Relational tax planning under risk-based rules

University of Pennsylvania Law Review

Volumn 156, Issue 5, 2008, Pages 1181-1262

  Raskolnikov, Alex ^a  

^a School of Law ^*

Author keywords

[No Author keywords available]

Indexed keywords

EID: 46449136702     PISSN: 00419907     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (468)

1. Daniel Shaviro defined risk-based rules as "rules that give the presence or absence of elements of economic risk a tax significance that is distinct from any effect that such risk has on fair market value or the accrual of economic gain or loss." Daniel Shaviro, Risk-Based Rules and the Taxation of Capital Income, 50 TAX L. REV. 643, 643 (1995)
2. [hereinafter Shaviro, Risk-Based Rules].
3. See, e.g., Robert H. Scarborough, Risk, Diversification and the Design of Loss Limitations Under a Realization-Based Income Tax, 48 TAX L. REV. 677, 679-81 (1993);
4. Deborah H. Schenk, Taxation of Equity Derivatives: A Partial Integration Proposal, 50 TAX L. REV. 571, 583-87 (1995);
5. Shaviro, Risk-Based Rules, supra note 1, at 645-50;
6. Alvin C. Warren, Jr., Commentary, Financial Contract Innovation and Income Tax Policy, 107 HARV. L. REV. 460, 460-61 (1993).
7. The wash sale rule, as well as all other rules and judicial doctrines mentioned in this Article, is more complicated than the discussion suggests. For a detailed analysis of the intricacies of the wash sale provision, see, for example, David M. Schizer, Scrubbing the Wash Sale Rules, TAXES, Mar. 2004, at 67
8. [hereinafter Schizer, Wash Sale Rules].
9. In fact, many did.
10. It is. I.R.C. § 1091 (a) (2000).
11. For an effort to offer one such approach to a particular type of relational tax planning, see Alex Raskolnikov, The Cost of Norms: Tax Effects of Tacit Understandings, 74 U. CHI. L. REV. 601, 665-77 (2007)
12. [hereinafter Raskolnikov, The Cost of Norms].
13. See, e.g., sources cited infra notes 74-75.
14. See, e.g., HARVEY S. ROSEN, PUBLIC FINANCE 283-305 (7th ed. 2005).
15. See Shaviro, Risk-Based Rules, supra note 1;
16. Lewis R. Steinberg, Commentary, 50 TAX L. REV. 725 (1995).
17. Frictions are "transaction costs incurred in the marketplace that make implementation of certain tax-planning strategies costly." MYRON S. SCHOLES ET AL., TAXES AND BUSINESS STRATEGY: A PLANNING APPROACH 9 (3d ed. 2005). The view of risk-based rules as pure frictions is not uniformly accepted.
18. See, e.g., David M. Schizer, Frictions as a Constraint on Tax Planning, 101 COLUM. L. REV. 1312, 1360 (2001)
19. [hereinafter Schizer, Frictions].
20. See Schizer, Frictions, supranote 10, at 1325-26.
21. Id.
22. See David A. Weisbach, An Economic Analysis of Anti-Tax-Avoidance Doctrines, 4 AM. L. & ECON. REV. 88, 99-103 (2002) (arguing that the sham transaction doctrine is a weak deterrent because it fails to stop numerous shelters and, therefore, has little effect on taxable income elasticity). Admittedly, not all risk-based rules give rise to continuous frictions. Yet, as David Schizer explains, the government often stumbles into effective discontinuous frictions simply by accident.
23. See Schizer, Frictions, supra note 10, at 1371-74.
24. Daniel N. Shaviro, Economic Substance, Corporate Tax Shelters, and the Compaq Case, 88 TAX NOTES 221, 223 (2000)
25. [hereinafter Shaviro, Tax Shelters].
26. Traditional tax law's distinctions between ownership and use, debt and equity, and fixed and contingent returns have proved to be no match to those who have conquered the power of the put-call parity theorem. See Warren, supra note 2, at 465-67;
27. see also Edward D. Kleinbard, Equity Derivative Products: Financial Innovation's Newest Challenge to the Tax System, 69 TEX. L. REV. 1319, 1327-30 (1991);
28. Schenk, supra note 2, at 574-79; Shaviro, Risk-Based Rules, supra note 1, at 651-56;
29. Jeff Strnad, Taxing New Financial Products: A Conceptual Framework, 46 STAN. L. REV. 569, 569-74 (1994);
30. Warren, supra note 2, at. 482-91;
31. David A. Weisbach, Tax Responses to Financial Contract Innovation, 50 TAX L. REV. 491, 496-507 (1995).
32. But see Mark P. Gergen, Afterword, Apocalypse Not?, 50 TAX L. REV. 833 (1995).
33. At least in one commentator's view, they produce nonoptimal risk allocations, misdirect resources, and generally amount to not much more than a "tax planning nuisance." Shaviro, Risk-Based Rules, supranote 1, at 695.
34. See id. at 723-24;
35. Strnad, supra note 15, at 604-05. In the words of one observer, Congress's repeated attempts to "'solve' the problem of financial asset taxation . . . has done little to stop the proliferation of tax-motivated financial transactions, while subjecting (at least part of) the tax law to mind-numbing complexity."
36. Steinberg, supra note 9, at 729;
37. see also Joseph Bankman, Commentary, 50 TAX L. REV. 787, 787 (1995) ("The income tax treatment of capital is a mess.").
38. See Bankman, supra note 17, at 792 (concluding that while David Bradford's proposed reform of income tax is an improvement over the current system based on risk-based rules, it is still less efficient than a consumption tax) ;
39. Shaviro, Risk-Based Rules, supra note 1, at 723-24 ("[T] he increasing difficulty of reaching capital under an income tax, by reason of the declining effectiveness of deterring tax planning through risk-based rules, strengthens the case for shifting to consumption taxation.");
40. Steinberg, supra note 9, at 729 & n.20.
41. See David A. Weisbach, Ironing Out the Flat Tax, 52 STAN. L. REV. 599, 616 (2000)
42. [hereinafter Weisbach, Flat Tax].
43. The troubling distinctions between owners and non-owners, debt and equity holders, and time-value and risky returns remain important See id. at 615, 628-29, 635. Even the transactional patterns are eerily familiar: a sale and repurchase,
44. see id. at 615, 628-29, 660-61, a sale and leaseback,
45. see id. at 628 n.41, 660-61,
46. a straddle, see id. at 616, and, more generally, an arbitrage based on different tax treatment of similar cash flows,
47. see id. at 625, 628-29. In what must come as a shock to those who hoped that a switch to a consumption tax would end the struggles with taxation of financial instruments once and for all, Weisbach concludes that "[t]he financial products rules under an income tax probably have greater potential to be coherent than those under the Flat Tax."
48. Id. at 663. By referring to Professor Weisbach's scrutiny as "moderate," I in no way mean to undermine the rigor of his analysis or the value of his contribution, both of which are high. Rather, as Weisbach himself pointed out, his incentives to find loop-holes in the Flat Tax were not nearly as strong as those incentives would be when the brightest minds of the American tax bar set out on a search for weaknesses in the new regime with hundreds of millions of dollars on the line.
49. Id. at 629.
50. See id. at 660-61. For instance, tax planning using forward straddles is possible because the Flat Tax is open and ignores financial transactions.
51. Id. at 616.
52. Similarly, mischaracterization of interest is much less of a concern in closed regimes like the European VATs. Id. at 629. Professor Shaviro anticipated these types of difficulties and advocated a consumption tax that makes "the form of one's asset transactions as irrelevant as possible - a consideration that may prove in some tension with transitional relief or proposals that use a graduated rate structure."
53. Shaviro, Risk-Based Rules, supra note 1, at 724.
54. Ownership is one example. See Weisbach, Flat Tax, supra note 19, at 615 (explaining why personal versus business ownership of assets is important both in a closed cash-flow consumption tax and an open system that, like the Flat Tax, combines a cashflow tax with a yield-exempt tax). The distinction between financial and other returns is another.
55. See id. at 627-28 (noting that tax planning based on overstating the interest component of purchase price is possible in open and closed systems).
56. For an academic discussion, see, for example, DAMD F. BRADFORD, FUNDAMENTAL ISSUES IN CONSUMPTION TAXATION (1996), proposing the "X-Tax," and David A. Weisbach, Does the X-Tax Mark the Spot?, 56 SMU L. REV. 201 (2003), which discusses Bradford's proposal.
57. For continuing political debates, see, for example, Heidi Glenn, "Cleanse the Code" Reform Effort Attracts Strange Bedfellows, 113 TAX NOTES 949 (2006), describing statements by various politicians and lobbying groups favoring different versions of a consumption tax.
58. See JOEL SLEMROD & JON BAKIJA, TAXING OURSELVES: A CITIZEN'S GUIDE TO THE DEBATE OVER TAXES 15 (3d ed. 2004) ("By 1999, all OECD nations except the United States had adopted VATs, which on average raised 6.4 percent of GDP."). For a similar proposal in the U.S. context,
59. see Michael J. Graetz, 100 Million Unnecessary Returns: A Fresh Start for the U.S. Tax System, 112 YALE L.J. 261, 281-99 (2002).
60. See, e.g., I.R.C. § 901 (k) (2000) (added in 1997);
61. I.R.C. § 901 (l) (Supp. IV 2004) (added in 2004);
62. I.R.C. § 1259 (2000) (added in 1997);
63. id. § 1260 (added in 1999).
64. For recent cases relying on the economic substance doctrine, see, for example, Coltec Industries v. United States, 454 F.3d 1340 (Fed. Cir. 2006)
65. and Black & Decker v. United States, 436 F.3d 431 (4th Cir. 2006). The point that the economic substance doctrine is a risk-based rule is not new.
66. See Shaviro, Tax Shelters, supra note 14, at 222.
67. Granted, the magnitude of risk produced by a 30-day waiting period is uncertain, and it differs depending on the volatility of the security in question. Similarly, the constructive sale rule that accelerates a taxable gain for taxpayers who eliminate "substantially all" of their economic exposure to an appreciated stock produces considerable ambiguity. Still, policymakers probably have a rough sense of the amount of risk involved. A 30-day waiting period is surely less risky than a 90-day one. Compare I.R.C. § 1091(a) (2000) (thirty days),
68. with id. § 246(c)(2) (ninety days). A test that triggers unfavorable tax consequences if taxpayers eliminate "substantially all" risk of loss is clearly less demanding than the one that applies even to those who merely "substantially diminish" that risk.
69. Compare STAFF OF J. COMM. ON TAXATION, 105TH CONG., GENERAI. EXPLANATION OF TAX LEGISLATION ENACTED IN 1997, at 177 (Comm. Print 1997) (interpreting § 1259 to include transactions that "have the effect of eliminating substantially all of the taxpayer's risk of loss and opportunity for income and gain with respect to the appreciated financial position" (emphasis added)),
70. with I.R.C. § 1092(c)(2) (2000) (using the "substantial diminution" standard). Comparisons are rough, but they at least appear to be reasonably plausible based on a general common sense of the drafter of a statute, regulation, or judicial opinion. But if we want to find and develop stronger discontinuous frictions, general common sense will no longer suffice. As David Schizer explained, to identify and evaluate these frictions one would need to be familiar with "a wide range of institutional details, including the securities and commodities laws, the state of financial technology, accounting, and broker-dealer regulations."
