Enlisting the Tax Bar


Tax shelters and aggressive planning derive in part from a structural imbalance in our tax system that has not been adequately explored: In important respects, the private tax bar outmatches their counterparts in government. Although a strong policy case can be made for remedying this mismatch, this Article emphasizes two institutional barriers that complicate any solution, rooted in the political economy of taxation and the economics and professional norms of the legal profession. First, although it would be enormously helpful to dramatically increase the staffing levels and pay of government tax administrators, this is a politically daunting task. Second, a fallback strategy is to look to the private bar for help, but they face a significant conflict. Both market pressure and professional norms motivate them to serve their clients, who generally do not have an interest in improving government tax enforcement.

In light of these two challenges, what can be done to mitigate the mismatch between the government and private bar? This Article offers two sets of proposals. First, a number of suggestions focus on the government, offering ways to improve recruiting and reinforce the expertise at their disposal without dramatically increasing funding or raising pay substantially across the board. For example, the government should focus on recruiting senior lawyers out of retirement (whose financial demands will be limited) and having them mentor junior lawyers directly from law school (whose private sector pay is high, but not nearly as high as it will become in later years). The government should also consider a loan forgiveness program for these recent graduates, and should also enlist academics to assist with discrete projects. The government should also retain private law firms to litigate tax controversies with extraordinary precedential value.

Second, this Article offers guidance about the right way (and the wrong way) to tap the expertise and information possessed by the private bar. It is more effective, whenever possible, to ask lawyers to help the government in a way that also helps their clients. Using this principle, this Article identifies promising opportunities that have been overlooked, and critiques unpromising initiatives that have attracted significant government support. For example, clients do not want their own tax deals shut down, but they feel differently about their competitor’s deals, so the government should make more systematic use of this opportunity. Likewise, although clients are less motivated to help the government identify bad transactions that are inadvertently permitted, they are highly motivated to identify good transactions that are inadvertently prohibited. As a result, the government can look to the private bar for help in narrowing overbroad anti-abuse measures. On the other hand, in asking tax advisors to disclose their clients aggressive transactions ñ in effect, to rat on their clients ñ the government is asking for something that clearly is not in the clients interest. This reality is likely to undercut this initiative, which has been one of the centerpieces of the government’s efforts to date.

Tax shelters have proliferated in the United States not only because of financial innovation, the globalization of capital markets, the increasing complexity of our tax system, the inadequacy of tax penalties, the lack of political support for tax reform, and the growing popularity of textualist interpretation ñ all factors that have attracted considerable attention in the literature.[1] Shelters also derive from a structural imbalance in our tax system that has not been adequately explored: In important respects, the private tax bar outmatches their counterparts in government. This imbalance is one of sheer numbers, of access to information, and, at least in some cases, of sophistication and expertise. The problem is evident not only in the low audit rate,[2] a familiar issue, but also in the way the government staffs drafting projects and litigation. Unfortunately, this mismatch helps breed the familiar equity and efficiency concerns associated with aggressive planning.[3]

Even though a strong case can be made for remedying this mismatch, this Article emphasizes two institutional barriers that complicate any solution, rooted in the political economy of taxation and the economics and professional norms of the legal profession. First, although a dramatic increase in the staffing levels and pay of government tax administrators would be good policy, this is a politically daunting step. The levels of pay necessary to be competitive with top private lawyers sound utterly outlandish to the typical voter, since current government salaries (which are 90% less than what top lawyers earn) are already multiples of the median household income. Given the unpopularity of the IRS, moreover, underfunded tax administration has an almost irresistible political advantage: It is an under-the-radar tax cut ñ appreciated by those who benefit from it ñ that purports not to be a tax cut at all, but an effort to cut fat from an unpopular federal bureaucracy.

Since dramatic infusions of resources for tax enforcement are unlikely, a fallback strategy is to look to the private bar, enlisting their help in combating aggressive tax planning. This brings us to the second institutional source of the mismatch between the government and private bar: the conflict faced by the private bar. Both market pressure and professional norms motivate them to serve their clients, who generally do not have an interest in improving government tax enforcement. As a result, appeals to the bar’s conscience[4] or their commitment to professionalism[5] will be of only limited use.

In light of these twin challenges ñ the political difficulties of increasing funding and salaries and the structural difficulties of relying on private lawyers ñ what can be done to mitigate the mismatch between the government and private bar? Although there are no perfect solutions, this Article offers two sets of proposals. The first group focuses on the government, offering ways to upgrade the government’s expertise without dramatically increasing funding or raising pay across the board. For example, the government should focus on recruiting senior lawyers out of retirement (whose financial demands will be limited) and having them mentor junior lawyers directly from law school (whose private sector pay is high, but not nearly as high as it will become in later years). The government ought to consider a loan forgiveness program for these recent graduates. In addition, the government should enlist academics to assist with discrete projects, and should retain private law firms to litigate tax controversies with extraordinary precedential value. This Article considers and rejects a more radical proposal for the government to rely on private attorneys general to aid in tax collection.

Second, this Article offers guidance about the right way (and the wrong way) to tap the expertise and information possessed by the private bar. It is more effective, whenever possible, to ask lawyers to help the government in a way that also helps their clients. Although this principle is straightforward ñ even obvious ñ it has not been invoked, at least as far as I know, to determine which functions the private bar can usefully perform in this context. Using this principle, this Article identifies promising opportunities that have been overlooked, and critiques unpromising initiatives that have attracted significant government support. For example, clients do not want their own tax deals shut down, but they feel differently about their competitor’s deals, so the government should make more systematic use of this opportunity. Likewise, although clients are less motivated to help the government identify bad transactions that are inadvertently permitted, they are highly motivated to identify good transactions that are inadvertently prohibited. As a result, the government can look to the private bar for help in narrowing overbroad anti-abuse measures. On the other hand, in asking tax advisors to disclose their clients aggressive transactions ñ in effect, to rat on their clients ñ the government is asking for something that clearly is not in the clients interest. This reality is likely to undercut this initiative, which has been one of the centerpieces of the government’s efforts to date.

The first part of this Article elaborates on the mismatch between the government and the private bar, explaining why the problem is so entrenched. Part II considers ways to hire better government lawyers, both for full time responsibilities and for discrete projects. Parts III and IV focuses on how to tap the expertise of private lawyers without hiring them directly. Part III explores initiatives in which the interests of the government and the client are aligned, at least to an extent, so that the bar faces less of a conflict in assisting the government. In contrast, Part IV explores initiatives in which the interests of the government and client are in conflict. These initiatives can succeed only if the government can detach lawyers from their client’s interests with sufficiently powerful carrots or sticks. This is possible, in some cases, but it is never easy. Part V is the conclusion.

I. The Entrenched Mismatch Between the Government and Private Bar

There is an old story about someone who loves Beethoven and is invited by his brother to hear a performance of the Ninth Symphony. Soon after agreeing enthusiastically, he is disappointed to learn that the symphony will be performed by his niece’s Middle School orchestra.

The point here, obviously, is that a symphony is beautiful only if played properly. Likewise, a set of tax rules will have their intended effect only if administered properly.

The U.S. tax law is filled with complex rules that aspire to capture economic nuance. Yet these rules can only be effective if they can be implemented ñ not by some idealized tax administrator, but by the flesh-and-blood administrators we have. In short, the quality of the people who draft rules, audit returns and litigate controversies is profoundly important to the integrity of our system.

This Part argues that there is a mismatch in expertise and numbers between the government administrators who seek to collect tax from sophisticated taxpayers and the private lawyers who represent these high net worth individuals and large corporations. (The balance is different for the average taxpayer, but that aspect of our tax system is beyond the scope of this Article.) This Part then offers a theory about why this mismatch is difficult to correct, grounded both in political economy and in the economics and professional norms of the legal profession.

A. The Mismatch

The aggressive tax planning of sophisticated taxpayers relies on three familiar failures of government tax administration. Although a broad phenomenon within our tax system, these failures are easiest to see with tax shelters ñ transactions that rely on a strained reading of the relevant tax provision to claim inflated deductions, to avoid including otherwise taxable income, and the like.

First, shelters (and, indeed, all aggressive planning) exploit poor drafting of statutes and regulations. The relevant rules are capable of being read (albeit aggressively) to allow, for example, tax losses with no corresponding economic losses. Drafters need to anticipate this sort of misreading more effectively. This task is especially important ñ and especially difficult ñ when judges focus on text, instead of on congressional purpose, and construe ambiguities against the government. Textualist judges cannot be counted on to ask, why would Congress allow such a generous result? Instead, they consider it the job of Congress or the Treasury, not the courts, to shut down abusive transactions. Meanwhile, the practice in Congress and Treasury is to act prospectively, grandfathering early deals. This means that drafters cannot afford to make mistakes ñ but, unfortunately, they often do. It is extremely difficult even for an experienced drafter, who has ample time, to anticipate all relevant contingencies. This is all the more true when language is cobbled together in the middle of the night ñ as is often the case ñ by drafters who do not have deep experience practicing in the relevant area.

Second, tax shelters obviously take advantage of poor auditing. Even if the taxpayer’s argument on the merits is weak, the tax will never be paid if the transaction is not challenged on audit. Taxpayers have been encouraged to play this audit lottery by the low percentage of high income returns that are audited, as well as by the unsophistication of some auditors and their auditing techniques.[6] Indeed, tax shelters often have extraneous pieces that are included solely to befuddle auditors. Facing a large and complicated return, auditors try to intuit what questions to ask, without really knowing where the bodies are buried. Some auditors are further weakened by a civil servantís nine-to-five mentality. The key to a successful audit [from the taxpayer’s perspective], the tax director of a large multinational once told me, is to give the auditor an office near the elevator: this way, the auditor will feel comfortable going home early, since he won’t have to pass anyone else’s office on the way out.[7]

Third, even if the auditor discovers the aggressive transaction, the government litigators who pursue the matter usually are more leanly staffed and less experienced than taxpayer’s counsel. The government must settle most of its cases and, when it litigates, it sometimes offers concessions that it should not offer[8] and, more generally, loses cases it should win. These lost cases have a troubling ripple effect, serving as precedent that aggressive taxpayers in turn use to justify other aggressive transactions. This prospect of adverse precedent is all the more reason for the government to avoid risky litigation, a reality that is not lost on aggressive taxpayers. They know that, even in the worst case in which their transaction is identified on audit, they have a good chance to settle on favorable terms (e.g., no penalties) or, if necessary, to litigate with some prospect of success.[9] So why not roll the dice?

