360 (1998) and the social security tax imposed on federal judges and challenged in United States. v. Hatter, 532 U.S. 557 (2001). Much more often, however, the cases that stir activity in our system of checks and balances involve narrow and technical tax issues, see United States v. Fior D’Italia , 536 U.S. 238 (2002) (the proper method for assessing taxes on employee tips), Gitlitz v. Comm’r, 531 U.S. 206 (2001) (the correct algorithm for determining basis in S Corporation stock), O’Gilvie v. United States, 519 U.S. 79 (1996) (inclusionary rules governing punitive damages received in tort suits). All these issues involve statutes adopted by Congress, interpreted by federal courts, and then reconsidered in legislative hearings, debates, and reports.

[33] Congress established the Joint Tax Committee under the Internal Revenue Code of 1926. I.R.C. §§ 8001-8005, 8021-8023 (2006). Ten legislators sit on the committee: 5 Members from the Senate Committee on Finance (3 majority and 2 minority) and 5 Members from the House Committee on Ways and Means (3 majority and 2 minority). Id. The committee’s statutorily prescribed duties include the obligation to investigate and report on: (1) the operation and effects of internal revenue taxes and the administration of such taxes; and (2) measures and methods for the simplification of such taxes. I.R.C. § 6405 (2006).

[34] These numbers sum to greater than 52% because several of the cases were subject to multiple audits. In fact, legislators proposed both a codification and an override of some cases. See discussion of the substance of the oversight, infra notes 133-96 and accompanying text.

[35] See generally JOHN W. KINGDON , AGENDAS, ALTERNATIVES, AND PUBLIC POLICIES (1984) (investigating factors that impact legislative agenda); ARNOLD, supra note 15, at 149-223 (same); George C. Edwards III & B. Dan Wood, Who Influences Whom? The President, Congress, and the Media, 93 AM. POL. SCI. REV. 327, 327 (1999) (same).

[36] Various scholars interested in the oversight process, however, offer specific hypotheses as to why Supreme Court decisions become salient in and around the Capitol. See Henshen & Sidlow, supra note 19, at 719-24 (exploring role of the President, interest groups, and the media in making Supreme Court cases salient in legislative debates); Michael E. Solimine & James L. Walker, The Next Word: Congressional Response toSupreme Court Statutory Decisions, 65 TEMP. L. REV. 425, 444 (1992) (suggesting case- related factors such as unanimous decisions and modes of analysis increase the salience of a Supreme Court case); Ignagni et al,supra note 21, at 348-60 (hypothesizing that public opinion and elections increase salience of Supreme Court decisions in congressional debates).

[37] KINGDON, supra note 35, at 41-43 (discussing legislative incentives when setting the agenda); Jack Walker, Setting the Agenda in the U.S. Senate: A Theory of Problem Selection, 7 BRIT. J. POL. SCI. 423, 433-45 (1977) (same).

[38] See Nancy C. Staudt, Redundant Tax and Spending Programs, 100 NW. U. L. REV. 1197, 1204- 05 (2006) (describing legislative committee system).

[39] Id.

[40] DAVID C. KING , TURF WARS: HOW CONGRESSIONAL COMMITTEES CLAIM JURISDICTION 105-20 (1997) (committees fight to retain politically salient issues and seek to make issues in their own jurisdictions salient); Staudt, supra note 38, at 1214-22 (exploring importance of committee jurisdictions for political maneuvering and noting legislators fight to protect jurisdictional borders).

[41] The case Newark Morning Ledger Co. v. United States, 507 U.S. 546 (1993), for example, involved a newspaper that sought to deduct the cost of customer lists that it obtained in a merger. The government objected to the deduction arguing it would cost the government close to a billion dollars in revenue. See Respondent’s Brief, Newark (No. 91-1135).

[42] Eskridge, supra note 2, at 361(discussing importance of interest groups in oversight process); see also Bryan D. Jones, Frank R. Baumgartner & Jeffrey C. Talbert, The Destruction of Issue Monopolies in Congress, 87 AM. POL. SCI. REV. 657, 960 (1993) (interest groups that testify at hearings are often aligned with the interests of the legislators).

[43] Scholars interested in the Court-Congress dynamic have looked to amicus briefs as a measure of interest group pressure on both the Court and on Congress. See Hettinger & Zorn, supra note 24, at 5 (using number of amicus briefs as proxy for level of interest group pressure on Congress); Ignagni et al supra 21 note, at 452-53 (same); Gregory Calderia & John R. Wright, Organized Interests and Agenda

Setting in the U.S. Supreme Court, 82 AM. POL. SCI. REV. 1109 (1988) (exploring impact of interest groups via amicus brief filings in Supreme Court decision-making).

[44] 441 U.S. 772 (1981).

[45] For a list of the twelve amicus curiae briefs, go to http://web2.westlaw.com/search/default.wl?docsample=False&rltdb=CLID_DB50013268&srch=TRUE&eq=search&dups=false&scxt=WL&sv=Split&blinkedcitelist=False&fn=_top&action=EditQuery&rs=WLW6.08&cxt=RL&method=ConcordTemplate&fcl=True&origin=Search&db=SCTBRIEFALL&mt=LawSchoolPractitioner&service=Search&qtrcc=QueryTemplate&qttab=QT_BRIEFBANK_BRIEF_NO_STATE&vr=2.0&rp=%2fsearch%2fdefault.wl&cd=y&query=(St.+Martin+Evangelical+Lutheran+and+South+Dakota)&cnt=DOC&cfid=1&tempinfo=SCT-BRIEF- ALL%7cTEMPLATE%7cKeyWord%3aInTxt%3dSt.+Martin+Evangelical+Lutheran+and+South+Dakota%7c&ssrc=10748&ss=CXT (last visited August 26, 2006).

[46] 423U.S.360(1998).

[47] For a list of the thirteen amicus curiae briefs, go to http://web2.westlaw.com/search/default.wl?docsample=False&rltdb=CLID_DB50013268&srch=TRUE&eq=search&dups=false&scxt=WL&sv=Split&blinkedcitelist=False&fn=_top&action=EditQuery&rs=WLW6.08&cxt=RL&method=ConcordTemplate&fcl=True&origin=Search&db=SCTBRIEFALL&mt=LawSchoolPractitioner&service=Search&qtrcc=QueryTemplate&qttab=QT_BRIEFBANK_BRIEF_NO_STATE&vr=2.0&rp=%2fsearch%2fdefault.wl&cd=y&query=(St.+Martin+Evangelical+Lutheran+and+South+Dakota)&cnt=DOC&cfid=1&tempinfo=SCT-BRIEF- ALL%7cTEMPLATE%7cKeyWord%3aInTxt%3dSt.+Martin+Evangelical+Lutheran+and+South+Dakota%7c&ssrc=10748&ss=CXT (last visited August 26, 2006).