71. Schizer, Frictions, supra note 10, at 1335. This, no doubt, is only a partial list.
72. Admittedly, even as far as continuous frictions go, risk is a particularly weak one. For instance, the passive loss rules that force taxpayers to sacrifice their time to gain tax benefits appear to be a more effective deterrent. See Shaviro, Risk-Based Rules, supra note 1, at 701.
73. Id. at 690.
74. See, e.g., Allen Kenney, SILO Shutdown: How the New Law Could Cripple the Industry, 105 TAX NOTES 638, 638 (2004) (reporting that additional revenue from shutting down lease-in, lease-out shelters is projected to be more than $25 billion by 2014, according to congressional budget estimates).
75. For one estimate, see SLEMROD & BAKIJA, supra note 23, at 33 tbl.2.3, 34-35 (reporting their estimates for the year 2000 and discussing why a large portion of capital gains escapes taxation). In fact, according to some estimates, net capital gains are the second largest component of the total adjusted gross income, following only wages and salaries.
76. See, e.g., id.
77. More precisely, market risk is an unwanted risk of price changes incurred by a passive investor (that is, an owner who has no inside information about the asset's value and cannot influence this value by personal efforts). Thus, market risk arises even if a particular asset is not traded in a liquid (or any) market. However, because most market-risk-imposing provisions do deal with fungible, actively traded assets, I use the term market risk to provide an intuitive reference. My use of this term should not be confused with its use in the corporate finance literature where market (or systematic, or undiversifiable) risk refers to the risk of owning a security tfiat cannot be eliminated by diversification. See RICHARD A. BREALEY & STEWART C. MYERS, PRINCIPLES OF CORPORATE FINANCE 168 (7th ed. 2003).
78. See I.R.C. § 246(c)(1)-(2) (2000)
79. See I.R.C. § 901(k), (l) (Supp. IV 2004).
80. See I.R.C. § 1259(c) (3) (2000).
81. See id. § 1259(a), (c) (1), (d) (market risk arises because taxpayers are forced to hold appreciated positions that are hedged less than they would have preferred).
82. See id. § 1260(a), (d) (market risk arises because taxpayers are compelled to gain less complete exposure to hedge funds than they would have liked).
83. See id. § 1092(a)(1) (market risk arises because taxpayers cannot substantially diminish unwanted risk of loss without triggering adverse tax consequences).
84. See id. § 1258(c)(1) (market risk arises because taxpayers can avoid a tax-increasing recharacterization only if they embed enough unwanted risk-based (non-time-value) return in their transaction).
85. See id. § 163(l) (market risk arises because corporate issuers are forced not to price their equity-flavored securities in a way that would assure repayment in equity, as they would have preferred).
86. See id. § 351 (g) (2) (A) (iii) (market risk arises because preferred stock whose repayment is too certain carries unfavorable tax consequences).
87. See id. § 465 (market risk arises because taxpayers lose investment-related deductions unless they make investments with their own money or fund investments with recourse (rather than nonrecourse) debt).
88. See Schizer, Frictions, supra note 10, at 1323 n.29.
89. See I.R.C. § 1259 (2000) (applies only to appreciated financial positions);
90. id. § 1260 (applies only to gain from constructive ownership transactions).
91. See id. §§ 1091, 1092.
92. See id. § 1258.
93. Martin D. Ginsburg, Making the Tax Law Through the Judicial Process, A.B.A. J., Mar. 1984, at 74, 76. For instance, if the government makes it exceedingly easy to characterize corporate securities as debt (and difficult to treat them as equity), corporations will significantly increase their interest deductions by issuing equity-like securities viewed as debt for tax purposes. If the government goes to the opposite extreme (making debt characterization difficult and equity treatment easy), taxpayers will start claiming many more tax benefits associated with dividend payments, such as a dividends-received deduction and foreign tax credit. For a more detailed study of the same dynamic using an example of a risk-based rule applicable to contingent payment debt instruments,
94. see David M. Schizer, Sticks and Snakes: Derivatives and Curtailing Aggressive Tax Planning, 73 S. CAL. L. REV. 1339, 1377-92 (2000).
95. This is exactly what Congress did when it enacted I.R.C. § 163(l). Of course, making the distinction between debt and equity irrelevant (e.g., by repealing the corporate income tax) would solve the problem as well.
96. The so-called passive loss rules are an example of anti-abuse rules that are not risk based. Rather, they exact the price from taxpayers who wish to claim tax benefits in a coinage of time. See id. § 469 (2000);
97. Schizer, Frictions, supra note 10, at 1324;
98. Shaviro, Risk-Based Rules, supra note 1, at 701. Note that there is no general correlation between the type of symmetrical rule and the type of related anti-abuse rule. Each rule may be risk based or not. Passive loss rules are an example of non-risk-based anti-abuse rules protecting symmetrical risk-based ownership and debt/equity rules.
99. See I.R.C. § 1001 (2000).
100. See id. § 351(a).
101. Id.
102. Compare W. Coast Mktg. Corp. v. Comm'r, 46 T.C. 32 (1966) (treating the purported sale of stock of a newly incorporated business as a sale of assets because incorporation and sale were too closely linked),
103. and Rev. Rul. 70-140, 1970-1 C.B. 73 (same),
104. with Culligan Water Conditioning of Tri-Cities, Inc. v. United States, 567 F.2d 867, 870 (9th Cir. 1978) (finding, based on "somewhat murky" facts, that the requirements of § 351 were met where the transfer of assets to the corporation occurred six months prior to the sale of its stock),
105. and Weikel v. Comm'r, 51 T.C.M. (CCH) 432 (1986) (refusing to link the steps where assets were "dropped down" more than two months prior to signing of the documents for the sale of stock and four months prior to the sale's closing, distinguishing West Coast Marketing and Rev. Rul. 70-140 by highlighting considerable uncertainty of the second-step sale at the time of the first-step incorporation, and relying on Vest v. Comm'r, 57 T.C. 128 (1971),
106. rev'd on other grounds, 481 F.2d 238 (5th Cir. 1973)).
107. See generally 1 BORIS I. BITTKER & JAMES S. EUSTICE, FEDERAL INCOME TAXATION OF CORPORATIONS AND SHAREHOLDERS ¶ 3.09 [2] (7th ed. 2006). The opposite scenario presented the same issue until Congress intervened in 1954 and, again, in 1986. A taxpayer wanted to sell her company, but the buyer wanted to purchase the company's assets. The obvious solution was for the taxpayer to liquidate the corporation, receive its assets, and sell them to the buyer. If the two steps were too interlinked, however, they were combined.
108. Compare Comm'r v. Court Holding Co., 324 U.S. 331 (1945) (affirming the Tax Court judgment linking the steps),
109. with United States v. Cumberland Pub. Serv., 338 U.S. 451 (1950) (upholding the Court of Claims decision declining to combine the steps on similar facts). Separating the steps to gain a benefit of a tax-free distribution of corporate assets exposed the taxpayer to a risk of holding these assets until their ultimate sale without the protection of a corporate shield, See BITTKER & EUSTICE, supra, ¶ 10.05[5] [a].
110. See BITTKER & EUSTICE, supra note 52, ¶ 10.08.
111. These benefits included a bailout of corporate earnings at the tax-preferred capital gains rate, a step-up in tax basis of business assets, a recognition of loss built into the stock of a liquidated entity, and so on. See id. ¶ 12.64[1] [a].
112. Specifically, the government argued that the transaction should be treated as a nontaxable reorganization or a taxable dividend of all accumulated earnings and profits. Neither characterization would give rise to recognition of loss or increase in basis of corporate assets, and the latter would create additional taxable income for the tax-payer. See, e.g., Simon v. Comm'r, 644 F.2d 339, 341-43 (5th Cir. Unit B Apr. 1981);
113. Moffatt v. Comm'r, 42 T.C. 558, 573-82 (1964);
114. Rev. Rul. 76-429, 1976-2 C.B. 97;
115. BITTKER & EUSTICE, supra note 52, ¶ 12.64[1][c].
116. See, e.g., Swanson v. United States, 319 F. Supp. 959, 960 (E.D. Cal. 1970) (explaining that taxpayer operated as sole proprietor "for a time");
117. Kind v, Comm'r, 54 T.C. 600, 605 (1970) (recognizing a seven-month gap between liquidation of a flower shop and incorporation of its successor);
118. Mathis v. Comm'r, 19 T.C. 1123, 1129 (1953) (documenting a nine-month gap between liquidation of a lumber company and incorporation of its successor).
119. See, e.g., Pridemark, Inc. v. Comm'r, 345 F.2d 35, 38-42 (4th Cir. 1965).
120. See, e.g., Swanson, 319 F. Supp. at 960 (explaining that the taxpayer decided that operating an unincorporated business following the liquidation "was unsatisfactory, primarily because it required him to risk all his personal assets in his new ventures");
121. Kind, 54 T.C. at 605 (finding that the taxpayer operated an unincorporated business for seven months "without the protection of a corporate umbrella").
122. See, e.g., Pridemark, 345 F.2d at 38-39, 41-42.
123. I introduce the concept of business risk through a series of examples because it is, essentially, a catchall category. Business risk is an unwanted business-related uncertainty other than market risk, credit risk, see infra note 76, and counterparty risk,
124. see infra text accompanying note 73.
125. See, for example, United States v. Cumberland Pub. Serv., 338 U.S. 451 (1950)
126. and Comm'r v. Court Holding, 324 U.S. 331 (1945), discussed in note 52, supra.
127. The term is "reorganization." See BITKER & EUSTICE, supra note 52, ¶ 12.64 [2] [b] (referring to courts' use of three different types of reorganizations to characterize a liquidation followed by a reincorporation), The "immediately after the exchange" clause of § 351 is another example of a statutory term interpreted to impose business risk.
128. See BITTKER & EUSTICE, supra note 52, ¶ 12.64[1] [c].
129. See id.
130. The step transaction doctrine, for instance, applies when the steps are "interdependent," or when a court is convinced that "the end result" of a sequence of steps was predetermined before the first step was made. For a discussion, see id. ¶ 12.61 [3].
131. The Acquirer's stock received by the Seller will be "restricted, " and the Acquirer will need to file a registration statement with the SEC before this stock can be sold to the general public. See McDonald's Rests, of 111., Inc. v. Comm'r, 688 F.2d 520, 522 (7th Cir. 1982);
132. Penrod v. Comm'r, 88 T.C. 1415, 1419 (1987);
133. Heintz v. Comm'r, 25 T.C. 132, 138 (1955).
134. Compare McDonald's Rests., 688 F.2d at 525 (holding that, because there was too much certainty regarding the second step, the transaction was a taxable cash sale),
135. and Heintz, 25 T.C. at 14243 (holding that, because there was too much certainty regarding the second step-even though the planned public sale of the Acquirer's stock by the Seller fell through and the Acquirer arranged for a private sale later - the Seller was not entitled to the benefits of a tax-free reorganization),
136. with Penrod, 88 T.C. at 1437 (finding that there was not enough certainty regarding the second step, respecting separate steps, and giving the Seller the benefit of nonrecognition of gain in the first step). Similar issues arise in the so-called bootstrap acquisitions, where certainty of the second-step sale of a corporation determines whether a pre-sale dividend paid by that corporation to the selling shareholder is respected or treated as part of the purchase price.
137. Compare Steel Improvement & Forge Co. v. Comm'r, 314 F.2d 96, 98 (6th Cir. 1963) (treating the dividend as part of the purchase price),
138. with Walker v. Comm'r, 544 F.2d 419, 422 (9th Cir. 1976) (respecting the dividend form).