The mismatches in each of the three phases described above ñ in drafting rules, in auditing returns, and in litigating cases ñ all arise, to a significant degree, from a common source: the limited resources allocated to tax administration. This underfunding itself has three manifestations. First, the government is chronically understaffed. At the drafting stage, there are simply not enough people to plug holes that are constantly being detected in the tax law ñ let alone to think proactively about how to improve the system. Statutory gaps and ambiguities endure for years, while only a few technical corrections are enacted every session.[10] Likewise, the Treasury often takes years to issue draft regulations on an issue, and then still more time to finalize them. There are simply not enough people to do this work. The same is true for auditors, who must comb through piles of documents in a treasure hunt for inappropriate items. It is not unusual for one or two auditors to sit in a room with two or three times as many representatives for the taxpayer. This pattern repeats in litigation, where a young Assistant U.S. Attorney, who is devoting a small fraction of her time to a case, opposes a team of lawyers for the taxpayer.

The second manifestation of tight budgets is that government tax administrators are not of uniformly high quality or deep experience. Many are relatively new to the profession, signing up for government service as a way to accelerate their careers. The elephant in the room is the enormous pay differential between government and the private sector. A junior lawyer in the I.R.S. will earn half what a junior lawyer at a firm makes, and the gap grows significantly wider as a lawyer becomes more senior, so that senior government officials earn less than ten percent of what some senior private attorneys earn.[11] The government offers offsetting advantages, of course, such as work that is extremely interesting, the power to shape the tax system, the satisfaction of serving the country, and, as noted above, the chance to build a professional reputation quickly. The reality is, though, that many of the government’s best people stay only a few years, leaving ultimately for more lucrative private sector responsibilities. In addition, many of the nation’s best tax lawyers never serve in government at all.

In addition to being understaffed and underpaid, the government is at a third disadvantage, relative to the private bar, which is as much structural as financial: they suffer from an imbalance in information. In a self-reporting system, the initiative lies with taxpayers. They have the freedom to structure their affairs, inevitably choosing the tax efficient course.[12] They have the further advantage of controlling the flow of information to the government, offering only what the government requests (and, even then, typically construing the government’s request in a self-interested way). Without a significant investment in gathering information, the government may have to wait years before discovering a new tax planning technique. There is also the risk that, by the time they do discover it, they will feel constrained not to challenge it, given the heavy volume of transactions that already have taken advantage of it, and the political clout behind those transactions.[13]

Readers who are not tax experts may wonder whether the mismatch described here is unique to tax administration. Is there a parallel mismatch between SEC lawyers and the securities bar that represents large public companies, or between prosecutors and elite white collar lawyers, or between EPA lawyers and the elite environmental bar? Although the question is beyond the scope of this paper, it is worth a brief comment. I suspect that such a mismatch does exist in other areas of law, but that the tax mismatch may be especially significant for three reasons. First, government lawyers are not backstopped by private attorneys general as they are, for instance, in the securities field by the plaintiff’s bar. Second, tax rules generally are written narrowly and precisely ñ unlike some criminal statutes, for instance, which cover more conduct than is really culpable, and leave it to prosecutorial discretion to determine who is prosecuted. As a result, the tax authorities are more likely to face conduct that violates the spirit, but not the letter, of the regime, and thus the lawyering challenge of using not-quite-on-point rules to stop it. Third, the tax regime ñ for capital, especially ñ may be more malleable than other regimes, such that good lawyering (and, correspondingly, mismatches in lawyering between the government and private bar) are especially significant. In other regimes, unalterable physical realities often backstop legal distinctions, so that there is only so much a lawyer can do to change the result. Even if one country’s environmental laws are more favorable than another, a lawyer can’t move a factory from one jurisdiction to another. But a tax lawyer can easily shift certain types of income from one jurisdiction to another without changing anything substantive; for example, does it matter which corporate entity ñ the Delaware subsidiary or the Cayman Islands subsidiary ñ holds the business’s intangibles? Likewise, if environmental laws treat cars and trucks differently, there is only so much a lawyer can do to classify a car as a truck. Yet a tax lawyer would not say the same thing about the debt-equity distinction. In the taxation of capital especially, frictions are weak, so that the quality of lawyering alone makes an enormous difference.

B. Why the Mismatch Is So Entrenched

Since the mismatch between government and private lawyers is an important cause of aggressive tax planning, it would be eminently sensible to correct the mismatch. The government could raise more revenue, and could do so more fairly and efficiently.

Aggressive taxpayers would lose their edge over conservative ones, and also would see less advantage to investing time and resources in tax planning. Indeed, with better drafting and enforcement, the government could cut rates without losing revenue. Thus, although additional government expenditures on tax administration are admittedly a form of social waste, they foreclose another form of waste ñ private expenditures on tax planning. As long as spending another dollar on administration reduces at least a dollar of planning waste, the expenditure is socially worthwhile.[14] For example, if the government invests in writing a well-crafted regulation to resolve an ambiguity in the law, this up-front expenditure saves the tax bar countless hours of analysis about how to handle the ambiguity, and also heads off aggressive strategies that take advantage of it.[15]

Yet although the advantages of better administration may seem irresistible, they are consistently resisted. This Section offers a theory about why the mismatch between the government and private bar is so deeply entrenched. In essence, there are two main paths away from the status quo, but neither is easy. The first, involving greater government expenditures on administration, is blocked by political constraints, since overly-lean enforcement functions as a tax cut with an almost irresistible political advantage: it can be presented not as a tax cut at all, but as an effort to wring the fat from the unpopular I.R.S. bureaucracy. The second path away from the status quo, involving greater public responsibilities for private lawyers, is obstructed by the professional norms and economics of the legal profession, since private lawyers face a significant conflict in helping the government combat aggressive planning. Although some progress on each dimension is possible, this is a difficult problem to solve.

1. Politics: The Constraint on Enlisting the Public Bar

Before explaining the political obstacles to better tax administration, it is worth emphasizing that there are, of course, political factors that counsel in favor of this goal. Making the tax system more efficient and equitable is good policy, which obviously should have some influence on political actors (although one has to be naive to think such considerations are dispositive). For a political actor focused on advancement, additional revenue is the mother’s milk of a career, since it can pay for profile-raising tax cuts or programs (but, of course, only if the political actor is able to steer this revenue in her preferred direction). Likewise, I caught the tax cheats is an appealing reelection slogan, especially when tax fraud is salient in the public mind.

For a number of reasons, however, it is politically difficult to introduce substantial increases in the staffing or salaries of government tax lawyers. For one thing, taxes on sophisticated taxpayers ñ or, at least, taxes on investment returns ñ are controversial. Within the academy and the body politic, a significant group believes that investment returns should not be taxed (a position with which I have a great deal of sympathy). Likewise, there are concerns that high marginal rates on wages discourage economic growth. Although no one would defend poor enforcement as a first-best method of reducing taxes (since rate cuts are always more efficient and equitable), some might opt for poor enforcement as a second-best means of cutting a tax that can’t be cut more directly.

Indeed, hidden tax cuts are a deeply entrenched feature of our political life. For example, the difference between an effective loophole-closing measure and an ineffective one tends to be quite technical, so that the general public and the media often cannot discern the difference. Indeed, only a small cohort of trained experts can truly tell whether the shelter is being shut down, and these experts have developed their expertise in the service of taxpayers ñ often, the very taxpayers who have been using the shelter. Given how hard it is for the public to monitor these technical issues, it is tempting for self-interested political actors to propose half-measures that are not fully effective. These partial responses offer their sponsors the best of both political worlds, allowing them to score points with the general public for shutting down the shelter while, ironically, also scoring points with the affected taxpayers for doing so in a reasonable way that leaves them other deals to do.[16] Similarly, passing rules that seem to respond to tax planning, while providing insufficient funding for the effective administration of these rules, allows political actors to claim credit with voters without actually impeding, and thus offending, aggressive taxpayers.[17]

While the political support for taxes is unenthusiastic at best, the political support for tax enforcement and administration is even more ambivalent. No one sails to reelection on the platform, more money for the I.R.S! Indeed, consider what would happen if the Washington Post began publishing stories about the millions spent on hiring elite tax lawyers, while lunch money for public school children was being cut. The reaction of members of Congress is easy to predict. Needless to say, U.S. voters have little affection for the I.R.S. Perhaps there is an element of national mythology at work here. The tax collector is unlikely to be a hero in a nation founded in a rebellion against British taxes. The rhetoric of limited government and individual initiative is pervasive in U.S. public life. This anti-tax collector feeling is certainly reinforced by the bad experiences almost everyone has had, at some point, with the lower level I.R.S. bureaucracy. Obviously, there is a political constituency that wants to stoke this anti-I.R.S. feeling: high income taxpayers who benefit from lax enforcement. As noted above, providing insufficient funding for tax administration is a way to cut taxes that can be presented, not as a tax cut, but as a good government initiative to trim waste from an unpopular federal bureaucracy. As the Republicans showed in hearings on I.R.S. misconduct in the late 1990ís, tax administration is an easy political target.[18]

Raising the salary of government tax lawyers, so that pay is more competitive with the seven figure salaries earned by top private lawyers, is an especially hard sell. The essential problem is that these private lawyers earn so much more than the median household income, which currently is just over $46,000.[19] A salary three times this level for senior government lawyers seems quite high to the median voter, but is only a small fraction of what a top lawyer earn in private practice. Likewise, it is difficult to raise the salary of government tax lawyers, without also raising the salary of other government experts who are also underpaid. To pose the most extreme case, should leading tax administrators really earn as much as (or more than) the President?20

This is not to say that government enforcement cannot be improved at all. The case on the merits is powerful, as is public opposition to tax fraud, as well as the incentives of public actors to raise revenue. The point is that significant political constraints limit what can be done, requiring incrementalism, careful thought, and compromise. In light of these limits, it is also worth considering whether private lawyers can be induced to supplement these efforts. Put another way, the mismatch between the government and the private bar is likely to endure, at least to a degree. The question, then, is whether the government can somehow make use of the private bar’s expertise, without hiring them directly.

2. Conflicts: The Constraint on Enlisting the Private Bar

In other words, can the private tax bar be enlisted to help the government? For example, can private lawyers be interrogated to discover where the loopholes are in particular rules? The reality is that it is quite difficult to force the private bar to do what they do not want do to.

At a minimum, there is a significant political problem. The opposition of bar associations, for instance, can be a formidable political hurdle.[21] This is not to say that it is insurmountable. Particularly in times of crisis, when the public eye is focused on a problem, additional mandates sometimes can be crammed down the throats of professional advisors. This is something securities lawyers have seen recently in Sarbanes Oxley and, to an extent, tax lawyers have seen as well in the government response to tax shelters. But in ordinary political conditions, the general public remains blissfully ignorant of the details of a technical regime such as the tax law, and experts represent one of the key constituencies that understand, and actually focus on, this body of rules. Under these circumstances, it is difficult to force experts to discharge a responsibility they do not want.