[48] 507 U.S. 546 (1993).

[49] For a list of the 21 amicus curiae briefs, go to http://web2.westlaw.com/search/default.wl?docsample=False&rltdb=CLID_DB50013268&srch=TRUE&eq=search&dups=false&scxt=WL&sv=Split&blinkedcitelist=False&fn=_top&action=EditQuery&rs=WLW6.08&cxt=RL&method=ConcordTemplate&fcl=True&origin=Search&db=SCTBRIEFALL&mt=LawSchoolPractitioner&service=Search&qtrcc=QueryTemplate&qttab=QT_BRIEFBANK_BRIEF_NO_STATE&vr=2.0&rp=%2fsearch%2fdefault.wl&cd=y&query=TI(united+%2f3+states+%26+unite d+%2f3+states+%2f3+shoe)&cnt=DOC&cfid=1&tempinfo=SCT-BRIEF-

ALL%7cTEMPLATE%7cParty%3aParty1%3dUnited+States%7cParty%3aParty2%3dUnited+States+Shoe%7c&ssrc=19&ss=CXT (last visited August 26, 2006).

[50] 532 U.S. 822 (2001). See Brief for the National Association of Manufacturers and the

Manufacturers Alliance/MAPI Inc. as Amicus Curiae Supporting Petitioner (No. 00-157).

[51] 532 U.S. 200 (2001). See Brief Amicus Curiae of the Major League Baseball Players Association in Support of Respondent (No. 00-203).

[52] KINGDON, supra note 35, at 61-64 (study of media suggests that salient issues do not originate with journalists but journalistic coverage accelerates its salience in national policy debates).

[53] Id. See also Lee Epstein & Jeffrey A. Segal, Measuring Issue Salience, 44 AM. J. POL. SCI. 66 (2000) (exploring media coverage as a useful measure of issue salience for elite actors).

[54] See Nancy Staudt, Agenda Setting in Supreme Court Tax Cases: Lessons from the Blackmun

Papers, 52 BUFF. LA W REV. 889 (2005) (citing memorandum in the Blackmun papers that indicate justices view tax cases as the “crud” and the types of issues that “put law clerks to sleep.”); Neil M. Richards, TheSupreme Court Justices and “Boring” Cases, 4 GREEN BAG 2d 401, 401-408 (2001) (article commenting on the fact that the Justices find tax cases boring).

[55] We collected data on the media coverage in the New York Times, Wall Street Journal, L. A.

Times, and Chicago Tribune within five days of the decision date.

[56] 485 U.S. 505 (1988).

[57] Stuart Taylor, Jr., Congress Can Tax Municipal Bonds, High Court Rules, N.Y. TIMES,April 20, 1988, at A6.

[58] Id.

[59] Id.

[60] See, e.g., S. Res. 463, 100TH CONG ., 134 CONG . REC. 11672,. (1988) (“Resolved, That it is the sense of the Senate that Federal laws regarding the taxation of State and local bonds should be not be changed in order to increase Federal revenues,” notwithstanding the Supreme Court’s decision in South Carolina v. Baker ruling that Congress may impose such a tax); H.R. CONG . RES. 42, 101ST CONG., 135 CONG . REC. 7909, (1989) (“Resolved by the House of Representatives (the Senate Concurring) that we hereby memorialize that the Congress of the United States to take steps to safeguard the tax-free status of state and local bonds.”).

[61] 420 U.S. 141 (1974) (IRS authority under I.R.C. §§ 6601 and 7602 to issue “John Doe” summons to bank or other depository to discover the identity of a person who has had bank transactions suggesting possibility of liability for unpaid taxes sustained).

[62] Warren Weaver, High Court Grants I.R.S. Wide Access to Bank Files, N.Y. Times, Feb. 20,

1975, at A1.

[63] See Tax Reform Act of 1976, Pub. L. No. 94-455, § 1205 (modifying third party summons rules applicable to banks and other parties).

[64] We created a dichotomous variable and coded it as 1 if coverage occurred and as 0 if not. As expected the Wall Street Journal’s coverage was far more extensive than that found in the other newspapers. Our data indicates that the Wall Street Journal published articles on 38% of the cases, the New York Times reported on 8%, the L.A. Times covered 15%, and the Chicago Tribune reported on 12% of the decisions decided between 1954-2004.

[65] 408 U.S. 125 (1972).

[66] Id. at 127-28.

[67] Id. at 143-51.

[68] Id. at 135.

[69] 26 U.S.C. 2036(b).

[70] 465 U.S.C. 330 (1984).

[71] Id.

[72] Id. at 345-53 (describing the majority decision as “ill-advised and inequitable,” and arguing “the Court’s decision today rejects a longstanding principle of taxation, and creates in its stead a new and anomalous rule of law. Such action is best left to Congress.”).

[73] See Deficit Reduction Act of 1984, Pub. L. No. 98-269, § 4609 (codifying decision to allow taxation of benefits obtained from loan with below-interest terms or on gift terms).

[74] Eskridge, supra note 2, at 388-89.

[75] Spiller & Tiller, supra note 23, at 507-19; see also Lori Hausegger & Lawrence Baum, Inviting Congressional Action: A Study of Supreme Court Motivations in Statutory Interpretation, 43 AM. J. POL. SCI. 162 (1999) (empirical exploration of various rationales for judicial invitations).

[76] We do find this to be an interesting and important feature of the oversight process, however, and continue to investigate the rational underlying judicial invitations to review. See Nancy Staudt & Emerson Tiller,Judicial Invitations for Review (work-in-progress).

[77] See Hettinger & Zorn, supra note 24, at 5 (hypothesizing Supreme Court voting patterns a role in the oversight process); Ignagni et al, supra note 21, at 362 (same).

[78] The empirical findings in the literature are as distinct as the theories presented. Marshall,

Eskridge, Solmine &Walker, and Hettinger & Zorn investigate statutory decision-making and find that a unanimous Court is more likely to avoid subsequent judicial overrides than a divided Court. Eskridge, supra note 2, at 388-89; Solimine & Walker, supra note 36, at 444; Hettinger & Zorn, supra note 24, at 5 Ignagni et al, however, find just the opposite: Congress is more likely to override unanimous decisions given that these decisions are more of a threat to Congress as a political institution vying for power in inter- branch conflicts. Ignagni et al, supra note 21, at 362.

