UUnniivveerrssiittyy ooff CChhiiccaaggoo LLaaww SScchhooooll CChhiiccaaggoo UUnnbboouunndd Journal Articles Faculty Scholarship 2011 DDoo AAccccoouunnttiinngg RRuulleess MMaatttteerr?? -- TThhee DDaannggeerroouuss AAlllluurree ooff MMaarrkk ttoo MMaarrkkeett Richard A. Epstein M. Todd Henderson Follow this and additional works at: https://chicagounbound.uchicago.edu/journal_articles Part of the Law Commons RReeccoommmmeennddeedd CCiittaattiioonn Richard A. Epstein & M. Todd Henderson, "Do Accounting Rules Matter? - The Dangerous Allure of Mark to Market," 36 Journal of Corporation Law 513 (2011). This Article is brought to you for free and open access by the Faculty Scholarship at Chicago Unbound. It has been accepted for inclusion in Journal Articles by an authorized administrator of Chicago Unbound. For more information, please contact [email protected]. Do Accounting Rules Matter? The Dangerous Allure of Mark to Market Richard A. Epstein* M. Todd Henderson** This Article examines the relative strength of two imperfect accounting rules: historical cost and mark to market. The manifest inaccuracy of historical cost is well known and, paradoxically, one source of its hidden strength. Because private parties know of its evident weaknesses, they look elsewhere for information. In contrast, mark to market for hard-to-value assets has many hidden weaknesses. In this Article we show how it creates asset bubbles and exacerbates their negative collateral consequences once they burst. It does the former by allowing banks to adopt generous valuations in up- markets that increase their lending capacity. It does the latter by forcing the hand of counterparties to demand collateral, even when watchful waiting and inaction is the more efficient course of action when downward cascades generated by mark-to-market accounting might trigger massive sell-offs at prices below true asset value. The fears of private suits and regulatory sanctions on counterparties can compound the problem. Mark to market generates the functional equivalent of bank runs for which the functional equivalent of the automatic-stay rule in bankruptcy is the appropriate response. I. INTRODUCTION: HISTORICAL COST VERSUS MARK-TO-MARKET ACCOUNTING..... 514 II. THE S&L CRISIS AND THE RISE OF MARK-TO-MARKET ACCOUNTING................... 515 III. THE FRAGILE THEORETICAL CASE FOR MARK-TO-MARKET ACCOUNTING............ 519 IV. THE IMPORTANCE OF VALUATION ......................................... 522 A. Private Valuation .................................... ....... 522 B. Government Valuation ............................................ 525 V. THE MANY SENSES OF VALUATION ............................... ..... 528 A. Threshold Issues ...................................... ..... 528 B. Valuation Methods Revisited. ................................... 530 C. Valuation in Roiled Markets ............................ ....... 532 *Richard A. Epstein, Laurence A. Tisch Professor of Law, New York University Law School, Peter and Kirstin Bedford Senior Fellow at the Hoover Institution, and a Senior Lecturer at the University of Chicago Law School. **M. Todd Henderson, Professor of Law, the University of Chicago Law School. Thank you to Daniel Currell, Richard Holden, Alan Jagolinzer, and Saul Levrnore for helpful notes. Sean Sharp provided able research assistance. HeinOnline -- 36 J. Corp. L. 513 2010-2011 514 The Journal of CorporationL aw [Vol. 36:3 VI. Do ACCOUNTING RULES MATTER? .............................. .... . . 535 A. TheoreticalS tartingP oints................................................ 535 B. Accounting Rules Matter...................................... 536 1. They Ma tterf or Regulation........................................................................... 536 2. They Matterf or Private Valuation ..................... ........ 539 C. Default Rules Matter Too............................ ......... 547 D. What About the Government? ......................................5.4..8. VII. CONCLUSION...................................................... 549 "There is no such uncertainty as a sure thing. "f I. INTRODUCTION: HISTORICAL COST VERSUS MARK-TO-MARKET ACCOUNTING There is nothing like profound financial dislocation to spur inquiry into the first principles of political economy. In some cases, that inquiry is over grand issues about the relationship of state to market. In other cases, it is about seemingly smaller theoretical issues that loom very large in practice. This Article is about one such issue, the question of whether mark-to-market accounting is the proper technique to apply to hard-to-value assets. That question raises a tension between two techniques of valuation, each with its own imperfections. The first technique uses historical cost as the benchmark for valuing certain assets. In essence the method starts with cost and then makes certain formal adjustments to estimate the value of an asset that has not been sold. For example, the normal rules for valuing a real estate improvement start with its cost. Thereafter it reduces that basis by an allowance for depreciation, which is granted by regulation wholly without regard to the actual changes in value of the underlying asset. In the end, a fully depreciated asset will be carried on the books as if it were worth zero, even if it still has positive value in use or salvage. 