Even if the political will can be mustered, there is the further problem of ensuring compliance. Will the private bar actually do what they have been asked to do? At first blush, this is a surprising problem to raise. After all, lawyers are as likely as anyone to obey the law ñ more so, really. But the point here is that there is a difference between actually complying ñ in the sense of truly giving the government the help they are seeking ñ and merely going through the motions. Indeed, tax lawyers are a creative bunch, and they know how to avoid providing help, while seeming to do what is asked of them. Needless to say, you can count on tax lawyers to find ambiguities in almost any government instruction. Of course, if the government had perfect information and ample staff to monitor whether lawyers were complying with these directives, then lawyers presumably would comply more faithfully. But obviously that is not the case. Indeed, the reason private lawyers are being asked to help is that government capabilities are limited.

The broader point, then, is that if the tax bar doesn’t want to help, they will find ways to avoid helping. This brings us to an essential question: When, if ever, will the tax bar be willing to help? Do they ever have the incentive to assist the government? I believe the answer is yes in certain special circumstances, which the government should exploit to the fullest. But before we get to these exceptional circumstances, we should begin with the general answer, which is a resounding no. Most of the time, private lawyers can not be counted on to help the government for two self-evident reasons.

First, and most importantly, lawyers are eternally mindful of who pays their bill. They worry about keeping their clients, and about attracting new ones. Clients obviously want to pay as little tax as possible. They certainly do not want their paid advisors to help the government raise their tax bill. A tax lawyer who is known to help the government collect more tax from his client is unlikely to remain a tax lawyer for long. More generally, lawyers are looking to advance the interests of their client, and, ordinarily, a client’s interests are adverse to those of the government.

Professional norms supply a second reason why lawyers will be reluctant to help the government. In our adversarial system, lawyers generally owe duties to their own client, but not to the other side. In tax planning, the other side is the government, and even when a lawyer is giving advice about planning, she focuses on what would happen if the matter is litigated. Will the client’s strategy succeed if tested in court? How can the client’s position be strengthened? To a tax advisor who is thinking through these issues, the government is, quite frankly, the enemy. The client, meanwhile, looms very large in the lawyer’s professional obligations. Stirring rhetoric is often used to describe the attorney-client relationship ñ a sacred trust, and the like. The lawyer is supposed to advocate for a client’s interest zealously (within the bounds of the law, of course), and the lawyer is supposed to keep a client’s confidences. The details of a client’s transactions are a client’s property, and a lawyer generally is not supposed to disclose them to anyone. Given these client-focused professional norms, it is not a comfortable thing to provide help to the government. Although a tax lawyer may owe a separate duty to the tax system ñ a point discussed below ñ her primary allegiance is to her client.

To reinforce this point, consider the contrast between tax shelters and the corporate scandals of the last decade. At first blush, there are obvious similarities. They each involve the manipulation of systems of financial reporting (whether accounting or tax), usually through strained readings of the relevant rule, paired with complex structures designed to obscure economic reality. Yet a key difference is who is being deceived. It is shareholders in the case of corporate scandals, versus the tax authorities in the case of shelters. In the corporate context, it is a matter of concern that multiple layers of gatekeepers failed to protect shareholders, including research analysts and boards of directors.[22] Yet at least there are multiple layers of gatekeepers in the corporate context; there are no such private gatekeepers for the tax system. For example, a research analyst would rarely think she was obligated to notify the I.R.S. upon discovering that a company is engaging in tax shelters.

So far, in assessing the difficulty of arranging collaborations between the bar and the government, I have emphasized problems with the bar’s incentives. But it is also worth observing a problem on the government’s side. If they do not have to pay for the bar’s time, the government is less likely to use it efficiently. The government may be tempted to assign the bar time-consuming tasks that are not cost justified. As with any unfunded mandate, the costs do not fall on the government. The point should not be overstated, since there are political costs in wasting the bar’s time. But, as we shall see, there is a regrettable element of makework in some of the initiatives the government has proposed.

To sum up, then, it is difficult for the government to enlist the tax bar’s help, given the bar’s incentives and, to a degree, the government’s as well. But the operative word is difficult, not impossible. Careful thought is required, and there obviously are some things that do not work, but there also are some that do.

Part II considers various ways the government can hire private lawyers directly, consistent with the political constraints described above. Parts III and IV discuss a range of ways the government can tap into the expertise of private lawyers without hiring them, while at the same time taking account of, and hopefully mitigating, the conflict discussed above.

II. Enlisting Government Lawyers

To sum up, the government does not have enough people working on tax administration, and the people it has are not all up to the task. There are also political constraints on what can be done about this problem. This Section considers a range of ways to enhance the capabilities of government enforcement, including strategies for recruiting top talent for full time positions and new sources of expertise who could be retained for discrete projects.

These government-focused strategies have the important advantage of (largely) avoiding the conflict that arises when the private bar’s assistance is sought. Yet in steering clear of the Scylla of conflicts, they veer closer to the Charibdis of political constraints, described above; thus, the effort here is to focus on ideas that either work within the government’s pay scale or can be justified by politically plausible rationales. These ideas are meant to be doable ñ or at least not too much of a stretch ñ in light of the political constraints described above. The Section then concludes by considering (and rejecting) a more radical step in which private attorneys general would be enlisted to backstop government enforcement.

A. Full-Time Recruits

1. A Great Place to Start ñ and Finish

In recruiting top talent, the government should target people at stages in their career in which the pay gap is least significant. An important category is recent law school graduates. Although the pay gap is still significant, it is substantially less than the gap becomes once these graduates become partners at top firms (e.g., a 50% pay cut instead of a 90% pay cut).[23] The government also can offer several non monetary benefits that private practice does not offer. Because government staffing is so lean, these junior lawyers will bear significant responsibilities very quickly, developing invaluable skills and reputations. Their hours will also be more manageable. They also will have the satisfaction of performing a valuable public service. Admittedly, the fact that these young lawyers will have a long career ahead of them ñ which is likely to include service to private clients ñ may introduce some conflict issues, as they may steer clear of offending potential employers; but this traditional concern about revolving doors should not be overstated. After all, potential clients should be more impressed by (and thus more likely to hire) government lawyers who effectively advance the interests of their client (i.e., the government) than those who do favors for the other side. To his credit, Donald Korb, the Chief Counsel of the I.R.S., has recognized the government’s comparative advantage with younger lawyers, and has launched the Great Place to Start initiative, designed to recruit them while they are in law school.[24]

Congress would be wise to complement this effort with a commitment to repay the student loans of a select group of highly capable graduates who enter government service. This is necessary because tuition at top law schools is more than $40,000 per year. When room and board is added in, students are spending over $200,000 in out-of- pocket costs on their legal education, much of which is financed by debt. They may also have student loans from college. This reality significantly undercuts the otherwise promising strategy of targeting recent law school graduates. Even though the pay gap is narrower for recent graduates, their loans render them less able to afford any gap at all. Congress could solve this problem by forgiving a fraction of a student’s loans for every year she works for the government (coupled with a commitment from the student to work a minimum number of years). For example, if a student could emerge debt-free after working six years for the Treasury department, government service would become extremely attractive. Of course, this is functionally similar to a pay increase, but structuring a raise in this way has political advantages. The government isn’t committed to give this pay increase across the board, since a principle can be offered to distinguish these needy junior lawyers from other employees. Also, a sympathetic story can be told to voters about why this program is necessary.

While junior lawyers should be an important component of the government’s hiring strategy, they need more senior lawyers to mentor them. Another promising pool is made up of elite lawyers retiring from law firm practice. The pay differential should be less significant for this group as well, since they will have accumulated significant savings; indeed, if they were prepared to accept no salary income as retirees, they should be satisfied with the government’s pay scale. They may also be motivated by a desire to give back to the tax system. They also have deep experience. Mentoring a team of young and highly motivated junior lawyers could be quite an attractive proposition, and could leverage their expertise without placing excessive demands on their time. They could slow down from the demands of a top private practice, while doing work that is interesting and meaningful.

2. First Response Team

While the focus above is on remaining (largely) within the government’s existing pay scale, there is much to be said for increasing the pay scale significantly for a small cohort of top people. What if the government hired two dozen partners from leading firms at market rates ñ or, for that matter, even at half or a quarter of what they have been earning in practice? Their expertise could be spread across the various regimes that are relevant to sophisticated taxpayers, including financial products, international tax, partnerships, and corporate tax.

With this expertise, the government stands a better chance of winning more precedent-setting litigation and of writing tighter regulations. This group could also serve as a first-response team. As soon as they learn of an aggressive transaction (e.g., by reading one of Lee Sheppard’s columns, by receiving an anonymous tip, or by reviewing disclosure received from a taxpayer), they could invite the taxpayer and their advisor to defend the transaction, and then make a judgment about whether to take special action to shut the deal down quickly (e.g., through a notice issued immediately), or to allow the government to consider the deal through its usual (slower) process.[25] They can also review draft regulations and statutes to screen them for potential loopholes, and they can give advice to government litigators about strategy.

Although this proposal involves a significant pay increase, it limits the increase to a small number of government lawyers, so that the budgetary impact is a rounding error within the tax enforcement budget. The cost would easily be justified by increases in revenue and significant reductions in the social waste from tax planning. Admittedly, there is the problem of inequity within the ranks of government employees. The pay scale for this team could breed resentment among other government lawyers, and could also become the mantra for disgruntled advocates of other programs: If we can afford to pay a salary of $X to a government tax lawyer, they will say, then surely we can afford my program. This proposal is not free of political issues, then, the benefits would be quite significant if it was adopted.

B. Project-Based Recruits

In addition to recruiting full-time employees, the government can also reinforce tax administration by adding part-time employees (and, in some cases, volunteers) to assist with particular matters. This Section considers some potentially promising options that should be relatively affordable and conflict-free.

1. Legal Academics

An important source of experts, who will not face the same incentive problems as private lawyers, are legal academics.[26] Tax professors have expertise in both legal and policy analysis, and many have practice experience. Assuming they do not have private sector clients ñ and many do not ñ they do not face client conflicts in assisting the government with law reform or litigation.

To my mind, academics are a significantly underutilized resource. In part, the reason is that academics sometimes focus on research that does not have immediate practical application. This research is still valuable, of course, in illuminating fundamental questions about the ideal system, in exposing hidden premises, and the like.

But it is regrettable that some academics believe this is the only research that advances their reputations within the academy, and that more grounded projects are thought to be less helpful or even counterproductive. Academics should devote some of their time to research that has a practical payoff, critiquing specific cases, rulings, or statutory provisions and offering reform alternatives that are capable of implementation. For example, Alvin Warren has written a thoughtful article on the challenges of taxing a company on transactions in its own stock. Dan Shaviro has written a valuable article on the interest allocation rules, and David Weisbach has recently advised the President’s commission on tax reform. These sorts of projects can have an enormously positive impact on our system.

To do this sort of research effectively, academics should view interactions with the government as valuable field research, exposing them to novel issues and enriching their understanding. In exchange for access, some academics would share their expertise free of charge or, certainly, for relatively modest levels of funding. An added benefit to academics is that they can use the relationship to educate the government about their ideas, thereby increasing the probability that their proposals will be implemented.

