Shareholder opportunism in a world of risky debt

Harvard Law Review

Volumn 123, Issue 5, 2010, Pages 1151-1213

  Squire, Richard ^a  

^a FORDHAM LAW SCHOOL   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77955499947     PISSN: 0017811X     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (240)

1. For a firm's unsecured creditors to bear the full brunt of a contingent debt that is triggered when the firm is insolvent, the firm's equity investors must enjoy limited liability, which is why those investors are referred to here as "shareholders," implying a corporation. But limited liability is a feature of most of the other widely used modern business entities, including the limited liability company, limited liability partnership, and Delaware statutory trust.
2. See Henry Hansmann, Reinier Kraakman & Richard Squire, Law and the Rise of the Firm, no. HARV. L. REV. 1333, 1397 (2006). Only the common law partnership continues to hold equity investors fully liable for firm debts.
3. One possibility is that a court would deem this hypothetical contract an improper "ipso facto" arrangement. Provisions of the Bankruptcy Code deny enforcement of ipso facto clauses in specific contexts.See n IJ.S.C. § 365(e)(I)(A) (2006) (executory contracts).
4. id. § 541(c)(I)(B) (debtor interests in property). And several courts have drawn from these provisions and from legislative history a principle that ipso facto clauses generally are against public policy.
5. See Michael J. Di Gennaro & Harley J. Goldstein, Can Ipso Facto Clauses Resolve the Discharge Debate?: An Economic Approach to Novated Fraud Debtin Bankruptcy, 1 DEPAUL BUS. & COM. LJ. 417, 443 (2003) (collecting cases). Alternatively, a court might deem the contract a deliberate fraudulent transfer.
6. See II U.S.C. § 548(a)(1)(A) (empowering bankruptcy trustees to invalidate obligations incurred "with actual intent to hinder, delay, or defraud" creditors.
7. See II U.S.C. § 548(a)(i)(B)(i). The application of fraudulent transfer doctrine to contingent debt is discussed in section III.B.
8. The claimant will actually pay something less than $10 due to the time value of money and her expectation that she will not to be paid in full if her claim is triggered.
9. The implicit assumption here is that the contingent debt is not large enough itself to cause the debtor's insolvency, which therefore must arise from an independent source, such as a business downturn that diminishes thevalue of the debtor's assets. The alternative possibility - a contingent debt big enough to cause insolvency by itself - is addressed in section LB.2
10. See Lucian Arye Bebchuk & Jesse M. Fried, The Uneasy Case for the Priority of Secured Claims in Bankruptcy, 105 YALE L.J. 857, 864 (1996) (describing why many creditors do not protect themselves from the risk that theirdebtor will use a secured loan to subordinate their claims).
11. Elizabeth Warren & Jay Lawrence Westbrook, Contracting Out of Bankruptcy: An Empirical Intervention, 118 HARV. L. REV. 1197, 1226 (2005) (arguing that many creditors are unlikely to find it cost-effective to enforce protective loan covenants).
12. See Marcel Kahan & Edward Rock, Hedge Fund Activism in the Enforcement of Bondholder Rights, 103 Nw. LT. L. REV. 281, 296-301 (2009).
13. The authors argue that activist hedge funds can overcome the underenforcement problem, id. at 301-02, but they note that incentive-based obstacles remain, id. at 309
14. See Richard Squire, The Case for Symmetry in Creditors' Rights, 118 YALE L.J. 806, 840-41 (2009) (making the same point with respect to the secured loan).
15. Two sets of such rules are discussed in Part III.
16. ACCOUNTING FOR CONTINGENCIES, Statement of Fin. Accounting Standards No. 5, para. 8 (Fin. Accounting Standards Bd. 1975).
17. Id. para. 10.
18. An exception applies to guarantees, which the Financial Accounting Standards Board has decided should be disclosed even if the contingency risk is remote. Id. para. 12.
19. see also GUARANTOR'S ACCOUNTING AND DISCLOSURE REQUIREMENTS FOR GUARANTEES, INCLUDING INDIRECT GUARANTEES OF INDEBTEDNESS OF OTHERS, Interpretation No. 45 (Fin. Accounting Standards Bd. 2002) (elaborating on accrual and disclosure requirements for guarantees).
20. See Marcel Kahan & Michael Klausner, Antitakeover Provisions in Bonds: Bondholder Protection or Management Entrenchment?, 40 UCLA L. REV. Q31, 940-41 (1993).
21. Yedidia Z. Stern, A General Model for Corporate Acquisition Law, 26 J. CORP. L. 675, 706-08 (2001).
22. Squire, supra note 8, at 819-20
23. These figures make for a leverage ratio of 4:1, a conservative level that permits analysis of a contingent debt that is relatively large but notbig enough in itself to cause insolvency. Higher leverage increases the likelihood that the contingent debt will be sufficient to cause insolvency, a possibility addressed in section LB.2.
24. In theory, a contingent debt contract might generate efficiencies that benefit the claimant, and the debtor might be able to capture the value of those efficiencies in the premium. The most obvious efficiency - assumingthe claimant is risk-averse - would be to smooth out returns on the claimant's investment portfolio. However, the model already imputes such a benefitby assuming that Claimant does not apply a risk-based discount when valuating her uncertain recovery on the contract. A second possibility would be that the contract induces the debtor to try to prevent the triggering event, such as by monitoring the "reference" entity. This potential benefit is perhaps most likely with certain types of guarantee.
25. In the world of corporate debt this would make Debtor a high credit risk; the discussion later considers the implications if Debtor were safer. Also, setting the downturn risk equal to the contingency risk permits analysis of the full range of positive internal correlations. For example, if Debtor's downturn risk were 10% but the contingency risk were only 5%, then the highest possible internal correlation would be 0.69, See infra note 20.
26. This simple model can be formalized with the following terms, which will be used for all formulae in this Article: A: initial value of Debtor's assets 11: amount Debtor owes Unsecured Creditor C:face value of debt to C aimant V: Debtor's total liabilities after contract is signed, or U + C d: discount factor (between o and i) used to determine value of Debtor's assets in down-turn p(R): probability that contingent debt is triggered p(D): prprobability that Debtor suffers an asset downturn.
27. The baseline loss equals the difference between the face amount of Unsecured Creditor's claim and his expected recovery if there is no contract with Claimant, and hence the only risk he faces is the downturn risk. In general terms, it can be expressed as: B=U - [p(not D)*U + p(D)*dA] Because p(nol D) equals I - p(D), this formula simplifies to: (1) B=p(D)*[U-dA]. Under the numerical assumptions specified above, the baseline loss is $ 2.50.
