All Senators Should Watch Their Claims About Ending Bailouts

Click here for a printer-friendly version of this article

In the debate over reform of America’s financial regulation, perhaps the most important sticking point has been how to handle failing non-bank institutions such as insurance companies, hedge funds, and bank holding companies (the corporate parents of banks). Many Republicans argue that the proposed “resolution authority” for “systemically significant” firms would institutionalize bailouts. Other left-of-center experts contend that neither the House nor the Senate measure would have prevented the recent crisis, or stand much of a chance of heading off a future emergency. This has not stopped the President from calling some of these assertions “cynical and deceptive” – and boasting that the bill sponsored by Senate Banking Chairman Chris Dodd would end bailouts forever, not perpetuate them.

Such diametrically opposed interpretations of the same provision are standard fare when governments establish ambiguous, extrajudicial procedures to resolve failing businesses. A better idea would be to focus on changes to the bankruptcy code so these matters can be left to judges and courts of law.

First, this paper will quickly review the Lehman Brothers bankruptcy and why it should not be cited as the “reason” why the bankruptcy process is ill-suited for resolving a large financial institution. This section also touches on why AIG did not follow Lehman into bankruptcy.

Second, to gain a better understanding of why it’s problematic to create a new and separate resolution process outside of the bankruptcy code, this paper will review some of the other cases where extrajudicial resolution have been proposed (and used). In particular, this section of the commentary will focus on the GSEs and how the FDIC currently facilitates the sale of failed banks.

Lehman Brothers

For those who believe the U.S. needs a new resolution mechanism to deal with large interconnected firms, the panic that surrounded Lehman Brother’s failure 19-months ago is often cited as “Exhibit A.” Put simply – it shouldn’t be.

Former Treasury Secretary Hank Paulson and Fed Chairman Bernanke stated at the time that they lacked the authority to “save” Lehman. In many ways, these statements started the discussion about what new resolution authority had to be added to the government’s tool kit to prevent a future Lehman-type failure.

Mr. Paulson and Mr. Bernanke’s argument went like this. Lehman was insolvent and the firm’s management had been clearly told that the federal government was not going to bail them out. The market had been watching Lehman’s demise over the course of many months and should have understood that a bankruptcy filing was possible or even likely. This might not have been clear to some people since federal money had been available to facilitate the purchase of Bear Stearns by JP MorganChase. In the case of Lehman, the Treasury did not have the authority (pre-TARP) to inject capital directly in the firm and the Fed only had authority to provide a liquidity backstop. Basically, the Fed could loan Lehman money (see Bear Stearns) but only against collateral that they deemed sufficient – in terms of quality and amount. In Lehman’s final days, the Fed simply did not believe that it had sufficient collateral.

What’s important to remember about the Lehman case? The panic that ensued was not because the bankruptcy process failed, but because the Reserve Fund “broke the buck” leading short-term credit markets to approach lock-up.

The Reserve Fund was one of the largest and oldest money market funds – and it owned Lehman’s commercial paper. When the value of these assets plunged, the Fund’s net asset value dipped below par. This was a major event and it sparked a run on all money market mutual funds. Investors simply did not understand or see this as a largely idiosyncratic event, deciding instead to flee all funds. In turn, the entire commercial paper market froze – as money market funds were also normally big buyers of short-term paper. Then, money market funds started hoarding cash rather than investing any money back into the short-term paper or repo market. As the short-term paper market seized up, large financial institutions in turn started pulling back the credit lines that they had extended to their other clients. Clients (from corporations to hedge funds) then started selling assets to raise the necessary cash to run their businesses (absent any other “open” short-term credit market). The fire sale was officially on – and the entire system was deleveraging all at once. From the perspective of the White House, the Treasury or the Fed, the situation clearly looked like a run on the entire financial system.

The problems only intensified when Lehman Brothers’ senior bond holders discovered that they would receive less than 10 cents on the dollar. (This was partly due to derivatives counterparties going to the front of the line as the firm entered bankruptcy.) The interest rate offered on unsecured short-term loans to financial institutions (measured by LIBOR) was simply too low to justify such large potential losses. As a result, the interbank lending market came to a standstill and the LIBOR spread over Treasury bills shot up to record levels.

AIG

This same week, AIG crumbled. No one in the Administration wanted to “save” AIG and/or face the criticism that if the government stepped in to help AIG now, that its actions could not be reconciled with how Lehman was just handled. Nevertheless, the Fed decided – with very little time to inspect AIG’s books – to keep the firm alive through an emergency loan.