[79] See James M. Verdier, The President, Congress, and Tax Reform: Patterns Over Three Decades, 499 ANNALS AM. ACAD. POL. & SOC. SCI. 114, 117-18 (1988) (noting economic forces have helped to place tax reform onto the legislative agenda but have not impact the substance of the reform); KINGDON, supra note 35, at 113-15 (noting the state of the economy places budgetary issues on the agenda).

[80] Verdier, supra note, at 79, 117-18.

[81] Economic Recovery Tax Act of 1981, Pub. L. No. 97-34, 95 Stat. 172 (1981).

[82] Balanced Budget and Emergency Deficit Control Act of 1985, Pub. L. No. 99-177, 99 Stat. 1937 (1985).

[83] Staudt, supra note 54, at 911-13 (exploring Supreme Court certiorari memoranda and noting that amount of money at issue plays a role in determining importance of case); see also Sup. Ct. R. 10 (outlining standards for making certiorari decisions and indicating case must have important and national implications).

[84] Verdier, supra note 79, at 117-19; see also JOINT ECONOMIC COMMITTEE, CONSTANT CHANGE: A HISTORY OF FEDERAL TAXES (Comm. Print Sept. 12, 2003) (discussing divergent approaches taxation and economic stimulation); JOINT ECONOMIC COMMITTEE, DEFICITS, TAXATION , AND SPENDING (Comm. Print April 2003) (same).

[85] Verdier, supra note 79, at 117-19.

[86] 485 U.S. 212 (1988).

[87] See generally J. DWIGHT EVANS, A CRITICAL ANALYSIS OF TA XATION OF BUSINESS HEDGING AND THE CASE FOR COMPREHENSIVE CONGRESSIONAL LEGISLATION 1-13 (1994) (Tax Foundation Background Paper #7).

[88] 138 Cong. Rec. S15567 (Sept. 19, 1992) (statement of Senator Dole) (“the Supreme Court’s decision in Arkansas Best Corp. versus Commissioner has caused serious disruptions in the ability of farmers, ranchers, and other businesses both in agriculture and elsewhere in our economy to use hedging transactions to reduce the risks of doing business.”).

[89] See generally HAROLD J. SPAETH & JEFFREY A. SEGAL, MA JORITY RULE OR MINORITY WILL:

ADHERENCE TO PRECEDENT ON THE U.S. SUPREME COURT (1999) (justices are ideologically driven in the decision-making process); JEFFREY A. SEGAL & HAROLD J. SPAETH , THE SUPREME COURT AND THE ATTITUDINAL MODEL REVISITED (2002) (same); Jeffrey A. Segal & Chad Westerland, The Supreme Court, Congress, and Judicial Review, 83 N.C. L. REV . 1324 (2005) (same); Jeffrey A. Segal, Separation of Powers Games in the Positive Theory of Congress and the Courts, 91 AM. POL. SCI. REV. 28 (1997) (justices make decisions according to their own political and ideological preferences regardless of competing or conflicting congressional viewpoints).

[90] See generally, Kenneth Shepsle & Barry R. Weingast, The Institutional Foundations of Committee Power, 81 AM. POL. SCI. REV. 85 (1987) (exploring legislative decision-making and noting role of committees in effectuating political preferences); Hettinger & Zorn, supra note 24, at 4 (“in every instance, existing separation-of-powers models predict that, in an instance where Congress is unhappy with a decision of the Court, it will always act to overturn the offending decision.”)

[91] Although we hypothesize that political preferences will play a role in legislative oversight of Supreme Court, we also expect that capturing this dynamic will be a difficult task. For a discussion of the problems exploring the ideological component of taxation, see Nancy Staudt, Lee Epstein & Peter Wiedenbeck, The Ideological Component of Judging in the Taxation Context (forthcoming publication in WASH . L. REV. (2007)).

[92] For a discussion of these scores, see KEITH T . POOLE & HOWARD ROSENTHAL, CONGRESS: A POLITICAL-ECONOMIC HISTORY OF ROLL-CALL VOTING (1997); Keith T . Poole, Estimating a Basic Space From a Set of IssueScales, 42 AM. J. POL. SCI. 954 (1998); to review the actual data, see Keith Poole, Common Space Scores Congresses 75-108, Mar. 9, 2005, available at http://voteview.com/basic.htm.

[93] Lee Epstein, Andrew Martin & Jeff Segal, The Judicial Common Space (forthcoming publication in J. L. ECON . & ORG. (2006)).

[94] See Kenneth A. Shepsle & Barry R. Weingast, POSITIVE TH EORIES OF CONGRESSIONAL INSITUTIONS, 19 LEGIS. STUD . Q. 149, 158-74 (1994) (describing modern political science theories of congressional decision-making power); KEITH KREHBEIL, INFORMATION AND LEGISLATIVE ORGANIZATION 49-52 (1991) (arguing chamber majority is the important body for policy-making purposes); GARY COX & MATHEW D. MCCUBBINS, LEGISLATIVE LEVIATHAN: PARTY GOVERNMENT IN THE HOUSE 83-134 (1993) (arguing party in control of chamber is key decision-making body); Barry R. Weingast & William J. Marshall, The Industrial Organization ofCongress; or, Why Legislatures, Like Firms, are Not Organized as Markets, 96 J. POL. ECON. 132, 142 (1988) (arguing power lies with congressional committees).

[95] For a discussion of the three competing models of power in Congress, see FOREST MALTZMAN, COMPETING PRINCIPALS: COMMITTEES, PARTIES, AND THE ORGANIZATION OF CONGRESS 9-32 (1997).

[96] This is the distance between the Senate majority party in 2003 and Supreme Court in 1969. This distance is relevant because a Court decision rendered in 1969 could be reviewed and acted upon by Congress in 2003 (or any other year post-1969). Keith Poole, Common Space Scores Congresses 75-108, Mar. 9, 2005, available at http://voteview.com/basic.htm.

[97] This is the distance between House Ways and Means Committee in 2004 and the Supreme Court in 1956. Keith Poole, Common Space Scores Congresses 75-108, Mar. 9, 2005, available athttp://voteview.com/basic.htm.

[98] See Keith Poole, Common Space Scores Congresses 75-108, Mar. 9, 2005, available at http://voteview.com/basic.htm (for the congressional scores ) and http://epstein.wustl.edu/research/JCS.html (for the judicial scores).