28. y equals p(D) if we assume (as we do in Figure 2) that Debtor is insolvent only if a downturn occurs. A discrete expected value for the product of the differences between each dummy variable and its mean - that is, a discrete covariance for X and Y - can be obtained for each possible distribution of probabilities across the model's four outcomes, producing a one-to-one correspondence between each possible probability distribution and each correlation coefficient represented along Figure 2's x-axis.
29. In terms of Figure 1, a higher internal correlation means higher probabilities of the outcomes (insolvent, triggered) and (solvent, not triggered), and lower probabilities of the outcomes (insolvent, not triggered) and (solvent, triggered).
30. The transfer is computed in two steps. First, the premium is calculated, which by assumption equals Claimant's expected recovery on her contractwith Debtor. Assuming that A ≥ V and that Claimant is unsecured, the formula for the premium is: P = p(not D\R)*C + p(D\R)*[C/V]*d[A + P]. Solved for P, this equation becomes: (3) P = [p(not D\R)*CV + p(D\R)*dA*C]/[V-p(D\R)*dC]. The wealth transfer away from Unsecured Creditor is defined as Unsecured Creditor's expected recovery without the contract between Debtor and Claimant minus his expected recovery when the contract is in place. It can be expressed as: T = p(D)*dA - p(D\not R)*d[A + P] -p(D\R)*d[A + P]*[U/V]. Because Unsecured Creditor's recovery is not affected by the contract when Debtor does not fall insolvent, those outcomes are excluded from the equation. Because p(D) equals p(D\not R) plus p(D\R), the equation simplifies to: (4) T =p(D\R)* d[A - (A + P)*(U/V)]- p(D\not R)*dP.
31. The figure's diagonal line is created by plugging the outcome probabilities corresponding to each correlation level into equations (3) and (4) in note 22, supra, and then dividing T by B as calculated by equation (1) innote 19, supra. As described in note 20, supra, each value for the correlation coefficient in Figure 2 corresponds to a discrete distribution of probabilities across the model's four possible outcomes.
32. A correlation coefficient of 1.0 means that only the outcomes (solvent, not triggered) and (insolvent, triggered) can occur, with the probability of outcome (solvent, not triggered) being 90% and of outcome (insolvent, triggered) being 10%. The opposite extreme, meing a correlation coefficientof -1.0, would require that only the outcomes (solvent, triggered) and (insolvent, not triggered) could occur, which is impossible given the assumptions that the triggering event and the asset downturn are each only 10% likely. Rather, the lowest possible correlation occurs when the outcome (solvent, triggered) is 10% probable, the outcome (insolvent, not triggered) is 10%probable, and the outcome (solvent, not triggered) is 80% probable, which yields a correlation coefficient of-0.1.
33. The nominal value of the transfer at this correlation level is $0.15.
34. In nominal terms, T is $1.43 when the internal correlation is perfect.
35. The fact that an uncorrelated contingent debt produces only a trivialwealth transfer is robust to a broad range of numerical assumptions consistent with the model's four-outcome structure. By the four-outcome structure, what is meant are the parameters that Debtor starts out solvent, an asset downturn renders Debtor insolvent, Claimant pays a premium equal to herexpected recoverj' on the contract, and - in this first case - the face value of the contingent debt is not greater than Debtor's starting equity value. Thus, if we hold constant the assumption that the amount owed Claimant is$25 and that the downturn and contingency risks are 10% each, but toggleall other variables, no combination of alternative assumptions that isconsistent with these parameters causes an uncorrelated contingent debt totransfer more than $ 0.06.
36. Given the assumptions used in Figure 2, the formula for the increase in Debtor's expected equity value attributable to the contract is: G = p(not D\not R)*(A + P - U) + p(not D\R)*(A +P- V) -p(not D)*(A- V); with P calculated using equation (3) in note 22, supra. Because by definition p(not D) equals p(not D\not R) plus p(not D\R), the formula simplifies to:(5) G = pfnot D\not R)*P + pinot D\R)*(P - C). For example, at a correlation of 1.0, G equals $1.37, as contrasted with a T (per equation (4) in note 22, supra) of $1.43. The $0.06 difference between the gain to Shareholders and the loss to Unsecured Creditor is attributable to the decline in value of the premium assets held by Debtor when Debtor suffers a downturn.
37. This result is calculated by using equations (3) and (4) in note 22, supra, and assuming that p(not D\R) is 90% and p(D\R) is 10%, making the probability of the triggering event 100%. The nominal amount of the transfer is $ 0.33.
38. See sources cited supra notes 13-14.
39. For example, the ratio between the wealth transfer amounts would fall from 4.3 to 3.7 if we assumed that Debtor's assets lost 90% of their value in a downturn rather than 40%. This is because a deeper devaluation destroys more Debtor wealth when the premium is larger, and a fixed claim fetches a larger premium than a contingent claim. Thus, increasing the devaluation percentage raises the relative dilutive effect of the fixed debt. The ratio would further drop to2.5 if we also assumed that Debtor's assets started out worth $150 rather than $125. This occurs because a fixed debt harms unsecured creditors primarily by increasing the debtor's debt-to-equity ratio, and the impact would be relatively larger when the initial ratio is smaller. This ratio, however, is completely insensitive to the debt's face value. If we were to use the other assumptions in Figure 2 but cut the face values of Claimant's fixed and contingent claims to $ 10 each, or double them both to $ 50, the ratio would remain 4.3.
40. For an explanation of the model's general, four-outcome structure, see supra note 27.
41. The only way to get the wealth transfer ratio below 2.0 given the model's general structure is to assume a much smaller initial value for the amount owed Unsecured Creditor, and thus an unusually low leverage ratio. Andthis assumption, in turn, requires a further assumption that Debtor's assets suffer a deeper devaluation in a downturn, for otherwise Debtor is not insolvent when a downturn occurs. For example, if we assume that Unsecured Creditor is owed $50 rather than $ 100, and a downturn causes Debtor's assets to depreciate 70% rather than 40%, the wealth transfer produced by a $25 contingent debt with a perfect internal correlation is only 1.6 times greater than the transfer produced by a $ 25 fixed debt.
42. In particular, it would increase from 4.3 to 4.9.
43. It would fall to 3.4.
44. This result is robust to a wide range of available numerical assumptions. For example, the point at which the contingent debt line crosses the ixed debt line in Figure 2 remains between 0.2 and 0.4 if we assume that Debtor's assets depreciate 90% rather than 40% in a downturn, that the initial value of those assets is $150 rather than $125, or that Unsecured Creditor is owed $ 80 rather than $100. An important exception occurs with a change in the face value. As is seen in Figure 3, doubling the face value causes the contingent debt to transfer more wealth than the fixed debt at all available correlation levels.