2. Bar Associations

Although academics can be helpful, private lawyers obviously have an even deeper reservoir of knowledge about aggressive tax planning. The challenge, of course, is to find ways for them to offer this knowledge free (or, at least, relatively free) of the conflict they face in helping the government.

To an extent, bar associations can provide this sort of setting. For example, the New York State Bar Association’s Tax Section writes influential reports about regulatory projects and other reforms. It is a matter of pride for the group that they focus on systemic concerns, not taxpayer interests. The government obviously is well aware of these reports, and relies on them for technical advice. As Professor Rostian has emphasized, the ABA and the NYSBA have taken positions on tax shelters that run contrary to the interests of their clients.[27] An important reason why they can do this, which she does not mention, is the protection offered by collective authorship. Although reports have principal authors, these authors speak for the group, not for themselves. As a result, it is safer to expose client abuses in these collectively written reports, as opposed to solo-authored articles. This safety in numbers phenomenon is at least a partial remedy for the conflict

Of course, the idea that bar associations are vehicles to help the government should not be overstated, since they sometimes serve as interest groups to champion the interests of lawyers and their clients.28 Indeed, another way to interpret the bar association’s position on tax shelters is as an effort to protect the interests of the elite bar. The comparative advantage of elite lawyers is in giving accurate predictions about what a court will do and, in the case of tax shelters, this often means saying no. If government enforcement is weak, though, taxpayers may be less interested in accurate advice (the tax bar’s comparative advantage), and more interested in a less expert (and less principled) advisor who will simply facilitate the aggressive transaction.29 In response to this competition (from accounting firms, among others), it is rational for a self-interested elite lawyer to want government enforcement to improve as a way to raise the value of their services, and thus to help them beat back the competition.

3. Law Firms

In addition to relying on private lawyers to volunteer their services through bar associations, the government can also retain elite practitioners for discrete projects instead of as full-time employees, just as corporate clients do. This course has the advantage of using elite lawyers most efficiently ñ only when their services are especially valuable. This approach also avoids the need to raise government salaries across the board or to tolerate significant inequities in the pay scale for in-house government lawyers. Although the legal fees could be substantial ñ itself a separate political issue, to be sure ñ it is possible that private lawyers will offer a discounted rate in return for the prestige of working on an especially important matter. In any event, the economic benefits of having top talent involved should justify the cost, for all the reasons mentioned above. Obviously, the government entered into this sort of arrangement with David Boies, a top antitrust litigator, to represent the U.S. government in its suit against Microsoft.[30]

There is still the issue of conflicts, though. Will a lawyer feel comfortable representing the U.S. on tax issues, or will she (and her partners) be concerned about losing private clients? Can a private lawyer represent the U.S. faithfully without also worrying about the interests of their other clients?

The context that seems freest of conflicts ñ and thus most promising ñ is tax litigation. In cases with significant precedential value, the government would be well advised to hire a leading practitioner, and this lawyer will have clear incentives to win. After all, success or failure is highly visible, so that the lawyer will have strong reputational interests in providing vigorous advocacy. Of course, if a lawyer helped develop an aggressive strategy, she is not the right person to litigate the issue for the government. The same probably is true if her partner developed the strategy. Given the large pool of private sector talent, though, it ought to be possible to retain a tax litigator who has not had contact with the relevant issue before.

In addition to litigation, elite law firms could advise on regulatory projects and statutory drafting. They could be asked, for example, to draft a regulation. The government could then review the work, but this input from leading lawyers could move the process along more quickly. It is possible, moreover, that lawyers would work for a discounted rate, or even for free, if they were given credit for their work.[31] A practitioner could find it valuable, in attracting clients, to be able to say that she drafted a relevant regulation. In a sense, the government would be paying in prestige, rather than in cash.

However, the conflict issue seems more difficult in this context than in litigation, in part because the scope of the project is less contained. A lawyer’s other clients will not be interested in the details of a litigation (e.g., what discovery motions are filed, what the opening statement will be, etc.), but they will be very interested in the progress of a regulatory project. Can a lawyer share this insight with clients? Correspondingly, if a lawyer learns about planning opportunities through work for clients, can the lawyer use this knowledge in helping the government to tighten up the relevant provision? If they do, clients will become more wary of consulting them in the future. Likewise, there is a risk that a practitioner will deliberately leave gaps in a regulation he drafts for the government, and then will steer his clients through these loopholes. To an extent, though, this risk is mitigated if government lawyers (and even other private practitioners) review the language before it is finalized.

C. The Promise and Pitfalls of Private Attorneys General

The problems of government resource constraints and private sector incentives are not unique to the tax system. For example, the S.E.C. does not have a large or highly compensated staff. Neither does the E.E.O.C. As noted above, an important difference is that in these regulatory areas, and others as well, the government authorizes private actions. As a result, plaintiffs lawyers, working for contingent fees, are supposed to pick up some of the slack.

The tax system does not have private actions under current law but, at least in theory, it could. Congress could authorize qui tam actions, in which private citizens sue to collect taxes for the federal government, and keep a share of the recovery as a contingent fee. Before briefly surveying the merits of these private tax actions, I should state at the outset that this idea is politically implausible, at least for the foreseeable future. Not only is tax enforcement unpopular, but so too are trial lawyers, and this idea combines both.

Even so, private tax actions do address the problems identified in this paper, at least to a degree. Potentially lucrative contingent fees could attract a corps of extremely capable lawyers, who would be more of a match for the private bar than government lawyers currently are. The allegiance of these lawyers would be to the tax base, not to taxpayers. The prospect of being sued is likely to deter some aggressive planning.

Although this idea has advantages, it has many serious problems as well, which, on balance, leave me skeptical. This idea is worth an article of its own, but I will use only a couple of paragraphs to highlight four problems First, as with other class action or plaintiffs suits, there is the concern of frivolous strike suits. It is commonly observed that plaintiff’s lawyers sue corporations every time the stock price declines, and that it is often cheaper to pay them off than to litigate.[32] By analogy, it obviously is problematic for taxpayers to feel pressured to settle, even if their transaction is uncontroversial. This problem might be ameliorated, to a degree, by requiring private tax collectors to secure approval from the government before bringing suit, and again for settlements. But government supervision is a solution only if government officials have an incentive to say no to frivolous cases, and the time and expertise to discern the true value of a case. I doubt they would, so I view the problem of frivolous suits as extremely serious.

Second, there is the risk of collusion between taxpayers and these private tax collectors. For example, if a taxpayer can secure a final judgment in settling a private suit ñ and thus is insulated from other suits, including by the government ñ there is room for a deal at the government’s expense. The taxpayer offers the private tax collector a settlement, complete with a generous fee, for very little work; in return, the taxpayer is protected from further trouble. Indeed, taxpayers are even happier if, in settling, they assure no further action ñ not just on the narrow issue that was settled ñ but on a more broadly defined set of issues that are related in some way to the one that was settled. The potential for abuse is easy to see.[33] Government or judicial oversight of settlements helps address this problem, but, again, only if it is effective. It is worth emphasizing a reason why government oversight is unlikely to be effective: the premise behind creating a class of private attorneys general is that the government is overstretched. If private attorneys general require significant government oversight, they have much less to add to the system.

Third, there is a tradeoff between protecting taxpayer privacy and arming private tax collectors with the information they need to identify underpayments and bring suit. Under current law, taxpayer returns are confidential, and the I.R.S. does not share them with anyone ñ not even with other parts of the federal government, such as the SEC.34

Under these conditions, private suits obviously are not possible. The government would have to allow some disclosure. This is not necessarily a problem in principle, since the privacy concerns that apply to an individual seem less compelling for a public company. It may be that these returns, or designated excerpts from them, could be publicly disclosed. Private tax collectors might then be given special discovery power (perhaps under government or judicial supervision), subject to a confidentiality obligation. Yet this is a logistically challenging road. Another alternative is to give private tax collectors access to the special tax shelter disclosures mentioned above, and to authorize these private actions, and whatever further discovery is necessary, only for such transactions. Needless to say, this would be extremely controversial.

Finally, another potential problem with private tax actions is the somewhat counterintuitive problem of over enforcement. In the current environment, it is perhaps hard to imagine too many resources being devoted to tax enforcement, since our current system is now so overstretched. But just as we do not want to underinvest in enforcement, we also do not want to over invest. For example, a private tax collector would have the incentive to spend up to one dollar to collect a dollar fee. If the contingent fee were one third, private collectors would be willing to spend up to one dollar to collect three dollars in tax, which is almost surely too much.[35] To an extent, the problem can be mitigated by mandating that fees cannot exceed a fixed percentage of the tax collected (e.g., five percent). But even then, the government should be wary of surrendering control over the number of suits brought, or the minimum level of tax that must be at issue in order to justify a suit. This problem is likely to be solvable, but only with careful thought and a fairly elaborate administrative apparatus supervising the private tax collectors; but, of course, having to field an elaborate system of supervision undercuts the original advantage of private suits, i.e., that they supplement scarce government enforcement resources.

III. Enlisting Private Lawyers: Find (or Create) Common Interests with the Client, and the Lawyer Will Follow

Now that we have considered various ways in which the government can hire advisors to work for them directly ñ either as full-time employees, on a project basis, or as a team of private attorneys general ñ the next question is whether the government can draw on the expertise of private lawyers without hiring them. Given the political constraints on hiring elite lawyers directly, it would be desirable to tap into their expertise and information some other way.

A range of ideas are considered in the next two Parts, and some are more promising than others. This Part identifies a range of contexts in which the interests of the government align with the interests of the client. Once this occurs, the tax bar is motivated to help the government, since the bar is advancing client interests in the process. The next Part moves to the harder set of cases in which the client’s interests do not support cooperation with the government.

A. Finding Common Interests With Taxpayers

Since lawyers have ample reason to champion their client’s interests, the government can take advantage of the private bar’s expertise in situations in which the client wants to cooperate with the government. This Section considers situations in which the client is already motivated to cooperate. Once the client wants to help, the lawyer obviously will do so dutifully ñ and, perhaps, even cheerfully.

1. Taxpayer’s Incentive to Kill a Competitor’s Deal

The most straightforward example ñ but one that has not received enough systematic attention ñ is for the government to take advantage of rivalries among taxpayers. For instance, if one investment bank (Innovator) has developed a tax advantaged strategy, and a competing investment bank (Competitor) is not able to offer it ñ either because they think it does not work, or because they are entering the market too late ñ Competitor has the incentive to kill Innovator’s deal. Competitor will want its lawyer to notify the government of this new strategy, and to explain its vulnerabilities. This might be done discreetly (e.g., by calling or even sending an anonymous envelope to the I.R.S.), or it might be done openly (e.g., by publishing an article in Tax Notes). They are more likely to be discreet in order to avoid damaging their reputation (e.g., with the rival investment bank’s clients), but, either way, the government gains access to valuable information and expertise.