[99] J. COMM. ON TAXATION , 104TH CONG ., BACKGROUND AND ISSUES RELATING TO TAXATION OF U.S. CITIZENS WHO RELINQUISH CITIZENSHIP, Sec. III (Comm. Print Mar. 20, 1995) (citing to Comm’r v. Glenshaw Glass, 348 U.S. 426 and General American Investors v. Comm’r, 348 U.S. 434).

[100] J. COMM. ON TAXATION , 105TH CONG ., DESCRIPTION AND ANALYSIS OF CERTAIN REVENUE-RA ISING PROVISIONS CONTAINED IN THE PRESIDENT’S FISCAL YEAR 1998 BUDGET PROPOSAL (Comm. Print Mar. 11, 1997) (citing to Comm’r v. P.G. Lake, Inc., 356 U.S. 260 and Comm’r v. Gillette Motor Transp., 364 U.S. 130).

[101] See, e.g., J. COMM. ON TAXATION , 99TH CONG ., DESCRIPTION OF BILLS RELATING TO THE TAX TREATMENT OF MORTGAGE RELATED AND OTHER ASSET BACKED SECURITIES (S. 1959 AND S. 1978) AND ENVIRONMENTAL ZONES (S. 1839)(Comm. Print Jan. 30, 1986) (citing to Comm’r v. National Alfalfa Dehydrating & Milling Co., 417 U.S. 134 (1974)); S. Res. 99-313, 99TH CONG. 2nd(1986) (citing Comm’r v. Idaho PowerCo., 418 U.S. 001 (1974)).

[102] RevenueActof1964,Pub.L.No.88-272,78Stat.99(Feb.26,1964).

[103] TaxReformActof1969,Pub.L.No.91-172,83Stat.487(Dec.30,1979).

[104] EconomicRecoveryTaxActof1981,Pub.L.No.97-34,95Stat.172(Aug.13,1981).

[105] TaxReformActof1986,Pub.L.No.99-514,100Stat.2085(Oct.22,1986).

[106] Congressional Budget and Impoundment Control Act of 1974,Pub.L.No.93-344,88Stat.297- 339 (1974).

[107] Balanced Budget and Emergency Deficit Control Act of 1985, Pub. L. No. 100-119, 101 Stat. 754 (1985).

[108] Budget Enforcement Act of 1990,Pub.L.No.101-508,104Stat.1388(1990).

[109] We use atwo year window because it accounts for the two year election cycles in the House of Representatives.

[110] BalancedBudgetActof1997,Pub.L.No.105-33,111Stat.251,§10205.

[111] 507U.S.546(1993).

[112] Respondent’sBrief,supranote45.

[113] 137 CONG . REC. E218 (1991)(statement of Representative Donnelly) (arguing for overturning the outcome in Newark Morning Ledger Co., supra note 45).

[114] H.R. 103-111, 103D CONG. § 5 (1994) (bill proposing codification of Newark Morning Ledger Co.).

[115] The dependent variable (i.e. the action that we seek to explain) is the occurrence of congressional oversight; we coded cases with oversight as a 1 and cases that Congress did not address as as 0. Because we have a dichotomous dependent variable, we use a logit model for testing the correlation between our explanatory variables and oversight of Supreme Court cases. For a useful discussion of logit models, see J. SCOTT LONG, REGRESSIONAL MODELS FOR CATEGORICAL AND LIMITED DEPENDENT VARIABLES 34-113 (1997).

[116] Cases such as Comm’r v. Lo Bue, 351 U.S. 243 (1956) and Cheek v. United States, 498 U.S. 192 (1991) for example, sparked just one comment while Bob Jones v. United States, 461 U.S.574 (1983), a case that permitted the Internal Revenue Service to deny charitable status to educational institutions with racially biased policies led to 57 legislative responses; Regan v. Taxation With Representation of Washington, 461 U.S. 540 (1983) a case upholding the law that denied charitable status to lobbying organizations, generated more than 40 mentions in the legislative history; and legislators raised the issues involved in Comm’r v. Soliman, 506 U.S. 168 (1993) a case denying a home office deduction to a medical doctor, nearly 30 different times in committee reports and floor debates. The extent of the commentary, however, does not necessarily lead to subsequent congressional action.

To explain this difference in oversight activity we use a use a negative binomial regression model to analyze our data. We chose this model over the Poisson model due to over-dispersion of the data. See Long, supranote 115, at 217-249 for a useful discussion of various count models and for an explanation of the statistical advantages of the negative binomial regression model. The dependent variable (i.e. the action that we seek to explain) is the number of congressional responses to each individual case. This variable is continuous and ranges from 0 to 57 congressional references to a single case in the hearings, debates, and

soforth.

[117] We must ignore all these variables because they exist only when oversight takes place. Thus we cannot compare cases that Congress commented upon and that Congress ignored with the help of these variables as they perfectly predict outcomes. We are able to investigate these factors, however, when we explore the speed of oversight, see infra notes 122-33 and accompanying text, and the type of oversight that takes place, see infra notes 134-96 and accompanying text.

[118] See supra note 115 for description of the model.

[119] See supra note 116 for description of the model.

[120] To estimate these probabilities we rely the software CLARIFY; we determined the average probability of oversight by setting all the variables to their means. See Michael Tomz, Jason Wittenberg, & Gary King,CLARIFY: Software for Interpreting and Presenting Statistical Results. Version 2.0Cambridge, MA: Harvard University, June 1, 2000, available at http://gking.harvard.edu; see also Gary King, Michael Tomz & Jason Wittenberg, Making the Most of Statistical Analyses: Improving Interpretation and Presentation. 44 AM. J. POL. SCI. 347 (2000).

[121] To investigate this finding further, we created a model that has dichotomous dependent variable (coded as 1 if Congress adopted legal reform in response to the Supreme Court case and 0 if no reform was associated with the response) and an explanatory variable equal to the number of reviews each case received in the oversight process. We estimated this simple bivariate logit model and found that as the legislative references to a case increased—the likelihood of a legislative enactment also increased at statistically significant levels. As far as substantive impact, on average a case that is commented upon in Congress has a 16% change of being overridden, modified, or codified but when the frequency of oversight is at the maximum –the probability increases to 91%. To estimate these probabilities we again used the CLARIFY software. See Tomz et al, supra note 120 (providing a link to the website where CLARIFY is available). High levels of oversight, however, do not necessarily guarantee subsequent legislative reform.

[122] See also, Eskridge, supra note 2, at 345 (study indicates that the two-thirds of overrides occur withint five years of Court decision).