45. Louis K.C. Chan et al., On Portfolio Optimization: Forecasting Covariances and Choosing the Risk Model, 12 REV. FIN. STUD. 937, 942 (1999). The result was obtained by randomly selecting 500 stocks from the New York andAmerican stock exchanges for each year from 1968 through 1998, and calculating an average pairwise correlation for all sets. See id.
46. To be sure, the correlation between two companies' stock returns is not the same as the correlation between one's stock performance and the other's insolvency risk, even though the two measures of correlated risk are likely to be similar. The cited figure nonetheless suggests that the internalcorrelations on most contingent debts found in the economy probably are significantly greater than zero.
47. See Chan et al., supra note 37, at 943-44.
48. Many contingent debt contracts do not specify a face value, but rather expose the debtor to a range of possible liability amounts based on the performance of the underlying "reference" asset. For example, the liability generated by a put option depends on how far the price of the reference stock drops below the strike price. If the liability range on a contingent claim straddles the value of the debtor's equity, then the probability-weighted mean value of the liability determines whether a graph of the wealth transfer relative to the correlation level looks like Figure 2 or Figure 3.
49. The formula for the premium that Claimant pays when V > A is: P'=p(not D\R)*[C/V]*[A + P] + p(D\R)*[C/V]*d[A + P]; which solved for P' gives: (6) P'= AC*[p(not D\T) + p(D\T)*d] /[V- p(not D\T)*C - p(D\T)*dC]. The formula for the wealth transfer when V > A, defined again as the difference between Unsecured Creditor's expected recovery7 without a contract with Claimant and his recovery when there is one, is: T' = p(D)*dA + p(not D)*U - p(D\not R)*d[A + P'] - p(D\R)*d[A+P']*[U/V] - p(not D\not R)*U - p(not D\R)*[A + P']*[U/VJ; which can be simplified as: (7) T' = p(D\R)*d[A - (A + P')*(U/V)] + p(not D\R)*[U - (A + P')*(U/V)] - p(D\not R)*dP. The value on the y-axis line in Figure 3 is calculated by dividing T by B as computed in equation (1) in note 19, supra.
50. The correlation coefficient in Figure 3 is calculated using equation (2) in note 20, supra, except that Y (the variable corresponding to the insolvency risk) is now set at 1 (representing insolvency) rather than o (representing solvency, as in Figure 2) in the outcome where a downturn does nooccur but the contingent debt is triggered.
51. This ratio falls if we further increase the contingent debt's face value.
52. See infra pp. 1187-88.
53. To be sure, this larger expected recovery translates into a larger premium. But as was true for an unsecured claim, at positive correlation levels the larger premium is insufficient to counter act the claim's dilutive effect on the unsecured creditors.
54. For a secured claim with full priority over all of Debtor's assets, assuming C ≤ dA, the formula for the premium is:(8) P" = p(R)*C.Assuming that A > V, the formula for the transfer is:(9) T" = - p(D\R)*[dP"- C]-p(D\not R)*dP". Plugging the other assumptions used in Figure 2 into these formulae gives a transfer for a perfectly correlated contingent claim of $2.35, as contrasted with a transfer of $1.43 when the claim was unsecured.
55. Based on equations (8) and (9) in note 46, supra, the nominal transfer produced by the secured fixed debt is $1.00, as compared with a transfer of $0.33 when the debt was unsecured. This change represents a marginal increase of $0.67, as compared with the marginal increase of $0.92 for the perfectly correlated contingent debt.
56. For a review of this literature, see Squire, supra note 8, at 863-65.65.
57. See WILLIAM A. KLEIN & JOHN C. COFFEE, JR., BUSINESS ORGANIZATION AND FINANCE 345-46 (10th ed. 2007) (noting that higher leverage increases the volatility of equity returns).
58. Yakov Amihud, Kenneth Garbade & Marcel Kahan, A New Governance Structure for Corporate Bonds, 51 STAN. L. REV. 447, 464 (1999) (noting that an increase in equity return volatility tends to enrich shareholders at the expense of creditors).
59. See Alan Schwartz, Security Interests and Bankruptcy Priorities: A Review of Current Theories, 10 J. LEGAL STUD. 1, 11 (1981).
60. Asset substitution is conceptualized here as a situation in which an indebted firm faces two possible future states, one in which it does not suffer an asset downturn, and the other in which it does. The firm owns one asset, A, and exchanges it for asset B, where B relative to A has both a higher variance of future values and a lower expected value. It can be shown that the exchange of A for B, if it increases the firm's expected equity value, also increases the variance of that value.
61. See Jeffrey N. Gordon, Deutsche Telekom, German Corporate Governance,and the Transition Costs of Capitalism, 1998 CoLUM. Bus. L. REV. 185, 195-96 (noting that asset substitution increases the volatility of equity returns).
62. New fixed debt (unless matched by a proportionate infusion of new equity capital) always boosts equity volatility. New contingent debt typicallydoes so as well, but an exception can occur when the contingency and insolvency risks are both high and the internal correlation is deeply negative. See infra note 60.
63. The net effect of the contract that creates the perfectly correlated contingent debt will be a slight increase in equity volatility, because thepremium will augment the debtor's equity value and thus give shareholders more to lose in a downturn. But this volatility increase is not caused by the triggering of the debt itself, the effects of which are borne entirely bythe debtor's unsecured creditors.
64. This is an "all other things being equal" statement, meaning that it holds constant the probability that the assets will depreciate and permits variation only in the correlation between this probability and the contingency risk.
65. The expected (or "mean") value of Shareholders' equity stake in Debtor is the sum of the probability-weighted equity values in each of the model's four outcomes. For the assumptions used for Figure 2, the equity value in the two outcomes in which Debtor suffers a downturn is always zero. The formula for the expected equity value, taking into account the remaining two outcomes, is: (10) 5 = p(not D\not R)*(A +P-C) + p(not D\R)*(A + P - V); with P calculated using equation (3) in note 2 2, supra. When thecorrelation is at its lowest possible level (-0.1), S is $22.25. t the highest possible correlation (1.0), S is $23.87.
66. Equity volatility is conventionally defined as the standard deviationof returns on equity. Because Debtor's initial equity value is the same ($25) regardless ofthe internal correlation on the contingent debt, a reduction in the standard deviation of the final equity value necessarily translates into a lower standard deviation of equity returns. The standard deviation is calculatedusing the expected value of equity as provided by equation (10) in note 55,supra, and the equity value in each of the model's four possible outcomes. The equity value is A + P - U in outcome (solvent, not triggered), A + P - Vin outcome (solvent, triggered), and zero in the two insolvency outcomes.s'The expected value of Shareholders' equity stake given the assumptions usedfor Figure 3 is:(11) S' = p(not D\not R)*(A + P'-C); with P' given byequation (6) in note 41, supra. The standard deviation is calculated using S' and the equity value in each of the model's four possible outcomes. The equity value is A + P - U in outcome (solvent, not triggered), and zero in the other three outcomes.