This dynamic is not unique to the financial sector or to those who develop shelters. Potential users can be a resource too. Consider two rival consumer products companies, of which one has an aggressive tax director (Aggressive Co.) and the other has a conservative tax director (Conservative Co.). If Aggressive Co uses a tax shelter to reduce its tax cost, and then passes some of this cost on to consumers in the form of lower prices, Conservative Co finds itself at a competitive disadvantage. Assuming that Conservative Co is unwilling to adopt the same tax strategy, they have a strong incentive to stop their competitor from using it. They would be willing to pay their lawyer to place a discreet call to the government.

Although the government already benefits from this dynamic to a degree,36 the government should take more systematic advantage of it in at least two ways, neither of which, to my knowledge, has been discussed in the literature. First, the government should be more aggressive in offering bounties to informants, including former employees of the taxpayer, or employees of the taxpayer’s rival.37 Second, a concerted effort should be made to ask companies about their competitors tax planning. For example, auditors can be instructed to ask something like, Are there aggressive practices common in your industry that you do not use and, if so, what are they? How do they work? And how can we identify them? If companies are rewarded for producing this information ñ for instance, by developing a better rapport with their auditor in showing that they abstain from unsavory practices ñ they are likely to be forthcoming, thereby giving the government a useful source of information.

This is not to say that competitors will always be forthcoming. For example, they will hesitate if they fear an overbroad government response. If they are engaging in conduct similar enough to be swept up in an overbroad response, they will not share the information.

In cases where the government does get this sort of information, it obviously needs to respond ñ something that, unfortunately, it does not always do. Indeed, it is not uncommon for the government to wait months or even years before responding to a planning technique that has been openly discussed.[38] In other words, in addition to finding new sources of information, the government needs to make better use of the information it already has. Ironically, tips from competitors serve another function as well ñ as a source not just of information, but also of motivation. In some cases, the government is slow to react to an abusive transaction for political reasons. Countervailing political pressure from competitors can motivate (or even embarrass) the government into action.

2.Taxpayer’s Incentive to Assist in Law Reform: Narrowing Overbroad Rules Just as taxpayers have the incentive to help the government kill competitor’s deals, they also have the incentive to help the government make overly harsh rules more lenient. In other words, even if taxpayers are reluctant to tell the government about a bad transaction that the law technically allows, they are very motivated to tell the government about a good transaction that the law inadvertently prevents.[39]

When the government targets an abuse, therefore, it is well advised to propose a rule that is broader than necessary. To be clear, it may not wish to announce that it is doing this deliberately, as such an announcement could trigger a political outcry. Such an announcement is unnecessary, though. The tax bar will see that the rule is overbroad ñ whether or not it is announced as such ñ and the government can then rely on the tax bar, whose clients will want them to help narrow it. In this way, the bar and the government have a similar goal. They cooperate to modify the rule in order to permit desirable transactions.

Admittedly, the incentives do not align perfectly. The government wants to be sure that, in narrowing the rule, it still blocks all variations of the abusive transaction. The bar is less worried about that. Indeed, there is the risk that the bar will propose language that spares not only a meritorious transaction, but also an abusive one.[40] The government needs to be sensitive to this risk, and to take the bar’s advice with a grain of salt. But the opportunity for collaboration is there, nevertheless.

B.Creating Common Interests With Taxpayers Through the Penalty Structure

If taxpayers truly feared the consequences of overly aggressive positions, they would have a strong interest in carefully complying with the law. Obviously, this has not been the case in recent years. Penalties have not been stiff enough to compensate for the low probability of detection. The government has moved in the right direction by toughening the penalty structure through recent legislation, but more needs to be done.[41]

The guiding principle ought to be increasing marginal deterrence.42 Taxpayers ought to feel that, by taking more risky positions, they significantly increase the penalty they would incur if caught. This means that the penalty for fraud needs not only to be high in absolute terms, but also to be substantially higher than the penalty for negligence.

By moving from a position that is debatable (or merely sloppy) to something that is clearly over the line, taxpayers should become exposed to a newly daunting level of risk.

While this is a familiar point, there is an important implication that has not been sufficiently appreciated: By tweaking the penalty structure in the right way, the government enlists help from the private bar. As long as the penalty structure is sufficiently daunting, clients will want their lawyers to help them steer clear of overly aggressive structures. This, of course, is exactly what the government should want tax lawyers to do. The private bar is, in effect, doing the government’s work, monitoring transactions to make sure they are not too aggressive. In other words, the right penalty structure aligns the interests of the taxpayer with those of the government, and thus allows the government to rely on the private tax bar to police the system. To achieve this goal, though, the penalty structure must be crafted with care. The rest of this section considers variations in the penalty structure, with a particular focus on the role of tax opinions.

1. The Conflict Inherent in Tax Opinions

What if the tax lawyer, in providing a written opinion that a transaction is taxed in a particular way, is able to offer a get out of penalties free card? This generally was thought to be the case, at least until recently.[43] As long as a taxpayer received an opinion indicating that a position was more likely than not to succeed, the taxpayer would claim to have reasonable cause to believe in her position, and thus would not expect penalties to be imposed.[44] Likewise, the tax director of a corporation can expect to keep her job, even if a tax strategy proves unsuccessful, as long as she secured an opinion from an outside firm. In authorizing tax lawyers to provide penalty protection (and thus job insurance for the tax director as well), the government is, in effect, enlisting the tax bar to monitor their clients, so that they will make sure their transactions comply with the law.

As an effort to enlist the tax bar’s help, the practice of allowing penalty protection through tax opinions has a distinct advantage over, say, a mandate requiring tax lawyers to rat on clients who are engaged in overly aggressive transactions. In acting as a source of potentially compromising information on clients, the lawyer is not offering the client anything of value. In contrast, in offering a tax opinion for a transaction, the advisor is, indeed, providing something of value ñ protection from penalties (and, in the corporate context, job security for the tax director). At first blush, this privileging of tax opinions seems to strengthen the hand of the tax lawyer, giving them leverage to extract concessions from the client. Structure the deal this way, the lawyer can say, or I will not give you the opinion. This at least creates the basis for cooperation between lawyer and client ñ the possibility of common interests, as the lawyer works with the client to ensure that the transaction is not too aggressive, and rewards them for doing the right thing.

But it is a familiar point that, in a competitive environment, the tax lawyer’s hand is not strong enough. If lawyers have this valuable carrot to offer, taxpayers will want to gobble it up, even if they haven’t earned it. Give me the opinion or I will take my business elsewhere is the implicit (or sometimes even the explicit) message that clients send.[45] The result is a familiar race to the bottom, which the government has sought to reverse through a range of recent initiatives.

2. Strict Liability

The cleanest way to avoid this race to the bottom ñ but, unfortunately, one that has not been adopted on a broad scale[46] ñ is to strip away the penalty protection function from tax opinions. Under this approach, the fact that a taxpayer receives an opinion of counsel is no longer relevant in determining whether penalties should be imposed; rather, the penalty determination depends only on how aggressive the transaction is. As a result, the tax opinion offers a prediction, not inherent protection. It plays an important role in forecasting what the taxpayer’s exposure actually is, but only if it is honestly and carefully done. If you win the case ñ because your deal is not too aggressive ñ you avoid penalties. But if you lose, you are penalized. If your lawyer advises you properly on the chances of winning, you know what the risks are, and you know what an appropriate settlement is as well. But an opinion is no longer of inherent value. Instead, it becomes only as valuable as the advice it gives.

Under this approach, endorsed by the New York State Bar Association Tax Section, the lawyer’s incentive is exactly what it should be: to tell the client, in plain terms, exactly how aggressive a transaction is.[47] There is no longer any race to the bottom, since the lawyer cannot deliver anything of value by offering an overly favorable conclusion. By retooling the function of opinions in this way, the government enlists the help of the private bar: An expert is deployed for each transaction, in real time, to offer honest advice against taking overly aggressive positions.

This approach, paired with an adequate penalty structure, would go a long way toward incentivizing the bar to help the government ñ not out of altruism, but out of fealty to client interests. This is not to say, though, that the approach is uncontroversial. Are we really prepared to impose penalties when people believed in good faith, based on the advice of counsel, that they were complying with the law? If a lawyer offers an opinion in good faith, shouldn’t that opinion in turn demonstrate good faith on the part of the client? These arguments have political resonance, even though, in my opinion, they are unpersuasive. After all, many favorable opinions are not offered in good faith ñ that is precisely the problem. For those that are offered in good faith, I would leave it to the judge, or to the government official considering a settlement, to make a judgment about the appropriate penalty. Or, as a compromise, the penalty structure could be segmented, so that an information asymmetry charge of, say, ten percent is automatically imposed (i.e., under strict liability) and an additional 15% penalty would be imposed for bad faith.[48] But I realize that this approach is unlikely to be enacted in a comprehensive way. For political reasons, we are destined to allow tax opinions to provide some version of penalty protection. How can we do this without triggering the familiar race to the bottom?

3.Turning Competition Among Tax Advisors to the Government’s Advantage

Although every tax lawyer feels pressure to give an opinion that a competitor would give, it is not accurate to describe the problem as a single and unified race to the bottom. Rather, there are a number of different races occurring at different tiers of the profession. For example, the fact that a second tier accounting firm will give a favorable opinion is of real interest to another second tier accounting firm, but it is not particularly important to Sullivan & Cromwell. On the other hand, S&C will be extremely interested to know if another top law firm, such as Clearly Gottlieb, is willing to give a favorable opinion. Another way to make this point is that it is a matter of pride ñ and, indeed, self interest ñ to refuse to give some opinions. This is the way tax advisors build their reputations and signal the quality of their advice. If a firm says no to many deals, then the fact that the say yes has real meaning (and thus can be the subject of premium billing).

This sorting process ñ and the competition among advisors that animates it ñ represents a possible brake on the race to the bottom, as long as the government takes proper advantage of it. Lawyers will want conservative reputations, and clients will want to hire such lawyers, if the government treats opinions of conservative lawyers more favorably than opinions of aggressive ones. If the market comes to believe that an opinion from a top tier conservative firm is a necessity, and an opinion from a more aggressive competitor is valueless, then conservative advisors will have more bargaining power to hold clients to a higher standard. Again, the private bar would be doing the government’s work.