[123] Recently Virginia Hettinger and Christopher Zorn examined the duration period between the time the Supreme Court decided a case and Congress responded—but they limited their analysis to override activity.See Hettinger & Zorn, supra note 24, at 5. We discuss how our findings differ, infra notes 133-35 and accompanying text.

[124] Judith D. Singer & John B. Willet, It’s About Time: Using Discrete-Time Survival Analysis to Study Duration and the Timing of Events, 18 J. EDUC. STAT. 155 (1993). Following Singer and Willet, we replicated each observation (i.e. each case) in our database until the point at which Congress commented upon the case. Accordingly, cases that were subject to oversight within a year of the decision show up in the dataset just once, but cases that lasted 30 years before oversight took place, show up 30 times, and cases that are never subject to oversight are replicated 51 times. By using this method, we are able to account for the time-varying variables such as politics, economic, and legal reform. See Id.

[125] Sees upra notes89-98 and accompanying text for discussion of the three political models.

[126] Each period-dummy represents a year. Because we examined congressional activity over the course of 51 years (1954-2004), Supreme Court cases decided in the first year had a total of 51 different periods in which it could be subject to oversight. Cases decided later, say in 1995, were subject to oversight in a possible 10 different periods. Accordingly, each period represents the time that each case could be reviewed.

[127] The period-dummies did not achieve statistical significance in any of our models indicating that there is no time period in which Congress is particularly likely to act on a Supreme Court case.

[128] We used the software CLARIFY to determine probabilities. See Tomz et al supra note 120 (providing information on link to software).

[129] For a discussion of the role of the media in the civil rights context, see Lee Epstein & Jeffrey A. Segal, Measuring Issue Salience, 44 AM. J. POL. SCI. 66 (2000) (exploring media coverage as a useful measure of issue salience for elite actors).

[130] See supra notes 14-24 and accompanying text for discussion of aggregation problems in the extant literature.

[131] Spiller & Tiller, supra note 23, at 507-19 (suggesting the when Court and Congress are aligned two branches work together for purposes of policy-making).

[132] Our political variables capture the ideological position of the median member of each branch of government, but we observe individual legislative comments. This means the distance measure, while capturing the political relationship between Congress and the Court as a general matter may not be an accurate description of the distance between the political views of the individual legislators making comments on the floor and the Supreme Court. For example, the House generally maybe extremely conservative, but if the member making the comment is left-leaning—our measures will not accurately depict the circumstances in which the Court-Congress dialogue takes place.

[133] Hettinger & Zorn, supra note 24, at 5; Ignagni et al, supra note 21, at 362; Eskridge, supra note 2 at 388-89.

[134] Hettinger & Zorn, supra note 24, at 5; Ignagni et al, supra note 21, at 362 (finding Congress is more likely to override unanimous decisions).

[135] Seesupranotes14-18andaccompanyingtext.

[136] 348U.S211(1955).

[137] 536U.S.238(2002).

[138] 390U.S.62(1968).

[139] 397U.S.301(1970).

[140] 383U.S569(1966).

[141] 506U.S.168(1993).

[142] 531 U.S. 206 (2001). Additional cases subject to modifying legislation include, Comm’r v. Duberstein, 363 U.S. 278 (1960); United States v. Cannelton Sewer Pipe Co., 364 U.S. 76 (1960); Comm’r v. Lester, 366 U.S. 299 (1961); Braunstein v. Commissioner, 374 U.S. 65 (1963); Haynes v. United States,390 U.S. 85 (1968); Bob Jones Univ. v. Simon, 416 U.S. 725 (1974); Comm’r v. Standard Life & Accident Ins., 433 U.S. 148 (1977); United Statesv. LaSalle Nat’l Bank, 437 U.S. 298 (1978); Farmar v. United States, 464 U.S. 206 (1984); United States v. Am. Coll. of Physicians, 475 U.S. 834 (1986); and Comm’r v. Lundy, 516 U.S. 235 (1996).

[143] Legislators proposed overrides in the following cases but did not succeed. United States v. Acri,348 U.S. 211 (1955); Corn Prods. Ref. Co. v. Comm’r, 350 U.S. 46 (1955); Automobile Club of Mich. v. Comm’r, 353 U.S. 180 (1957); Lisbon Shops v. Koehler, 353 U.S. 382 (1957); United States v. Pioneer Am. Ins. Co., 374 U.S. 84 (1963); United States v. Catto, 384 U.S. 102 (1966); United States v. Equitable Life Assurance Soc., 384 U.S. 323 (1966); Northeastern Penn. Nat’l Bank & Trust Co. v. United States, 387 U.S. 213 (1967); Marchetti v. United States, 390 U.S. 39 (1968); Jewett v. Comm’r, 455 U.S. 305 (1982); Central Tablet Mfg. Co. v. United States, 417 U.S. 673 (1974); Diedrich v. Comm’r, 457 U.S. 191 (1982); South Carolina v. Baker, 485 U.S. 105 (1988); Colonial Am. Life Ins. v. Comm’r, 491 U.S. 244 (1989); Cottage Sav. Ass’n v. Comm’r, 499 U.S. 554 (1991); United States v.Brockamp, 519 U.S. 347 (1997);

United Dominion Industries, Inc. v. United States, 532 U.S. 822 (2001); and United States v. Fior D’Italia, Inc., 536 U.S. 238 (2002).

[144] S.R. 1910, 86TH CONG . 2ndSess. (1960) (citing Berra v. United States, 351 U.S. 151 (1956) in discussion of false and fraudulent claims of for deductions).

[145] 397U.S.72(1970).

[146] 132CONG.REC.S7580(1970)(statementofSen.Laxalt)(discussionofIRSwarrants).

[147] 512U.S.26(1994).

[148] 476U.S.558(1986).

[149] 139 CONG . REC. S10625 (statement of Sen. Gorton) (1994) (arguing retroactive taxation is a violation of taxpayer’s constitutional rights); 139 CONG . REC. S14243 (1986) (statement by Sen. Domenici) (arguing that retroactive taxation is unfair).

[150] Twelve (31%) of the cases involved the corporate income tax, 15 cases (40%) involved the individual income tax, and the remaining 11 cases involved a variety of other code provisions, including the gift and estate tax, charitable organizations, and so forth.

[151] Taxpayers prevailed in 15(39%)of the tax cases subject to a negative review.

[152] The Court issued a unanimous decision in 12(31%)of the cases subject to negative review.