67. When the downturn risk and contingency risk have their lowest possible negative correlation in this example, Debtor's risk of insolvency is 0%; when they are perfectly correlated, it is 10%. By contrast. Debtor's risk of insolvency is a constant 10% across all correlation levels in Figure 2, because in that figure Debtor is insolvent if and only if a downturn occurs.In Figure 3, Debtor's insolvency risk actually decreases from 20% to 10% asthe internal correlation rises from its lowest to its highest possible level.
68. The formula for the premium in this example is P as given by equation(3) in note 22, supra. The only structural difference between this case andthe one from Figure 2 is that Debtor is now solvent when a downturn occurs but the triggering event does not occur, and this change does not affect the premium because the contingent debt has no value to Claimant unless it becomes payable. The formula for the expected value of Shareholders' equity stake in this example is:(12) S" = p(not D\not R)*(A + P-C) + p(not D\R)*(A + P - V) + p(D\not R)*(d(A + P)) - U). The standard deviation is calculated using 5" and the equity value in each of the example's four possible outcomes. The equity value is A + P - U in the outcome (no downturn, not triggered) A + P - V in the outcome (no downturn, triggered), d(A + P) - V in the outcome (downturn, not triggered), and zero in the outcome (downturn, triggered).
69. This example is exceptional for a second reason: a contract for a contingent debt with such a high probability can cause the absolute level of the debtor's equity volatility to be lower than it would be without the contract. This effect occurs because such a contingent debt can have a negativecorrelation that approaches -i.o, in which case the contract can neutralizemuch of the equity volatility that would otherwise result from fluctuationsin the debtor's asset values. Such a contract is unlikely to be found inthereal world, however, because a negative internal correlation on a contract with such a large face value transfers wealth away from shareholders.
70. This observation implies that the calculations in section I.B understate value transfers by assuming a premium equal to Claimant's expected recovery on her claim rather than the expected burden of the claim on Shareholders, which is lower whenever the internal correlation is positive, and indeed is zero when the correlation is perfect. The extreme case - in which Claimant pays no premium at all - is addressed in Figure 4 in section III.B.
71. Overinvestment is also a social cost of other types of shareholder opportunism, including asset substitution and misuse of the secured loan. SeeBebchuk & Fried, supra note 6, at 899 (secured borrowing).
72. Clifford W. Smith, Jr. & Jerold B. Warner, On Financial Contracting: An Analysis of Bond Covenants, 7 J. FIN. ECON. 117, 118-19 (1979) (asset substitution).
73. Although correlation-seeking will usually increase the debtor's insolvency risk, this consequence is not inevitable. For example, insolvency risk increases whenever a firm incurs a contingent liability that is itself large enough to cause insolvency. However, if we hold the face value of this liability constant, reverse correlation-seeking actually reduces insolvencyrisk by decreasing the chances of a downturn-induced insolvency as anindependent event. Thus, Debtor's insolvency risk in Figure 3 actually decreases from 20% to 10% as the internal correlation moves from its lowest to its highest level.
74. This possibility corresponds to the third case described in section I.C. See supra p. 1179.
75. See Gregor Andrade & Steven N. Kaplan, How Costly Is Financial (Not Economic) Distress? Evidence from Highly Leveraged Transactions that Became Distressed, 53 J. FIN. 1443, 1471 (1998) (noting that financial distress destroys wealth by causing firms to liquidate assets ra ther than make capital expenditures).
76. Mark J. Roe, Bankruptcy and Debt: A New Model for Corporate Reorganization, 83 COLUM. L. REV. 527, 538 (1983) (noting that distraction of managers is one of the "heavy costs of financial distress").
77. These costs are illustrated by the impact of collateral calls on AIG in September 2008, described in the next Part. Cf Ronald J. Gilson & Charles K. Whitehead, Essay, Deconstructing Equity: Public Ownership, Agency Costs, and Complete Capital Markets, 108 COLUM. L. REV. 231, 250 (2008) (identifying the reluctance of various creditor groups to deal with an insolvent firm as a social cost of financial distress).
78. See Robert K. Rasmussen, The Ex Ante Effects of Bankruptcy Reform on Investment Incentives, 72 WASH. U. L.Q. 1159, 1185 (1994) (noting that asset substitution is more likely when a firm is insolvent)
79. Squire, supra note 8, at 820-21 (same).
80. See U.S. DEP'T OF THE TREASURY, FINANCIAL REGULATORY REFORM, A NEW FOUNDATION: REBUILDING FINANCIAL SUPERVISION AND REGULATION passim (2009)[hereinafter TREASURY" REPORT], available at http://www. financialstability.gov/docs/regs/FinalReport-web.pdf. A potential illustration of systemic risk is the fact that the September 2008 bankruptcy of Lehman Brothers precipitated a run on money market funds because one fund - the Reserve Primary MoneyFund - owned large amounts of Lehman Brothers commercial paper. Daisy Maxey, Expense Tally for Reserve Primary Since "Breaking Buck": $16.6 Million, WALL ST. J., June 13, 2009, at B6. This run caused money market fundsto stop buying commercial paper, thereby denying firms a key source of short-term financing.
81. The inadequacy of monetary damages in this context is the reason thatthe standard con tractual remedy for a covenant violation is to accelerate the debtor's repayment obligation. Kalian & Rock, supra note 7, at 302.
82. See F.H. Buckley, The. Bankruptcy Priority Puzzle, 72 VA. L. REV. 1393, 1440-41 (1986). As an alternative, creditors might try to protect themselves by demanding a secured claim, but that arrangement creates its own set of costs, mostly by producing an opportunistic wealth transfer to the secured creditor's advantage.
83. See Bebchuk & Fried, supra note 6, at 870-71.
84. Squire, supra note 8, at 820-21.
85. For a summary of the various bailout programs and amounts paid on behalf of the largest recipients, see Adding Up the Government's Total Bailout Tab, N.Y. TIMES, Feb. 4, 2009, http:// www.nytimes.com/interactive/200q/02/04/ business/20090205-bailout-totals-graphic.html.