1 A simple example illustrates the point. An asset that costs $100 with an assigned 20- year life could have its basis adjusted downward under the "straight-line" method by $5 each year, so that at the close of the second year it is carried at $90. A tax deduction is given for a $10 reduction in value, whether the market value of the asset is $70 or $110.2 The difference between this "adjusted" basis of $90 and the market value is taken into account as income (or loss) only on the disposition of the asset.3 Although in the interim one could fairly describe the asset as mispriced, the approach could be justified on the ground that the administrative costs of accurate annual pricing are too high relative to the gains from more precise valuation. As a matter of general theory, improvements in real estate will always depreciate so T Attributed to the Scottish poet Robert Burns. 1. See generally I.R.S. Pub. 946 (Oct. 25, 2010), available at http://www.irs.gov/pub/irs-pdflp946.pdf (describing depreciation methods-including strict line-used under 26 U.S.C. § 179 (2006)). 2. See id. 3. See id. HeinOnline -- 36 J. Corp. L. 514 2010-2011 2011] Do Accounting Rules Matter? 515 that on average these adjustments tend to reduce the gaps between market value of the asset and its value on the books. Historical cost accounting for corporate shares does not have that downward directional bias, which reduces the reliability of the method. The market for shares is often thick so that there is no need to rely on historical cost accounting at all, as values can be continually and accurately updated. Not surprisingly, the alternative to this system of historical cost requires the revaluation of unsold assets to market on a periodic basis. Accountants typically refer to this system as "fair-value accounting." Tax lawyers and others prefer the equivalent term "mark to market." Mark- to-market accounting became popular4 after thrifts were accused of hiding "bad" assets by using historical-cost accounting in the years leading up to the S&L crisis in the 1980s.5 The business reality turns out to be more complex, for historical-cost accounting can only "hide" true values from people who do not want to discover them.6 In this Article we shall explore the tensions between these two areas from historical, economic, and legal perspectives to support the proposition that for all its manifest weaknesses, the hidden virtues of historical cost accounting could render it the sounder approach to accounting issues with hard-to-value assets. Part II reviews some of the historical events that led up to the mark-to-market system. Part III examines the theoretical vulnerabilities of the theory of mark-to-market accounting. Part IV deals with the importance of valuation issues, as it applies to both private and government actors, in setting the appropriate accounting rule. Part V addresses the many different approaches that can be taken toward this critical valuation problem. Part VI then seeks to assess the extent to which the choice of accounting rules matters in light of the previous analysis. II. THE S&L CRISIS AND THE RISE OF MARK-To-MARKET ACCOUNTING The evolution of the S&L crisis during the 1980s played out as follows. Thrifts used deposits to fund long-term, fixed-rate mortgage loans. When interest rates are stable and thrifts do not need to compete aggressively on price for depositors, this business model is harmonious. But, in the 1970s, inflation drove up interest rates, which induced new competitors-like money market funds-to enter the market. To attract depositors in this environment, thrifts had to pay higher rates on deposits. At this point the thrifts faced this deadly combination: long-term assets delivering low rates of return (since the contracts were made when rates were low) paired with the higher rates of interest needed to attract deposits to fund these assets. This asset-liability mismatch drove many thrifts into insolvency.7 Historical-cost accounting gave no explicit warning of the impending 4. See LAWRENCE J. WHITE, THE S&L DEBACLE: PUBLIC POLICY LESSONS FOR BANK AND THRIFT REGULATION 225-29 (1991) [hereinafter WHITE, THE S&L DEBACLE] (arguing for the switch to market value accounting following the S&L crisis). 