Conclusion

In the mid-2000s, the tax shelter industry had begun to slow down. By then a significant number of well-respected professional organizations and individuals had been drawn into the criminal justice system. To fight shelter activity, the government deployed every weapon in its regulatory arsenal—strengthening provisions in the Internal Revenue Code, enacting new regulations, challenging claims on audit, offering time-limited amnesties to taxpayers, bringing enforcement actions against firms, and ultimately launching criminal investigations and prosecutions of firms and individual tax practitioners. As evidence of the role of prestigious accounting firms and law firms mounted, the abusive tax shelter market emerged as an episode of collective wrongdoing across a major sector of professional services that historically had traded on its honesty, integrity and public-mindedness.

Using criminal prosecutions to address wrongdoing among a sector of lawyers is rare but not unprecedented. As the shelter episode suggests, the line between criminal and noncriminal conduct, especially in the white-collar context, is not always clear. Even wrongful conduct that can easily be characterized as criminal is frequently addressed through civil mechanisms. In choosing to invoke criminal processes, the government sought to communicate a clear message that the wrongdoing violated fundamental norms against assisting tax evasion. Turning to the criminal regime, however, had significant costs. The most high-profile prosecutions ended in embarrassing defeats, suggesting to observers that the government had been overly aggressive in its tactics or—of even greater concern to officials—had attempted to prosecute behavior that, even if wrongful, was arguably not criminal. The individual prosecutions that went to trial, moreover, emphasized individual culpability and obscured the organizational factors that led to the rise of the shelter industry.

By suggesting that tax shelters were the product of individual wrongdoing, the criminal cases gave a misleading account of the tax shelter episode. In the most egregious cases of wrongdoing among accounting firms, tax shelters were not the result of rogue behavior by one or a small handful of partners operating outside official organizational channels. To the contrary, shelter activity was encouraged, approved, or at the very least acquiesced in, by firm leadership, which often devoted significant organizational resources to the effort. Firms fostered internal cultures in which tax shelter development and sales were valorized. As in other cases of organizational misconduct, accounting firms socialized their constituents to engage in wrongdoing.¹ An account of the institutional dynamics serves as a necessary counterpoint to viewing the events through an individualistic lens.

In addition, the tax shelter episode holds important lessons for lawyers in general and the tax bar in particular. In spearheading the tax shelter industry, tax lawyers and accountants used their expertise to devise and sell complex transactions with neither business purpose nor economic substance. These strategies enabled taxpayers to escape billions of dollars in taxes. The tax professionals involved, moreover, did not hesitate to invoke the elite standing of the organizations in which they practiced to deflect questions about the propriety of the strategies they were promoting and investigations into their activities.

In misusing their expertise and status, tax professionals betrayed an implicit bargain between the legal and accounting professions and the state. This bargain has been at the heart of the societal understanding of the role of the legal profession in the United States since the late nineteenth century and continues to be the basis of its cultural and legal authority. Under this bargain, professionals were granted authority to regulate terms of entry, impose ethical rules, and control the institutional arrangements under which they practiced without interference from external regulators. In return, professionals would use their knowledge for the betterment of clients and society, and eschew using their expertise solely to pursue their own financial gain.

This agreement has been particularly important in the tax practice field. Given the structure of the tax code and the system’s dependence on self-reporting, high-wealth individuals have available a range of mechanisms to eliminate their tax liability. They also have the resources to retain professional expertise to assist them in this goal. Tax lawyers, therefore, have been expected to advise clients against engaging in transactions whose sole or primary purpose is tax minimization. Tax advisors perform a distinctive function in the tax system, which puts them in a position to affect taxpayers’ sense of civic duty to meet their tax obligations. Encouraging clients to engage in tax compliance may not only affect their specific behavior. It may also contribute to tax morale over all, fostering the view that the tax system is fair and that the burden of taxes falls equally on wealthy individuals and on salaried employees subject to automatic withholding.

Reviving professional norms that tax professionals can draw on to guide clients to engage in tax compliant behavior is no easy task. It will require lawyers to pursue new avenues of professional engagement. During much of the twentieth century, tax lawyers took for granted that formal and informal mechanisms of self-regulation—at the organized bar level and inside firms—could buffer their practice from the influence of market forces. The rise of the tax shelter industry suggests that reliance on traditional accounts of professional self-regulation is outdated. The organizations in which tax lawyers practice are too large and too vulnerable to intensified competitive forces to permit their constituents to engage in the creation and elaboration of professional ideals without regulatory and organizational support.² To strengthen its professional authority and independence, the tax bar may need to participate in regulatory processes that strengthen advisors’ capacity to practice consistent with professional ideals. They may also need to implement organizational structures and processes that facilitate the elaboration of norms favoring compliance. We offer some suggestions along these lines, but in the end it is up to tax lawyers to identify and strengthen the sites for engagement and the norms that might animate tax practice in the twenty-first century.

Criminal Prosecution

Historically, government authorities have rarely invoked criminal sanctions against lawyers in connection with their practice of law. When they do, they seek to articulate an account of the wrongdoing that ties it to more general social norms. In the first decade of the twenty-first century, when corporate counsel engaged in backdating stock options awarded to corporate executives, authorities attempted through prosecutions to communicate their view that the behavior constituted securities fraud. In the same vein, the tax shelter prosecutions sought to communicate the view that designing and promoting tax elimination strategies was assisting clients in tax evasion. Bringing charges against the tax professionals involved connecting the dots between their hypertechnical interpretations of the tax law and the commission of tax fraud. Although there are vigorous debates in the tax community about how much economic substance a strategy needs in order for its tax benefits to be recognized, the requirement cannot simply be elided or wished away. To have economic substance, a transaction has to result in meaningful economic consequences to the client, a feature that was absent from the strategies being promoted. The prosecutions also connected the dots between aggressively promoting abusive tax strategies and involvement in a range of illegal conduct to elude detection by the IRS. This involved, in the most egregious instances, recommending that clients use grantors’ trusts to hide their shelter activities, improperly resisting enforcement actions, fabricating documentation after the fact, intentionally misleading IRS examiners, and lying in formal proceedings.