86. See, e.g., Brady Dennis & Robert O'Harrow, Jr., A Crack in the System, WASH. POST, Dec. 30, 2008, at A1.
87. BERKSHIRE HATHAWAY INC., 2002 ANNUAL LETTER TO SHAREHOLDERS 15 (2003).
88. See, e.g., Dennis & O'Harrow, supra note 73 (blaming the fall of AIG entirely on its derivatives sales).
89. Am. Int'l Group, Inc., Annual Report (Form 10-K), at 179 n.12(d) (Feb. 28, 2008) [hereinafter AIG 2007 10-K Report].
90. Dennis & O'Harrow, supra note 73.
91. Robert O'Harrow, Jr. & Brady Dennis, Downgrades and Downfall, WASH. POST, Dec. 31, 2008, at A1.
92. See Christopher L. Peterson, Fannie Mae, Freddie Mac, and the Home Mortgage Foreclosure Crisis, 10 LOY. J. PUB. INT. L. 149, 149 (2009).
93. See id. at 161-62.
94. See Dennis & O'Harrow, supra note 73.
95. A credit rating reflects a debt obligation's perceived default risk and hence suggests the minimum yield that investors will require in order tobe willing to buy the obligation. This particular rating is from Standard & Poor's Financial Ratings. Fitch Ratings and Moody's Investors Service also downgraded AIG's long-term debt rating during the same period. Am. Int'l Group, Inc., Annual Report (Form 10-K), at 14 (Mar. 16, 2006) [hereinafter AIG 2005 10-K Report].
96. Id. at 15.
97. AIG insiders have told reporters that Financial Products stopped selling credit default swaps in 2005 because executives were concerned at that point about the company's concentrated exposure to the subprime housing market.
98. See, e.g., Michael Lewis, The Man Who Crashed the World, VANITY FAIR, Aug. 2009, at 98, 137-38.This explanation is contradicted by the fact that executives permitted the AIG parent to make heavy purchases of subprime mortgages in 2006 and 2007.
99. See AIG 2007 10-K Report, supra note 76, at 105.
100. This figure is from 2007. AIG 2007 10-K Report, supra note 76, at 122 n.(b). AIG did not disclose its notional swap exposure for 2 005. However, the figure is unlikely to have changed significantly in the interim given that the company stopped selling swaps in 2005 but did not unwind the positions it had already incurred.
101. Id. at 122.
102. See Joshua D. Coval et. al., The Economics of Structured Finance 6 (Harvard Bus. Sch., Working Paper No. 09-060, 2008), available at http://www.hbs.edu/research/pdf/09-060.pdf.
103. Id. at 3-4.
104. See id. at 23 (noting that ratings analysts relied on models that the analysts knew would "break down" if housing prices stopped rising (internal quotation mark omitted)).
105. See id. at 20 (noting that exposure on diversified swaps was driven entirely by systematic risk, an example of which is housing price changes hat cannot be diversified away by investing across multiple regions).
106. AIG 2005 10-K Report, supra note 82, at 24.
107. In 2007, AIG (other than Financial Products) held $3.7 billion in subprime mortgages of pre-2005 vintage. AIG 2007 10-K Report, supra note 76, at 105 (including securities classified by the company as both "Alt-A" and "subprime," both categories being subprime in the sense that neitherwas guaranteed by government-sponsored entities such as Fannie Mae and Freddie Mac). Adding in the 2005 to 2007 vintage holdings produces a total of $44.9 billion. Assuming that all of the pre-2005 vintage holdings were purchased before 2005, this marks a twelvefold increase. If some of these earlier vintages instead were purchased during the reference period, the ratio is even higher.
108. Id. at 104.
109. Id. at 105 (consisting of the sum of $22.5 billion in AAA Alt-A securities and $18.5 billion in AAA subprime securities, divided by the sum of $23.7 billion in total Alt-A securities and $21.2 billion in total subprime securities, which results in 91%).
110. These securities consisted of $21 billion in "prime" but nonetheless nonguaranteed residential mortgage-backed securities, $4 billion in "other housing-related" securities, and $23 billion in commercial mortgage-backed securities (CMBS).
111. Id. at 104. Along with AIG's subprime investments, these assets lost value in 2007.
112. Id. And this trend accelerated in 2008. Am. Int'l Group, Inc., Annual Report (Form 10-K), at 161 (Mar. 2, 2009) [hereinafter AIG 2008 10-K Report].
113. AIG 2007 10-K Report, supra note 76, at 97.
114. AIG reported losses in 2008 of $99 billion. AIG 2008 10-K Report, supra note 95, at 36. As noted above, AIG's equity value at the end of2007 was $96 billion. In addition, another $5 billion in losses were recorded for the first quarter of 2009. Am. Int'l Group, Inc., Quarterly Report (Form 10-Q), at 5 (May 7, 2009). These reported losses for 2008 and early 2009 almost certainly understate the actual damage to the company's intrinsic value as a result of the housing market collapse because they fail to exclude benefits that AIG received from its bailout.
115. U.S.C. §§ 362(b)(6)-(7), 362(b)(i7), 546(e)-(g), 555, 556, 559, 560 (2006).
116. see also Edward R. Morrison & Joerg Riegel, Financial Contracts and the New Bankruptcy Code: Insulating Mar kets from Bankrupt Debtors and Bankruptcy Judges, 13 AM. BANKR. INST. L. REV. 641, 651 (2005) (noting that amendments in 2005 to the Code's definition of "swap agreement" widened the exemptions to cover essentially all derivatives). To be precise, counterparties are exempt from provisions relating to "constructive" fraudulent transfers, butnot from those directed at intention al fraud. 11 U.S.C. § 546 (e)-(g) (cross-referencing 11 TJ.S.C, § 548(a)(1)(A)).
117. Although AIG eventually ranout of collateral when its credit rating was again downgraded in September2008, the amount of collateral it was ableto post before it received its first bailout. loan covered most of the losses on its counterparties' underlying securities. By August 31, 2008, AIG had posted $22 billion to its swap counterparties, almost all of whichwent to those that had purchased default protection on subprime securities.AIG 2008 10-K Report, supra note 95, at 3, 146. Continuing deterioration inthe value of AIG's own mortgage-backed assets led the rating agencies to downgrade AIG again on September 15, 2008, which in turn led to further collateral calls of approximately $11 billion.
118. Id. at 4, 146. AIG was unable to meet these collateral calls in full until it received a loan from the New York Federal Reserve on September 22, 2008. Id. at 5.
119. These unsecured creditors were owed $108 billion at the end of 2007. AIG 2007 10-K Report, supra note 76, at 89.
120. AIG owed these policyholders more than $420 billion in 2007. Id. at 131.
121. See id. at 13-14, 16 (describing state-law restrictions on the investment activity of AIG's insurance subsidiaries).