5. See id. at 83-85 (describing the negative effects of the thrift's accounting practices). 6. See infra Part H (detailing the history of the S&L crisis). Everyone knows that historical-cost accounting figures are inaccurate and do not reflect economic value, and therefore professional investors will look at market values when spending their money or that of their clients. 7. See, e.g., MICHAEL R. DARBY, MACROECONOMIC SOURCES OF THE U.S. SAVINGS AND LOAN CRISIS, IN THE SAVINGS AND LOAN CRISIS: LESSONS FROM A REGULATORY FAILURE, THE MILKEN INSTITUTE SERIES ON FINANCIAL INNOVATION AND ECONOMIC GROWTH, Vol. 5, Part 3, at 105-07 (2004). See also, MARTIN Lowy, HIGH ROLLERS: INSIDE THE SAVINGS AND LOAN DEBACLE 146-52 (1991) [hereinafter LowY, HIGH ROLLERS] (detailing what one judge called the "looting of Lincoln"); WHITE, THE S&L DEBACLE, supra note 4, HeinOnline -- 36 J. Corp. L. 515 2010-2011 516 The Journalo f CorporationL aw [Vol. 36:3 meltdown, because thrift financial statements did not reflect the losses from the change in interest rates. As we discuss, these tranquil balance sheets could have fooled no one, since everyone knew-or could easily calculate-the values (both real and as reported) of all assets and liabilities. Even though this asset-liability mismatch happened relatively quickly, market observers could see the day of reckoning approaching.8 For example, Lincoln Savings and Loan Association, of the Keating Five scandal, collapsed in 1989, but its problems were well known for many years before that.9 But the peculiarities of thrift regulation provided cover for the deterioration in assets by carrying them on the books for more than their actual value. In 1987 and 1988, that regulatory forbearance let Lincoln invest in nearly $2 billion in Arizona real estate, most of which proved worthless. The bailout of Lincoln eventually cost the taxpayers over $3 billion.10 As this incident reveals, the historical-cost accounting system has two dominant features. The first is that it helps forestall regulatory action to take over or shut down a bank, thrift, or other financial institution. If regulators are duty-bound to put a bank into receivership once it is found critically undercapitalized or insolvent, positive financial statements supply valuable cover precisely because they do not accurately reflect the bank's economic position. Accounting regulation becomes a powerful tool in the cause of regulatory discretion in cases where no discretion should be allowed. The S&L crisis did not reach fever pitch because the markets believed thrift balance sheets. It came to a boil because of excessive forbearance by regulators.1' The root cause of this was political tampering. Thus in the Keating Five scandal, five senators intervened with regulators on behalf of their donor and friend, thrift owner Charles Keating.12 The resulting delay of regulatory intervention took place under the prevailing accounting rule, which provided a patina of solvency when there was not even a glimmer of hope.13 The phenomenon can be generalized. Too often, government officials practice regulatory forbearance in response to political pressure from bank owners or out of a desire to avoid shutting down institutions that supply valuable services to well-connected constituents. One important change put in place after the S&L crisis was a nondiscretionary system of regulatory intervention for undercapitalized banks, under the so-called prompt corrective action regime. 14 The combination of mark-to-market and prompt corrective action was designed in part to take politics out of the government valuation process.15 at 85. 8. See, e.g., LowY, HIGH ROLLERS, supra note 7, at 91 (citing criticism from pundits and academics). 9. See, e.g., WILLIAM K. BLACK, THE BEST WAY TO ROB A BANK IS TO OWN ONE 193 (2005) (describing the situation at Lincoln as "an easy call" and how "[a]nyone with experience knew it would be a catastrophic failure."). 10. See Lowy, HIGH ROLLERS, supra note 7, at 146-52. 11. See WHrrE, THE S&L DEBACLE, supra note 4, at 83-85. 12. The Senators were Alan Cranston (D-CA), Dennis DeConcini (D-AZ), John Glenn (D-OH), John McCain (R-AZ), and Donald W. Riegle, Jr. (D-MI). See Philip Shenon, S Senators Struggle to Avoid Keating Inquiry Fallout, N.Y. TIMES, Nov. 22, 1989, at B8 (describing the debacle). 13. See Joni J. Young, Getting the Accounting "Right ": Accounting and the Savings and Loan Crisis, 20 ACCT., ORGS, & Soc'Y 55, 65 (1995) ("Accounting provided a way of "seeing" these organizations as not requiring regulatory intervention.") (internal citation omitted). 14. Federal Deposit Insurance Act of 1950 § 38, 12 U.S.C. § 1831o (2006). 15. See, e.g., Frederic S. Mishkin, How Big a Problem Is Too Big to Fail? A Review of Gary Stern and HeinOnline -- 36 J. Corp. L. 516 2010-2011 2011] Do Accounting Rules Matter? 