Creating failures in the market for tax planning.

Also, as individuals face more risk being associated with tax planning activity, they may be less inclined to invest in building up reputations for their skills in the area, reputations which might improve their prospects. They thus should be deterred from such activities. The inability to profitably acquire and exploit such reputations, as well as their employers' difficulty in compensating them for their "performance" if that performance could lead to sanctions, should together provide individuals with an incentive to pursue other, possibly more lucrative, areas. It is not likely that firms will want to cultivate such reputations either. (48) The weaker the seller's reputation, the more of a lemons discount the buyer of tax planning methods will demand.

C. Discussion and Summary

In the preceding subsection, we considered how to make individuals, who are presumably less able to bear risk do so. We briefly discussed the possibility that firms might in effect be required to provide information on their methods for free.

Of course, a great deal more could be done along both these lines: individuals could be made to certify techniques, firms could be made to disclose their use in "plain language" in their public filings as a condition to being able to use them, contracts for tax planning could be denied enforcement on public policy grounds, and so on.

But these proposals are ultimately largely infeasible. One of the main difficulties faced by the government is one we assume away. Notwithstanding some commentators' beliefs that all tax planning activities are undesirable and that there is no "right" to engage in tax planning, (49) many others believe in a principled distinction between tax planning and tax shelter activity. Nobody knows exactly where such a distinction might be and what might be the principle at issue--many possible principles have been tried on for fit, with no consistent success either in the courts or with commentators. (50) Approaches that discourage "too much" tax planning will therefore be infeasible, and those that target "tax shelters" have been hindered by the difficulty of drawing a principled distinction between tax shelters and tax planning.

Indeed, in this regard the recent proposal by Presidential candidate Barack Obama to deny patents for "inventions designed to minimize, avoid, defer, or otherwise affect liability for Federal, State, local, or foreign tax" (51) notwithstanding that patents are generally available for business methods, including tax planning methods. There has been considerable criticism of the proposal, on several grounds. Given how much difficulty the courts and legislators have in distinguishing between tax planning that is expressly allowed and tax planning that is disfavored--that is, planning that constitutes a "shelter"--it seems inconceivable that the patent office could readily make such a distinction. Thus, the proposal seems more likely to invite litigation than serve any other purpose. (52)

VI. CONCLUSION

We have presented an argument that governments, when designing their tax policy, should be aware of two distinct types of social costs: the cost associated with lost revenue and the cost that arises from taxpayers' search for new methods to reduce their tax burden. Inevitably, reducing one of these costs comes at the expense of increasing the other; the government faces a tradeoff. By recognizing these costs and the tradeoff the government faces, we can better understand current tax policy. Moreover, a wider recognition of the tradeoff described above, and a systematic consideration of how to disrupt markets in tax planning activities, should lead to better tax policy.

Philip A. Curry*

Claire Hill**

Francesco Parisi***

* Assistant Professor of Economics, Simon Fraser University

** Professor and Director, Institute for Law and Rationality, University of Minnesota Law School

*** Professor, University of Minnesota Law School. We wish to thank the participants at The Future of Tax Shelters symposium and particularly, David Weisbach, for helpful comments. We also wish to thank Kristin Hickman, Brett McDonnell, Gregg Polsky, and Bruce Shnider for useful discussions, and Emily Kraack for useful research assistance.

(1) In the literature, some commentators distinguish between tax planning and tax shelters. For our purposes, we assume that all costs of both activities are wasteful social costs. In this regard, David Weisbach argues that tax planning is "almost always positively bad for society." See David A. Weisbach, Ten Truths About Tax Shelters, 55 TAX L. REV. 215, 222 (2002). Weisbach does discuss arguments to the contrary; those arguments carve out (without precise specification) a category of legitimate tax planning activities. Id. at 220 n.1. We consider in Part V how the possibility of some tax planning activity being deemed legitimate might affect our analysis.

(2) For a review of the optimal taxation literature, see Andres Erosa & Martin Gervais, Optimal Taxation in Infinitely-Lived Agent and Overlapping Generation Models: A Review, 87 J. FED. RES. BANK RICH. Q. 23 (2001).

(3) Paul A. Samuelson, The Pure Theory of Public Expenditure, 36 REV. ECON. & STAT. 387 (1954).