Although this insight is familiar, there obviously are a number of ways to operationalize it that, to my knowledge, have never been discussed in the literature. One is for the government to keep a list of advisors who have given favorable opinions for shelters that have been rejected in court. Similarly, this roster can include lawyers who have given favorable opinions for so-called listed transactions (i.e., transactions designated suspect by the I.R.S.) Of course, individual lawyers already have developed reputations with individual government lawyers, either for trustworthiness or for unreliability, but a list strengthens the effect of these reputations in two ways. First, it institutionalizes these reputations so that they survive turnover in the government’s ranks. Second, a list publicizes reputations, so that clients learn how the government views their lawyer. The list can be used simply to embarrass these advisors ñ and this may be the only feasible use. In theory, the government could give the list more teeth, but this is much more difficult. For instance, the government could disqualify any lawyer or firm that appears on the list from giving opinions that provide penalty protection for a period of time, with escalating sanctions for repeat offenders. By targeting the firm, as well as the individual, the government would mobilize firms to police their partners more diligently. But although this step might have a useful disciplining effect, it would not be straightforward to impose, since the government presumably would also have to provide a means to appeal the sanction, and standards would have to be developed to decide what sort of advice is beyond the pale.

At the other end of the spectrum, the government can look for ways to reinforce the reputations of conservative practitioners, however defined. For example, the government could create an expedited revenue ruling process for these lawyers. This could be quite attractive, since the current ruling process’s glacial pace renders it useless in many contexts. In the same spirit, the government could rely on a (rebuttable) presumption that lawyers who have served in government are conservative, as a way to encourage government service. Again, though, there may be need for a process to appeal the government’s decision to deny this benefit to practitioners who are deemed to be insufficiently conservative, and there is always the concern that inappropriate factors will be considered in awarding some lawyers a special status that is denied to others.

4. Mandates and Penalties: Circular 230

Are there other ways, then, to encourage the bar to give opinions only for deserving transactions? One way that lawyers have blessed undeserving transactions was to give only a partial opinion, which made unrealistic factual assumptions (e.g., in giving this opinion, we rely on your representation that there is a realistic possibility to make a pretax profit) or declined to address important anti-abuse doctrine (e.g., in giving this opinion, we do not consider the application of the doctrines of economic substance and business purpose). Quite appropriately, the government has responded by requiring advisors to address all the relevant legal issues, including anti-abuse doctrines that might be relevant, and to scrutinize the facts, so that they do not rely on patently unrealistic representations.[49] Circular 230ís various technical problems are explored elsewhere, so I will not detail them here,[50] except to say that this is a case in which the government has not been careful enough with the bar’s time. For example, they should have used an opt-in regime, in which lawyers affirmatively state when an opinion can be used for penalty protection. Instead, they created an opt-out regime, in which lawyers have to say when their advice cannot be so used. The result has been a frenzy of legending, so that every email sent by a tax lawyer ñ including ones accepting a dinner invitation ñ automatically say that they cannot be used for penalty protection. Highly compensated tax advisors have spent time debating such questions, Michael Schler has observed, as whether the legend must be above the signature in an email, or may be below the signature . . . .[51] I doubt anyone would defend this use of the bar’s expertise.

Putting aside these excesses, though, Circular 230 is, at its core, an effort by the government to enlist the tax bar’s help. In requiring lawyers to verify facts and address all relevant issues, the government is relying on them to monitor clients, and to give opinions (and thus the potential for penalty protection) only when they are deserved. This initiative has both the promise and the problems associated with tax opinions, as discussed above. In asking lawyers to monitor their clients, the initiative is strengthened by the fact that lawyers have something of value to give to their clients ñ so that they can feel, to a degree, as if they are serving client interests ñ but, at the same time, the inherent conflict remains: clients want opinions even when they don’t deserve them.

The new rules make it more difficult for lawyers to give opinions for undeserving transactions, to be sure, but much depends on whether the penalties for noncompliance are adequate. For example, a lawyer can be barred for life from practice before the I.R.S, a step that would be very embarrassing and could lead to disbarment in some states; but, of course, the I.R.S. has not sought to impose this sanction yet, even in egregious cases, leaving the bar to wonder whether they can safely ignore this possibility. In any event, the function of the penalty provision is to drive a wedge between the lawyer’s interest and the client’s, focusing lawyers on the need to protect themselves. In this way, the penalty provision is less like the initiatives discussed in this Part ñ which assume that the lawyer’s loyalty will remain with the client and try to align the client’s incentives with the government ñ and is more like the initiatives discussed in Part IV, which try to detach the lawyer’s loyalty from the client.

5. Separating Deal Practice from Opinion Practice

In still another effort to ensure that tax opinions are rendered honestly, the government no longer allows lawyers to give opinions on certain reportable transactions[52] ñ a category that is designed to describe tax shelters, but is widely known by tax lawyers to be both under- and over-inclusive[53] ñ if they (or their firm) have billed enough hours to developing the transaction. In fashioning this category of disqualified advisor, the government presumably was targeting shelter promoters. But the definition sweeps more broadly so that, in effect, a lawyer who is implementing the transaction cannot give an opinion about it. As a result, Michael Schler has observed, if the taxpayer desires penalty protection, it must use one firm to draft the documents for the transaction, and another firm to give an opinion on the tax consequences of the transaction.[54]

One advantage of this division of roles ñ that is, of having a deal counsel and a separate opinion counsel ñ is that, from the perspective of the deal counsel, the regime feels a lot like the strict liability regime recommended above: the incentive of deal counsel is to give honest advice, since offering an overly optimistic opinion is not an option for them anyway. Notice, though, that the same benefit is available by disqualifying all lawyers, not just deal counsel, from giving penalty protection through an opinion, as recommended above.

In theory, there might be an advantage in preserving the penalty protection function of opinions, but cordoning it off so that it is available only through a separate set of opinion counsel. This division of rules is appealing, though, only if we think opinion counsel is more independent than deal counsel, and thus is more likely to give an unbiased opinion. The argument for this independence is that opinion counsel are not personally invested in the success of the deal, and they may also not have a preexisting client relationship with the taxpayer.

Of course, even if we accept this premise, so that we think opinion counsel can make a valuable contribution, the current rules constrain them in bizarre ways. For example, they are not allowed to suggest material modifications in the structure; if they do, they too become disqualified. As Michael Schler has observed, this means opinion counsel can either (1) suggest the changes, helping the taxpayer on the merits but hurting the taxpayer on penalty protection, or (2) refrain from suggesting the changes and give the opinion nonetheless, hurting the taxpayer on the merits but helping the taxpayer on penalty protection.[55]

Although this sort of glitch is probably fixable, and it is worth preventing actual shelter promoters (as opposed to advisors on regular business deals) from giving opinions that provide penalty protection, I am skeptical about the value, in general, of looking to separate opinion counsel, because I doubt that they are more independent. Although opinion counsel may sort among conservative and aggressive, as noted above, there is no reason to think they are more likely to do so than deal counsel. Indeed, opinion counsel have their own reasons to engage in a race to the bottom. If they develop a reputation for readily blessing deals, they are more likely to be hired. Advisors may be particularly eager for this work as an opportunity to get a new client, since they may hope that, if they make a good impression, they will be the deal counsel next time. But even if they expect only to drum up more business as opinion counsel, this will be appealing because such work is interesting and relatively easy. Opinion counsel do not have to negotiate deal terms, or draft deal documents in the middle of the night. They have to understand the structure and the relevant law ñ with the assistance of deal counsel, obviously ñ and then write up the analysis. If they can bill premium rates, the engagement is quite attractive. In addition, if opinion counsel has made a regular practice of issuing aggressive opinions, the incremental risk from giving one more aggressive opinion is modest. For all these reasons, the incentives of opinion counsel strike me as no better, on average, than the incentives of deal counsel (aside, of course, from shelter promoters). The only benefit here is in removing the opinion-writing function from the lawyer doing the deal. A much cleaner way to do that, though, is the strict liability approach discussed above.

6. Disclosing the Opinion as a Condition of Penalty Protection

The race to the bottom among opinion writers is a very difficult problem, then, and it does not become easier in requiring a separate opinion counsel to write it. But there is a worthwhile measure that should be implemented ñ and, thus far, the government has done so only partially: The government should say that an opinion cannot be used for penalty protection unless it is attached to the tax return.[56]

Instead of requiring disclosure of the opinion itself, the government has begun requiring disclosure about the transaction as a condition for penalty protection in some cases. Specifically, if the transaction is otherwise reportable,[57] the reasonable cause exception is not available if the transaction was not disclosed.[58] This is a step in the right direction, since the opinion cannot offer penalty protection if the taxpayer tries to hide the transaction.

It would be even better, though, to require disclosure of the opinion itself. Under the revised Circular 230, the opinion has to lay out the issues presented in the transaction, serving, in a sense, as a roadmap for the government to challenge the transaction. This step would significantly reduce the government’s costs in analyzing the transaction, and also would deter aggressive planning, since taxpayers would have to expose the transaction’s vulnerabilities in order to secure penalty protection.[59] Of course, taxpayers may be concerned that rogue auditors will misuse the opinion, and some procedural mechanism, such as input from a centralized office (such as the Office of Tax Shelter Analysis), is needed to address this concern. But however the opinion is used by the government, it is much less valuable to the taxpayer if it is not persuasive: even if the opinion protects the taxpayer from penalties, it makes them more likely to have to pay the underlying tax. If an opinion must be turned over in this way, taxpayers will be less satisfied with a document that offers an overly optimistic bottom line. Rather, the taxpayer will want to be sure the opinion is right, and that the transaction actually works. As with the strict liability regime discussed above, this approach creates exactly the right incentive for the private bar. They will be asked to give honest advice about how to comply with the law.

Given these useful effects, this requirement should be imposed not just for reportable transactions ñ an imperfect category, as noted below ñ but for any transaction for which the taxpayer wants penalty protection. This is critical, since a substantial volume of shelters are not reportable transactions, but are simply deals that have no business purpose or economic substance.

IV. Enlisting the Private Bar: Drive a Wedge Between the Lawyer and Client

In considering ways to enlist the tax bar’s help for the government, the prior Part focused on tasks that would, at least to an extent, be in the client’s interest. This Part, in contrast, considers ways to pressure the private bar for help, even if giving this help is not in their client’s interest. In a sense, the goal here is to detach the lawyer from the client, something that is very hard to do. As a result, the initiatives in this Part are generally less promising than those in the prior Part, although some are more appealing than others.

A. Sanctions

Given that lawyers are strongly predisposed to pursue client interests, the government will have to create a powerful counterweight to distract them from this mission. One obvious reason for a lawyer to favor the government over client interests is self interest ñ and, in particular, fear of various penalties. For example, the government recently indicted a tax lawyer for aiding and abetting tax fraud in connection with KPMG’s tax shelter business.[60] Circular 230 contains new penalties as well.[61] There have also been a number of malpractice actions brought against lawyers who gave favorable opinions in shelter cases.[62] In theory, the government could create a similar system in which it looks to lawyers to bear a portion of the understatement penalty imposed when a deal they blessed is successfully challenged.