[153] Amicuscuriae filed brief in 14(38%)of the cases subject to a negative review in Congress.

[154] The justices invited a congressional response to 5(13%)of the cases.

[155] The media reported on 23(60%)of the decisions subject to a negative review.

[156] From a normative perspective, scholars question congressional codification proposals as a waste of time. ADAM HARRIS KURLAND , TINKERING WITH THE EVIDENCE RULES (2000) (codifying a Supreme Court decision is a wholly empty and unnecessary gesture). Of course, a proposed amendment to overrule a Supreme Court decision based on statutory grounds (as opposed to constitutional grounds) is another matter entirely, and is certainly proper role). Interest groups, however, often argue for codification in an effort to solidify an outcome. The Cato Institute for example, has recently argued for “codifying the Supreme Court’s ruling in NLRB v. Mackay Radio & Telegraph Co., 304 U.S. 333 (1938) that employers have an undisputed right to hire permanent replacement workers for striking workers in economic strikes.” CHARLES W. BA IRD, CATO HANDBOOK FOR CONGRESS, available at http://www.cato.org/pubs/handbook/hb106/hb106-33.pdf.

[157] 347U.S.81(1954).

[158] 519U.S.79(1996).

[159] 435U.S.561(1978). J.COMM.ONTAXATION,97THCONG.,ANALYSISOF SAFE-HARBOR LEASING (Comm. Print 1982).

[160] 364U.S.361(1960).

[161] 460U.S.370(1983).

[162] J. COMM. ON TAXATION , 107TH CONG ., BACKGROUND ON PRESENT LA W RELATING TO TAX Shelter (Comm. Print 2002) (referring to several cases including Snow v. Comm’r, 416 U.S. 500 (1974); Comm’r v. Nat’lAlfalfa Dehydrating and Milling Co., 417 U.S. 134 (1974); Frank Lyon Co. v. United States, 435 U.S. 561(1978); Hillsboro Nat’l Bank v. United States, 460 U.S. 370 (1983)(combining United States v. Bliss Dairy, Inc., 645 F.2d 19 (9th Cir. 1981)) Cheek v. Unites States, 498 U.S. 192 (1991)).

[163] Cases that were expanded include: United States v. City of New Britain, 347 U.S. 81 (1954); United States v. Scovil, Harris v. Comm’r, 348 U.S. 214 (1955); Enochs v. Williams Packing & Navigation Co., 370 U.S. 1 (1962);United States v. Midland-Ross Corp., 381 U.S. 54 (1965); United States v. Correll, 389 U.S. 299 (1967); United States v. Hilton Hotels Corp., 397 U.S. 580 (1970); Woodward v. Comm’r, 397 U.S. 572 (1970); Donaldson v. UnitedStates, 400 U.S. 517 (1971); United States v. Chicago, Burlington & Quincy R. R. Co., 412 U.S. 401 (1973); Fausner v. Comm’r, 413 U.S. 838 (1973); St. Martin Evangelical Lutheran Church v. South Dakota, 451 U.S. 772 (1981);United States v. Goodyear Tire & Rubber Co, 493 U.S. 132 (1989).

[164] See, e.g., H. Res. 98-432 (1984) (discussing and expanding Harris, 348 U.S. 214); S. Res. 97-144, 97th Cong . (1981) (expanding Midland-Ross, 381 U.S. 54); H. Res. 97-201, 97th Cong. (1981) (expanding Correll); J. COMM. ON TAXATION , 103D CONG ., EXPLANATION OF PROPOSED INCOME TAX TREATY AND PROPOSED PROTOCOL BETWEEN THE UNITED STATES AND THE KINGDOM OF THE NETHERLANDS (Comm. Print 1993) (discussion of proposal to expand United States v. Goodyear, 493 U.S. 192 (1989)).

[165] Cases that alleged caused confusions were: Comm’r v. Lo Bue, 351 U.S.243 (1956); Tank Truck Rentals v. Comm’r, 356 U.S. 30 (1958); Comm’r v. Lincoln Sav. & Loan Ass’n, 403 U.S. 345 (1971); Comm’r v. Shapiro, 424 U.S. 614 (1976); G.M. Leasing Corp. v. United States, 429 U.S. 339 (1977); Central Ill. Pub. Serv. Co. v. United States, 435 U.S. 21 (1978); United States v. Vogel Fertilizer Co., 455 U.S. 16 (1982); Bufferd v. Comm’r, 506 U.S. 523 (1993); United States v. Williams, 514 U.S. 527 (1995); Comm’r v. Schleier, 515 U.S. 323 (1995). In two cases, O’Gilvie v. United States, 519 U.S. 79 (1996) and Laing v. United States, 423 U.S. 161 (1976), legislators had begun tax reform efforts in response to an appellate court decision and when the Court decided the case in way that aligned with legislative preferences the outcome was codified. Finally, not every positive response entailed a legislative initiative. In three cases, Members of Congress used the case a means to signal that they believed the justices got the law exactly right. See, e.g., 132 CONG . REC. S7580 (statement of Sen. Laxalt) (discussion of Colonnade Catering Corp. v. United States, 397 U.S. 72 (1970); J. COMM. ON TAXATION , 101ST CONG., TAX RULES RELATING TO PUERTO RICA UNDER PRESENT LAW AND UNDER STATEHOOD , INDEPENDENCE, AND ENHANCED COMMON WEALTH STATUS (Comm. Print 1989) (discussing United States v. Ptasynski, 462 U.S. 74 (1983)).).

[166] Fifteen cases involved the corporate income tax, 10 cases involved the individual income tax, and the remaining 8 cases involved a variety of other provisions.

[167] Thetaxpayerprevailedin11(33%)ofthecases.

[168] Amicuscuriaefiledbriefsin12ofthecases(38%).

[169] Justicesinvitedcongressionalreviewin4(12%)ofthecases.

[170] Themediacovered20(61%)ofthecasesthatCongressreviewedinfavorableterms.

[171] 348U.S.351(1957).

[172] 507U.S.546(1993).

[173] See,e.g.,UnitedStatesv.Cal.EasternLineCo.348U.S.351(1957);UnitedStatesv.Speers,382 U.S. 66 (1965).

[174] 485U.S.212(1957).

[175] 138 CONG . REC. S15567-02 (statement of Sen. Dole). Legislators proposed codification after critical comments to: Thor Power Tool v. Comm’r, 439 U.S. 522 (1979); United States v. Lee, 455 U.S. 252 (1982); UnitedStates v. Am. Bar Endowment, 477 U.S. 105 (1986).