Despite the powerful communicative role played by criminal prosecutions, using this tool had enormous costs. As criminal defendants, the tax professionals charged enjoyed important constitutional protections. Consequently, they had grounds to challenge the prosecutions at each step of the process. For example, the indictments of more than a dozen of the KPMG partners and managers involved in the firm’s shelter activities, including tax leaders who had driven its participation, were dismissed on the ground that prosecutors had interfered with their right to counsel.

Analogous problems plagued the other prosecutions. After several months of trial, the prosecution of Paul Daugerdas and his colleague Donna Guerin, along with Denis Field of BDO and two Deutsche Bank employees, resulted in conviction of four of the five defendants. The judge was forced to set the convictions aside, however, because a juror had engaged in misconduct that affected the defendants’ rights to an impartial jury and fair trial. While Daugerdas was convicted in a second trial, Field was acquitted after his lawyer argued in her summation that despite his energetic support of BDO’s tax product initiatives, he was not sufficiently involved in the deployment and marketing of shelters to know that they were abusive. The prosecution and trial of four Ernst & Young employees involved in shelter promotion met a similar fate. Four months of trial resulted in their convictions, but the convictions of two were reversed on appeal because of due process violations. The appeals court ruled that the evidence was not sufficient to establish a connection between the individual defendants’ deeds and the core wrongdoing that the prosecution had established.

These setbacks came at significant cost to the government, which expended substantial resources to prosecute and try the cases; they also provoked criticisms of the government’s tactics and strategy, deflecting attention from the wrongful conduct underlying tax shelter activities. In the worst-case scenario, defendants whose indictments had been dismissed or convictions reversed could portray themselves as the innocent victims of overzealous prosecutors. Lost in the drawn-out processes of criminal investigations, indictments, motions to dismiss, trials, convictions, and appeals, was the idea that the tax professionals who had been most deeply involved in designing and marketing abusive tax shelters deserved the moral opprobrium that criminal sanctions were intended to communicate.

The individual prosecutions, moreover, obscured the organizational factors that contributed to the pervasiveness of the wrongdoing. It is true that KPMG and Ernst & Young—as well as Deutsche Bank—entered into deferred-prosecution and non-prosecution agreements that required them to catalog the wrongful acts in which the firms had engaged. But because the charges against these organizations were never brought to trial, details about the participants who led organizational efforts to promote tax shelter activity—and the organizational resources and structures that were diverted for that purpose—never emerged.

Instead, most of the prosecutorial energy was devoted to trying individuals for abetting tax evasion. Under the heightened constitutional requirements that apply to mens rea in tax prosecutions, the government had to establish that each defendant was aware of his or her legal obligations but acted in violation of them.³ The government’s proof, therefore, had to focus on showing what each defendant knew and did despite that knowledge. The effect was to produce a powerful narrative that individuated responsibility for wrongdoing and omitted accounts of organizational complicity.

The Dark Side of Organizations

Popular accounts portrayed the tax shelter industry as a story about individual wrongdoers—greedy tax professionals intent on hijacking their organizations to enable tax evasion. The firms involved promoted this characterization because it deflected attention from organizational complicity in the wrongdoing. The prosecution of individual wrongdoers, which served as a final exclamation point to the abusive tax shelter story, reinforced this explanation.

The preceding chapters suggest that organizational factors played an important role in shelter activities and their rapid spread. As social scientists have emphasized, organizations are more than mere collections of people. An organization’s structures and dynamics shape how its constituents frame information and how they allocate responsibility for the negative effects of organizational actions. Many harmful actions are “not the volitional products of individual evildoers but rather essentially organizational products that result when complex social forces interact.”⁴ As sociologist Diane Vaughan notes, harms caused by organizations are often a “routine by-product of the system itself.”⁵ Researchers focusing on organizational crime have argued that the organizational dynamics that lead to wrongdoing are the effects of the distribution of power and division of tasks inside organizations. The very features of organizations that render them highly successful social systems—hierarchy and specialization—also give rise to the development of organizational pathologies.⁶

Vaughan proposes that exploring the interactions among three features of an organization sheds light on the “dark side of organizations.” These features are the environment in which an organization functions, its structure, and the interplay of cognitive and decision-making dynamics inside the organization. An organization’s environment provides the underlying conditions for the initial decision among organizational leaders and constituents to embark on wrongful conduct. External factors function to ratify and entrench this initial decision. Wrongdoing becomes institutionalized—routinized and normalized—by being embedded in an organization’s structure. As Vaughan observes, “Structures, processes, and tasks are opportunity structures of misconduct because they provide a) normative support for the misconduct, b) the means for carrying out the violation and c) concealment that minimizes detection and wrongdoing.”⁷

Structural features such as the allocation of authority and responsibility, the processes for decision making and review, and the division of tasks within an organization allow constituents to contribute to organizational wrongdoing while remaining ignorant of the effects of their actions and believing that they are engaged in a morally neutral or even beneficial activity. Together, environment and structure influence the cognitive and decision-making dynamics within an organization, shaping how actions and decisions are framed. According to Vaughan, “the origin of routine social non-conformity is in the connections” among the environment, organizational characteristics, and the cognitive practices of the individuals within them.⁸

The Environment

Organizations do not act in a vacuum, but react to and shape in a recursive process the economic, regulatory, and political environments in which they function. The major accounting firms that launched the shelter industry were operating in a highly competitive environment. Among the challenges they faced were the difficulty of differentiating their services and their reliance on a traditional fee structure—the hourly rate—that gave them little room to increase profits. Subject to comparable market forces, corporate law firms were in an escalating competition for clients and partners who would generate profits. Even as accounting and law firms found themselves in an increasingly competitive struggle, the economy around them was booming. In the late 1990s and 2000s, professionals at these firms compared their lot to that of professionals at financial institutions. They longed for a “Goldman-Sachs type practice”⁹ that would generate substantially more income.