122. Patrick Collins, Note, HMO Eligibility for Bankruptcy: The Case for Federal Definitions of 109(b)(2) Entities, 2 AM. BANKR. INST. L. REV. 425, 431 (1994) (noting that state insolvency proceedings generally give policyholders priority of claim).
123. AIG 2008 10-K Report, supra note 95, at 166. At the end of 2007, AIG's outstanding obligations under this program were $76 billion, two-thirds of the proceeds from which were invested in subprime assets. AIG 2007 10-K Report, supra note 76, at 108.
124. Press Release, Am. Int'l Group, Inc., AIG Discloses Counterparties to CDS, GIA and Securities Lending Transactions (Mar. 15, 2009), available at http://www.businesswire.com/news/home/20090315005036/en.
125. See id.
126. See Serena Ng et al., In Battle with U.S. over Pay, AIG Chief Meets His Match, WALL ST. J., Dec. 24, 2009, at Al.
127. See, e.g., O'Harrow & Dennis, supra note 78.
128. Dennis & O'Harrow, supra note 73.
129. Id. (internal quotation mark omitted).
130. Id.
131. See, E.G., RICHARD A. POSNER, A FAILURE OF CAPITALISM: THE CRISIS OF '08 AND THE DESCENT INTO DEPRESSION (2009).
132. Indeed, the company's sales of credit default swaps accounted for only about i% of the market. Houman B. Shadab, Guilty by Association? Regulating Credit Default Swaps, 4 ENTRE PRENEURIAL BUS. LJ. 407, 417 (2010).
133. Dawn Kopecki & Shannon D. Harrington, Banning "Naked" Default Swaps May Raise Corporate Funding Costs, BLOOMBERG, July 24, 2009, http://www.bloomberg.com/apps/news?pid=20601208&sid=aoWiVTivaq2A.
134. During the nine quarters ending in the third quarter of 2009, FannieMae lost a total of $120.5 billion, and Freddie Mac lost $67.9 billion. Rebecca Christie & Jody Shenn, U.S. Treasury Ends Cap on Fannie,Freddie Lifeline for 3 Years, BLOOMBERG, Dec. 25, 2009, http://www.bloomberg.com/apps/news?pid=20601070&sid=aflZdeooJA6I.
135. Fed. Hous. Fin. Agency, Data as of December 10, 2009 on Treasury andFederal Reserve Purchase Programs for GSE and Mortgage-Related Securities, http://www.fhfa.gov/webfiles/ 15315/TSYFed121009.pdf (last visited Jan. 31,2010) [hereinafter Fed. Hous. Fin. Agency 2009 Data].
136. Fed. Nat'l Mortgage Ass'n, Annual Report (Form 10-K), at 1 (Feb. 26,2009) [hereinafter Fannie Mae 2008 10-K Report].
137. Id. at 15.
138. For example, at the end of 2007, mortgage-related assets constituted82% of Fannie's total assets. See id. at 81, 124 (noting total assets in 2007 of $879 billion, and total mortgage assets of $723 billion). This figure is comparable to the 2000 proportion, which was 90%.
139. See Fed. Nat'l Mortgage Ass'n, Annual Report (Form 10-K), at 20 (Mar. 31, 2003) (noting total assets in 2000 of $675 billion, and a net mortgage portfolio of $608 billion).
140. Fannie Mae 2008 10-K Report, supra note 116, at 80 (noting net losses in 2007 of $2 billion).
141. Christie & Shenn, supra note 114.
142. Fannie Mae 2008 10-K Report, supra note 116, at 81.
143. Id. at 20-21.
144. Id. at 25.
145. Fed. Hous. Fin. Agency 2009 Data, supra note 115.
146. See, e.g., Steve Forbes, In-credit-able!, FORBES, Dec. 28, 2009, at 15 (arguing that Fannie and Freddie's "implicit government guarantees" enabled them to "borrow cheaply and leverage up on a scale no private company could').
147. At the end of 2007, Fannie Mae's total liabilities were § 835 billion, and its shareholder equity was §44 billion. Fannie Mae 2008 10-K Report, supra note 116, at 81. Dividing the first number by the second gives a leverage ratio of 19.
148. See Peter Eavis, Fannie, Freddie May Have Further To Fall, WALL ST. J., Feb. 29, 2008, at C2 (noting that Fannie's leverage ratio at the end of2007 was "at a level far above that of other financial institutions").
149. At the end of 2001, Fannie's total liabilities were $791 billion, and its shareholder equity was $ 23 billion. Fed. Nat'l Mortgage Ass'n, Annual Report (Form 10-K), at 63 (Dec. 6, 2006) [hereinafter Fannie Mae 2004 10-K Report]. Dividing the first number by the second gives a leverage ratio of 34.
150. E.g., Jody Shenn, Fannie, Freddie Subprime Spree May Add to Bailout,BLOOMBERG, Sept. 22, 2008, http://www.bloomberg.com/apps/news?pid= 20601109&sid=ayoKkt47a3S4.
151. see also Peter Eavis, Fannie and Freddie's Home Inequity, WALL ST. J., Dec. 28, 2009, at Cio (asserting that Fannie and Freddie played "a central role in the housing crash with their huge purchases of dangerous Alt-A mortgages"); Forbes, supra note 125, at 15 (characterizing Fannie as "bingeing" on subprime mortgages in the years prior to the 2008 crash).
152. See Fed. Nat'l Mortgage Ass'n, Annual Report (Form 10-K), at 93 (Feb. 27, 2008) (noting that Fannie's total holdings of subprime and Alt-A mortgage-backed securities at the end of 2007 were $74 billion, of which$ 14 billion were of a pre-2005 vintage). As noted in note 92, supra,subprime and Alt-A mortgages are lumped together for purposes of the analysis here.
153. Fannie lost $59 billion in 2008. Fannie Mae 2008 10-K Report,supra note 116, at 80. Of this amount, $7 billion was attributable to reduced cash flows and defaults on subprime and Alt-A securities in the company's investment portfolio.
154. Id. at 103. In addition, the company lost approximately $4 billion on trades in mortgage-backed securities, most of which involved subprime andAlt-A securities.
155. See id. at 107. Thus, total losses on subprime and Alt-Asecurities were $11 billion, or 19% of $59 billion.
156. Fannie's total mortgage portfolio was §917 billion in 2004.
157. see Fannie Mae 2004 10-K Report, supra note 128, at 63, and shrank to §728 billion in 2007.
158. see Fannie Mae 2008 10-K Report, supra note 116, at 81.
159. This figure is listed on Fannie's balance sheet as "Fannie Mae MBS held by third parties." Fannie Mae 2008 10-K Report, supra note 116, at 81.