139. See generally BITTKER & EUSTICE, supra note 52, ¶ 8.07 [2] [a].
140. For an example of cooperative tax planning in international context, see Koehring Co. v. United States, 583 F.2d 313, 315-17 (7th Cir. 1978), describing how a U.S. company relied on its long-term U.K. partner to circumvent the controlled foreign corporation provisions.
141. For a detailed discussion, see BITTKER & EUSTICE, supra note 52, ¶ 10.21 [3] [a].
142. See Granite Trust Co. v. United States, 238 F.2d 670, 674 (1st Cir. 1956) (allowing the loss);
143. Comm'r v. Day & Zimmermann, Inc., 151 F.2d 517, 519-20 (3d Cir. 1945) (same);
144. BITTKER & EUSTICE, supra note 52, ¶ 10.21 [3] [a] (referring to the Day & Zimmermann decision as "surprising" but conceding that it "now seems sanctified by the passage of time").
145. See Stein v. Comm'r, 36 T.C.M. (CCH) 992,995 (1977) (disallowing the loss).
146. For instance, consider an isolated sale-repurchase agreement (or "repo"). Imagine a company that needs to borrow $2 million, and has $2 million worth of municipal bonds that it can pledge as collateral (but does not want to sell). The company has a close relationship with a bank. The policy of the company's president, however, is not to borrow. Besides, the bank's lending limit to a single customer is only $800,000. One solution would be for the company to sell the bonds to the bank for $2 million and simultaneously enter into a contract to buy them back for the same price, allowing the bank to keep the tax-exempt interest on municipal bonds. This transaction will probably be treated as a secured loan for tax purposes, however, and the interest received by the bank on the municipal bonds will be taxable. An alternative solution would be for the parties not to enter into a repurchase agreement. Instead, the bank will retain the right to sell the bonds back to the company for $2 million, but the company will not be able to compel the bank to do so. Without this right, the company will assume a risk that the bank will decide to keep the bonds. See Citizens Nat'l Bank of Waco v. United States, 551 F.2d 832, 843 (Ct. Cl. 1977) (refusing to treat such a transaction as a secured loan).
147. Charles J. Goetz & Robert E. Scott, Principles of Relational Contracts, 67 VA. L. REV. 1089, 1091 (1981).
148. For instance, the Wash Buyer may agree to purchase the loss security only if the Taxpayer gives her a free put option to resell it back to the Taxpayer for the same price. An at-the-money put option does not defeat the tax strategy, and it eliminates the Counterparty's exposure to the Taxpayer's defection. Cf. Rev. Rul. 85-87, 1985-1 C.B. 268 (holding that § 1091 disallows a loss from a sale of stock if the taxpayer simultaneously sells an in-the-money put option on the same stock and there is no substantial likelihood that the put will not be exercised).
149. For a definition of opportunistic behavior, see OLIVER E. WILLIAMSON, THE ECONOMIC INSTITUTIONS OF CAPITALISM 47-49 (1985) ("By opportunism I mean self-interest seeking with guile."). For a discussion of holdups,
150. see Benjamin Klein, Why Hold-Ups Occur: The Self-Enforcing Range of Contractual Relationships, 34 ECON. INQUIRY 444 (1996)
151. [hereinafter Klein, Why Hold-Ups Occur].
152. For an explanation of quasi-rents, see Benjamin Klein, Robert G. Crawford & Armen A. Alchian, Vertical Integration, Appropriable Rents, and the Competitive Contracting Process, 21 J.L. &: ECON. 297, 298-302 (1978).
153. "[S]trategic behavior' is concerned with influencing another's choice by working on his expectation of how one's own behavior is related to his." THOMAS C. SCHELLING, THE STRATEGY OF CONFLICT 15 (1980 ed.). Relational contractors have developed several ways to control strategic bargaining and asymmetric information problems.
154. See, e.g., Victor P. Goldberg, Price Adjustment in Long-Term Contracts, 1985 WIS. L. REV. 527, 531 (arguing that parties "enter into [long-term] agreements to achieve the benefits of cooperation," not "because of their concern for the future course of prices");
155. Klein, Why Hold-Ups Occur, supra note 74, at 458-59 (explaining how contractors shift private enforcement capital to alleviate contracting difficulties);
156. Oliver E. Williamson, Credible Commitments: Using Hostages To Support Exchange, 73 AM. ECON. REV. 519, 524, 531 (1983)
157. [hereinafter Williamson, Credible Commitments] (describing how hostages assure contractual performance). More generally, transactors adopt various institutional frameworks (or governance structures) to overcome the difficulties of imperfect bargaining.
158. See Oliver E. Williamson, Transaction-Cost Economics: The Governance of Contractual Relations, 22 J.L. & ECON. 233, 234-35, 247-54 (1979)
159. [hereinafter Williamson, Transaction-Cost Economics]. Yet the bargaining problems cannot be eliminated entirely. Thus, the term "counterparty risk" should be viewed as referring to the residual risk that remains after the parties have reduced uncertainty to the extent possible.
160. See infra text accompanying notes 259 & 265-270. Yet another type of uncertainty arises whenever a taxpayer transacts with third parties who may default on their binding contractual obligations. This is credit risk. Occasionally, taxpayers who accept it escape unpleasant tax consequences. For example, § 163(l) grants interest deductions to issuers of certain debt-equity units only if they make debt and forward components separable, accepting credit risk of public investors who may default and refuse to perform under the forward. For a discussion,
161. see Schizer, Frictions, supra note 10, at 1330-31 (explaining that because the corporate issuer will have no assurances that public investors will make payments if an issuer's stock price declines, "naked" publicly traded forwards are "unworkable"). To the extent that credit risk is conceived of as arising because of counterparties' possible insolvency or bankruptcy (which, one may assume, counterparties do not deliberately bring upon themselves), it is very similar to business risk. However, credit risk also reflects a possibility that a solvent counterparty will simply breach its contract with the taxpayer. While existence of an enforceable contractual obligation makes this type of risk different from counterparty risk, an opportunity for strategic actions by the counterparty is present in both cases. Thus, credit risk is better understood as a hybrid between business and counterparty risks. I will not discuss it further because analysis applicable to each of the "basic" risks applies to credit risk to the extent it reflects each basic type in any given case.
162. See I.R.C § 302(b) (2000) (defining several tests that must be met in order for a corporate distribution not to be treated as a dividend).
163. See id. § 302(b)(3). Since 2003, dividends are often taxed at a lower capital gains rate, so the disparity between the dividend and sale treatments is not as drastic as it has been for years. However, the entire amount of the dividend is includable in gross income (assuming the company has sufficient accumulated earnings and profits,
164. see id. § 316(a)), while only the excess of the amount received in a full redemption over the shareholder's basis in the redeemed stock is generally subject to tax,
165. see id. § 1001(a). Thus, a sale treatment continues to be tax-preferred for individual share-holders.
166. See id. §§ 302, 318;
167. see also United States v. Davis, 397 U.S. 301 (1970). A limited exception to this draconian rule is discussed below.
168. See infra text accompanying note 223.
169. For a detailed description, see Raskolnikov, The Cost of Norms, supra note 6, at 625-27.
170. The market practice reverted to fully documented repurchase agreements to clarify the uncertain legal status of repos highlighted by a high-profile bankruptcy. See MARCIA STIGUM, THE REPO AND REVERSE MARKETS 218-21 (1989) (describing the 1982 Lombard-Wall bankruptcy and the development of a new standardized repo agreement).
171. See First Am. Nat'l Bank of Nashville v. United States, 467 F.2d 1098, 1101 (6th Cir. 1972);
172. Union Planters Nat'l Bank of Memphis v. United States, 426 F.2d 115 (6th Cir. 1970);
173. Am. Nat'l Bank of Austin v. United States, 421 F.2d 442, 453-54 (5th Cir. 1970).
174. But see Am. Nat'l Bank of Austin v. United States, 573 F.2d 1201, 1205-07 (Ct. Cl. 1978) (holding for taxpayer on similar facts).
175. The charity may use the money to purchase a taxpayer's yacht, see, e.g., Blake v. Comm'r, 697 F.2d 473, 474 (2d Cir. 1982), or a building owned by the taxpayer,
176. see, e.g., Carrington v. Comm'r, 476 F,2d 704, 709 (5th Cir. 1973), or the charity may simply cooperate with the taxpayer in liquidating the corporation that issued the donated securities,
177. see, e.g., Grove v. Comm'r, 490 F.2d 241, 243-44 (2d Cir. 1973);
178. DeWitt v. United States, 204 Ct Cl. 274, 279-89 (1974);
179. Palmer v. Comm'r, 62 T.C. 684, 685-90 (1974).
180. The charity's cooperation is important because a sale of appreciated securities by a taxpayer produces a taxable gain, while a sale of the same securities by a taxexempt charity has no tax consequences. If a charity sells the securities and uses the proceeds to, say, purchase a building from the donor, the end result is that the donor parts with the building (which he essentially contributes to the charity) and sells the securities for cash without recognizing gain on that sale.
181. Among the charitable donation cases cited above, Blake is the only government win. The IRS's attempt to rely on this case without repealing a revenue ruling acquiescing in the earlier taxpayer-favorable decisions prompted a stiff rebuke from the Tax Court. SeeRauenhorst v. Comm'r, 119 T.C. 157, 169-73 (2002).
182. The related party rules make sure that the purchaser is not the seller's close relative, see I.R.C. § 267 (2000), but beyond that anything is fair game, or at least it so appears on the face of the statute. For a discussion of the relevant cases and a proposal to add a broad related-party rule to the wash sale regime,
183. see Schizer, Wash Sale Rules, supra note 3, at 75-76.
184. The court announced, for instance, that the wash sale loss was allowed only if "there was no express or implied agreement to buy back" the loss security, without offering any support for this proposition.
185. Stein v. Comm'r, 36 T.C.M. (CCH) 992, 994 (1977) (emphasis added). Section 1091 provides that a loss is disallowed if, within the thirty-day period, "it appears that. . . the taxpayer has acquired . . . or has entered into a contract or option to so acquire, substantially identical stock or securities." I.R.C. § 1091(a). While the "it appears" clause suggests that the provision should be read broadly, it was clear that the taxpayer had nothing close to an enforceable contract to repurchase the security for the same price. The Stein court cited no authority supporting its conclusion that an "implied agreement" is a "contract" within the meaning of § 1091.
186. See, e.g., I.R.C. §§ 246(c)(4), 1058(b)(3), 1092(c)(2), 1259(c)(3).
187. See id. § 246(c)(1)(B) (waiting period suspended while the "taxpayer is under an obligation . . . to make related payments with respect to positions in substantially similar or related property");
188. id. § 901 (k) (1) (A) (ii) (same); Treas. Reg. § 1.246-5 (1995) (defining positions in "substantially similar or related property" to include certain portfolios of stocks).
189. See, e.g., supra note 26.
190. See articles cited supra notes 2 and 15.
191. Several tax rules are designed to impose business risk. The term "reorganization" and the "immediately after the exchange" requirement of § 351 are prime examples. Only those corporate sales that qualify as "reorganizations" are eligible for a desirable deferral of gain. In order for the exchange to qualify, however, at least a portion of consideration received by the seller must come in the form of buyer's stock. That is, the seller cannot completely cash out. Ownership of stock involves business risk. For a discussion of the "immediately after the exchange" requirement, see supra text accompanying notes 51-52.