At the same time, firms providing tax services faced a lax regulatory environment. Among other problems, Treasury officials were preoccupied with dealing with a hostile Congress and attempting to modernize IRS operations. Limited by outdated information gathering systems, demoralized personnel, and few enforcement resources, the agency did not seem particularly eager to ferret out and challenge highly complex tax strategies. For some tax professionals involved in shelter activity, the fact that shelters were low on the government’s priority list meant that they were not going to be caught if they engaged in promoting highly questionable deals. Others likely interpreted the government’s inattention to shelters as a signal that shelters did not raise serious legal or ethical problems.

This second interpretation was especially plausible given the contested nature of tax law itself. As chapter 2 describes, the income tax, unlike some other areas of law, has significant gray areas that make the line between tax avoidance and evasion unclear. Because of fundamental definitional difficulties, there are a large number of scenarios about which knowledgeable and well-intentioned tax practitioners vigorously disagree. At a high level of abstraction, the requirement that tax-favored transactions have “economic substance” is uncontroversial. Nevertheless, courts, tax lawyers, and scholars are not of one mind about the most important considerations that comprise economic substance, or what amount of economic substance a tax shelter must have to be upheld by a court. Viewed after the fact, many of the shelters promoted by large accounting firms clearly lacked economic substance. At the time, however, the ambiguities inherent in tax law—magnified by organizational factors—enabled many tax professionals to convince themselves that the shelters they were promoting were legitimate.

The view among many tax professionals that shelter activity was unproblematic was reinforced by the fact that other prestigious firms were involved. For some participants “everyone else is doing it” meant that as a practical matter their firm could engage in wrongdoing without being penalized. For many others, however, the fact that other elite accounting firms—KPMG, Ernst & Young, Arthur Andersen, BDO Seidman—were marketing tax shelters was a clear indication that there was nothing illegal or unethical about the activity.

Institutionalization

At KPMG, Ernst & Young, BDO, and other firms, shelter activity became embedded in the structures and processes of the organization. Tax leaders at these firms devoted substantial organizational resources to developing and marketing tax shelters. They aligned organizational incentives, including compensation and advancement at the firm, with revenue production tied to promoting tax shelters.

On one level, the alignment of formal mechanisms with shelter activity was a signal to tax professionals at the firms that if they wanted to succeed they needed to participate in this activity. In some instances, participants believed with good reason that their positions at the firm were at stake. But the message worked at a different level too. By using the power of their positions to encourage tax shelter activity, tax leaders were communicating their view that these activities were appropriate. Tax leaders, who functioned as role models, fostered a culture in which involvement in tax shelter activity was not only not problematic but was consistent with high standards of client service and a commitment to the success of the firm. The institutionalization of shelter activity thus did more than simply dictate the terms of a rational calculus of costs and benefits, where participating in tax shelter work promoted one’s career and staying clear led to career setbacks. Explicitly organizing business units around shelter work valorized it and shifted its normative valence. Tax shelter practice was no longer seen in negative terms; it was embraced as a worthwhile activity.

As tax shelter activity became intertwined with organizational processes, review mechanisms intended to prevent questionable shelters from going to market proved ineffective. In some instances, firms relied on informal processes that were deficient. At Ernst & Young, the same professionals who had designed the shelters and stood to gain from their approval conducted the review. While in hindsight this process seems highly problematic, the partners involved were well-respected tax lawyers with significant practice experience. Given their reputations as upstanding professionals who had always safeguarded the firm’s interests, their judgment was presumed to be trustworthy. Under traditional professional norms, they were expected to exercise appropriate discretion to prevent clients and the firm from engaging in questionable activity.

The formal processes implemented at some firms to impose appropriate safeguards also proved to have significant weaknesses. Most obviously, they were embedded in a larger institutional structure that emphasized shelter promotion. For all that Larry DeLap, the head of the Department of Professional Practice—Tax at KPMG, tried to maintain some semblance of independence, he reported to tax leaders who did not hesitate to make their enthusiasm for tax shelters known. For his part, Mark Watson, who had been charged with steering BLIPS through the approval process at KPMG, was well aware that the product was getting some “high level attention.”

A further weakness with the approval process at KPMG, illustrated by the BLIPS review, was the division of substantive issues among subject matter experts. Whether BLIPS had economic substance became a narrow technical question. Those assigned the task of resolving the issue struggled to create a scenario, however implausible, where a client might actually purchase the strategy to make money. Economic substance was resolved by recourse to some imagined set of facts. Treated as a technical question, the economic substance problem became decoupled from the broader question of whether the strategy would perpetrate tax fraud by allowing clients to claim losses for transactions that did not reflect economic reality.

As participants struggled to resolve this issue, responsibility for the problem moved from one participant to the next. Eventually the question of whether BLIPS had economic substance was “deferred to implementation.” As the process unfolded and responsibility shifted, other participants could reasonably believe that they no longer were responsible for deciding the issue. Despite Mark Watson’s grave reservations about BLIPS, he repeatedly assured himself that more senior tax partners with greater expertise were accountable for the final decision. He also believed he could rely on Deutsche Bank and its law firm Shearman & Sterling, which, he assumed, would never participate in activities that helped clients engage in tax evasion. Steven Rosenthal, who shared Watson’s concerns, repeatedly insisted during the review process that he was not opining on the question of economic substance.