517 Unfortunately, history has a way of repeating itself During the housing bubble, Representatives and Senators resisted calls for more regulation of mortgage giants Fannie Mae and Freddie Mac because they helped make housing cheaper for low-income families. For instance, when problems with Fannie and Freddie surfaced for all to see, Senator Charles Schumer said: "I think Fannie and Freddie over the years have done an incredibly good job and are an intrinsic part of making America the best-housed people in the world ... if you look over the last 20 or whatever years, they've done a very, very good job."l6 This incident-and countless others like it, known and unknown-shows that regulatory tampering will exist regardless of the accounting rule. The second feature of historical-cost accounting is that it gives a bank a longer window in which to conduct additional (risky) lending in order to bring the bank back from the brink of insolvency. This go-to-Vegas strategy is possible only if the bank's leverage ratio frees up money to lend. The difference between the stated value of assets and of liabilities determines that ratio. If under historical-cost accounting, assets are reported at inflated values and liabilities are relatively fixed, there is greater capacity to lend. The new regime created a heads-I-win, tails-you-lose mentality.17 Not surprisingly, failing thrifts took advantage of mispriced assets under the historical-cost accounting method to make additional risky loans that were in the interests of bank shareholders but not the taxpayers, who eventually had to bail out depositors of the failed thrifts.'8 The regulatory wiggle room provided by historical-cost accounting let the thrifts gamble with taxpayers' money. These events did not go unnoticed. The Financial Accounting Standards Board (FASB), 19 which operates under the aegis of the Securities and Exchange Commission, is the designated organization for establishing standards of financial accounting that nongovernmental entities use to prepare financial reports. In May 1993, in the wake of the S&L crisis, the FASB promulgated new accounting rules that spurred the move toward mark-to-market accounting.20 The new rule required certain equity and debt Ron Feldman's "Too Big to Fail: The Hazards of Bank Bailouts", 44 J. ECON. LIT. 988, 999-1003 (2006) (discussing the importance of the combination of these two rules). 16. Regulatory Reform of the Government-Sponsored Enterprises: Hearing on S. 2591 Before the S. Comm. On Banking, Housing, and Urban Affairs, 109th Cong. (2005) (statement of Sen. Charles E. Schumer, member of Senate Comm. On Banking, Housing, and Urban Affairs). 17. See, e.g., MARTIN MAYE, THE GREATEST-EVER BANK ROBBERY: THE COLLAPSE OF THE SAVINGS AND LOAN INDUSTRY 75 (1993) ("As S&L accounting was done, winners could be sold at a profit that the owners could take home as dividends, while the losers could be buried in the portfolio 'at historic cost,' the price that had been paid for them, even though they were now worth less, and sometimes much less."). 18. See id. See also Lowy, HIGH ROLLERS, supra note 7, at 146-52 (conducting a case study of Lincoln S&L and its eventual failure). 19. See Commission Statement of Policy Reaffirming the Status of the FASB or a Designated Private- Sector Standard Setter, 68 Fed. Reg. 23,333 (May 1, 2003) (describing the FASB's ability to set generally accepted accounting principles). See also AM. INST. OF CERTIFIED PUB. ACCOUNTANTS, RULES OF PROF'L CONDUCT R. 203 (May 1979). From 2002 until present, FASB has operated under the authority of the Public Company Accounting Oversight Board (PCAOB), which is also under the SEC. 20. See FIN. ACCOUNTING STANDARDS BD, STATEMENT OF FINANCIAL ACCOUNTING STANDARDS NO. 115: ACCOUNTING FOR CERTAIN INVESTMENTS IN DEBT AND EQUITY SECURITIES (1993), available at http://www.fasb.org/pdflfasl 15.pdf (addressing accounting and reporting for investments in equity securities with determinable fair values and all debt securities). HeinOnline -- 36 J. Corp. L. 517 2010-2011 518 The Journalo f CorporationL aw [Vol. 36:3 securities to be reported at "fair value" instead of adjusted historical cost.21 This mark-to- market process requires that the firms attach values to assets held on the books in order to improve the "transparency" of the firm for the benefit of both government regulators and present and future trading partners.22 Over the next decade, the use of mark-to-market increased in response to market demand for additional information about asset values.23 But, more importantly, the new regime worked hand-in-glove with leverage limit triggers under a federal legal regime that mandates government intervention as banks approach insolvency. The conventional