[176] Gift Tax Relief Legislation Before the Senate Committee on Finance, 98thCong. 2d Sess. (summary of proposal to overrule or limit Dickman v. Comm’r, 465 U.S. 330 (1984). In other contexts, legislators would agree on one approach and then later on another. Cases such as Peurifoy v. Comm’r, 358 U.S. 59 (1958); Commarano v. United States, 358 U.S. 498 (1959); United States v. Powell, 379 U.S. 48 (1964); Comm’r v. Clay Brown, 380 U.S. 563 (1965); Bingler v. Johnson, 394 U.S. 791 (1969); United States v. Byrum, 408 U.S. 125 (1972); Comm’r v. Idaho Power Co., 418 U.S. 1 (1974); Comm’r v. Kowalski, 434 U.S. 77 (1977); INDOPCO, Inc. v. Comm’r, 503 U.S. 79 (1992); and Newark Morning Ledger Co. v. United States, 507 U.S. 546 (1993) sparked these mixed responses. In other contexts, legislators argued the case should be construed narrowly or broadly for purposes later legislation such as in Comm’r v. Glenshaw Glass, 348 U.S. 426 (1955). And finally, legislators would at times, simultaneously overrule aspects and codify aspects such as in Comm’r v. Hansen, 360 U.S. 446 (1959) andRowan Cos. v. United States, 452 U.S. 247 (1981). Other cases that engendered mixed responses include: Comm’r v.Glenshaw Glass, 348 U.S. 426 (1955; Corn Prods. Ref. Co. v. Comm’r, 350 U.S. 46 (1955); Commarano v.United States, 358 U.S. 498 (1959); Comm’r v. Brown, 380 U.S. 563 (1965); Comm’r v. Idaho Power Co.,

418 U.S. 1 (1974); Comm’r v. Kowalski, 434 U.S. 77 (1977); Thor Power Tool v. Comm’r, 439 U.S. 522 (1979); Arkansas Best Corp. v. Comm’r, 485 U.S. 212 (1988); INDOPCO, Inc. v. Comm’r, 503 U.S. 79 (1992); Newark MorningLedger Co. v. United States, 507 U.S. 546 (1993); United States v. Cal. Eastern Line Co. 348 U.S. 351 (1955); Tank Truck Rentals, Inc. v. Comm’r, 356 U.S. 30 (1955); Peurifoy v. Comm’r, 358 U.S. 59 (1958); Comm’r v. Hansen,360 U.S. 446 (1959); United States v. Powell, 379 U.S.

48 (1964); United States v. Speers, 382 U.S. 266 (1965); Bingler v. Johnson, 394 U.S. 791 (1969); United States v. Byrum, 408 U.S. 125 (1972); Rowan Cos. v. United States, 452 U.S. 247 (1981); United States v. Lee, 455 U.S. 252 (1982); Dickman v. Comm’r, 465 U.S. 330 (1984); United States v. Am. Bar Endowment,

477 U.S. 105 (1986).

[177] Fifteen(71%)ofthecasessubjecttoamixedreviewweredecidedbyadividedvoteonthe Supreme Court.

[178] Amicuscuriaefiledbriefsin14(74%)ofthecasessubjecttoamixedreview.

[179] Our data indicate that 10 (47%) case involved the corporate income tax, 6 cases (28%) involved the individual income tax; the justices invited review of 6 cases (28%); and the media reported on 10 cases (48%).

[180] 348U.S.121(1954).

[181] 132CONG.REC.S14562(statementofSen.Thurmond).

[182] 532U.S.200(2001).

[183] 150CONG.REC.S129-02(2004)(statementofSen.Byrd).

[184] For a discussion of the responsibilities of the Joint Tax Committee, see supra notes 41-42 and accompanying text.

[185] For example, the Joint Tax Committee reported on presidential budget proposals and cited cases such as Comm’r v. P.G. Lake , 356 U.S. 260 (1958) and Comm’r v. Gillette, 364 U.S. 130 (1960) noting that the proposals did not imply any changes to the case outcomes. See J. COMM. ON TAXATION , 105TH CONG., DESCRIPTION AND ANALYSIS OF CERTAIN REVENUE-RA ISING PROVISIONS CONTAINED IN THE PRESIDENT’S FISCAL YEAR 1998 BUDGET PROPOSAL (Comm. Print March 11, 1997) (citing to P.G. Lake and Gillette).

[186] 132 CONG . REC. S14562 (statement of Sen. Thurmond). Legislative comments were also made with regard to See Comm’r v. Stern, 357 U.S. 39 (1958)(understanding interaction between state and federal tax laws);Aquilino v. United States, 363 U.S. 509 (1960)(same); Reisman v. Caplin, 375 U.S. 440 (1964)(understanding rules of privilege in proceedings against Ralph and Joseph Bernstein); Comm’r v. Estate of Bosch, 387 U.S. 456 (1967)(discussion of confusion in estate tax laws and state rules); Comm’r v. Gordon, 391 U.S. 83 (1968)(discussion of tax shelters); Leary v. United States, 395 U.S. 6 (1969)(discussion on term “knowingly”); United Statesv. Mitchell, 403 U.S. 190 (1971)(discussion of property rights and state and federal law); First Sec. Bank of Utah v. Comm’r, 405 U.S. 394 (1972)(discussion of § 482 rules); Consumer Life Ins. Co. v. United States, 430 U.S. 725 (1977)(discussion of taxation of life insurance); Massachusetts v. United States, 435 U.S. 444 (1978)(discussion of e-rate program); New York Telephone Co. v. New York State Dept. of Labor, 440 U.S. 519 (1979)(preemption issues); Comm’r v. Tufts, 461 U.S. 300 (1983)(tax shelters); United States v. Rodgers, 461 U.S. 677 (1983)(federal lien rules); United States v. Arthur Young & Co, 465 U.S. 805 (1984)(auditing and accounting rules); Comm’r v. Groetzinger, 480 U.S. 23 (1987)(unrelated business income rules); Comm’r v. Asphalt Prods. Co., Inc., 482 U.S. 117 (1987)(penalties and negligence—Congress overrules lower court ruling with no impact on Supreme Court); United States v. Stuart, 489 U.S. 353 (1989)(treaties); Centennial Sav. Bank v. United States, 499 U.S. 573 (1991)(summary of rules relating to financially troubled banks); Freytag v. Comm’r, 501 U.S. 868 (1991)(appointment powers); United States v. Irvine, 511 U.S. 224 (1994)(confusing role of state rules in federal tax controversies); Drye v. United States, 528 U.S. 49 (1999)(liens and inheritances); Church ofScientology of Cal. v. IRS, 484 U.S. 9 (1987) (Scalia nomination proceedings).