Later, as the implementation process moved forward, Rosenthal and Watson saw Deutsche Bank documents that made clear that the loan in BLIPS was a loan in name only. After their efforts to persuade David Brockway to revisit the decision to approve BLIPS proved unsuccessful, they saw no choice but to give up. Each believed, with good reason, that the decision to market BLIPS had been made at a higher level at the firm, and that they had done all they could.

Organizational dynamics also contributed to a diffusion of responsibility for the decision to market BLIPS. KPMG’s review process allowed participants to lose sight of the bigger question of whether their actions were implicated in tax fraud and to convince themselves that they bore no responsibility for the outcome. This phenomenon is consistent with frequently observed processes in organizations that engage in misconduct. As social psychologist John Darley notes, the diffusion and fragmentation of information and responsibility in an organization, which are by-products of the division of labor and specialization in organizational settings, will often lead to wrongdoing. These processes have often been observed in organizations whose actions end up harming others. They include Ford’s decision to continue marketing the Pinto model even after internal revelations that the car’s fuel tank had a potentially fatal design defect,¹⁰ and NASA’s disastrous decision to launch the Challenger space shuttle despite warnings that some components had not been proven safe in a launch at cold temperature.¹¹ The pervasive presence of these dynamics in professional firms—ostensibly organized as partnerships in which members are expected to grasp and be accountable for the bigger picture—underscores the importance of understanding how information becomes disaggregated and responsibility is diffused.

If disaggregation and diffusion marked the review process at KPMG, different informational dynamics were at the core of the review process implemented at Jenkens & Gilchrist. Once the decision was made to have Daugerdas open a branch office in Chicago, the firm instituted a review process in which a tax lawyer in Texas would sign off on opinions emanating from Chicago. This system was consistent with similar review processes already instituted at the firm. The assumption was that it would be sufficient to protect the firm from the risks associated with Daugerdas’s tax practice.

Some tax lawyers at the firm suggested to management that effective oversight required someone familiar with the highly technical subject of tax shelters in general and with the specific transactions for which opinions were being issued. One problem was that no one in the firm had worked on tax shelters. Michael Cook, who agreed to review Daugerdas’s strategies, was presented with opinions that were based on representations about transactions and clients’ motivation for entering into them. He was not in a position, however, to determine if those representations accurately described a client’s genuine intentions. The client’s purpose was crucial in assessing whether tax losses from the transaction were likely to be upheld, since clients had to be motivated by a legitimate business purpose or reasonable hope of economic gain. While Cook could review the technical aspects of the strategies, he did not have access to the very information he needed to distinguish legitimate tax strategies from abusive ones.

Like Mark Watson at KPMG, Cook was assigned a nearly impossible task. The firm had already implicitly endorsed Daugerdas’s shelter practice by deciding to hire him after a debate on its propriety. Daugerdas likely assumed that he would routinely receive approval for transactions that followed this template. In addition, Cook effectively stood between his partners and a projected additional $6 million to divide among themselves (including himself, of course). The structure that was implemented, in which a single lawyer with limited authority inside the firm was asked to review dozens of complex potentially lucrative transactions, was a recipe for failure.

In both the KPMG and Jenkens cases, geographic distance coupled with communication technologies exacerbated the weaknesses of the process. Mark Watson was acutely aware that tax leaders of his firm were monitoring his decision-making process. Many of them, including Jeff Stein and John Lanning, the vice chair of Tax Services, were not located at Washington National Tax, where Watson worked, but in KPMG’s New York office. During the review, they remained largely invisible, announcing their presence occasionally by email to urge approval. Emails, like other communication technologies, provide only an incomplete account of the sender’s intent. KPMG tax leaders used them to augment their power by turning them into communications by disembodied and omniscient overseers.¹² When the time came time to approve BLIPS, Larry DeLap wrote Watson an email, copying the tax leaders, pressing him to sign off. Because DeLap was in California, Watson did not have the opportunity to voice his concerns in a face-to-face conversation or to determine whether DeLap had similar hesitations. Had their offices been in close proximity, they might have talked informally and developed a strategy to push back against senior tax leadership. At KPMG, email functioned to rally the converted and to isolate and weaken those who had reservations or counseled a more cautious approach.

At Jenkens & Gilchrist, the geographic isolation of the Chicago office from the main office in Dallas, coupled with the failure to have a long-term Jenkens lawyer in the same office as Daugerdas, created similar challenges. In the modern age, Daugerdas’s location in Chicago, far from the firm’s Dallas headquarters, would present no obstacle to effectively practicing as a member of the firm. It would, however, mean that Daugerdas would not be subject to the kind of informal norms that people can impose in regular face-to-face relationships with one another. For all intents and purposes, Daugerdas ran his own practice just as he had before; only the name of the law firm on the door was different. This may have reinforced his sense that he could operate more or less in an unfettered fashion. Reviewing Daugerdas’s tax strategies back in Austin, Cook was significantly hampered by his inability to gauge Daugerdas’s aggressiveness and his lack of any contact with the clients who purchased them. At both KPMG and Jenkens, opportunities for face-to-face conversations during the review process might have opened up avenues for different choices and courses of action.

Organizational Commitment Bias

As these examples suggest, environment and organizational structure interacted to shape the interpretations, decisions, and actions of the tax professionals who participated in the tax shelter market. Two other types of cognitive and decision-making dynamics occurred repeatedly as firms became invested in participating in the tax shelter industry. These dynamics, which have been observed in individuals, are intensified in the organizational context.