Obviously, though, such sanctions are only effective if lawyers actually expect to be punished. A malpractice judgment is embarrassing, but there are a range of reasons why malpractice claims are unlikely to succeed in this context, including the proliferation of arbitration clauses.[63] Criminal indictments are an even more powerful disciplining device. However badly a client may want a result, lawyers will not go along if they believe they are risking their professional licenses, their solvency, and even their freedom.[64] On the other hand, if people think that such indictments are a fluke, they are not effective. Likewise, some who are at risk of being indicted may not realize it, since, human nature being what it is, people sometimes deceive themselves into believing their bad behavior is justifiable. An unfortunate consequence of this sort of sanction is that risk tolerant lawyers will have an advantage over risk averse ones, since some number of risk tolerant lawyers will push the line without being caught.

Another risk with sanctioning the lawyer is to make sure that such sanctions do not create bad incentives for the client. The obvious example is malpractice. Clients will be more likely to engage in shelters if they believe they can pass on their penalties or other costs to their advisors.[65]

B. Disclosure

In a truly unusual development, the government has recently required tax advisors to keep lists of clients who engage in tax shelters, which they must turn over to the government upon request, and also to disclose the details of these deals in some cases.[66] (A parallel disclosure regime has also been imposed on clients).[67]

At first blush, it is easy to see why the government wishes to try this. Tax advisors have much more information than the government about what taxpayers actually are doing. By gathering this information, the government hopes to fashion more effective rules, closing loopholes and shutting down abusive practices. Indeed, this regime seems, at first glance, to be an intriguing example of what colleagues at Columbia call experimentalism or new governance. The essence of this approach is for the decisionmaker to gather information, propose tentative rules, gather further information to see whether these rules are effective, and revise the rules. Under this view, good governance depends on rolling standards and constant efforts to gather information.[68]

Yet as one looks more closely at this disclosure initiative, it inspires less confidence. In assigning the tax bar this function, the government offers only sticks, without carrots. Failure to comply carries penalties, to be sure, but faithful and meaningful compliance carries no discernible reward. Most notably, the lawyer is not advancing client interests in providing this disclosure. Quite the contrary, the lawyer is in effect being asked to rat on the client. Unlike in the case of writing opinions, where the lawyer must comply with the mandates of Circular 230 but is, at least, providing something of value to the client, the bar here is being drafted to work for the government, and the government alone.

Not surprisingly, then, the tax bar is highly motivated to undermine the effectiveness of this effort. The goal is to be able to say they are complying, but without providing truly helpful information. Taxpayers and their advisors use aggressive readings of the relevant filters to conclude that they do not have to disclose some shelters. At the same time, the bar also reads the filters fastidiously to conclude that they must disclose transactions that are not at all objectionable. In some cases, this is an excess of caution, since the filters are vague, and there is little to be lost, from the client’s perspective, in disclosing these deals. But there is also a bit of gamesmanship at work here. By flooding the government with paper, lawyers make it all the more difficult for the government to find the truly useful disclosure. There is no penalty, after all, for adding hay to the haystack, in order to make the needle harder to find. Time will tell, of course, just how useful this initiative proves to be for the government. But to my mind, it is a questionable use of scarce resources and political capital.

C. Patriotism and Professional Duty to the System

Instead of relying on the bar’s fear of sanctions, the government could try appealing to their professional pride and patriotism. Although there is no consensus on the point, many commentators believe that tax lawyers owe a duty not just to their clients, but also to the system. One formulation, advanced by Bill Simon, is that the tax system gives tax lawyers their livelihood, so they in turn owe a duty to nurture it. My own view relies less on this narrow reciprocity, and more on the gratitude that any citizen should feel for the freedom and security the U.S. government provides; if we feel grateful, we should want to preserve the government’s lifeline, the tax system.

Perhaps in some cases, this sort of altruistic motive will guide tax advisors ñ after all, life is about more than billing hours. But it is naive to rely too heavily on these motives. For one thing, some take a different view of their patriotic duty, focusing on the fact that the government wastes money, and that I.R.S. agents can be petty tyrants. Under this view, it is a patriotic contribution to keep resources in the more dynamic private sector and to protect clients from bureaucratic bullies.

In any event, however a tax lawyer views our tax system, she is unlikely to be guided by sentiment alone, or by a professional duty that is unenforceable. Professor Lavoie urges legal academics to use their bully pulpits, inculcating students with a sense of their professional duty to the system, so that they will be faithful to this imperative after they graduate.[69] I certainly favor an emphasis on professional responsibility in the tax curriculum ñ indeed, in the entire law school curriculum ñ but I don’t think we should overstate the staying power of these lessons.

Advisors are most likely to follow their better angels when they are in an institutional setting that frees them to do so. Indeed, the goal of this paper is to identify the most promising settings ñ for instance, when they serve in government, when they participate in bar association activities, or, of course, when their client’s interests point in this direction. But outside of these safe settings, the patriotism of tax lawyers generally is too slender a reed to support cooperation with the government.

IV. Conclusion

The mismatch between the government and the private bar contributes significantly to tax shelters and to the inefficiency and inequity that aggressive planning brings to our system. This Article has offered two reasons why the mismatch is so hard to remedy. First, it is politically difficult to fund a truly robust public infrastructure for tax administration. Second, the private bar does not have incentive to pick up the slack; on the contrary, their clients ordinarily view tax administrators as the enemy, and thus do not want the bar to be of help

An important challenge, then, is to identify contexts in which the client and the government have common interests, so that the tax lawyer can do good and do well at the same time. These contexts are few and far between, but they do exist, and this Article has emphasized some that have not received sufficient attention. There are times, also, when a lawyer’s loyalties can be detached from the interests of private clients, if the carrot or stick applied directly to them is strong enough.

At the end of the day, though, the private bar can serve only a limited function. Raising the quality of the government’s efforts is vital. This Article has offered a range of steps that should not attract too much political opposition. Any opposition that does arise should be resisted strenuously, because there truly is no substitute for competent tax administration. Starving this infrastructure is a poor way to cut taxes, as this path encourages wasteful planning, benefits aggressive taxpayers more than conservative ones, and ultimately breeds disrespect for the law. It would be far wiser to cut rates. Indeed, if we paired rate cuts with more robust investments in tax administration, and thus more effective efforts to curtail aggressive planning, we might well find a revenue neutral path to a more efficient and equitable tax system.

[*] Dean and the Lucy G. Moses Professor of Law, Columbia Law School. Copyright David M. Schizer 2006. All Rights Reserved. Helpful comments were received from Marvin Chirelstein, Michael Doran, David Hariton, Mitchell Kane, Edward Kleinbard, David Miller, Alex Raskolnikov, Chuck Sabel, Michael Schler, Bill Simon, Alvin Warren, Diana Wolman, and Mark Yecies, Larry Zelenak. Please direct comments to David.Schizer@law.columbia.edu

[1] The literature on tax shelters is massive. Joe Bankman wrote an important early piece on the subject. See Joseph Bankman, The New Market in Corporate Tax Shelters, 83 Tax Notes 1775 (June 21, 1995). Volume 54 of the SMU Law Review and Volume 55 of the Tax Law Review have each published interesting symposia on the subject.

[2] See, e.g., David Cay Johnston, Corporate Risk of Tax Audit is Still Shrinking, I.R.S. Data Show, N.Y. Times, Apr. 12, 2004, at C1 (audit rate fell for firms with over $ 250 million in assets from 33.7% in 2002 to 29% in 2003).

[3] For a discussion of these equity and efficiency problems, see David M. Schizer, Frictions as a Constraint on Tax Planning, 101 Colum. L. Rev. 1312 (2001).

[4] Richard Lavoie, Deputizing the Gunslingers: Co-Opting the Tax Bar Into Dissuading Corporate Tax Shelters, 21 Va. Tax Rev. 43 (2001) Because Professor Lavoie believes that Most lawyers involved in corporate tax shelter activity are deeply troubled by the role they play, he says they will help combat shelters if the government, in effect, gives them an excuse to do so. They need a reason to raise the inadvisability of shelters with clients and, armed with the right mandates or doctrines (e.g., a duty to verify the facts recited in a tax opinion), they will do so. In addition, Lavoie would impress practitioners with a sense of obligation to uphold the tax laws through education, including in law school. Although I agree that many tax lawyers find shelters to be objectionable, and that it is valuable to encourage lawyers to feel a personal obligation to protect the tax system, the focus of my analysis is less on appeals to conscience, and more on appeals to self interest.

[5] Tanina Rostain, Sheltering Lawyers: The Organized Tax Bar and the Tax Shelter Industry, Yale J. Reg. (forthcoming 2006). Professor Rostain emphasizes the ways in which the ABA and NYSBA Tax Sections have supported government efforts to rein in tax shelters. Yet she does not focus on the merits of various anti-tax-shelter measures, but on the role of bar associations in this process and the associated lessons to be learned about professionalism.

[6] See generally Alex Raskolnikov, The Self-Adjusting Tax Penalty, 106 Columbia Law Review (forthcoming 2006).

[7] This conversation took place in 1997, and the tax director in question would prefer not to be named.

[8] For instance, in the Compaq case, the government stipulated to the fact that the taxpayer owned the relevant stock, thereby giving up a promising argument that could have won the case. See Daniel N. Shaviro, Economic Substance, Corporate Tax Shelters, and the Compaq Case, 88 Tax Notes 221, 223 (2000). The government presumably did not make this concession carelessly. They must have hoped to win on economic substance or business purpose, preferring these theories to ownership because they would have more general application.

[9] See generally Bankman, supra note 1.

[10] In addition to lean staffing, political considerations contribute to this problem, including the fact that a loophole-closing provision can be tarred as a tax increase. Another problem with technical corrections is that interest groups sometimes view them as a wedge to revisit substantive provisions, sucking up more staff time and possibly leading to new potential loopholes.

[11] See Interview with Donald Korb, Chief Counsel, I.R.S., Section of Taxation, Fall 2005, available on www.fedbar.org. The gap between government and private sector pay has grown wider over time. For instance, thirty years ago, the starting salary at the I.R.S. and private law firms was approximately the same.

[12] In other work, I have referred to this phenomenon as the taxpayer’s planning option. See David M. Schizer, Sticks and Snakes: Derivatives and Curtailing Aggressive Tax Planning, 73 S. Cal. L. Rev. 1339 (2001).

[13]See IRS Acting Chief Counsel Challenges Wall Street Rule at TEI Conference, Tax Analysts, Sep. 22, 2003 (using the phrase Wall Street Rule to describe idea that a deal that has become sufficiently common is politically difficult to challenge).

[14] It is a fallacy to say that the government should spend up to a dollar on tax enforcement to collect a dollar of revenue. The better test is to spend up to a dollar on tax enforcement to avoid a dollar of social waste from tax planning. For a discussion of this point, and of the marginal efficiency cost of funds framework from which it derives, see generally Joel Slemrod & Shlomo Yitzhaki, The Costs of Taxation and the Marginal Efficiency Cost of Funds, 43 Intíl Monetary Fund Staff Papers 172 (1996).