160. Fannie Mae 2004 10-K Report, supra note 128, at 63.
161. Fannie's total assets were $905 billion in 2002, Fannie Mae 2004 10-K Report, supra note 128, at 63, slightly higher than the 2007 totalof 879 billion, Fannie Mae 2008 10-K Report, supra note 116, at 81.
162. Fannie's equity value was $ 32 billion in 2002, Fannie Mae 2004 10-K Report, supra note 128, at 63, and $ 44 billion in 2007, Fannie Mae 2008 10-K Report, supra note 116, at 81. Dividing the value of the outstanding Fannie mortgage-backed securities listed in the text by these amounts provides the stated ratios.
163. See Fannie Mae 2008 10-K Report, supra note 116, at 109 (noting credit-related expenses of $30 billion, consisting of changes in loss reserves for guarantees on outstanding mortgage-backed securities, losses due to purchases of loans from Fannie Mae trusts that were nonperforming, and foreclosure expenses for these loans).
164. id. at 80 (noting losses of $ 60 billion).
165. After the mortgage-linked contingent debts, the biggest source of losses for Fannie in 2008 was its derivatives positions, most of which involved interest rate swaps that were meant to hedge against the risk to Fannie's mortgage investments of an increase in interest rates. These positions cost the firm $ 15 billion. Id. at 104-06.
166. Of course, Fannie's willingness to expand its issuances of guaranteed mortgage-backed securities may have itself contributed to the housing bubble.
167. See Shenn, supra note 129.
168. See Fed. Home Loan Mortgage Corp., Annual Report (Form 10-K), at 1 (Mar. n, 200a) [he reinafter Freddie Mac 2008 10-K Report].
169. See id. (noting that Freddie's purpose is to provide "liquidity, stability and affordability to the U.S. housing market").
170. Id. at 1, 5.
171. Id. at 58 (noting a 2008 net loss of $ 50.1 billion and a 2007 equity value of $26.7 billion).
172. Christie & Shenn, supra note 114.
173. Fed. Hous. Fin. Agency 2009 Data, supra note 115.
174. Freddie's total assets were $794 billion at the end of 2007. Freddie Mac 2008 10-K Report, supra note 140, at 58. Fannie was 11% larger,with $ 879 billion in assets. Fannie Mae 2008 10-K Report, supra note 116, at 81.
175. At the end of 2007, Freddie held $153 billion in subprime securities, a total that includes securities the company characterized as subprime, Alt-A, and MTA (which stands for Moving Treasury Average, the index that determines the underlying mortgages' adjustable rate).
176. See Freddie Mac 2008 10-K Report, supra note 140, at 93, 95. By contrast, Fannie held §74 billion in subprime securities at that point, asdescribed in note 130, supra.
177. Peter J. Wallison, The Price for Fannie and Freddie Keeps Going Up, WALL ST. J., Dec. 30, 2009, at A17 (stating that Fannie and Freddie were "binging" on subprime and Alt-A mortgages between 2005 and 2007).
178. see also Forbes, supra note 125, at 15.
179. As noted in note 143, supra, Freddie lost $ 50.1 billion in 2008. Of this, $ 16.6 billion was from impairment of the value of thecompany's subprime portfolio. Freddie Mac 2008 10-K Report, supra note 140,at 80.
180. See id. at 58 (noting 2007 total liabilities, defined as total assets minus shareholders' equity, of $ 768 billion, and shareholders' equity of $ 27 billion).
181. see also FREDDIE MAC, 2005 INFORMATION STATEMENT AND ANNUAL REPORT TO STOCKHOLDERS 19 (2006), available at http://www.freddiemac.com/investors/ar/pdf/ 2005annualrpt.pdf (noting 2001 total liabilities of $ 621 billion and shareholders' equity of $ 20 billion).
182. Compare FREDDIE MAC, 2002 INFORMATION STATEMENT AND ANNUAL REPORT TO STOCKHOLDERS 6 (2004) [hereinafter FREDDIE MAC, 2002 ANNUAL REPORT], available at http://www.freddiemac.com/investors/ar/pdf/2002annualrpt.pdf, with Freddie Mac 2008 10-K Report, supra note 140, at 127.
183. Freddie's 2002 shareholder equity value was $31.3 billion. FREDDIE MAC, 2002 ANNUAL REPORT, supra note 151, at 27. As noted in note 143,supra, its 2007 equity value was $26.7 billion.
184. Freddie Mac's 2008 losses attributable to its mortgage-linked guarantees included a $ 16.4 billion increase in its provision for lossesand $ 1.6 billion in losses on loans purchased from the underlying mortgage pools. Freddie Mac 2008 10-K Report, supra note 140, at 82-83. As noted in note 149, supra, Freddie's losses on its subprime portfolio were $16.6 billion.
185. As noted previously, Fannie Mae so far has received $ 849 billion in bailout funds.
186. see supra p. 1192, and Freddie Mac has received $570 billion,see supra p. 1196.
187. As noted previously, Fannie had $2.12 trillion in contingent debt at the end of 2007, see su pra p. 1194, while Freddie had $1.38 trillion, see supra p. 1197
188. See TREASURY REPORT, supra note 68.
189. Wall Street Reform and Consumer Protection Act, H.R. 4173, 111th Cong, (as passed by House of Representatives, Dec. 11, 2009), available at http://financialservices.house.gov/Key-Issues/Financial-Regulatory-Reform/ FinancialRegulatoryReform/III-hr-finsrv-4173-full.pdf.
190. TREASURY REPORT, supra note 68, at 47.
191. Id.
192. Id. In this way, the section on derivatives is analytically no different from the report's ear lier sections that blame the collapse of large bank holding companies on problems of "excess lever age.".
193. See id. at 29.
194. No variant on the word "correlation" appears in the section of the bill that addresses derivatives, and two of that section's most important provisions - those involving capital and margin requirements - could be set based entirely on notional amounts, at the discretion of regulators. The closest that the section on derivatives seems to come to addressing risk correlations are its provisions that permit regulators to impose position limitson individual parties. But the bill does not direct regulators to take riskcorrelations into account when setting limits, and instead seems concerned primarily with positions that might interfere with price discovery. H.R. 4173 §§ 3113, 3203.
195. Id. §§ 1103-04.
196. Id. §§ 3107, 3204.
197. see also TREASURY REPORT, supra note 68, at 48 (identifying stricter counterparty capital requirements as a regulatory goal).
198. See H.R. 4173 §§ 3103-3106, 3204. Because of exemptions contained in the bill, only about half of all derivative contracts would needto be exchange-traded.