192. One of the key insights of transaction cost economics is that easily available alternatives eliminate the threat of opportunism. See, e.g., Williamson, Transaction-Cost Economics, supra note 75, at 239 ("The crucial investment distinction is this: to what degree are transaction-specific (nonmarketable) expenses incurred. Items that are unspecialized among users pose few hazards, since buyers in mese circumstances can easily turn to alternative sources, and suppliers can sell output intended for one order to other buyers without difficulty.").
193. The holdup problem is one of the central issues studied by contract and organizational theorists. See, e.g., Introduction to CASE STUDIES IN CONTRACTING AND ORGANIZATION 7 (Scott E. Masten ed., 1996);
194. Alan Schwartz & Robert E. Scott, Precontractual Liability and Preliminary Agreements, 120 HARV. L. REV. 661, 685-87, 693-702 (2007).
195. To be sure, taking on counterparty risk while transacting with total strangers (as in the Seller's sale of the Target to the Acquirer) is a risky proposition, so we may expect taxpayers to use it only as a last resort Yet this point should not be overstated. Robert Scott has identified many cases where transactors deliberately entered into unenforceable contracts without an expectation of future dealings and with no tax benefits in sight. See Robert E. Scott, A Theory of Self-Enforcing Indefinite Agreements, 103 COLUM. L. REV. 1641, 1644 (2003). Of course, relational tax planning that involves friends (like the Wash Seller and Wash Buyer) or long-term business partners (like the Liquidator and 21-Holder) is a different story. The Taxpayer knows the Counterparty well, has an ongoing relationship with her, and has a good reason to believe that defection is unlikely.
196. Treas. Reg. § 1.170A-5(a)(4) (as amended in 1994).
197. Treas. Reg. §§ 1.483-1 (a)(1), 1.1271-1 (a)(1) (1994).
198. Treas. Reg. § 48.4217-1 (as amended in 1985).
199. Treas. Reg. § 1.382-3(a) (1) (i) (as amended in 1993).
200. Treas. Reg. § 1.483-1 (a)(1) (1994);
201. Treas. Reg. S 1957-1 (b) (2) (as amended in 1997);
202. Treas. Reg. § 20.2036-1(a) (as amended in 1960);
203. Treas. Reg. S 301.6111-2(c)(1) (2003).
204. Treas. Reg. § 1.355-7(h) (1) (i) (A) (2005).
205. Treas. Reg. § 1.507-2(a)(8)(iv) (A) (3) (i) (as amended in 1981).
206. Treas. Reg. § 1.6011-4(b) (3) (ii) (as amended in 2003).
207. A review of dozens of regulatory references to informal understandings reveals that all but a few of these provisions have never been invoked by the government in litigation, formal guidance, or even informal guidance. For rare exceptions, see, for example, Koehring Co. v. United States, 583 F.2d 313, 317 (7th Cir. 1978)
208. and Guynn v. United States, 437 F.2d 1148, 1150 (4th Cir. 1971).
209. Note that in Koehring, the government's case that a U.S. taxpayer had an implied agreement with a foreign co-owner of a joint company was greatly helped by the fact, among others, that the foreign party's directors referred to their participation as "nominal" in the minutes of the Board of Directors meeting. 583 F.2d at 316.
210. For example, in I.R.S. Priv. Ltr. Rul. 92-18-067 (Jan. 31, 1992), the IRS refused to find that donors reserved present interest in paintings - even diough they retained them in their home until the donee-museum opened in the following year and even though the parties expected that the donors would do so - because the donee had a formal right to take control of the paintings. This ruling essentially ignored the statement in the regulations that a taxpayer-donor will be treated as reserving present interest if she "has an understanding, arrangement, agreement, etc., whether written or oral, with the charitable organization which has the effect of reserving to, or retaining in, such donor a right to the use, possession, or enjoyment of the property." Treas. Reg. § 1.170A-5(a) (4) (as amended in 1994).
211. See BITTKER & EUSTICE, supra note 52, ¶ 11.11 [3] [d] (describing the government's struggle with regulations enacted under § 355(e) to clarify when taxpayers will be treated as having a plan to dispose of a newly received corporate stock, but which have been called "a parody" and "the Matrix").
212. Thus, while the regulatory admonitions appear to deter relational tax planning through ex ante rules (rather than ex post standards), it is unclear whether they actually do so. The distinguishing feature of rules is that they give content to legal commands before individuals act. See Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 DUKE L.J. 557, 560 (1992). Whether the regulatory references to informal understandings actually have specific content depends on whether dieir meaning is (relatively) certain. If it is not, the apparent rule-like provisions are actually standards whose content will be determined only ex post.
213. See id. at 601. Because the references in question have been used so rarely, it is unclear whether they establish a rule or a standard. I refer to them under the rubric of an ex ante approach to highlight the distinction from court-created doctrines whose content is even less certain, and thus, mat are even more standard-like. This distinction also emphasizes the difference between regulatory (ex ante) and judicial (ex post) solutions.
214. Congress and the Treasury know how to refer to tacit understandings well, the argument goes. They have done so on many occasions. If a given provision includes no reference to oral, implicit, or not legally binding arrangements, it must be because the government does not object to their use in this particular context.
215. Quite possibly, this kind of argument was precisely the reason why the Wash Seller in the real friendly wash sale case was so confident in his strategy mat he voluntarily revealed to the IRS the existence of an informal agreement to repurchase the loss security. See Stein v. Comm'r, 36 T.C.M. (CCH) 992,994 (1977).
216. In some cases the detection may be fairly easy, such as where taxpayers transact on non-arm's-lengdi terms. A repurchase of a loss security from a Wash Buyer at the initial sale price, as well as a repurchase of municipal bonds by a repo seller, are examples of such transactions. Even here, however, the parties may make detection difficult by selling for the market price and making compensating side payments not reflected in any discoverable written agreement. On many other occasions, things will be even less clear. For instance, the fact that the Acquirer assists the Seller in selling the Acquirer's stock, or that a charity disposes of the donated securities exactiy as the donor would have wished, may suggest - but certainly does not prove (as the government has learned on many occasions) - the existence of an informal understanding.
217. TIFD III-E, Inc. v. United States, 459 F.3d 220, 236 (2d Cir. 2006).
218. Frank Lyon Co. v. United Sutes, 435 U.S. 561, 573 (1978).
219. BORIS I. BITTKER & LAWRENCE LOKKEN, FEDERAL TAXATION OF INCOME, ESTATES AND GIFTS ¶ 4.3.3 (Supp. 3 2004).
220. See, e.g., McDonald's Rests, of III, Inc. v. Comm'r, 688 F.2d 520, 524 (7th Cir. 1982);
221. King Enters, v. United States, 418 F.2d 511, 516 (Ct. Cl. 1969).
222. See, e.g., True v. United States, 190 F.3d 1165, 1180 (10th Cir. 1999) (rejecting in part the lower court's summary judgment in favor of the IRS because the taxpayer's intent regarding transfers among his controlled companies was unclear);
223. Citizens Nat'l Bank of Waco v. United States, 551 F.2d 832, 841 (Ct. Cl. 1977) (upholding a transaction entered into "with no thought or purpose of tax evasion," even though when all was said and done the bank ended up lending funds to the taxpayer and receiving tax-exempt interest on that loan);
224. Vaughn v. Comm'r, 81 T.C. 893, 910 (1983) (explaining that the Tax Court established an "independent purpose test" for intrafamily installment sales).
225. See, e.g., Granite Trust Co, v. United States, 238 F.2d 670, 677 (1st Cir. 1956) (respecting a transitory sale of corporate stock to a long-term customer);
226. Citizens Nat'l Bank of Waco, 551 F.2d at 843 (respecting the form of a repo entered into by a bank and its long-term client); Richard Hansen Land, Inc. v. Comm'r, 65 T.C.M. (CCH) 2869, 2874 (1993) (upholding a taxpayer-favorable treatment of a family trust);
227. Bowen v. Comm'r, 78 T.C. 55, 85 (1982) (approving an inter-spousal installment sale).
228. See, e.g., True, 190 F.3d at 1180-81;
229. see also Comm'r v. W.F. Trimble & Sons Co., 98 F.2d 853 (3d Cir. 1938) (allowing a loss on a wash sale of a security from one wholly owned corporation to another, followed by its repurchase slightly more man a month later).
230. Estate of Craft v. Comm'r, 68 T.C. 249, 263 (1977). For a discussion of this opinion, see Raskolnikov, The Cost of Norms, supra note 6, at 635-36.
231. For a detailed explanation, see Raskolnikov, The Cost of Norms, supra note 6, at 613-22.
232. See id. at 674.
233. A Wesdaw search of the Federal Tax Cases Combined database reveals 186 cases invoking the step transaction doctrine and 142 cases discussing the substance-over-form doctrine (without also mentioning the step transaction doctrine).
234. The government's recent inquiries into the schemes that may have allowed wealthy foreigners to escape as much as one billion dollars in the U.S. dividend with-holding tax are likely to uncover some such strategies already described in the academic literature. For the reference to the government interest,
235. see Anita Raghavan, IRS Probes Tax Goal of Derivatives, WALL ST. J., July 19, 2007, at Cl. For the description of the strategies,
236. see Raskolnikov, The Cost of Norms, supra note 6, at 618-22.
237. The following table provides a partial list of one-way risk-based rules and the implicated tax policy considerations, together with cross-references to specific examples: Risk-Based Rule Tax Policy Issue Affected Reference Taxation of capital vs. labor income (tax rate ap-I.R.C. § 83 plicable to certain compensation based on whether it is subject to a substantial risk) § 163 (l) Double taxation of corporate income n.39 § 246(c) (1), (2) Double taxation of corporate income n.32 § 351 (g) Double taxation of corporate income n.40 § 465 Taxation of risky vs. riskless returns n.41 § 901 (k), (l) Worldwide taxation of U.S. residents n.33 § 1091 Realization requirement nn.3-5 § 1092 Realization requirement n.37 § 1258 Taxation of risky vs. riskless returns n.38 § 1259 Realization requirement nn.34-35 § 1260 Realization requirement n.36 Substance-over. A variety of issues, including: form doctrine, Realization requirement nn.49-52 step transaction Double taxation of corporate income nn.53-61 doctrine Tax ownership n.71
238. A risk-bearing loss (or a risk premium) arises because for a concave utility function, the expected income and the certainty equivalent diverge. See ROSEN, supra note 8, at 267-69.
239. See Shaviro, Risk-Based Rules, supra note 1, at 674 ("Requiring taxpayers to bear undesired risk as the price of receiving favorable tax treatment creates excess burden.").
240. See Louis Kaplow, The Optimal Probability and Magnitude of Fines for Acts That Definitely Are Undesirable, 12 INT'L REV. L. & ECON. 3, 9 (1992);
241. A. Mitcheil Polinsky & Steven Shavell, The Optimal Tradeoff Between the Probability and Magnitude of Fines, 69 AM. ECON. REV, 880, 880 (1979).
242. For instance, a rule that determines whether a particular outlay is a (deductible) business expense or a (nondeductible) personal expenditure cannot be avoided by accepting risk. Yet an entrepreneur who deducted some questionable items worries about violating this rule just like an investor who hedged her appreciated stock position a bit too perfectly worries about triggering a constructive sale (i.e., violating a risk-based rule). That is, both taxpayers are concerned about taking aggressive positions. In contrast, a taxpayer who deducts an uncontroversial business expense has nothing to worry about. But a taxpayer who clearly satisfies the wash sale rule and is entitled to a loss deduction still incurs a risk-bearing loss because she is forced to accept 31 days' worth of unwanted market exposure. This difference highlights a unique cost imposed on taxpayers by risk-based rules - the cost of incurring market risk.