The first is a phenomenon described among economists as the “sunk cost” problem. Once an individual invests resources in a course of action, it is very difficult to reverse course, even when continuing will likely result in harmful or costly outcomes. The problem of sunk costs is exacerbated in the organizational context. As Darley notes, “[i]t is hard enough for an individual to reserve a personal decision, even when no one else knows of the decision. In organizational settings, the decisions are far harder to reverse.”¹³ The employee who originally advocated for the decision is likely to suffer adverse reputational and career-related consequences; the organization risks embarrassment because costly resources have been wasted in implementing the decision.

The difficulty of altering course came up frequently at the firms involved in the tax shelter industry. KPMG, Ernst & Young, and other firms routinely continued to sell a particular shelter, even after a decision had been made that promotion of the shelter should cease. The continued sales were justified on the grounds that clients had already signed a retainer agreement. Only PwC, responding to intense public embarrassment from the exposure of its tax shelter activities on Capitol Hill, decided to exit the shelter market, unwind the transactions, and refund fees to clients.

The cognitive dynamics associated with the problem of reversing the decision to market shelters are reflected in the interpretative gloss that Michael Kerekes, a member of the tax shelter group at BDO, gave to Notice 2000-44, which was issued by the IRS in August of 2000. The notice made clear that the agency intended to challenge the tax benefits claimed from Son of BOSS transactions. In analyzing the notice, Kerekes correctly recognized that it applied to the transactions then being marketed by BDO. His memo focused exclusively, however, on the effective date of the listing requirements that applied to BDO. It failed to address the heart of the notice, which was the IRS’s position that Son of BOSS shelters did not have economic substance. As the memo emphasized, “it does not discuss whether the Notice should be viewed as having any effect on the substantive law relating to the transactions.”¹⁴

A more complete reading of the notice would have concluded that it raised a serious question about all the Son of BOSS transactions that BDO had already sold. It also would have addressed the further question of whether the firm should consider unwinding the transactions, or, at the very least inform clients who had already bought the strategy that there was a significant risk of challenge. Not only did Kerekes’s memo not consider these issues, it went so far as to argue that the memo did not apply to future transactions for which the firm already had an engagement letter. This gave the firm a green light to implement transactions that were already in the works without listing them with the IRS. Coupled with the failure to discuss the central issue raised by the notice, the strained logic of Kerekes’s memo suggests that he was so wedded to promoting Son of BOSS shelters that he was incapable of revisiting the question of whether they were legitimate tax reduction strategies.

Jenkens & Gilchrist also exhibited a stubborn commitment to an ultimately disastrous course of action despite gaining information that should have led it at least to consider whether to reverse course. Paul Daugerdas had projected revenues of about $6 million a year when he was negotiating moving to the firm. In 1999, however, his practice produced about $28 million in income, or almost five times the amount that the firm anticipated. This figure was more than 13 percent of the firm’s total revenues. In 2000, he generated almost $91 million in income, or almost a third of Jenkens’s revenues.

This vast increase in expected revenues raised two potential issues. First, could a tax practice relying on mass products and standard routines legitimately generate such massive revenues, or did this order of magnitude suggest that the firm should look more closely at Daugerdas’s practice? Second, if the firm continued to support the practice, transactions of this volume would require a much more extensive oversight process than the firm had established. Jenkens management never confronted either issue. Reexamining the propriety of Daugerdas’s practice and the effectiveness of the firm’s monitoring system would have required admission that the firm may have erred in its initial decisions. In addition, of course, it could have jeopardized the substantial flow of income from tax shelter practice. While the need to revisit the firm’s earlier decisions may seem obvious from the outside, not doing so seemed reasonable to those immersed in the situation.

Beyond illuminating the organizational dimensions of professional activity, the shelter industry has implications for the status and authority of tax advisors in the American tax system. In particular, it raises the question whether tax lawyers can advise clients consistent with the professional norms to which they have traditionally claimed adherence, given that the organizations in which professionals practice tend to magnify rather than buffer the competitive forces to which they are prey. Allegiance to these ideals is not simply a reflection of a self-serving ideology that rationalizes the material benefits and social status that lawyers enjoy—although lawyers have certainly invoked these ideals for this purpose. The assumption that tax professionals will encourage lawful conduct by their clients is incorporated into the fabric of the American tax system.

The Role of Tax Advisors in Promoting Compliance

The elite American bar has long justified the prerogatives it has been granted by society by invoking the conception of the independent counselor. Under this ideal, articulated by the legal philosopher Lon Fuller more than half a century ago, in the lawyer’s office “the lawyer’s quiet counsel takes the place of force.”¹⁵ For Fuller, as for elite lawyers who espoused this view, a lawyer’s capacity to counsel clients to engage in law-abiding behavior promoted freedom in a democratic society. This was particularly true for wealthy clients with the strongest incentives and greatest resources to evade legal requirements. Rather than rely on intrusive oversight and enforcement mechanisms to elicit their compliance with legal norms, society could rely on wealthy individuals voluntarily to obey legal mandates in accordance with their lawyers’ advice. As the Supreme Court emphasized three decades later, the role of a lawyer in encouraging legal compliance was the rationale for giving the attorney-client privilege a wide scope in the realm of client counseling.¹⁶

In tax practice, the importance of advice in promoting compliance is reflected in the function of tax opinions to abate penalties that taxpayers would otherwise be required to pay for understatement of taxes. Under the applicable standards, the IRS prohibits an opinion author from advising a client about the likelihood of audit or detection of the strategy. Instead, the lawyer is required to opine that more likely than not a court would uphold the tax benefits of the transaction were it challenged by the IRS. In other words, the lawyer is expected to give his or her expert opinion as to whether the tax treatment favored by the client is consistent with the law. Moreover, the provision that a taxpayer can avoid penalties by reasonably relying on a more-likely-than-not opinion assumes that tax lawyers are motivated to advise clients against engaging in abusive tax strategies.