On the heels of Lehman – and the aforementioned ripple effects – AIG was simply viewed as a much larger and more interconnected firm. And, given that nobody really saw the money market panic coming after Lehman filed for bankruptcy, the government was skittish about allowing some other externality to occur (this time). The lesson here is that only by guaranteeing Lehman’s debt (and indirectly the Reserve Fund’s holdings) could the original panic have been prevented. Had the commercial paper market not seized up, the AIG situation may have played out very differently. It’s really impossible to know, one way or the other.

What can be concluded, however, is that it is impossible for anyone to claim that a new piece of legislation would have prevented the recent crisis unless the measure explicitly would have allowed for a creditor-bailout (or the guarantee of a failing firm’s debt).

Government Sponsored Enterprises (GSEs)

It is important to recognize that so-called “resolution authority” is really nothing new. For example, the government has the power to take control of failing housing GSEs and insured depository institutions outside of the normal bankruptcy system. Yet in both of these cases, the extrajudicial process creates legal uncertainty that makes investors wary or increases the ultimate costs to taxpayers.

Equally important to recognize is that there is ambiguity here too. No law said the government must bail out GSE creditors. The GSEs were simply ineligible for liquidation or reorganization under the Bankruptcy Code because they were “instrumentalities” of the federal government (see page 609; 46 of 79 of the pdf). In lieu of bankruptcy, their regulator was permitted to appoint a conservator to take control of the enterprises.

While it may seem obvious in retrospect that “taking control” would be akin to a bailout, that was not the common understanding prior to their failure despite the judgment of the rating agencies. As is the case with the measure before the Senate this week, this extrajudicial resolution mechanism allowed for a range of contradictory interpretations of what would happen in the event of a GSE failure.

Everyone knows by now that the credit rating agencies made a number of colossal misjudgments that contributed to, and were exposed by, the financial crisis. The ratings on many structured credit products, particularly collateralized debt obligations (CDOs), were egregiously wrong. Taken to their logical limit, these ratings implied that all assets, however dodgy, could earn a AAA rating if they were securitized enough times. The agencies also badly missed on the large broker-dealers. Merrill Lynch and Lehman Brothers were rated A+ by S&P until June of 2008. This is equivalent to a five-year default probability of less than 1%, despite the fact that both firms had gross leverage ratios over 30-to-1 and held far more questionable commercial and residential mortgage assets than Tier 1 common equity.

Yet the agencies’ ratings of the GSEs – Fannie Mae and Freddie Mac – were exactly right. S&P rated Fannie and Freddie’s senior debt AAA thanks to “the strong explicit and implicit U.S. government support these securities hold in the marketplace.” As the Congressional Budget Office (CBO) explained to lawmakers in 2001, “S&P ratings for GSEs are explicitly intended to reflect ‘risk to the government,’ the risk that a GSE, while retaining all the privileges of government sponsorship, will require an infusion of cash from the federal government to meet its obligations (emphasis in the original).” The rating agencies were right – despite accumulating losses leading to a negative net worth of more than $125 billion, senior creditors to Fannie and Freddie suffered no losses.

At various times, it was in the interests of GSE executives, their supporters, and their critics in government to claim that they’d be treated no differently than other firms in the event of failure. Treasury Secretary Snow spoke of a “market misperception of an implied guarantee” on GSE securities. Congressman Barney Frank was more blunt, telling creditors “there is no guarantee, there is no explicit guarantee, there is no implicit guarantee, there is no wink-and-nod guarantee.” Fannie Mae CEO Franklin Raines also disputed the existence of an “implied guarantee” in a 2004 open letter to mortgage bankers where he compared the GSE charter to the thousands of other charters in existence.

This impasse was especially difficult to resolve during the good years, when the GSEs were profitable and had no difficulty meeting commitments to creditors. Claims about the hypothetical actions government officials would take in the event of sudden and unexpected losses were easy to dispute. The GSEs even scoffed at critics’ contention of a funding advantage, which Fannie CFO Tim Howard referred to as the “so-called subsidy.” When companies are liquid and facing little perceived risk, it is very difficult to prove (beyond doubt) that their cost of funding derives from creditor expectations of a bailout. If “systemically significant” firms are able to fund themselves at a lower cost after passage of the Dodd bill, expect them to present an alternative explanation for that advantage.

FDIC bank sales

The executive branch is also empowered to take juridical resolution actions with respect to insured depository institutions. However, the Federal Deposit Insurance Corporation (FDIC) generally acts to facilitate sales of failed banks, not take control of them. No one believes the FDIC could have handled the liquidation of Washington Mutual (WaMu) or Wachovia, let alone Citibank. Moreover, pointing to the “success” of the FDIC’s extrajudicial resolution process fails on two additional counts.