243. At least, this much is true of a typical (short) waiting period. For such periods, "the past rates of return on any stock conform closely to a normal distribution." BREALEY & MYERS, supra note 31, at 187. For longer intervals such as one year, the distribution would be skewed and would better approximate a lognormal distribution.
244. Id. at 187 n.2.
245. See, e.g., I.R.C § 246(c)(1) (2000) (forty-five days);
246. id. § 901(k) (fifteen days);
247. id. § 1091(a) (thirty days);
248. id. § 1259(c)(3) (sixty days).
249. See, e.g, id. § 246(c);
250. id. § 901 (k), (l).
251. See id. § 1091.
252. For a detailed discussion of enforcement and other costs, see Joel Slemrod & Shlomo Yitzhaki, Tax Avoidance, Evasion, and Administration, in 3 HANDBOOK OF PUBLIC ECONOMICS 1423, 1447-49 (AJan J. Auerbach & Martin Feldstein eds., 2002).
253. The cost of losing limited liability was recognized by several courts. See cases cited supra note 58.
254. Of course, the same argument applies to a taxpayer undertaking a liquidationreincorporation sequence.
255. It also does not include the expected tax liability for the same reason this liability is excluded from the social cost of market risk.
256. For instance, one may argue that the mode of analysis that relies on utility and social welfare functions is inappropriate in the first place.
257. See, e.g., Frank B. Cross, Law and Trust, 93 GEO. L.J. 1457, 1491 (2005) ("'[V]oluntarily risking vulnerability. . . promoters] the attribution of trustworthiness,'" while "[e]xposing oneself to risk . . . provides a prime mechanism for initiating trust."
258. (quoting Svenn Lindstold, Trust Development, the GRIT Proposal, and the Effects of Conciliatory Acts on Conflict and Cooperation, 85 PSYCHOL. BULL. 772, 789 (1978))).
259. Kenneth J. Arrow, Gifts and Exchanges, 1 PHIL. & PUB. AFF. 343, 357 (1972).
260. See Oliver E. Williamson, Calculativeness, Trust, and Economic Organization, 36 J.L. & ECON. 453, 463 (1993) (introducing the term "calculative trust" and explaining why it is "a contradiction in terms").
261. See G. Richard Shell, Opportunism and Trust in the Negotiation of Commercial Contracts: Toward a New Cause of Action, 44 VAND. L. REV. 221, 255 (1991).
262. See Brian Uzzi, Social Structure and Competition in Interfirm Networks: The Paradox of Embeddedness, ADMIN. SCI. Q., Mar. 1997, at 35, 43.
263. See id.
264. See Ranjay Gulati, Does Familiarity Breed Trust? The Implications of Repeated Ties for Contractual Choice in Alliances, 38 ACAD. MGMT. J. 85, 107 (1995).
265. Margaret M. Blair & Lynn A. Stout, Trust, Trustworthiness, and the Behavioral Foundations of Corporate Law, 149 U. PA. L. REV. 1735, 1757 (2001).
266. See id.
267. See Brian Uzzi, Embeddedness in the Making of Financial Capital: How Social Relations and Networks Benefit Firms Seeking Financing, 64 AM. SOC. REV. 481, 490 (1999).
268. See Larry T. Garvin, Credit, Information, and Trust in the Law of Sales: The Credit Seller's Right of Reclamation, 44 UCLAL. REV. 247, 341-42 (1996).
269. Brian Uzzi, The Sources and Consequences of Embeddedness for the Economic Performance of Organizations: The Network Effect, 61 AM. SOC. REV. 674, 677 (1996).
270. See, e.g., Garvin, supra note 147, at 342
271. (citing Albert O. Hirschman, Against Parsimony: Three Easy Ways of Complicating Some Categories of Economic Discourse, 74 AM. ECON. REV. (PAPERS & PROC.) 89, 93 (1984));
272. Lawrence E. Mitchell, The Importance of Bang Trusted, 81 B.U. L. REV. 591, 600 (2001).
273. See, e.g., Gulati, supra note 143, at 94
274. (citing Aravind Parkhe, Strategic Alliance Structuring: A Game Theoretic and Transaction Cost Examination of Interfirm Cooperation, 36 ACAD. MGMT. J. 794 (1993), for the proposition that a history of cooperation reduces perceptions of expected opportunistic behavior).
275. See, e.g., Cross, supra note 137, at 1503 (arguing that trust is multidimensional, meaning that a relationship may contain both trust and distrust).
276. More precisely, this joint surplus also includes the cooperation gain, see infra text accompanying note 162, and is reduced by the tax cost incurred by the Counterparty, if any.
277. c) (which is a benefit to the Counterparty) over the payment made by the Counterparty to the Taxpayer at the inception of the trade.
278. It may appear that this analysis holds only if there is a market for Counterparties who compete against each other for the Taxpayer's business, eliminating any possible rents. It is safe to assume that no such market exists, yet the analysis holds. The market is absent because not all potential Counterparties are the same. The Taxpayer trusts some a lot, others a litde, and many not at all. The stronger the trust, the lower the probability of defection and risk-bearing loss. (Probability of defection affects a risk-bearing loss because if the probability is very small (or very large), the expected income and the certainty equivalent converge.
279. See ROSEN, supra note 8, at 267-69.) Thus, the Taxpayer will choose the most trusted Counterparty as her relational tax planning partner. This suggests that the Counterparty is in a position to capture rents or quasi-rents, depending on the source of trust. (For a distinction between rents and quasi-rents,
280. see, for example, Klein, Crawford & Alchian, supra note 74, at 298-99.) The Counterparty, it appears, is like a monopolist who can set the price that the Taxpayer must pay in order to elicit the Counterparty's cooperation, capturing the amount equal to the additional cost that the Taxpayer will incur if she has to deal with a less-trusted partner. Yet the monopoly here is probably bilateral. Intuitively, if the Counterparty is the Taxpayer's best friend, it is quite likely that the reverse is true as well. If the Counterparty "overplays" his hand, no other equally trusting Taxpayer is standing in the wings to replace the one who went elsewhere. If so, the Counterparty's bargaining power is limited, and he will deal with the Taxpayer even if the benefit captured by the Counterparty is diminishingly small.
281. I assume that tax planning is viewed as similar enough to other informal interactions that the level of trust is the same in all these settings. For further discussion, see infra text accompanying notes 169-170.
282. while a preference change is conceivable, predicting its timing, underlying reasons, and direction remains an extremely challenging task. See, e.g., Robert E. Scott, The Limits of Behavioral Theories of Law and Social Norms, 86 VA. L. REV. 1603, 1605, 1633-37 (2000) (explaining why abandoning a basic assumption of the rational choice theory - that preferences are exogenous and stable - undermines the theory's capacity to generate testable hypotheses). Given the pervasive uncertainty surrounding preferences, there is no reason to assume that their change - if any - is caused by any specific event, especially one as minor as a relational tax planning episode.
283. An even "tougher" strategy would be to break up the business relationship entirely upon any Counterparty defection. Needless to say, many intermediate strategies exist. For a much more detailed discussion of retaliatory threats, see SCHELLING, supra note 75, at 35-43.
284. See, e.g., id. at 17-18 (explaining why abandoning narrow rationality may be a superior negotiating strategy);
285. Avery Katz, When Should an Offer Stick? The Economics of Promissory Estoppel in Preliminary Negotiations, 105 YALE L.J. 1249, 1298 (1996) ("Parties who can establish a credible reputation for stubbornness, spite, or even irrationality will increase tfieir bargaining power ....").
286. Minor defections (deviations from contract terms or requests for their modifications) are an everyday occurrence in commercial relationships. The voluminous literature describing how businesspeople resolve their disputes without resorting to law-yers and legal arguments provides plenty of evidence that these (relatively minor) disputes arise all the time. See, e.g., Lisa Bernstein, Merchant Law in a Merchant Court: Rethinking the Code's Search for Immanent Business Norms, 144 U. PA. L. REV. 1765, 1787-88 (1996) ;
287. Russell J. Weintraub, A Survey of Contract Practice and Policy, 1992 WIS. L. REV. 1, 22 (reporting that on average only 17.1% of respondent firms never ask to modify their contracts, while over 40% ask to do so at least one to five times a year). Restraint in responding to a counterparty's defection is a winning strategy. In a prisoner's dilemma tournament, a disproportionate reaction to defections made "nice" but "unforgiving" strategies less successful than tit-for-tat, a strategy that punishes a defection by one - and only one - defection of its own.
288. See ROBERT AXELROD, THE EVOLUTION OF COOPERATION 35-36 (1984).
289. See, e.g., M. Bensaou & Erin Anderson, Buyer-Supplier Relations in Industrial Markets: When Do Buyers Risk Making Idiosyncratic Investments?, 10 ORG. SCI. 460, 475-76 (1999) (finding that supplier reputation has no effect on the magnitude of relationspecific investments reflective of the level of trust in the buyer-supplier relationship);
290. Uzzi, supra note 148, at 677-80 (noting that in the New York garment industry, general reputation for reliability plays very litde role, while bilateral relationships are extremely valued).
291. If the Taxpayer responds irrationally, all bets are off. The most intuitive reaction is revenge and loss of the existing benefits of informal cooperation. But an irrational Taxpayer may draw exactly the opposite conclusion. She may decide, for example, that once the Counterparty betrayed her at a reliance level of 50, he is less likely to do so again. Predicting reactions of irrational individuals is impossible, which, no doubt, is why economics largely shies away from doing so.
292. The Counterparty is also likely to benefit from the Taxpayer's decision to engage in relational tax planning, although the Counterparty's gain is likely to be much smaller. The Taxpayer's decision to rely on the Counterparty reveals the Taxpayer's trust. But the characteristic that signals reliability and leads to transaction cost savings is trustworthiness. The Counterparty learns nothing about it from the Taxpayer's trusting behavior, so he derives no direct benefit. Scholars have argued, however, that trust building takes place through reciprocal acts of trusting. See sources cited supra notes 137, 149. To the extent the Taxpayer's decision to rely on the Counterparty facilitates the Counterparty's decision to respond in kind, the Taxpayer's reliance indirecdy facilitates trust building. Because this effect is more remote than that considered in the text, I will ignore it in the remainder of the discussion. This does not change the analysis. One can simply think of the cooperation gain as being slightly larger than otherwise to reflect the Counterparty's share.
293. Formal contracting is expensive, yet the parties incur its cost because trust-based reliance is even more cosdy at the margin. With total trust, reliance would be costless and no formal contracts would exist.
294. See Klein, Why Hold-Ups Occur, supra note 74, at 459 n.22 ("[C]ontract terms are set to minimize real costs (and not hold-ups).").
295. See id. at 449-50.
296. Strong support for this suggestion comes from the story told by Thomas Palay, discussed in detail below. See infra text accompanying notes 182-195. Palay documented an extraordinary level of cooperation achieved by some shippers and rail carriers who were forced to avoid extremely restrictive regulations imposed under the Interstate Commerce Act. As soon as the rail industry was deregulated, however, shippers and carriers who had been doing business without any formal documentation started documenting (some of) their dealings. Thus, even for the parties that had already reached a very high level of trust, it was cheaper to formalize some of their contractual relationships, putting less pressure on informal enforcement.