In addition to securing their clients’ compliance, tax advisors have a role in promoting positive tax morale—the belief that citizens should meet their tax obligations because others are paying their share. A distinctive feature of the tax regime is that taxpayers’ sense of obligation is especially sensitive to expectations of how other taxpayers will behave. Research indicates that high rates of compliance with tax obligations are significantly correlated with what is called high “tax morale”: “the intrinsic motivation to pay taxes based on citizens’ sense of obligation to their state.” This sense of obligation is quite fragile, because it is not simply an internally derived attitude. Rather, it is highly sensitive to the attitudes and behavior of others, as well as influenced by perceptions of government.¹⁷

As one researcher puts it, tax compliance is “a social act.” A sense of intrinsic obligation is the product of “conditional cooperation.” Research indicates that if a taxpayer perceives that there is a high percentage of other taxpayers who avoid paying their fair share, she is less likely to comply with her own obligations. In those circumstances, she doesn’t want to be a “sucker” who incurs the cost of helping support a cooperative scheme for free riders who benefit from it without contributing to it. In addition, less tangibly, lower compliance in this situation may reflect the desire to conform to norms of social behavior. In any event, tax morale, and thus compliance, decreases as people believe tax evasion to be more common. Research on tax morale, in other words, suggests a certain paradox: a taxpayer’s sense of categorical obligation to pay taxes is conditioned on the perception that a critical mass of other people share the same attitude.

Taxpayers’ sensitivity to others’ behavior can create a particular challenge for tax law because research suggests that most people believe that other taxpayers are less committed to paying their fair share of taxes than they are. As a result, any given taxpayer contemplating compliance may begin with an attitude of suspicion that potentially could disrupt conditional cooperation. As one scholar describes it, “The systematic misperception that other people hold norms and views that are less supportive of honest taxpaying could lead to a vicious cycle of people adapting their own ethics and behavior to these perceived norms and thus contributing themselves to the invidious culture.”¹⁸

In this respect, tax compliance can be likened to an assurance game, in contrast to a prisoner’s dilemma. In the prisoner’s dilemma, noncooperation is the dominant strategy for an individual regardless of what other people do. A prisoner’s dilemma scenario reflects the following preference ordering: “(I) I do not contribute, but enough others do; (II) we all contribute; (III) no one contributes; (IV) I contribute, but not enough others do.”¹⁹ The absence of communication or enforceable agreements to cooperate, reflected in the prisoner’s dilemma, leads a participant to reason that if sufficient people cooperate to provide a common benefit, it is preferable not to cooperate and to free-ride on their efforts. Alternatively, if not enough people cooperate, it’s best not to do so, thereby avoiding the cost of participating without receiving the benefits of shared participation. In the first instance, we can analogize a participant’s reasoning to greed, in the second to fear, both born of self-interest.

As the scholar Daphna Lewinsohn-Zamir suggests, failure to cooperate in some cases may be due neither to greed nor to fear but to what she calls “hopelessness.”²⁰ In this scenario, a person’s highest-ranked preference may be that everyone contributes so that all will enjoy a common benefit. This person may not cooperate, however, because “[b]oth the knowledge that one’s own contribution will have a miniscule effect on the desired outcome and the fear that not enough others will contribute may create a feeling of hopelessness: People reason that, regardless of what they choose to do, the collective goal will not be achieved.”²¹ If they somehow had assurance that others would cooperate, they would, too.

In an assurance game, an individual’s preference ordering is as follows: “(I) everyone contributes; (II) no one contributes; (III) I do not contribute, but others do; (IV) I contribute, but others do not.”²² In contrast to a prisoner’s dilemma, there is no dominant strategy in an assurance game. That is, an individual’s choice is dependent upon what others do. If others cooperate, the individual will also do so, because, unlike the preference ordering in a prisoner’s dilemma, the individual prefers mutual cooperation. If others do not cooperate, she will not, because she would incur the cost of cooperating without achieving the common benefit. Either alternative is an equilibrium, since no one has an incentive to change her behavior when she knows what other people will do.

The likelihood of cooperation in an assurance game “depends to a great extent on the existence and quality of information regarding the action that is likely to be taken by others (or likely to be expected by everyone to be taken by others).”²³ Thinking of tax compliance as an assurance game thus underscores that in this area of law compliance rests on a basic norm of obligation that is highly susceptible to being overridden by the perception of noncompliance by others.

A tax advisor can play a critical role in this assurance game, serving as both an explicit and implicit source of information for the client about other taxpayers’ behavior. Clients reasonably assume that a lawyer who is a tax specialist is familiar with a large number of other taxpayers. The lawyer may expressly tell the client that most people are taking advantage of certain tax provisions. Or she may communicate the idea that others treat tax law as rules that define the terms of a contest with the IRS. She also may implicitly convey such information through the way she discusses the tax code and the remarks she makes about the IRS. If the lawyer fosters the perception that other taxpayers lack a sense of civic obligation, she can lead the client to adopt a similar attitude in self-defense. By contrast, conveying a sense of respect for the tax law and of the good faith of other taxpayers can help secure the cooperation necessary for the tax system to work.²⁴

Historically, it was assumed that professional norms that encouraged tax lawyers to advise their clients to comply with tax laws were imparted informally through the organizations in which tax professionals worked and the organized tax bar activities in which they participated. This view was of a piece with the traditional account of professional self-regulation, under which lawyers were given broad discretion to determine the rules and conditions of practice that would foster their capacity to exercise their expert judgment to further societal aims. As the tax shelter episode illustrates, the organizations in which tax lawyers practice can no longer be trusted to buffer lawyers from competitive market forces or instill professional norms.