accounting wisdom concluded that the FASB had invented a better mousetrap, sufficient to avert a replay of the accounting problems that helped create the previous S&L crisis.24 Unfortunately, this bullish account ignores the downside of the story. Rules adopted in response to one crisis may not fit well with the challenges presented by the next; worse, new rules for old problems may help bring on new problems. For example, government regulation designed to solve the private monitoring problem took the form of "capital adequacy requirements," which through the so-called Basel accords, mandated banks hold a certain percentage of their capital in reserve to cover potential defaults on outstanding loans.25 Under the Basel rules, bank capital is not productive from the perspective of banks' shareholders,26 so banks can be expected to find ways to nevertheless deploy it.27 The banks did so primarily through exotic derivatives, like collateralized debt obligations (CDOs), which allowed them to originate loans (and take the fees associated with them), while shifting the risk off to special-purpose vehicles, and thus freeing up their capital to make more loans (and fees).28 These CDOs relied on Baroque structures to diffuse claims to the point where no one was sure who bore what risk, whether that risk was excessive, and, if a problem arose, what to do about it.29 A regulation designed to reduce excessive bank risk taking ended up increasing it, as well as the overall systemic risk in the market. This outcome is especially likely to occur 21. See id. 22. See, e.g., Justin Fox, Suspending Mark-to-Market Is for Zombies, THE CURIOUS CAPITALIST BLOG (Oct. 1, 2008, 6:25 PM), http://curiouscapitalist.blogs.time.com/2008/10/01/suspendingmarktomarket-isfor/ (discussing fair value accounting). 23. See, e.g., OFFICE OF THE CHIEF ACCOUNTANT DIvISION OF CORPORATE FINANCE, SECURITIES AND EXCHANGE COMMISSION, REPORT AND RECOMMENDATIONS PURSUANT TO SECTION 133 OF THE EMERGENCY ECONOMIC STABILIZATION ACT OF 2008: STUDY ON MARK-TO-MARKET ACCOUNTING 38-39 (2008) [hereinafter SEC, MARK-TO-MARKET REPORT], available at www.sec.gov/news/studies/2008/ marktomarketl23008.pdf (discussing fair-value accounting regarding derivatives transactions). 24. See id. at 169 ("[T]he suspension of fair value accounting to return to historical cost-based measures would likely increase investor uncertainty and adversely impact equity values by removing access to information at a time when that information is likely most useful to investors."). 25. See generally BASEL COMMITTEE, INT'L CONVERGENCE OF CAPITAL MEASUREMENT AND CAPITAL STANDARDS (1988), available at http://www.bis.org/publbcbscl I .htm (addressing the new standard). 26. Id. 27. Id 28. See, e.g., Carrick Mollenkamp & Serena Ng, Wall Street Wizardry Amplified Credit Crisis: A CDO CalledN orma Left 'Hairballo f Risk' Tailoredb y MerrillL ynch, WALL ST. J., Dec. 27, 2007, at Al (discussing fee generation by large banks in issuing credit derivatives on housing). 29. See id. See also Ash Bennington, The Most Complicated Mortgage Chart You've Ever Seen, CNBC.coM(Nov. 17, 2010), http://www.cnbc.com/id/40231732/TheMostComplicatedMortgageChartYouve_EverSeen (showing chart created by one homeowner to identify the parties involved in his mortgage). HeinOnline -- 36 J. Corp. L. 518 2010-2011 2011] Do Accounting Rules Matter? 519 where legal rules are designed to solve a problem unique to only one type of end user. This cautionary tale should guard us against the illusion that a smarter or more efficient regulator can nip large problems in the bud. The truth is often the opposite; the endless cycle of creating, enforcing, evading, and reregulating may dwarf the original problem. Each new round of imperfect rules creates new opportunities for capture and arbitrage, which only get larger if key private parties can collect and interpret obscure information, both public and private, more rapidly than government regulators. III. THE FRAGILE THEORETICAL CASE FOR MARK-To-MARKET ACCOUNTING One lesson that emerges from this historical account is that the risk of regulatory forbearance is always specific to government. This point influences the theoretical analysis that follows. More concretely, we believe the mark-to-market rules designed to trigger government action perversely triggered too much private action in the rapid inflation and ultimate implosion of the housing bubble. We believe the simple change in accounting convention made the current financial crisis much worse than it otherwise would have been. In contrast, the prevailing wisdom is that whatever the costs of mark-to-market accounting, the benefits of transparency are worth it. It is for just this reason that the SEC in its recent report rejected recommendations to suspend the