297. For a summary of the literature, see Raskolnikov, The Cost of Norms, supra note 6, at 604-05.
298. Lisa Bernstein and Robert Scott have each argued that contractors have good reasons to deviate from their formal agreements. See Bernstein, supra note 159, at 1796-98 (positing that contractors follow "relationship-preserving norms" to incorporate observable but unverifiable information, adapt to different stages in their relationship, simplify written contracts, and so on);
299. Robert E. Scott, A Relational Theory of Default Rules for Commercial Contracts, 19 J. LEGAL STUD. 597, 613 (1990) (arguing that parties follow a more flexible set of rules for informal enforcement in order to overcome "the inherent limitations of the legal enforcement mechanism"). In fact, Bernstein observed that contractors will not strictly follow the letter of their contracts "as long as they continue to trust one another and/or value potential future dealings."
300. Bernstein, supra note 159, at 1796 (emphasis added). My suggestion is that parties do this in order to continue to develop their mutual trust, or, in Schelling's words, to create a "tradition of trust."
301. SCHELLING, supra note 75, at 45.
302. The Taxpayer will have no incentive to describe the very strategy where she was betrayed because in that case the Taxpayer is the one relying on relational tax planning.
303. An alternative hypothesis is that once a Taxpayer requests Counterparty's assistance in skirting the tax rules, the Counterparty will view the Taxpayer as a less reliable business partner. See Eric A. Posner, Law and Social Norms: The Case of Tax Compliance, 86 VA. L. REV. 1781, 1789-90 (2000). For a rebuttal,
304. see Russell Hardin, Law and Social Norms in the Large, 86 VA. L. REV. 1821, 1822-24 (2000). Hardin's view seems more persuasive. It is also supported by empirical data. Moreover, in contrast with Posner's model, which focuses on taxpayers who fail to comply with tax law, some relational tax planning strategies are almost certainly legal. After all, numerous relational tax planners won in court.
305. Another distinctive trust-building benefit of relational tax planning is that it may produce a nonreplicable combination of benefits and payoffs. The private benefit at stake - the tax savings - is an artificial gain produced by a regulatory regime. As long as costs and benefits of informal business-related cooperation do not vary continuously (and there is no reason to expect that they always do), relational tax planning may provide a cost-benefit combination that is not available to the parties otherwise. To take just one example, the net tax benefit may simply be larger than any net private gain that would arise from non-tax-motivated informal cooperation. More generally, the tax planning benefit will just be different.
306. See CASE STUDIES IN CONTRACTING AND ORGANIZATION, supra note 94, at 7;
307. Williamson, Transaction-Cost Economics, supra note 75, at 242 (referring to "cosdy haggling").
308. See CASE STUDIES IN CONTRACTING AND ORGANIZATION, supra note 94, at 7;
309. Williamson, Transaction-Cost Economics, supra note 75, at 242 (referring to "efficient adaptations" that are not pursued due to strategic bargaining problems).
310. For a similar use of the term, see R.H. COASE, THE FIRM, THE MARKET, AND THE LAW 6 (1988).
311. In addition, cooperation strengthened by relational tax planning may lead to more avoidance of non-tax regulatory regimes. The magnitude of such carryover effect need not be large, however. There is no clear evidence, for instance, that the U.S. manufacturers who formed successful export cartels also colluded to fix prices in domestic markets, even though this kind of carryover is much easier (one would think) than the one just discussed, and even though enforcement in this area has not been particularly strong. See Andrew R. Dick, When Are Cartels Stable Contracts?, 39 J.L. & ECON. 241, 247-48 (1996)
312. [hereinafter Dick, Stable Contracts].
313. That is, the cooperation gain should be added to other costs in the social cost equation.
314. In this case, the cooperation gain should be subtracted from other costs in the social cost equation.
315. See, e.g., James A. Wilson, Adaptation to Uncertainty and Small Numbers Exchange: The New England Fresh Fish Market, 11 BELL J. ECON. 491 (1980) (a study of fresh water fishermen);
316. Comment, The Statute of Frauds and the Business Community: A Re-Appraisal in Light of Prevailing Practices, 66 YALE L.J. 1038 (1957) (a study of Connecticut manufacturers).
317. See, e.g., MARK CAREY ET AL., BOARD OF GOVERNORS OF THE FEDERAL RESERVE SYSTEM, THE ECONOMICS OF THE PRIVATE PLACEMENT MARKET, 4-5 (1993) (a study of the private placement debt market);
318. Scott E. Masten & Edward A. Snyder, United States Versus United Shoe Machinery Corporation: On the Merits, 36 J.L. & ECON. 33, 56-64 (1993) (a study of the shoe-making machine manufacturers and users);
319. Weintraub, supra note 159 (a study of a diverse sample of U.S. companies).
320. Relational contracting allows parties to better respond to uncertainty and information asymmetries, see Simon Deakin et al., "Trust" or Law? Towards an Integrated Theory of Contractual Relations Between Firms, 21 J.L. SOC'Y 329, 333 (1994);
321. act upon information that is observable (i.e., possible and wordiwhile for transactors to obtain) but not verifiable (i.e., not worthwhile for them to prove to a neutral arbiter in the event of a dispute), see Bernstein, supranote 159, at 1791-92;
322. Robert E. Scott, The Case for Formalism in Relational Contract, 94 Nw. U. L. REV. 847, 862 (2000);
323. enforce interior contractual provisions whose violation, while costly, is not harmful enough to justify bringing a law suit, see Lisa Bernstein, Private Commercial Law in the Cotton Industry: Creating Cooperation Through Rules, Norms, and Institutions, 99 MICH. L. REV. 1724, 1761 (2001);
324. David Charny, Nonlegal Sanctions in Commercial Relationships, 104 HARV. L. REV. 373, 394-95 (1990);
325. and more broadly, expand the self-enforcing range of their contractual relationships, see Klein, Why Hold-ups Occur, supra note 74, at 455-56.
326. Only when informal cooperation has macroeconomic effects (such as price changes for particular goods or services) can it be inferred based on econometric analysis. See, e.g., Gareth R. Jones & Michael W. Oustay, Interorganizational Coordination in the Airline Industry, 1925-1938: A Transaction Cost Approach, 14 J. MGMT. 529, 535 (1988) (inferring informal cooperation from the differences between the starting and final prices in airmail route auctions).
327. See Thomas M. Palay, Comparative Institutional Economics: The Governance of Rail Freight Contracting, 13 J. LEGAL STUD. 265, 266-67 (1984)
328. [hereinafter Palay, Rail Freight Contracting] (reporting the difficulties of obtaining information about informal cooperation designed to escape the constraints of the Interstate Commerce Act). Because commercial actors are much more forthcoming about violating the terms of their private agreements, this phenomenon is much better understood.
329. See, e.g., Bernstein, supra note 159, at 1799 (explaining that cotton traders routinely violate the terms of their contracts governing the use of weights);
330. Stewart Macaulay, Non-Contractual Relations in Business: A Preliminary Study, 28 AM. SOC. REV. 55, 61 (1963) (reporting that entrepreneurs view contract cancellations not as contractual breaches, but as an ordinary part of buyer-seller relationships);
331. Sally Falk Moore, Law and Social Change: The Semi-Autonomous Social Field as an Appropriate Subject of Study, 7 LAW & SOC'Y REV. 719, 725 (1973) (describing how jobbers and union representatives "regularly" accede to violations of the union contract).
332. See Thomas M. Palay, Avoiding Regulatory Constraints: Contracting Safeguards and the Role of Informal Agreements, 1 J.L. ECON. & ORG. 155, 156-57 (1985)
333. [hereinafter Palay, Avoiding Constraints].
334. See id. at 156 n.2.
335. In fact, the subjects' willingness to talk without involving the public affairs department was one of Palay's selection criteria. See id.
336. Palay conducted his interviews between October and December of 1979. Id. Deregulation was very much in the air at that time. The airline industry had already been deregulated in 1978.
337. See Airline Deregulation Act of 1978, Pub. L. No. 95-504, 92 Stat. 1705. Congress was considering an influential report describing the severe problems of the rail freight industry under the existing regulatory scheme.
338. See James M. Macdonald & Linda C. Cavalluzzo, Railroad Deregulation: Pricing Reforms, Shipper Responses, and the Effects on Labor, 50 INDUS. & LAB. REL. REV. 80, 80 (1996). The ICC was reversing its restrictive regulatory stands.
339. See Palay, Avoiding Constraints, supra note 182, at 158 n.4. The Staggers Act that deregulated the railroad industry was adopted in 1980, and it was likely being actively discussed in late 1979.
340. See Staggers Rail Act of 1980, Pub. L. No. 96-448, 94 Stat. 1895.
341. See Palay, Avoiding Constraints, supranote 182, at 157-58, 162.
342. This was the case when one of the parties had to make a large transaction-specific investment, such as when a carrier needed to construct auto racks that could carry only the cars made by a particular manufacturer, see id. at 161, or when a shipper had to build a plant that could be serviced only by a specific railroad,
343. see id. at 160.
344. See Palay, Rail Freight Contracting, supranote 181, at 276-77. "[These] informal contracts provided a vehicle for adjusting contract prices or values even where the [Interstate Commerce Act] prohibited such practices."
345. Palay, Avoiding Constraints, supra note 182, at 168.
346. Palay, Rail Freight Contracting, supra note 181, at 277.
347. Id. at 276.
348. Id. at 277, 282, 285-86.
349. Causation could run in the opposite direction: because these transactors had a strong cooperative relationship, they were able to successfully "alter" the unfavorable regulatory regime once the need to do so arose.
350. See Palay, Avoiding Constraints, supra note 182, at 159-60 (providing examples of a shipper that could use any of three rail carriers and a carrier whose cars could be used to transport bulk items of several shippers).
351. See Laurence T. Phillips, Contractual Relationships in the Deregulated Transportation Marketplace, 34 J.L. & ECON. 535, 544-45, 558 (1991) (referring to thousands of longterm contracts filed with the ICC after the 1980 deregulation that contain price escalators, individualized rates, and service level guarantees).
352. This appears to have been Palay's view as well. See Palay, Avoiding Constraints, supra note 182, at 167-68.
353. See Raskolnikov, The Cost of Norms, supra note 6, at 628-29.
354. The repo market developed in the 1910s and 1920s. See STIGUM, supra note 81, at 81-87. The regulatory problem arose in 1922.
355. See First Nat'l Bank in Wichita v. Comm'r, 19 B.T.A. 744, 745 (1930).
356. STIGUM, supra note 81, at 218;
357. see also id. at 3-4, 193-96 (explaining mat for decades, the repo market was highly informal and was an integral part of the interconnected money market, which strongly valued trustworthiness).
358. See Raskolnikov, The Cost of Norms, supra note 6, at 616-18.
359. See id. at 622-25.
360. See id. at 680.
361. The motion picture distribution industry is one example. From its early days, film distributors and exhibitors supplemented their written agreements with an implicit understanding - a combination that maximized the surplus shared by the parties. See Roy W. Kenney & Benjamin Klein, How Block Booking Facilitated Self-Enforcing Film Contracts, 43 J.L. & ECON. 427, 430-32 (2000)
362. [hereinafter Kenney & Klein, Self-Enforcing Film Contracts] (describing an informal understanding among motion picture distributors and exhibitors that allowed exhibitors to replace underperforming films). When the Department of Justice intervened and forced the distributors to accept new ways of marketing their movies, litde changed in practice. Industry participants defied the unwanted requirements through a different kind of informal cooperation.