(4) See, e.g., Raymond G. Batina, Public Goods and Dynamic Efficiency: The Modified Samuelson Rule, 41 J. PUB. ECON. 389 (1990); Robin Boadway & Michael Keen, Public Goods, Self-Selection and Optimal Income Taxation, 34 INT'L ECON. REV. 463 (1993); Mario Nava, Fred Schroyen, & Maurice Marchand, Optimal Fiscal and Public Expenditure Policy in a Two-Class Economy, 61 J. PUB. ECON. 119 (1996); Dan Usher, Tax Evasion and the Marginal Cost of Public Funds, 24 ECON. INQUIRY 563 (1986).

(5) See Dan L. Burk & Brett H. McDonnell, Patents, Tax Shelters, and the Firm, 26 VA. TAX REV. 981 (2007) (discussing the desirability of impeding markets in tax planning methods).

(6) See generally MICHAEL PARKIN, MICROECONOMICS, 321-82 (7th ed. 2005).

(7) Philip A. Curry, Claire Hill & Francesco Parisi, Optimal Government Responses to Tax Planning: A Mathematical Model (draft on file with authors).

(8) While this assumption is commonly held and sufficiently realistic for our purposes, in the real world people have differing attitudes towards paying tax. Some people are more inclined to expend efforts in order to not pay tax, whereas others may feel that it is their civic duty to pay tax. See Claire A. Hill, Tax Lawyers are People Too, 26 VA. TAX. REV. 1065 (2007).

(9) Rent-seeking models study the economic behavior of actors outside the traditional productive, profit-maximizing framework, and they can provide a valuable key for the understanding of the behavior of individuals engaged in tax planning activities. In 1967 Gordon Tullock was the first to study to what extent self-interested parties would incur costs in the pursuit of "rents" (in our application, the rents would be given by tax savings). See Gordon Tullock, The Welfare Cost of Tariffs, Monopolies and Theft, 5 W. ECON. J. 224 (1967). Tullock's basic model was followed by other formulations by Becker, Krueger, Posner, Demsetz, Bhagwati, Tollison, and many others. See Gary S. Becker, Crime and Punishment: An Economic Approach, 76 J. POL. ECON. 169 (1968); Jagdish N. Bhagwati, Directly Unproductive, Profit-Seeking (DUP) Activities, 90 J. POL. ECON. 988 (1982); Harold Demsetz, Economics as a Guide to Antitrust Legislation, 19 J. LAW & ECON. 371 (1976); Anne O. Krueger, The Political Economy of the Rent-Seeking Society, 64 AM. ECON. REV. 291 (1974); Richard A. Posner, The Social Costs of Monopoly and Regulation, 83 J. POL. ECON. 807 (1975); Robert D. Tollison, Rent-Seeking: A Survey, 35 KYKLOS 575 (1982). Much of the literature focuses on how much effort each player expends and how the degree of rent dissipation varies with the value of the prize. Two quite different positions were reached during the early years of this debate. Most scholars (Becker, Krueger, Posner, Demsetz, and others) suggested that rent-seeking competition would generate equilibria similar to those generated by competitive markets, with a full dissipation of the available rents. Posner's full dissipation hypothesis became popular in the empirical literature and also had a strong appeal in the theoretical literature. In the subsequent years the literature analogized rents to profits, maintaining that both were likely to be competed away in the long-run equilibrium. According to this hypothesis, in a long-run equilibrium, expenditures in tax planning would thus yield the normal market rate of return. Gordon Tullock shook this conventional wisdom in the literature, showing that the full dissipation result would hold only under very narrow conditions. Gordon Tullock, Efficient Rent-Seeking, in JAMES M. BUCHANAN, ROBERT TOLLISON & GORDON TULLOCK, TOWARD A THEORY OF THE RENT-SEEKING SOCIETY, 97-112 (1980). According to Tullock, in most situations, some residual rent could be captured by the players and the rent would not be fully dissipated. According to this alternative hypothesis, in a long-run equilibrium, expenditures in tax planning could thus yield above-normal rates of return.

(10) As we discuss in the text, however, at a certain point the "rewards" may be negative, as the taxpayer's tax liability, use of tax planning techniques, or both become more likely to attract the government's attention.