The AIG Case: Moral Hazard on Steroids!

The AIG decision (actually, Starr International Co. v. The United States[1]) has shocked many but for the wrong reason. Some commentators have focused on the ingratitude of Maurice Greenberg, AIG’s former CEO and the “architect” of its international insurance business. In their view, he should have been thankful for the $85 billion loan extended by the Federal Reserve Board (which still left AIG’s shareholders holding 20% of their stock). Ultimately, AIG’s shareholders did much better than their Lehman counterparts (who received nothing), but these issues of comparative fairness and Greenberg’s alleged chutzpah go mainly to the cosmetics and should be left to the tabloids to debate.

Others, including the Wall Street Journal’s editorial page, have applauded the decision, because “the judicial branch of government has forcefully reminded the executive of its legal limits, even in a crisis.”[2] But the decision is hypertechnical and defines the authority of regulators with a hair-splitting narrowness. Even before this decision, federal regulators (more the SEC than the Federal Reserve) appear to have acted in an overly equivocal and cautious fashion in response to the 2008 crisis, and this decision will likely further chill their willingness to act decisively. Still, the real issue is incentives, and this decision creates terrible incentives. By most accounts, the 2008 crisis had its roots in the problem of moral hazard. Executives at financial institutions caused them to take on enormous and often undiversified risk, believing (or at least hoping) that the federal government would bail them out because they were “too big to fail.” Indeed, at the outset of the crisis, exactly this pattern prevailed. Bear Stearns was rescued by a Federal Reserve-arranged shotgun marriage with JP Morgan, and later Merrill Lynch was similarly absorbed by Bank of America, in each case with the Federal Reserve quietly playing matchmaker (as it has done many times in the past).

When Lehman neared insolvency in October, 2008, the Federal Reserve and the Treasury decided that it was time (at last) to draw the line, and they refused a bailout. But when AIG’s much larger failure loomed in quick succession, they realized that the resulting financial panic might have been uncontrollable. Still, they resolved to be tough, and they charged a high interest rate, took some points, and demanded 79.9% of AIG’s stock for their $85 billion loan. For this toughness, they have now been rebuffed in a harshly worded decision that relies on a dubious logic to confine their discretion. According to Court of Federal Claims Judge Thomas Wheeler, the Federal Reserve, “operating as a monopolistic lender of last resort,… imposed a 12 percent interest rate on AIG, much higher than the 3.25 to 3.5 percent interest rates offered to other troubled financial institutions such as Citibank and Morgan Stanley.”[3] But wait a second here! What makes the Federal Reserve a “monopolist”? It is a strange definition of “monopoly” that finds it to be created by the fact that no one else would make this risky a loan. Is the Red Cross a “monopolist” if only it rushes to a disaster scene to provide emergency relief? Nor it is strange that the Federal Reserve would charge less to other financial institutions that it knew much better. After all, the Federal Reserve regulates banks (but not AIG, an unregulated insurance company), and thus it both better understands their condition and can control them more easily. Instead, in the case of AIG, the Fed was in effect buying the proverbial “pig in a poke.”

Next, the Court dismisses the AIG board’s approval of these “harsh” terms “because the only other choice would have been bankruptcy.”[4] But why does the fact that you are effectively bankrupt entitle you to easier terms? Judge Wheeler repeats this theme several times, finding that:

“The weight of the evidence demonstrates that the Government treated AIG much more harshly than other institutions in need of financial assistance.”[5]

So what? Implicit here is a suggestion that central bankers are under some duty to be “fair” to all loan applicants. If such a duty were to be judicially created, even a failure to loan a troubled firm could be actionable (Lehman shareholders take note here, but the statute of limitations has likely run). Also, if the threat of bankruptcy vitiates the borrower’s consent to the lender’s harsh terms, all banks will face problems. The better view is that the real duty of central bankers is to safeguard the economy.

Equally troubling is the way that the Court goes well beyond its judicial role to play amateur historian. It finds “that AIG actually was less responsible for the crisis than other major institutions.”[6] Because no other financial institution was before the court, this is a strange comparative finding to reach on an effectively ex parte basis. Although this is revisionist history by any standard, its relevance is also unclear. Why does this court feel entitled to second guess the authorized decision maker? Why does it matter that AIG was better or worse in its risk-taking than other shadow banks? Nonetheless, based on this finding, the Court concludes that “[t]the Government’s justification for taking control of AIG’s ownership and running its business was entirely misplaced.”[7] Urgent and necessarily rushed decisions made by the Federal Reserve merit more deference.