363. See Roy W. Kenney & Benjamin Klein, The Economics of Block Booking, 26 J.L. & ECON. 497, 519 (1983) (describing how distributors almost never attended showings even though this was a valuable right conferred on them by the consent decree negotiated by the Justice Department). Even after the Supreme Court fundamentally changed the formal rights and obligations of the industry participants, they developed yet another unenforceable norm that allowed the business to continue largely unchanged.
364. See Kenney & Klein, Self-Enforcing Film Contracts, supra, at 433-34 (describing the legal changes and the new informal understanding, and arguing that, as a result of adopting the new custom, film license fees as a percentage of admission revenue remained largely unchanged). This new implicit understanding has survived for at least five decades.
365. See id. at 434. Thus, distributors and exhibitors succeeded in fostering informal cooperation whether it made sense based on the business realities or was needed to circumvent legal constraints.
366. The magnitude of these losses, it is worth remembering, is not equal to the size of the tax savings. For a discussion, see Louis Kaplow, How Tax Complexity and Enforcement Affect the Equity and Efficiency ofthe Income Tax, 49 NAT'L TAX J. 135, 139-40 (1996).
367. In a bilateral context, cooperation and defection have similar effects on the general level of trust in a business environment. However, the link between the actions of specific transactors and the general atmosphere is much weaker in the absence of norms. Information is not disseminated nearly as easily and expectations about specific patterns of behavior are not nearly as well set. Therefore, I have ignored the externality in a bilateral setting.
368. Note that unlike the cooperation gain, the specific externality arises even if a bilateral interaction is a one-shot deal. If a particular donor plans no further gifts to a given charity, she may well make future gifts to other charities. At the same time, the particular charity will receive gifts from other donors. All of these interactions will be made easier (less risky for the donor-Taxpayer) if the charity receiving the gift in question cooperates. Thus, as long as there is a group sharing an informal norm, any interaction in which group members follow this norm strengthens it, yielding the specific externality.
369. In fact, this appears to be exactly how the repo norm affected the money market in general. See supra text accompanying note 198.
370. See ROSEN, supra note 8, at 81-82.
371. Raskolnikov, The Cost of Norms, supra note 6, at 613.
372. See id. at 643-45.
373. Whether the resulting total deadweight loss is larger or smaller depends on the elasticity of taxpayers' response to the cost of relational tax planning.
374. See Raskolnikov, The Cost of Norms, supra note 6, at 629.
375. See id. at 622-25.
376. See id. at 605-07.
377. Tax-relevant norms are less inefficient than tax-driven ones because they are not tax motivated. While they produce allocative distortions, they are also likely to give rise to non-tax-related transaction cost savings typical of many non-tax-driven norms. For a more detailed discussion, see id. at 645-46.
378. See id. at 673-74.
379. B.
380. B > 0).
381. B > 0).
382. Note that this option is not available as a response to traditional tax planning. If taxpayers are allowed to hedge, they will eliminate the risk imposed by risk-based rules completely, making these rules nugatory. The effect will be particularly strong for financial assets because of a wide availability of hedging techniques.
383. For instance, the constructive sale provision may be strengthened by switching to the so-called straddle standard, that is, by triggering a constructive sale if a taxpayer substantially diminishes (rather than eliminates substantially all of) her risk of loss and opportunity for gain from an appreciated security. The step transaction doctrine may be strengthened by collapsing the steps unless more uncertainty exists regarding the later steps when the earlier steps are taken (compared to the current rules).
384. This is so because the price of the loss security in six months may be so high that the Wash Buyer may defect, betray the friendship, and refuse to resell the security to the Wash Seller for the original price.
385. I.R.C. § 1014(e) (2000).
386. More precisely, the Parent is allowed to maintain an interest in the redeeming company, but only as a creditor or if received by bequest or inheritance. See id. § 302(c) (2).
387. The analysis for business risk is similar to that for market risk, and I omit it for brevity.
388. That is, a relational backstop denies the tax benefit to those who comply with the letter of the law but have an informal understanding that reduces the intended degree of risk.
389. This backstop will need to be expanded to incorporate not only enforceable contracts, but formal ownership structures as well. For instance, relational tax planning between two wholly owned subsidiaries should clearly not entitle either entity to claim any tax benefits.
390. See supra text accompanying notes 15-16.
391. It may appear that implicit agreements simply cannot replicate complex financial instruments used by taxpayers to hedge market risk. That much is true, but it proves little. To engage in relational tax planning, taxpayers need not abandon the realm of formal contracts entirely. They only need to remove a critical term from the enforceable (and observable) written agreement. Plenty of evidence suggests that taxpayers have successfully done this on many occasions even where the underlying formal contracts were quite complex. See Raskolnikov, The Cost of Norms, supra note 6, at 651-65.
392. For instance, the backstop may be relatively weak. It may be triggered only if taxpayers have a contract that is not legally binding only because it violates some formality (e.g., it lacks consideration or does not comply with the statute of frauds). Strong cooperators do not need to enter into such contracts to cooperate successfully. A more robust provision would deny a tax benefit if taxpayers have an explicit agreement, even though it is not specific enough to give rise to a contract. Strong cooperators can transact without reaching such explicit agreements. An even more farreaching rule would apply to taxpayers who had any discussions about a particular transaction at all. Strong cooperators can successfully rely on each other without discussing specific deals. Thus, strong cooperators are beyond the reach of the relational backstop.
393. To simplify the discussion, I focus here only on legal issues. Another difference among taxpayers is their ability to hide aggressive positions. Incorporating that difference does not alter the analysis. Weak cooperators who are particularly good at hiding their informal understandings will simply be in the same position (and will act) as strong cooperators.
394. Kg., I.R.C. § 267(b) (2000);
395. id. § 707(b). These rules will continue to be needed because even in the alternative regime taxpayers will bear no risk at all in some cases (such as where the Taxpayer controls the Counterparty), and too little risk in others (such as in transactions between parents and children).
396. This conclusion may be unduly optimistic. See infra text accompanying notes 237-238.
397. Tax planning may be overdeterred not because (like a polluting widget-making factory) it is socially beneficial at a certain level, but because at some point the full social cost of deterring it (a cost that includes risk-bearing and other losses borne by those who are not deterred) combined with the cost imposed on innocent taxpayers exceeds the benefit from decreased deadweight losses of tax planning. For a detailed explanation of why setting expected penalties in excess of external harm produced by a particular conduct is inefficient, see, for example, Richard Craswell, Deterrence and Damages: The Multiplier Principle and Its Alternatives, 97 MICH. L. REV. 2185, 2195 (1999),
398. and A. Mitchell Polinsky & Steven Shavell, Punitive Damages: An Economic Analysis, 111 HARV. L. REV. 869, 877-87 (1998). Perhaps because tax noncompliance is widely viewed as producing no social benefit whatsoever, tax enforcement literature has been largely unconcerned with the problem of overdeterrence.
399. See Joel Slemrod & Shlomo Yitzhaki, The Cost of Taxation and the Marginal Efficiency Cost of Funds, 43 INT'L MONETARY FUND STAFF PAPERS 172, 182 (1996) (describing tax evasion models as implying that it is optimal to prevent all evasion and ignoring the social costs of deterrence).
400. I am making simplifying assumptions that (i) weak cooperators who enter into informal, not legally binding agreements reduce counterparty risk to the same degree as strong cooperators who enter into no agreements at all, and (ii) the marginal benefit from entering into informal agreements for strong cooperators is diminishingly small (because they already trust each other "fully").
401. This is so because loners are assumed to be appropriately deterred by the current market-risk-imposing rules, and the amount of market risk borne by loners will necessarily increase following a switch to the alternative regime.
402. See, e.g., George K. Yin, JCT Chief Discusses the Tax Gap, 107 TAX NOTES 1449, 1449 (2005) ("[A]ny consideration of major tax reform in this country must give primary consideration to issues of tax compliance and enforcement"). The Government Accountability Office named tax enforcement as one of its "High Risk Areas."
403. See Allen Kenney, Tax Enforcement Makes GAO's 2005 List of "High Risk Areas, "106 TAX NOTES 531, 531 (2005) ("Given the broad declines in IRS's enforcement workforce, IRS's decreased ability to follow up on suspected noncompliance, the emergence of sophisticated evasion concerns, and the unknown effects of these trends on voluntary compliance, IRS is challenged on virtually all fronts in attempting to ensure that taxpayers fulfill their obligations." (internal quotation marks omitted)
404. (quoting U.S. GOV'T ACCOUNTABILITY OFFICE, HIGH RISK SERIES: AN UPDATE 38 (2005), available at http://www.gao.gov/new.items/d05207.pdf) ).
405. This type of planning is akin to an exchange of "hostages" that can be used as a device to overcome opportunism and bounded rationality problems in contracting. See Williamson, Credible Commitments, supra note 75. at 524, 531. The analogy, however, is closer not to "hostages" but to "bonds" - a less effective commitment device. A unique feature of a hostage is that it is valuable to the party giving it, but not to the party receiving it.
406. See id. at 526-27. Bonds, in contrast, are equally valuable to both sides. Therefore, bonds are more likely to be expropriated than hostages.
407. See id. If the Wash Seller "sells" the loss security to the Wash Buyer and the security doubles in value during the extended waiting period, the increased value is as attractive to the Wash Buyer as it is to the Wash Seller.
408. While the agreements need not be unwritten in order to be unenforceable as a matter of contract law, the government is free to treat any written agreement as enforceable for tax purposes if it decides that this is needed to reach the desired level of deterrence. This may not be necessary, however, because nothing would stop tax planners from producing competing writings describing their informal agreements.
409. This will be the case if one security appreciates while the other declines in value. A party who "owns" an appreciated security (to be resold at what turns out to be a below-market price) and has an informal agreement to repurchase a depreciated security (at what turns out to be an above-market price) will be sorely tempted to defect on both trades.
410. The Webb-Pomerene Act of 1918, 15 U.S.C. §§ 61-66 (2000), and the Export Trading Company Act of 1982, 15 U.S.C. §§ 4001-4003, 4011-4021, granted antitrust immunity to U.S. manufacturing exporters. See Dick, Stable Contracts, supra note 174, at 245-48;
411. Margaret C. Levenstein & Valerie Y. Suslow, The Changing International Status of Export Cartel Exemptions, 20 AM. U. INT'L L. REV. 785, 789-90 (2005). The CapperVolstead Act of 1922, 7 U.S.C. §§ 291-292, and the Agricultural Marketing Agreement Act of 1937, 7 U.S.C. §§ 671-674, awarded antitrust immunity to agricultural cooperatives.
412. See Darren Filson et al., Market Power and Cartel Formation: Theory and an Empirical Test, 44 J.L. & ECON. 465, 466 (2001).
413. See Andrew R. Dick, Identifying Contracts, Combinations and Conspiracies in Restraints of Trade, 17 MANAGERIAL & DECISION ECON. 203, 206 (1996)
414. [hereinafter Dick, Restraints of Trade] (reporting that Webb-Pomerene cartels "collectively ... accounted for just 5% of total manufactured exports," even though cartelization varied significantly by industry);
415. Dick, Stable Contracts, supra note 174, at 242 ("[N]early one-quarter of Webb-Pomerene agreements collapsed within 2 years ....");
416. Filson et al., supra note 240, at 472-73 (reporting that out of 182 possible instances of cartelization, only 39 (or just over 20%) actually occurred, despite the government's assistance in enforcing the cartel's terms).