mark-to-market standard.30 Many modem writers on the subject adopt this conclusion. For example, a recent World Bank report concludes that mark-to-market rules are needed to combat various forms of opportunistic behaviors: Especially at large and complex financial institutions, individual managers have strong incentives to discover and to exercise reporting options that overstate their capital and understate their exposure to loss. This expands their ability to extract implicit subsidies that risk-taking can generate from implicit safety-net support.31 The report supports this conclusion by noting, correctly, that the opacity of many balance sheets is accentuated by the use of special purpose vehicles that deftly keep certain key transactions off the books, thereby creating further opportunities for "arbitraging the supervisory system."32 Notwithstanding these strong claims, we do not think that the matter admits to such an easy resolution. In our view, there is no clear-cut, first-best solution for the valuation of certain complex assets. In a world of imperfect information, the relevant inquiry is which of the two methods-historical cost or mark-to-market-produces fewer errors on average. The answer to that question is context dependent. It is common ground that the mark-to-market rules do not work well with tangible depreciable assets, with their low 30. Press Release, U.S. Securities and Exchange Commission, Congressionally-Mandated Study Says Improve, Do Not Suspend, Fair Value Accounting Standards (Dec. 30, 2008), availablea t http://www.sec.gov/ news/press/2008/2008-307.htm. To be precise, fair value accounting represents a broad class of rules that reject historical cost-based accounting. Mark-to-market is one powerful rule in this group. 31. Gerard Caprio, Jr. et al., The 2007 Meltdown in Structured Securitization: Searching for Lessons, Not Scapegoats 22 (World Bank Policy Research Grp., Working Paper No. 4760, 2008), available at http://papers.ssm.com/abstract-id=1293169. 3 2. Id. HeinOnline -- 36 J. Corp. L. 519 2010-2011 520 The Journalo f CorporationL aw [Vol. 36:3 turnover rates. It is also clear that these mark-to-market rules are fine for assets in thick markets-those with lots of buyers and sellers-that make tradable assets easy to value. At the current stage of empirical knowledge, however, the verdict is less clear for hard- to-value financial assets. If our view is correct, it reveals a great irony about the claim that mark-to-market serves as a bulwark against accounting manipulation: if the mark-to- market rules work best when informational asymmetries are low (i.e., market prices are more readily available), they don't do valuable work in those hard cases where they are really needed. In our view, mark-to-market accounting (and its cousin, mark-to-model) may exacerbate bubbles when markets move upward and exacerbate downturns once those bubbles burst. Due to their magnitude, these swings leave in their wake dysfunctional lending markets and the dissolution of major investment banks. The valuation techniques are too imperfect to provide reliable information, so their application makes the entire system more fragile. FASB recognized this problem when it recently modified the mark-to-market rules to allow firms greater flexibility in reporting asset values where ill-functioning markets are unlikely to produce accurate prices.33 Although we agree with the spirit of the FASB modification to FAS 157, we fear it does not go far enough. But, whether right or wrong, we believe it is important to highlight a crucial theoretical criticism of mark-to-market rules thus far missing from the debate. We believe that mark-to-market accounting prematurely forces counterparties to demand collateral, even when it would be efficient, both privately and socially, for the counterparty to refrain from such demands. The constant application of the same misguided rule thus leads to self-perpetuating devaluation cascades, which, given the interconnectedness of modem financial markets, result in widespread destruction of real economic value. The difficulties are not mere happenstance. As we show below, the initial mark-to-market reform had two opposing consequences. It helped to discipline regulatory action. It also sparked risky practices by private parties. There are two key points we address. The first is accounting rules in down markets must deal with a basic prisoner's dilemma in which counterparties must decide whether to demand additional collateral or to forbear because they predict a future increase in asset values. Using historical-cost accounting has this hidden virtue for private players: it preserves their option to forbear on demanding collateral, which in turn allows them to avoid the devaluation cascades that these collateral calls