But the real question is the impact of this ruling. One can argue that because the Court awarded no damages, Greenberg’s victory was hollow. Still, that misses a key point. Imagine now a future CEO of a troubled financial institution that has hit the rocks. Even as financial failure looms, this CEO can say to himself (or counsel may say to him): “The Government has to offer us the same rescue terms as they offer everyone else, and they cannot play favorites. If we can show that our firm was no more responsible than others for the crisis, they have no business taking us over and replacing management.” Not only are regulators crippled, but such a perception (which this decision certainly encourages) fuels the moral hazard problem. On appeal, this is what most needs correction.

In fairness, the Court is aware of the moral hazard problem, but dismisses it, finding

The Government’s unduly harsh treatment of AIG in comparison to other institutions seemingly was misplaced and had no legitimate purpose, even considering concerns about moral hazard.”[8]

This repeated focus on fairness and relative fault can be explained in one of two ways. First, it may betray Judge Wheeler’s special position as a Court of Federal Claims judge. Not an Article III Court, the Court of Federal Claims has a limited and possibly warped perspective, and it is experienced only in resolving contractual claims and monitoring the Government’s behavior as a contracting party. In that role, it may be appropriate to require the Government to be scrupulously fair. But when the Government acts as a regulator or enforcer in an emergency, there are different considerations at play (for example, can the Federal Reserve save the country?). Judge Wheeler seems oblivious (or at least insensitive) to the difference between these two contexts. A second possibility is that the Court may have been captured by the brilliant advocacy of David Boies, who clearly induced the judge to swallow hook, line and sinker his theme that AIG was a victim of harsh and unjustifiably retributive treatment. That leitmotiv percolates throughout the decision (as earlier noted), even though it seems legally irrelevant to a case that the Court ultimately determines based on its narrow reading of Section 13(3) of the Federal Reserve Act. But that narrow reading of Section 13(3) may flow from Boies’ extraordinary orchestration of the case..

Whatever the reason, the decision is hypertechnical. Ultimately, the key legal issue for the Court is whether Section 13(3) of the Federal Reserve Act authorizes the Federal Reserve to take equity for a loan. Nothing in the express language of Section 13(3) grants or denies the Federal Reserve the power to do so, but over a long history it had never previously done this.[9] Still, Section 13(3) does state that Federal Reserve loans are “subject to such limitations, restrictions, and regulations as the Board of Governors of the Federal Reserve system may prescribe.”[10] The Government argued that this language gave the Federal Reserve Board the authority to impose a “restriction” or “limitation” that it receive a 79.9% block of stock (to which “restriction” or “limitation” AIG’s board then consented). Arguably, this interpretation was entitled to some measure of Chevron deference, but the decision never discusses the appropriate standard of review.[11] To further backstop its argument, the New York Federal Reserve Bank added a special twist to the deal: the AIG stock was not held by the Federal Reserve or any regional bank thereof, but was placed instead in a special trust. Judge Wheeler rejected this defense as pure “form over substance,” but sometimes federal courts pay heed to form when reviewing regulatory actions.

Prior decisions had permitted the Federal Reserve to hold stock as collateral for a loan, even though Section 13(3) did not expressly authorize this.[12] Thus, the question became whether the Federal Reserve could receive directly property that it was otherwise entitled to seize on default. Arguably that is not that major an additional step, but Judge Wheeler would not take it. Judge Wheeler also skirted established limits on the jurisdiction of his court by finding that the challenged transaction was not an unconstitutional “taking” in violation of the Fifth Amendment (as to which he lacked jurisdiction), but rather an “illegal exaction” in violation of the Due Process Clause (as to which he found that he had jurisdiction).[13] At every juncture, the Court split hairs to the Government’s detriment.