417. Agricultural cartels were government enforced. See Filson et al., supra note 240, at 466.
418. George J. Stigler, A Theory of Oligopoly, 72 J. POL. ECON. 44, 44 (1964).
419. See supra notes 181-191 and accompanying text.
420. An additional imperfection of the alternative regime is that it will impose varying costs on taxpayers who have different opportunities (and desires) to cooperate. The same is true in the current system, however. Moreover, the existing market-risk-imposing rules are similarly imprecise because, for example, they fail to take into account variability in riskiness of various assets covered by a single waiting-period provision.
421. The main enforcement cost in the alternative regime will be related to policing the line between enforceable and unenforceable agreements. Modern contract law gives us a good idea about the issues involved.
422. A hard-nosed economist may point out that relational tax planning involving friends who have no ongoing business relations should lead to stronger personal friendships (just like it should create stronger business ties for long-term business partners). The cooperation gain in this case will reflect private and social benefits of a friendlier society. This may be true, yet I resist this extension because it is much harder to say anything definitive about the cooperation gain when personal friendships are concerned. The rational analysis of defections in business relationships offered above probably does not apply to friends. A relatively minor betrayal in a relational tax planning episode may well damage a friendship substantially, or even ruin it completely. If so, it is ambiguous whether the cooperation gain produced by relational tax planning among friends is, on balance, a private and social benefit or cost. As on other occasions when I encounter this type of uncertainty, I assume that the net benefit or cost is roughly zero.
423. See supra note 26.
424. Id.
425. Schizer, Frictions, supra note 10, at 1318, 1372-74.
426. Id. at 1318, 1374.
427. For simplicity, I ignore the fact that the wash sale provision allows some formal hedging.
428. This is because, in any other case, the Counterparty's decision will be affected by the value of the asset received as a "gift" from the Taxpayer and the temptation to retain it for personal gain. When the Counterparty is a decedent, this analysis clearly does not apply.
429. In fact, a variety of waiting periods apply in very similar settings. See supra note 129. For a discussion of the profound uncertainty surrounding the concept of tax ownership, see Alex Raskolnikov, Contextual Analysis of Tax Ownership, 85 B.U. L. REV. 431 (2005).
430. More precisely, this used to be the case before the repo market became formalized. See STIGUM, supra note 81, at 218-21.
431. To take a business risk example, consider the special provision designed to deter drop-and-sell schemes. If an owner selling her business incorporates it having already reached an informal agreement to sell the newly received stock, the risk is lower than if no such agreement exists. It is unclear, however, how long an owner must hold the unwanted stock without any implicit understandings about the stock's future sale in order to satisfy the requirement that she control the corporation "immediately after the exchange." I.R.C. § 351(a) (2000).
432. Admittedly, since it is unclear how courts and policymakers make these comparisons today, one can never rule out a possibility that their decisions already reflect the analysis offered below.
433. In fact, judges may even take into account the magnitude of the expected business loss. It may be reasonable to conclude, for instance, that this loss is relatively high when a taxpayer must close her business and later restart it anew, lower when she must operate the business without limited liability (both scenarios are possible in the liquidation-reincorporation case), and lower still when she must continue to own the business in an incorporated form for some time (the drop-and-sell scenario). The substance-over-form and step transaction doctrines are clearly broad enough to allow these types of inquiries.
434. See, e.g., Kornfeld v. Comm'r, 137 F.3d 1231, 1235 (10th Cir. 1998) ("[Where] the parties to the transactions in question are related, the level of skepticism as to the form of the transaction is heightened, because of the greater potential for complicity between related parties in arranging their affairs in a manner devoid of legitimate motivations." (internal quotation marks omitted) (quoting Gordon v. Comm'r, 85 T.C. 309, 325-26 (1985))).
435. This is not always true. For instance, as Lisa Bernstein and Barak Richman explain, the diamond trade is impossible without a high level of trust between the traders. See Lisa Bernstein, Opting Out of the Legal System: Extralegal Contractual Relations in the Diamond Industry, 21 J. LEGAL STUD. 115 (1992);
436. Barak D. Richman, How Community Institutions Create Economic Advantage: Jewish Diamond Merchants in New York, 31 LAW & SOC. INQUIRY 383 (2006). In this kind of environment, an increase in the cooperation gain may produce a discontinuous change from almost no contracting to an extensive relationship.
437. This is probably a conservative assumption. As transactors develop trust, the share of informal contracting tends to rise. See, e.g., Bensaou & Anderson, supra note 160, at 475 (finding that more multifaceted, intense, and eventful buyer-supplier relationships produce larger relation-specific investments indicative of stronger trust). Therefore, the more extensive the relationship, the less formal it is likely to be.
438. The parties have no incentive to reveal information about their additional relational tax planning because this information would tend to suggest that the cooperation gain is a social cost and courts should be particularly unforgiving. At the same time, self-serving evidence of non-tax-motivated informal contracting will be highly unreliable. The evidence will be self-serving because it will tend to show that relational tax planning facilitates socially valuable commercial exchanges, suggesting leniency from a court. Unlike in contract disputes, where one of the parties has an incentive to demonstrate the course of dealing or the course of performance while the other party has an incentive to dispute this showing, neither the Taxpayer nor the Counterparty will have a reason to assist the IRS in questioning the evidence of socially valuable non-tax-driven relational contracting that is certain to be produced by the Taxpayer in order to argue that the cooperation gain is positive for the society as a whole.
439. Even this fairly narrow description covers wash sales, dividends-received deduction and foreign tax credit schemes, norm-based and bilateral repos, see supra notes 71 and 82, the heardess relative example,
440. see supra text accompanying note 222, the Parent/Children example,
441. see supra text accompanying note 223, the drop-and-sell situation,
442. see supra text accompanying note 52, some liquidation- reincorporation cases,
443. see supra text accompanying note 54, and many other situations.
444. See, e.g., United States v. Cumberland Pub. Serv., 338 U.S. 451 (1950);
445. Comm'r v. Court Holding Co., 324 U.S. 331 (1945);
446. Walker v. Comm'r, 544 F.2d 419 (9th Cir. 1976);
447. Steel Improvement & Forge Co. v. Comm'r, 314 F.2d 96 (6th Cir. 1963).
448. The basic insight is that if there are ten alternative relational tax planning strategies, and if taxpayers will receive tax benefits of equal value by pursing any one, they are indifferent about which strategy to follow as long as the costs are the same. If, however, pursuing any of the nine strategies costs 10 (dollars or utility units) while following the tenth one costs only 2, the effectiveness of 90% of the rules designed to deter relational tax planning is undermined because taxpayers will simply reduce their taxes by following the tenth strategy. (Eventually their marginal cost may rise to the level of following the other nine, but by then a lot of tax planning will have occurred.) For a fuller discussion of marginal deterrence in tax administration, see Alex Raskolnikov, Crime and Punishment in Taxation: Deceit, Deterrence, and the Self-Adjusting Penalty, 106 COLUM. L. REV. 569, 609-12 (2006).
449. A relationship is "thick" if it is multifaceted, intensive, and potentially long lasting. For a similar use of the term, see Melvin Aron Eisenberg, The Limits of Cognition and the Limits of Contract, 47 STAN. L. REV. 211, 251 (1995).
450. See, e.g., Bensaou & Anderson, supra note 160, at 475.
451. It is worth noting that at some point the cooperation gain will start declining as the relationship becomes thicker. An assumption that marginal returns to trust decline as the trust grows appears plausible. If the Taxpayer already trusts the Counterparty "like a brother," there is not much room for improvement. Few commercial parties, however, are likely to approach this level of trust.
452. Opinions routinely emphasize that a particular counterparty is a taxpayer's long-term customer, employee, and so on. See, e.g., Granite Trust Co. v. United States, 238 F.2d 670 (1st Cir. 1956) (taxable liquidation case, 21-Holder is the Liquidator's long-term customer);
453. Comm'r v. Day & Zimmermann, 151 F.2d 517 (3d Cir. 1945) (taxable liquidation case, 21-Holder is the treasurer of the Liquidator);
454. Citizens Nat'l Bank of Waco v. United States, 551 F.2d 832 (Ct Cl. 1977) (isolated repo case, the Counterparty is a long-term bank of the Taxpayer).
455. See, e.g., Stein v. Comm'r, 36 T.C.M. (CCH) 992 (1977) (friendly wash sale case, the Counterparty is a Taxpayer's friend, but the two have no ongoing commercial relationship).
456. For example, assume that it is sufficient for the Taxpayer to part with (or hold) the asset under an informal agreement with an occasional business partner for two months. If the same transactors have extensive commercial dealings, the waiting period should be longer (say, three months). Obviously, the numbers are used just to illustrate the progression. Taking a more general approach relying on the current step transaction doctrine, assume that the one-off deal is scrutinized under the doctrine's least far-reaching binding commitment version. A more expansive interdependence test should then apply to a thick relationship.
457. See Comm'r v. Gordon, 391 U.S. 83, 96 (1968) (establishing the binding commitment test);
458. McDonald's Rests, of Ill., Inc. v. Comm'r, 688 F.2d 520, 524-25 (7th Cir. 1982) (enunciating the interdependence test). Note that this conclusion is independent of the level of trust between relational tax planners. While thick relationships usually correspond to high trust, an occasional business deal between long-term friends may also take place in a high-trust setting. The former should be scrutinized more for the reasons discussed in the text.
459. A norm environment is "thick" if its members follow many different norms in dieir interactions with each other.
460. The cooperation gain may be larger in a thick bilateral relationship (when parties follow only one or two norms in a thin norm environment, enhanced trust can smooth only a few interactions), and the bargaining cost is probably low in both settings. Yet the key difference is in the amount of risk. Sanctions for defection in a norm environment are stronger because a failure to follow a norm will be punished not only by the offended party, but by many other members of the group as well. This risk-reducing effect of norms almost certainly overshadows the difference in the cooperation gain. Continuing with our example, the holding period for a Taxpayer who follows a well-established but isolated norm should be, say, four months (that is, longer than the three months required for an extensive dyadic relationship). If the norm environment is thick, only holding the asset for half a year should be enough to capture the tax benefit. More broadly, the most far-reaching end result version of the step transaction doctrine should apply to norm-based exchanges. See McDonald's Rests., 688 F.2d at 524 (explaining the end result test).
461. Evidence of socially valuable norms is much less falsifiable than evidence of non-tax-motivated relational contracting because a court may demand independent verification of the existence of norms.
462. See, e.g., James M. Peaslee, Circular 230: Make Room for Informal Written Advice, 106 TAX NOTES 1457, 1457 (2005) (referring, in a letter to the IRS, to "some practitioners [who write] low-grade, canned tax opinions" that "assume a business purpose or profit motive where none exists"). The government has certainly taken notice.
463. See, e.g., Circular 230, 31 C.F.R. § 10.35(c)(1)(ii) (2007) ("[I]t is unreasonable [for a tax practitioner giving an opinion] to assume that a transaction has a business purpose.").
464. See, e.g., Long Term Capital Holdings, LP v. United States, 150 F. App'x. 40, 42-43 (2d Cir. 2005).
465. See, e.g., First Am. Nat'l Bank of Nashville v. United States, 467 F.2d 1098 (6th Cir. 1972);
466. Union Planters Nat'l Bank of Memphis v. United States, 426 F.2d 115 (6th Cir. 1970);
467. Am. Nat'l Bank of Austin v. United States, 421 F.2d 442 (5th Cir. 1970);
468. Am. Nat'l Bank of Austin v. United States, 573 F.2d 1201, 1205 (Ct. Cl. 1978).