are virtually certain to create. Bankruptcy is a good analogy, wherein the common situation the demand of a single unsecured creditor for security could trigger a rash of similar demands that can undermine or destroy the potential going- concern value of the firm.34 It is for that reason that the legal system developed the automatic stay to prevent the disintegration of the asset pool for the benefit of all creditors as a class.35 The mark-to-market rule works the opposite way, by forcing parties to demand collateral even where it might be efficient for them not to do so. 33. See Proposed FASB Staff Position FAS 157-d, FASB, available at http://www.fasb.org/ fasb-staff positions/prop fspfasl57-d.pdf (proposing amendments to FASB Statement No. 157). 34. See, e.g., DOUGLAS G. BAIRD ET AL., GAME THEORY AND THE LAW 232-38 (1994) (discussing automatic stay in bankruptcy in the context of game theory generally and prisoners' dilemmas specifically). 35. See id. HeinOnline -- 36 J. Corp. L. 520 2010-2011 2011] Do Accounting Rules Matter? 521 We believe parties with the same information about current and future valuations might make different collateral decisions if historical-cost rules were still in effect. The mark-to-market rules force more private decisive action because shareholders and regulators stand behind their decision. The situation is fraught with danger because the counterparty to a contract for an asset is not a single individual capable of making an independent decision. Typically that counterparty is a legal entity that owes fiduciary duties to its shareholders or investors. Accordingly the threat of (ex-post biased) litigation from the counterparty's stakeholders can easily prevent firm managers from taking steps that are consistent with an efficient ex ante bargain that they would have made with these same stakeholders. At the same time, there is no reason to think that regulators make better valuations, when they may also be subject to the same ex-post bias that plagues stakeholders in private ventures. The bank's rational, or at least defensible decision to value assets at more than market price could generate a legal penalty before bank managers have the opportunity to prove that their estimations are indeed better than the rest of the market. Once the bubble starts to burst, it is all too easy to condemn a bank decision that may well have been perfectly rational and efficient when made, solely because it appears opportunistic ex post. The fear of attracting regulatory wrath that could lead to stakeholder suits, either private or public, could make even the most informed firm reluctant to separate itself from the herd. Visibility begets liability, which in turn may prevent bank managers from practicing rational forbearance that makes sense for all the parties. Regulators standing behind firms making collateral decisions also can exert a counterproductive influence. If banks are required by law to mark assets of declining value to market, regulators may be forced to intervene decisively on any short-term dip in asset values, even for assets that may reasonably be expected to rise over the long run. The market value is only the summation of general views. The regulatory intervention necessarily removes those with positive expectations from the mix of players whose views can drive the market. The second key point is that the two different types of end users of accounting information-government regulators and private counterparties-may each have its own distinct optimal valuation rule. Mark-to-market was implemented not to solve a problem in private markets, but to remove regulatory discretion provided by historical-cost accounting. That rule, however, may have gone too far by eliminating efficient forbearance by market participants who, after all, were not fooled by historical-cost accounting the first time round.36 But government actors weren't fooled either.37 As one paper on the history notes, "all of the major participants-the industry, Congress, the Executive branch, and the regulators-were aware of the relative depth of the crisis."38 The problem for historical-cost rules was too much discretion for government regulators; the problem for mark-to-market rules was too little discretion for private actors. 36. See, e.g., BLACK, THE BEST WAY To ROB A BANK Is To OwN ONE, supran ote 9, at 193. 37. See supra note 8 and accompanying text (explaining that while the problems with historical-cost methods may not be present on the balance sheet they are obvious to informed observers). 38. R. Dan Brumbaugh, Jr. & Catherine J. Galley, The Savings and Loan Crisis: Unresolved Policy Issues, in THE SAVINGS AND LOAN CRISIS: LESSONS FROM A REGULATORY FAILURE 83, 89 (James R. Barth et al., eds., 2004). HeinOnline -- 36 J. Corp. L. 521 2010-2011