The Limits of Traditional Professional Self-Regulation

In the traditional account of professionalism, each profession was an island of self-regulation, which enjoyed a measure of insulation from both market pressures and state supervision. This insulation gave lawyers and accountants the freedom of action necessary to act consistent with professional ideals. Freed from subservience to market imperatives and government control, a profession, it was claimed, generated its own distinctive normative order. Law firms and accounting firms ostensibly served as the vehicle for socializing lawyers into that order, relying principally on informal organizational norms to regulate their members’ behavior. For most of the twentieth century, tax lawyers in elite law firms were especially effective in developing and sustaining a shared sense of the boundaries of ethical tax practice. Norm enforcement relied not so much on specific rules that a practitioner consulted when deciding how to behave, but on the cultivation of a certain expertise and orientation to tax practice.

As chapter 3 describes, several forces have undermined belief in the continued viability of the implicit professional bargain. Intensified market pressures have made financial performance an increasingly important consideration in law and accounting firms. Of necessity, lawyers must attend more to economic self-interest in their deliberations, creating the risk that traditional professional values will be subordinated or displaced. In tax practice, this development has led respectable mainstream organizations to provide more aggressive advice to clients and become involved in abusive shelter work. In addition, the tax practice community has become more fragmented, and there is less consensus about the norms that should guide practice. The result is increasing skepticism about relying on the legal and accounting professions to police themselves and to ensure that professionals act ethically. This raises doubts about the extent to which law and accounting firms can serve as sites in which shared professional ideals are articulated and enacted in daily practice. If tax lawyers are to revive the norms that animated their traditional counseling role, they will have to explore alternative avenues of engagement with the tax regime.

Hybrid Regulation

What Ted Schneyer has called “bar corporatism” may elicit the revival of professional norms by providing a middle ground between traditional self-regulation and outside regulation of tax practitioners.²⁵ Under this approach, a regulatory agency with expertise in the field oversees practice in a specialized area, guided by dialogue and negotiation with practitioners. Schneyer suggested that such a regime emerged twenty years ago with respect to banking practice in the wake of the collapse of the savings and loan industry. When questions arose about the conduct of lawyers that had represented failing savings and loans, the federal Office of Thrift Supervision (OTS) did not wait for bar disciplinary authorities to act. Rather, the agency itself undertook aggressive enforcement action under interpretations of the ethics rules it regarded as applicable to banking lawyers.

As part of its response, the OTS worked closely with the private bar in establishing remedies and standards of conduct. Some firms, for instance, negotiated consent agreements with the agency that set forth detailed procedures for representing banking clients. The agency worked closely with the American Bar Association (ABA) Business Section to establish standards for preparing third-party legal opinions and reform other areas of banking practice. Reflecting on the OTS approach, Schneyer comments, “If the regulators are unwilling to defer to the traditional regime of professional self-regulation, they seem equally reluctant to dictate standards for lawyers unilaterally, such as through their own rulemaking proceedings. Instead, informal dialogue with a designated bar representative—the ABA—is the order of the day.”²⁶

Bar corporatism holds promise as a form of regulation that can include both outside oversight and participation by the bar in setting practice standards. While the bar is not completely self-governing, it has the opportunity to participate in shaping the rules that will govern practitioners. The regulatory agency welcomes such participation because it operates on the assumption that individuals are motivated to some degree by a desire to fulfill their professional obligations, not simply to avoid incurring penalties. This approach also accepts the possibility that this desire can lead professionals to establish and informally enforce norms of behavior that further compliance with regulations.

The tax shelter crisis, however, suggests that the tax bar’s efforts to engage in a bar corporatist approach have been too narrow. Since Circular 230 was first enacted, the tax bar has collaborated with the IRS to articulate standards governing practice before the agency.²⁷ When the Treasury Department has considered revising these standards, it has solicited the views of the tax bar and given it an influential role in the process. On occasion, the tax bar has taken the initiative to propose more stringent requirements. At the height of the tax shelter crisis, for example, the organized tax bar mounted a vigorous campaign to strengthen regulations to deter shelter activity.²⁸

Historically, however, the bar and IRS have emphasized controlling the conduct of individual tax professionals. Regulatory efforts have not given attention to the organizational influences that shape the conduct of tax professionals working in firms. As the rise of the shelter industry illustrates, wrongdoing by professional firms is a product of interactions among its environment, structure, and the cognitive and decision-making dynamics generated by the specialization and diffusion of responsibility. In exploring new regulatory directions, tax lawyers and the IRS might focus on developing approaches that limit professional organizations’ incentives to engage in tax shelter activity. Such an approach would pay attention to the less visible organizational influences that gave rise to the shelter market. It would also empower law and accounting firms to resist client pressures. The current period of low demand for shelters is an opportune moment for the tax bar to explore initiatives that will permit them to resist client pressure when the demand for shelters arises again.

The tax bar might, for example, engage with the IRS to develop an amendment to Circular 230 that would render firms liable for involvement in abusive tax shelter by their members. One such approach would be to extend firm liability to any instance in which there is failure to take reasonable steps to ensure compliance with Circular 230 provisions. A second measure, which would create even more incentive to develop an adequate compliance program, would provide that a firm is presumptively vicariously liable for any of its practitioners’ violations of Circular 230. The firm, however, could either assert a defense or receive lenient treatment if it had in place an effective compliance system.

This second proposal mirrors the Organizational Sentencing Guidelines, which have prompted the development of extensive compliance programs in corporations. As two practitioners have observed, “Without question, the Guidelines’ greatest practical effect thus far is to raise the business community’s awareness of the need for effective compliance programs.”²⁹ The rules could also require certification of the effectiveness of a compliance program from the head of tax practice and managing partner of the firm, similar to the certifications currently required by top management of internal control systems.