Where are we left? Section 13(3) was repealed by the Dodd-Frank Act and replaced by an elaborate set of provisions for “resolution authority.”[14] These provisions are sufficiently complex and legally untested that the Government will be reluctant to use them. Instead in a crisis, the Federal Reserve may still seek to negotiate an emergency loan (in reality, a de facto bailout), but it will have even less authority after this decision to acquire stock or any form of control over a troubled borrower. As a result, financial managers may have increased incentive to accept high risk and anticipate a soft bailout at worse (“déjà vu all over again,” in Yogi’s words). The decision will be appealed, and they key issues are questions of law which the appellate court should review de novo. Much of the decision seems vulnerable.[15]

Let me conclude with a strange pairing: failing financial firms and Greece. Both want their creditors to bail them out on favorable terms without the creditors taking control or changing their operating policies. Judge Wheeler has gone some distance towards accomodating financial firms in this desire, and it remains to be seen what will happen with respect to Greece. But in both cases the risks are immense.


[1] Starr Int’l Co. v. United States, No 11-779C (United States Court of Federal Claims, June 15, 2015)(Slip Opinion).

[2] See “Washington’s Illegal Bailout,” Wall Street Journal, June 16, 2015 at A-14 (editorial).

[3] Starr Int’l Co., Inc. v. United States, supra note 1, at 2.

[4] Id at 6

[5] Id.

[6] Id.

[7] Id.

[8] Id. at 7.

[9] Id at * 55 (describing prior use of Section 13(3)).

[10] Id at 58. See also 12 U.S.C. § 343 (2006).

[11] Arguably, the Federal Reserve is entitled to Chevron deference in interpreting the authority granted it by Section 13(3). See Chevron U.S.A. Inc. v. Natural Resources Defense Counsel, Inc., 467 U.S. 837 (1984). The scope of Chevron is subject to unending debate, particularly among academics who have little else to do. Although the Federal Reserve is not a traditional administrative agency (for example, it is not subject to the “notice and comment” provisions of the APA), the same policy considerations arguably should apply. One reason that the Court may have ignored this issue is that a “smoking gun” email between two of the Federal Reserve’s outside counsel was introduced at the trial that characterized their client (the Federal Reserve) as being “on thin ice,” but predicting (wrongly) that no one would sue. This ranks as one of the all-time blunders by defense counsel, even if such emails are normally subject to the attorney/client privilege. Law school should teach more about such mistakes.

[12] See Lucas v. Federal Reserve Bank of Richmond, 59 F. 2d 617 (4th Cir. 1932).

[13] See Starr Int’l Co., Inc. v. United States, supra note 1, at 50-52. The Court found that the Tucker Act denied the Court jurisdiction over Fifth Amendment claims (unless based on money-mandating language in a statute), but that it had jurisdiction over an “illegal exaction” claim when based on an asserted statutory power. See Aerolineas Argentinas v. United States, 77 F. 3d 1564, 1573 (Fed. Cir. 1995).

[14] See Dodd-Frank Act, Tit. II §§ 201-217, codified at 12 U.S.C. 5381-5394. These provisions apply only to bank holding companies and certain other institutions determined to be “systemically important” (such as AIG). Thus, there are major institutions not subject to resolution authority.

[15] Although I will not predict the outcome on appeal, it is noteworthy that the Federal Circuit did grant a writ of mandamus to overrule Judge Wheeler earlier in the litigation (on a procedural issue). See In re United States, 542 Fed. Appx. 944 (Fed. Cir. 2013). The Federal Circuit could duck the Section 13(3) issue by finding (1) that the court lacked jurisdiction over the claim under the Tucker Act; (2) that AIG’s informed acceptance of the loan waived any flaw in the transaction; or (3) that the lack of damages made it unnecessary to review the legal issues, particularly when the statute has been repealed. Given the lack of damages (if upheld), the Circuit Court may feel it inappropriate to render an advisory opinion on a repealed statute.

The preceding post comes to us from John C. Coffee, Jr., the Adolf A. Berle Professor of Law at Columbia University Law School and Director of its Center on Corporate Governance.

1 Comment

  1. Connor Raso

    The comment that “urgent and necessarily rushed decisions made by the Federal Reserve merit more deference” is very interesting. In all the work on judicial deference (and especially Chevron), has there been any empirical analysis of whether courts are actually more deferential to agency decisions that are necessarily rushed? That seems like an interesting and important issue.

Comments are closed.