[15] Cf. Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557, 562ñ70 (1992).

[16] For a discussion of this dynamic, see generally Schizer, supra note 3.

[17] In light of this interest-group dynamic, as well as the broader political opposition to taxing capital, it is fair to wonder how significant the staffing issues emphasized in this Article really turn out to be. If political factors prevent a vigorous government response to shelters, then the question of how many government lawyers work on these issues and how much experience they have becomes less important. An extreme version of this claim is that, even if the government were populated only by the leading lights of the profession, nothing will change because the political leadership will stop them from doing anything meaningful. A corollary is that top notch people will refuse to serve if they are constrained in this way, since the job will not be sufficiently interesting. But in this extreme form, the claim is unpersuasive. A more balanced assessment is that the political winds tend to shift, depending upon which party is in power, whether a booming economy is already filling the government’s coffers, whether a war or other large scale initiative requires extra revenue, and the like. These fluctuating factors filter down to tax enforcement, determining whether the political appointees who manage the process will flog their staff to react forcefully or restrain them from doing so. In this sort of cyclical environment, the capacity of the staff to respond effectively obviously remains quite important.

[18] See House Comm. Gov. Oversight, 105-43, Internal Revenue Service Mismanagement and Ideas for Improvement, Apr. 14, 1997.

[19]This information comes from a press release from the US Census Bureau, Aug. 29, 2006. See http://www.census.gov/Press-Release/www/releases/archives/income_wealth/007419.html

[20] The President currently earns a salary of $400,000. Title 3, section 102 of USCA, from Pub. L. 108≠199, Jan 23, 2004.

[21] As Professor Rostain notes, the NYSBA and ABA supported some recent anti-shelter initiatives. See Rostain, supra note 5. But it is worth emphasizing that the NYSBA and ABA did not support all of the initiatives, and they supported some only on the condition that they be modified. For example, the organized bar opposed codifying the economic substance doctrine and, not surprisingly, this provision was not enacted.

[22] John C. Coffee, Gatekeepers (2006).

[23] See Robert Guy Matthews, It’s Taxing to Recruit Top Law Grads to IRS, But a New Push Betters Returns, Wall St. J., Oct. 10, 2006, at B1 (entry level pay at IRS is $65,000 or $70,000, which is approximately half the starting salaries at top private firms).

[24] Robert Guy Matthews, It’s Taxing to Recruit Top Law Grads to IRS, But a New Push Betters Returns, Wall St. J., Oct. 10, 2006, at B1.

[25] I am indebted to David Miller for this suggestion.

[26] I am indebted to Marvin Chirelstein and Alex Raskolnikov for this insight.

[27] See Rostain, supra note 5.

[28] Some pro-government tax lawyers have told me that they feel less at home in bar committees than they used to, and that the younger generation of bar leaders is more prone to taxpayer advocacy. This may be a symptom of a broader generational trend within the tax bar, which others have observed. See Bankman, supra note 1.

[29] Put another way, if all you need is an opinion ñ any opinion, which could provide penalty protection ñ you don’t need to hire one of the leading lights of the profession. This issue is discussed below in Part III.

[30] See generally David Boies, Courting Justice (2005). Of course, Boies charged the government a steeply discounted rate. See Daniel Okrent, Get Me Boies, Time, December 17, 2000, at http://www.time.com/time/poy2000/mag/boies.html (noting that Boies charged only $40 per hour, compared with his usual $750 per hour rate at the time).

[31] For example, David Hariton, a highly regarded practitioner at Sullivan & Cromwell, helped to draft the contingent debt regulations, offering his services pro bono. Likewise, Gordon Henderson, a distinguished practitioner from Weil Gotshal, responded to proposed regulations under Section 752 by submitting alternative language, and the government drew heavily on his language in finalizing the regulation. I am indebted to Mark Leeds for informing me of Mr. Hendersonís role.

[32] See generally Coffee, supra note 22.

[33] Cf. Kopel, supra note 62 (describing allegation that lawyers representing plaintiffs are colluding with defendant KPMG).

[34] See Anna Bernasek, Why Let the I.R.S. See What the S.E.C. Doesn’t?, New York Times, Feb. 5, 2006, at
BU 3.

[35] See supra note 14.

[36] When this paper was discussed at a session of approximately two-dozen leading practitioners, they immediately came up with three examples of transactions that had been shut down because of this sort of tip from a competitor: step-down preferred, as well as aggressive uses of Canadian income trusts and of commodities swaps. This conversation occurred on February 27, 2006 at a meeting of the Tax Club in New York City.

[37] In the latter case, the bounty arguably should not go to the individual employee of the taxpayer’s rival, but to the corporation itself, since the employee’s knowledge is a corporate asset.

[38] For example, commentators have understood for some time that the Achilles heel of tax-free hedging is the counterparty’s interest in borrowing the taxpayer’s stock, and potential problems under Section 1058. I discussed the point in detail in a 2001 article. See Schizer, supra note 3. The government waited nearly five years before responding. See Lee A. Sheppard, Should Share Lending Affect a Prepaid Forward Contract, Tax Notes, Jan. 9, 2006, at 12.

[39] See Schizer, supra note 3, at 1337.

[40] This pattern has played out with the new disclosure regime, discussed below. It was overbroad, and the bar has worked with the government to develop angel lists, but taxpayers analogize to transactions on this list in order to conclude that they do not have to disclose transactions that, in light of the purposes of the regime, should be disclosed.

[41] Raskolnikov, supra note 6.

[42] Id.

[43] See Rostain, supra note 5, at 117.

[44] Of course, the relevant regulation, 1.6664-4(e)(2)(B)(2) (as of April 2003), did not quite say this, but it was the conventional wisdom, nevertheless. But even under that regulation, a court recently imposed a penalty on a taxpayer that had an opinion. See Long Term Capital Management, 330 F. Supp. 2d 122 (D Conn 2004), affíd 150 F. Appíx 40 (2d Cir 2005). A recent change in the penalty structure, discussed below, imposes strict liability for reportable transactions that have not been disclosed.

[45] See Bankman, supra note 1, at 1782-83.

[46] Of course, narrow variations of a strict liability regime have been enacted for specified circumstances, and are discussed below. The opinion of a disqualified tax advisor does not provide penalty protection, and an opinion offers no penalty protection for a reportable transaction that was not disclosed.

[47] See N.Y. State Bar Assín Tax Section, Report on Corporate Tax Shelters, in NYSBA Applauds Some Anti-Corporate Tax Shelter Measures, Rejects Others, Apr. 23, 1999, Tax Notes Today, 1999 TNT 82-29.

[48] I am indebted to Edward Kleinbard for this suggestion.

[49] 10.35(b)(4)(ii).

[50] See, e.g., Jeffrey H. Paravano & Melinda Reynolds, The New Circular 230 Regulations — Best Practices or Scarlet Letter, Tax Management Memo, Aug. 22, 2005; see also Michael Schler, Effects of Anti-Tax Shelter Rules on Non-Shelter Tax Practice, Tax Notes, Nov. 14, 2005.

[51] See Schler, supra note 50.

[52] Transactions are reportable if they satisfy any one of five filters: (1) listed transactions; (2) transactions in which the advisor imposes a condition of confidentiality to protect the advisor’s planning strategy; (3) transactions in which the advisor’s fee is contingent on the success of the planning strategy; (4) transactions in which a gross loss exceeds certain thresholds; and (5) transactions that involve credits and short holding periods for assets. See Section 6111(a); Treas. Reg. 301.6112ó(1)(c), 1.6011-4, Notice 2005-22, 2005-1 Cum. Bull. 756.

[53] For example, many non abusive transactions involve a large loss, since taxpayers sometimes genuinely do lose money. At the same time, the filters leave out significant issues as well. For example, although losses under Section 165 have to be reported, losses under Section 162 do not. Schler, supra note 50.

[54]See Schler, supra note 50. The regime applies only if certain modest fee thresholds are reached –$250,000 for C corporations and $50,000 for other taxpayers ñ and only if a significant purpose of the transaction is tax avoidance, but this phrase applies to many transactions that are not shelters. See, e.g., Cottage Savings Assn. v. Commissioner, 499 U.S. 554 (1991).

[55] Schler, supra note 50.

[56] Michael Schler suggested this idea over three years ago. See Michael Schler, 10 More Truths About Tax Shelters, 55 Tax L. Rev. 325 (2002)

[57] For a discussion, see supra note 52.

[58] The government originally introduced this rule administratively, and then confirmed its authority to do so in the Jobs Act. See Treas Reg. 1.6664.4 (failure to disclose reportable transaction was a strong indication that the taxpayer did not act in good faith for purposes of accuracy related penalty); see I.R.C. 6664(d)(2)(B)-(C) (eliminating reasonable cause for undisclosed transactions and thus subjecting taxpayers who fail to disclose reportable transaction to 30% penalty).

[59] In serving as a source of information for the government, which could then use this information in reevaluating existing law, this proposal is a variation of the new governance or experimentalist approach to regulation developed by colleagues at Columbia. For a discussion, see infra note 68 and accompanying text.

[60] For a discussion, see Lynnley Browning, I.R.S Offers Deal to Firms Promoting Tax Shelters, N.Y.Times, Feb. 14, 2006, at C1.

[61] See generally Paravano & Reynolds, supra note 50.

[62] See, e.g., Nathan Kopel, Law Firm Offers An Unusual Fee in KPMG Case, Wall Street J., Jan. 27, 2006.

[63] Of course, if the government wants malpractice to be a more effective sanction, they can discourage the use of arbitration clauses (e.g., by saying that a tax opinion cannot provide penalty protection if the lawyer and client are bound by an arbitration clause). I thank Bill Simon for this observation.

[64] Correspondingly, offers not to prosecute can sometimes induce cooperation. The government has tried this tack with promoters, so far with mixed results. See Browning, supra note 60, at C1 (noting that government has offered to spare promoters from prosecution if promoters pay penalties and turn over information, but that some promoters are likely not to accept the deal).

[65] The question becomes more complicated if malpractice insurance covers these judgments, and lawyers are able to pass the cost of insurance on to clients in the form of higher rates. For instance, the end result may be that conservative clients subsidize aggressive ones (assuming the lawyer does not charge sufficiently different rates for standard and overly aggressive advice).

[66] See Section 6112(b), 6708, 6111, 6707.

[67] See Section 6011, 6707A.

[68] Indeed, this pattern is familiar to tax lawyers, since most new rules ñ whether statutes or regulations ñ are initially proposed in draft, and then are revised in response to comments. Likewise, informal authorities such as revenue rulings are sometimes revoked or superseded in response to new information or changed circumstances.

[69] See generally Lavoie, supra note 4.

Previously published by the Columbia Law School – The Center for Law and Economic Studies, October 2006

Dean and the Lucy G. Moses Professor of Law - Columbia Law School, USA