Consistent with responsive regulatory approaches, the IRS would not dictate the procedures that firms used, but would focus on outcomes.³⁰ Firms would be able to establish whatever procedures they believed were effective in ensuring compliance with Circular 230. This independence would provide the opportunity for the exercise of professional discretion, and would encourage the involvement of tax practitioners and other lawyers in the firm in establishing and enforcing practice norms. Some firms might choose to share the details of their compliance programs with the IRS, which could publicize those that it believed reflected best practices. The result ideally would be an evolving set of evidence-based compliance procedures.

The different structures of accounting firms and law firms would require different organizational approaches. As the previous chapters suggest, accounting firms engaged in coordinated shelter strategies that received an imprimatur of legitimacy from high-level executives. Transforming organizational incentives in accounting firms to discourage tax shelter activity would require attending to the hierarchical reporting structure of accounting firms and their historical tendency to meld revenue-generating and compliance functions.

One possible model might be the firm-wide changes E&Y implemented in response to the adverse publicity it received in 2003 as a result of the Permanent Subcommittee on Investigations’ investigation into the firm’s tax shelter activities. According to the nonprosecution agreement the firm entered into in 2013, E&Y had “implemented extensive changes to its governance and compliance procedures, and also substantially increased the number of its legal and tax quality and risk management personnel.” Its software-based Quality and Integrity Program, which monitored data entered by members of its tax practice regarding a range of tax engagement matters, was praised by the IRS as a model for an effective compliance program. E&Y also instituted a series of procedures and practices to ensure legal and ethical conduct among its tax professionals and firm-wide.³¹

Expanding the reach of Circular 230 would also create an incentive for law firms to respond to the challenges posed by their structures, which are less integrated and centralized than those of accounting firms. At law firms, the risk of individual lawyers’ involvement in tax shelters is exacerbated by a reluctance to establish constraints on profitable individuals and practices and the uneven distribution of management skills in firms. Work on abusive shelters by lawyers in corporate firms was likely to result from a combination of pressures for individual performance, the opportunity to avoid close firm oversight, and competitive pressures on firms that led them to look the other way when they suspected that high revenues might be the product of questionable activity. The prospect of firm liability could serve as one incentive for firms to take more robust steps to monitor behavior in the face of these tendencies.

Broadening the scope of Circular 230 to encompass firm liability for wrongdoing by members of a firm’s tax practice is one possible approach to address firms’ roles in causing misconduct, but there are others to be explored. Regulatory and professional initiatives that give attention to organizational factors would complement and enhance existing measures to discourage tax shelter activity, including enforcement of regulatory requirements and, in the most egregious circumstances, recourse to the criminal process. These approaches would seek to deter misbehavior by tax professionals who take a calculating approach to compliance with professional standards. Organizational approaches developed collaboratively by the tax bar and the IRS might also encourage the articulation of professional norms and elicit informal efforts to enforce them. They might also prompt interpretation of sanctions as expressions of support for and commitment to professional ideals and practices.

Conclusion

Lawyers’ and accountants’ involvement in the tax shelter crisis at the turn of the century differs from culpable professionals’ roles in other corporate scandals in recent years. Participants in shelter activity did not simply play a supporting role in the wrongdoing of others; they were major protagonists in the commission of fraud. Tax professionals developed hypertechnical interpretations of obscure provisions of the tax code, designed and marketed highly complex shelters, and sold lengthy opinions to purchasers claiming that the shelters would be upheld in court. This recent shelter episode also differs from earlier ones because accountants and lawyers were not sketchy salesmen trying to sell partnerships in dubious real estate deals to generate phony tax losses. They were partners in Big Five accounting firms and prestigious law firms, whose reputations gave shelter transactions a patina of respectability. Professionals at the heart of the tax system, in other words, were actively working to undermine it.

For the moment tax shelter activity abetted by elite professional firms in the United States appears to have declined. But, if the past is any indication, it is unlikely that the industry has been shut down for good. The tax code is likely to become more complex, and financial instruments more complicated. Avenues to evade tax liability have moved and flourished off-shore. Competitive pressures on accounting and law firms almost certainly will continue to intensify, and the IRS is never a favorite of budgetary committees on Capitol Hill. For these reasons, while it is not possible to know what form they will take or what provisions of the Internal Revenue Code they will seek to exploit, there is a risk that at some point in the future a new wave of abusive shelters is likely to emerge. With current demand for shelters ebbing, professional organizations have an opportunity to put in place robust procedures that insulate them from the temptation to pursue abusive shelter work, and their members from organizational pressures to engage in it. If they can do this effectively, the next wave of abusive shelters may not be as severe, and accountants and lawyers may succeed in steering clear of wrongdoing.

The stakes for tax professionals are substantial. Across almost every area of specialization, lawyers and other professionals are becoming increasingly subject to external regulatory mandates, and their authority to regulate their practice is eroding. The vulnerability of the organizations in which lawyers practice creates risks of increasing external oversight and encroachment. As the shelter episode illustrates, one danger is that a handful of professionals will engage in wrongdoing and subject themselves and their firms to criminal liability. A more pervasive concern is that the public will become increasingly skeptical of claims by the accounting and legal professions that they are capable of regulating themselves to serve the public good, and will conclude that firms in which tax professionals practice are “professional service” firms in name only. At that point, the public is likely to demand a more intrusive regulatory response that could eliminate meaningful opportunities for the exercise of professional judgment at elite firms. Moreover, the professional ideals that once drew lawyers and accountants to sophisticated tax practice will no longer have much traction. If tax professionals are concerned about such developments, it behooves them to reinvigorate the commitments that have traditionally animated tax practice.