[187] See, e.g., J. COMM. ON TAXATION , BACKGROUND AND ISSUES RELATING TO TAXATION OF U.S. CITIZENS WHO RELINQUISH CITIZENSHIP (Comm. Print 1995) (discussion General American Investors and James); 138 Cong. Rec. S3950 (statement of Sen. Pressler) (discussing Comm’r v. Acker, 361 U.S. 87 (1959)). The following cases also sparked comments involving statutory interpretation: American Auto Ass’n v. United States, 367 U.S. 687 (1961); Gurley v. Rhoden, 421 U.S. 200 (1975); Ivan Allen Co. v. United States, 422 U.S. 617 (1975); Slodov v. United States, 436 U.S. 238 (1978); National Muffler Dealers Assoc. v. United States, 440 U.S. 472 (1979);Regan v. Taxation With Representation of Washington, 461 U.S. 540 (1983); Comm’r v. Clark, 489 U.S. 726 (1989); Hernandez v. Comm’r, 490

U.S. 680 (1989); United States v. Cleveland Indians Baseball Co., 532 U.S. 200, (2001).

[188] S.Rep.No.94-938,at1000(1976)(discussionofBiscegliaandHaynes).

[189] 145CONG.REC.H7129(statementofRep.Lindy).

[190] SeeJ.COMM.ONTAXATION,104THCONG.,PRESENTLAW ANDBACKGROUNDINFORMATION ON FEDERAL TRANSPORTATION EXCISE TAXES AND TRUST FUND EXPENDITURE PROGRAMS, (Comm. Print 1996) (discussing cases and legislative response).

[191] The justices invited congressional review in 9 (18%) of the cases; they rendered a unanimous decision in 21 (43%) of the case; and the media covered 28 (57%) of the decisions that sparked a neutral response in Congress.

[192] Our dependent variable (i.e. the congressional activity we seek to explain) has four different categories, and thus we estimated a multinomial logit model. For a useful discussion of this model, see Long, supranote 115, at 138-186. A formal presentation of our findings would involved the coefficients and robust standard errors for three or our categories, using a forth category as the baseline or comparison groups. Id.Accordingly, even our relatively simple model would force us to present 3 different dependent variables, and 27 different independent variables. If we wanted to present every possible comparison, we would need to estimate the model four different times, leading to 12 dependent variables and 108 independent variables—an incomprehensible amount of data. Id.

[193] See Long, supra note 115, at 164-66 (suggesting that empiricists present the results of multinomial logit models in the form of probabilities for easier interpretation). We use the software CLARIFY to estimate our probabilities. See Tomz et al supra note 120 (providing information and link to software).

[194] In calculating these probabilities, we set all variables to their means.

[195] Gregory Calderia & John R. Wright, Organized Interests and Agenda Setting in the U.S. Supreme Court, 82 AM. POL. SCI. REV. 1109 (1988) (exploring interest groups participation via amicus brief filings in Supreme Court decision-making and noting parties’ strategy of filing—only when the opposition files).

[196] See supra note23 and accompanying text.

[197] See citations supra notes14-25 and accompanying text.

[198] See supra notes34-35 and accompanying text.

[199] See supra notes34-35 and accompanying text.

[200] See WILLIAM ESKRIDGE, ELIZABETH GARRET, & PHILIP FRICKEY, CASES AND MATERIALS ON

LEGISLATION: STATUTES AND THE CREATION OF PUBLIC POLICY 25 (2001)

[201] See, e.g., Gary W. Carter, The Commissioner’s Nonaquiescence: A Case for a National Court of Appeals, 59 TEMPLE L.Q. 879 (1986) (arguing problems associated with tax decision-making should lead Congress to pursue legislation authorizing National Court of Tax Appeals); Charles L.B. Lowndes, Federal Taxation and the Supreme Court, 1960 SUP. CT. REV. 222, 222 (“The thesis of this paper is simple: It is time to rescue the Supreme Court from federal taxation; it is time to rescue federal taxation from the Supreme Court); Oscar Bland, Federal Tax Appeals, 25 COLUM. L. REV. 1013 (1925) (criticizing the

federal tax litigation process as “clumsy and time-consuming” and arguing for a single court with national jurisdiction over all tax disputes).

[202] Carter, supra note 201 at 879; Lowndes, supra note 201 at 222; Bland, supra note 201 at 1013.

[203] For a discussion of congressional attempts to limit Supreme Court jurisdiction as well as lower federal court jurisdiction in various eras for a variety of reasons, see generally Maxim O. Mayer-Cesiano, OnJurisdiction-Stripping: The Proper Scope of Inferior Federal Courts’ Independence from Congress, 8 U. PA. J. CONST. LAW 559 (2006).

I.INTRODUCTION

II.A DESCRIPTION OF THE COURT-CONGRESS DYNAMIC

A. The Extant Literature Inadvertently Conceals Valuable Insights

B. A New Methodological Approach

C. Descriptive Statistics: The Frequency and Type of Congressional Oversight

III.SUPREME COURT CASES ON THE LEGISLATIVE AGENDA

A. Theory and Hypotheses: Why Do Legislator Review Court Cases?

1. Agenda Entrepreneurs

a. Legislators: Tax and Budget Committees Members b. Interest Groups

c. Journalists

d. Supreme Court Justices

2. Economic Crises

3. Political and Ideological Differences

4. Periods of Major Legal Reform

B. Summary of Hypotheses

IV.THE OCCURRENCE, FREQUENCY, AND TIMING OF CONGRESSIONAL OVERSIGHT

A. The General and Count Models

B. The Duration Model

V.THE SUBSTANCE AND DIRECTION OF CONGRESSIONAL OVERSIGHT

A. A Qualitative Analysis of Four Categories of Oversight

1. Negative Reponses and Overrides

2. Positive Responses and Codifications

3. Mixed Responses and a Combination of Overrides/Codifications

4. Impartial Responses and Neutral Legislative Enactments

B. A Quantitative Analysis of the Substantive Responses

1. Explaining Four Different Types of Congressional Responses

2. Explaining Different Types of Actual Statutory Enactments

3. Summary of Finding on Substantive Activity

VI. POSITIVE AND NORMATIVE IMPLICATIONS OF THE STUDY

Edward G. Lewis Chair in Law and Public Policy. From Northwestern University - School of Law, Chicago, USA