199. See Randall Smith & Sarah N. Lynch, How Overhauling Derivatives Died, WALL ST. J., Dec. 26-27, 2009, at BI.
200. AIG 2007 10-K Report, supra note 76, at 104-05, 121-23.
201. See H.R. 4173 §§ 3107, 3204.
202. See supra note 97.
203. See supra pp. 1187-88.
204. See Deborah Solomon, Pay Czar Targets Salary Cuts, WALL ST. J., Oct.6, 2009, at Ai (internal quotation marks omitted).
205. Deborah Solomon, Firms Back Plan To Change Pay Policies, WALL ST. J., Sept. 21, 2009, at A3.
206. see also Aaron Lucchetti, Morgan Stanley To Overhaul Pay Plan, WALL ST. J., Dec. 29, 2009, at Ci (noting that Morgan Stanley has adopted a pay policy that increases the degree to which senior executives will be paid indeferred stock).
207. In particular, the Federal Reserve's pay rules would bind U.S. bank holding companies, state member banks, and the U.S. operations of foreign banks, among others. See Proposed Guidance on Sound Incentive Compensation Policies, 74 Fed. Reg. 55,227, 55,228 n.i (Oct. 27, 2009) [hereinafter Proposed Guidance].
208. See Solomon, supra note 169 ("Instead of awarding large cash salaries, [the pay czar] is planning to shift a chunk of an employee's annual salary into stock that cannot be touched for several years⋯.")
209. see also Proposed Guidance, supra note 171, at 55,234 (indicating that banking organizations can comply with guidelines on management risk-taking incentives by paying senior executives "deferred incentive compensation in equity (such as restricted stock of the organization) or equity-based instruments (such as options to acquire the organization's stock)").
210. See sources cited supra note 172.
211. See, e.g., Proposed Guidance, supra note 171, at 55,232 (expressing concern with compensation that focuses on short-term revenues or profits because "[activities that carry higher risk typically yield higher short-termrevnue.").
212. Other scholars have also observed that paying bank executives in equity may induce banks to take risks that are excessive from a social perspective.
213. See Lucian A. Bebchuk & Holger Spamann, Regulating Bankers' Pay, 98 GEO. L.J. 247, 256 (2010).
214. Franklin R. Edwards & Edward R. Morrison, Derivatives and the Bankruptcy Code: Why the Special Treatment?', 22 YALE J. ON REG. 91, 97-99 (2005)
215. See supra pp. 1174-75.
216. See infra section III.B, pp. 1205-12.
217. While the future form of government-sponsored mortgage firms remainsunclear, the federal government appears likely to retain at least some rolein the underwriting of residential mortgage-backed securities.
218. See Nick Timiraos, Support Grows for Fan-Fred Plan, WALL ST. J., Dec. 14, 2009, at A7.
219. Jeffrey Rogers Hummel, Why Default on U.S. Treasuries Is Likely, LIBR. ECON. & LIBERTY, Aug. 3, 2009, http://www.econlib.org/library/Columns/y2009/ Hummeltbills.html.
220. See 11 U.S.C. § 548(a)(1)(A) (2006) (incorporating a prohibition on actual fraudulent transfers into federal bankruptcy law).
221. Robert A. Fogelson, Toward a Rational Treatment of Fraudulent Conveyance Cases Involving Leveraged Buyouts, 68 N.Y.U. L. REV. 552, 555 (1993) (indicating that the original fraudulent conveyance statute, the Statute of Elizabeth, prohibited only actual fraud).
222. 11 U.S.C. § 548(a)(i)(B)(i); UNIF. FRAUDULENT TRANSFER ACT §§ 4(a)(2), 5(a), 7A II.L.A. pt. 2, at 58, 129 (2006). Under the older Uniform Fraudulent Conveyance Act, which is still in effect in some states, a "fair consideration" test stands in for the "reasonably equivalent value" requirement.
223. Id. § 4 cmt. 2 (internal quotation marks omitted).
224. See 11 U.S.C. § 548(a)(i)(B)(ii).
225. See id. § 548(a)(i)(B)(ii)(I).
226. The most influential decision in this line of cases was written by Judge Richard Posner. See In re Xonics Photochemical, Inc., 841 F.2d 198 (7th Cir. 1988).
227. See id. at 200.
228. This result is calculated by taking equation (3) in note 22, supra, and setting P equal to $0.
229. This line is thus identical to the contingent debt line in Figure 2.
230. The question whether Claimant pays in full becomes even less important as the risk of the triggering event decreases. For example, if we were to assume in Figure 4 that the contingency and downturn risks were each 1% rather than 10%, the correlation level would still make more than a 50% difference in terms of Unsecured Creditor's expected losses. But the differencemade by the premium would shrink to less than 1%.
231. This result is calculated by taking equation (4) in note 22, supra, setting T equal to $0, and solving for P.
232. See 11 XJ.S.C. § S48(a)(i)(B)(ii)(I) (2006).
233. UNIF. FRAUDULENT TRANSFER ACT § 5(a), 7A U.L.A, pt. 2, at 129 (2006).
234. II U.S.C. § S48(a)(i)(B)(ii)(III); accord IJNIF. FRAUDULENT TRANSFER ACT § 4(a)(2)(ii), 7A U.L.A. pt. 2, at 58 (employing an essentially identical test).
235. See Asarco LLC v. Ams. Mining Corp., 396 B.R. 278, 399 n.140 (S.D. Tex. 2008) (noting that courts have had little opportunity to interpret the provision).
236. Although the statutory language suggests a subjective intent requirement, courts have held that the requirement can be satisfied upon a showingthat "the debtor could not have reasonably believed that it would be able to pay its debts as they matured." WRT Creditors Liquidation Trust v. WRT Bankr. Litig. Master File (In re WRT Energy Corp.), 282 B.R. 343, 415 (Bankr. W.D. La. 2001).
237. In the typical fact pattern, the challenged transaction is a conveyance of assets rather than the incurrence of an obligation, and the questionis whether the conveyance was fraudulent with respect to subsequent creditors. See, e.g., Asarco, 396 B.R at 400. Courts have, however, deemed an obligation to be fraudulent when the obligation itself is the debt that the debtor did not intend to be able to repay.
238. See, e.g., Pajaro Dunes Rental Agency, Inc. v. Spitters (In re Pajaro Dunes Rental Agency, Inc.), 174 B.R. 557, 593-94 (Bankr. N.D. Cal. 1994) (applying the corresponding provision in California's fraudulent transfer statute).
239. Chan et al., supra note 37, at g.43.
240. For claims that create a range of possible liability amounts, courtscould treat the expected mean, or perhaps the mean plus a standard deviation, as the relevant point for determining whether to subordinate.