First, the banks “resolved” by the FDIC are subsidiaries of bank holding companies, not the “banks” themselves. Given that most of the complexity surrounds cross-guarantees and interaction among subsidiaries, the FDIC experience facilitating sales of one subsidiary is hardly analogous to the proposed resolution mechanism that would empower the government to assume control of literally thousands of subsidiaries.

Second, the FDIC did not acquit itself very well in its assisted sale of WaMu to JPMorgan. Senior bond holders were subordinated to uninsured depositors and counterparties even though these claimants were legally entitled to be treated “equally and without preference.” The FDIC’s ability to (ignore standard bankruptcy procedures and) decide unilaterally which equal status creditors would receive 100 cents on the dollar and which would receive nothing likely spooked investors, whose analyses of maximum potential losses were shown to be premised on a legal stability that does not, in fact, exist in extrajudicial resolutions. Proponents of having new resolution authority outside of the bankruptcy code conveniently overlook the enormous losses incurred by senior unsecured creditors to WaMu and the contribution those losses had to the freeze in interbank lending.

That the Senate measure would expand the government’s discretion to pick winners and losers among similarly situated creditors has not gone unnoticed by the rating agencies. If the government can ignore the priority of claims to assign losses to creditors that it believes can bear them without further incident, potential losses are likely to be much higher than expected for unsecured senior creditors. The likely result is an increased funding cost for banks accessing the commercial paper and bond markets.

For other classes of creditors, the bill is likely to provide more certainty on the front end but at the expense of opening them up to nearly limitless liabilities after the fact. “Systemically significant” firms, it is argued, need to be resolved outside of bankruptcy because their failure would impose losses on creditors that would trigger a cascade of subsequent losses capable of imperiling the financial system. If a resolution mechanism does not lead to bailout payments to the resolved firm’s creditors or counterparties, the firm was, by definition, not systemically significant. Therefore, the supposition among supporters of the Senate bill is that bailouts will exist to prevent cascading losses from bringing down the financial system, but these bailouts will be financed from entirely within the financial system through (ex ante or ex post) assessments. The “beneficiaries” of the bailouts will pay for whatever support is provided, while the government conservator dismantles the failed institution.

Instead of the predictability of bankruptcy and the clear legal protections (and definitions) that it offers, a new and separate resolution mechanism would create a form of “jackpot justice” where certain classes of creditors get subordinated to others based on the judgments of executive branch officials. Resolution authority could combine the worst of both worlds: taxpayer bailouts and crisis-inducing “runs” by creditors seeking to exit their investment so as to avoid the unpredictability of the resolution, including the future liability for any funds extended immediately following the failure.¹ Alternatively, some other “well-connected” creditors might look for exactly these sorts of situations – perhaps because they believe they’ll be able to manipulate their government connections to get their creditor class a handsome payout.²

Improving the Bankruptcy Code Makes Sense Moving Forward

Rather than accepting resolution as a fait accompli, the Senate should work to enhance the bankruptcy code’s ability to handle the failure of a large, complex firm. This probably requires removing the counterparties’ exemption from the automatic stay that normally bars creditors from demanding immediate repayment from bankruptcy filers. The current stay allows repo and derivatives counterparties to terminate their transactions, take possession of the failed firm’s collateral, and “jump ahead” of even senior secured creditors in the order of payment priority in bankruptcy. This eliminates the normal incentives to monitor the borrowing of troubled companies and permits businesses to become more risky. As Harvard Professor Mark Roe explains:

Someone had to lose money when Lehman failed. But had the super-priorities not been in place when Lehman built its capital structure and derivatives portfolio, the incentives for Lehman’s derivatives and repo counterparties’ to want a more stable Lehman would have been greater. And Lehman itself would have been incentivized to keep to a safer capital structure, in order to encourage the counterparties to keep dealing with them.

Bankruptcy reform may also obviate the need for many parts of the controversial title of the (Senate) bill on derivatives³. Were derivatives owners not provided these “super priorities,” they would take the credit risk of their counterparty far more seriously. But since an exhaustive credit analysis of each counterparty in deep and liquid trading markets is impractical, the bankruptcy change would either force over-the-counter (OTC) derivatives to exchanges charged with managing such exposures, or dramatically shrink the size of the market.

This bankruptcy change may not be sufficient in and of itself. There is a strong case also for the creation of a debtor-in-possession financing mechanism that places the Treasury in a senior, secured position so that asset “fire sales” can be avoided while the judge sorts through claims with no loss to taxpayers. The difference between DIP and an ex post bailout fund is that the assets of the estate would secure the liquidity extended by Treasury, and the resolution of the firm (including possible reorganization and disposition of all assets) would be decided through normal bankruptcy procedures.

These changes would remove uncertainty about the treatment of creditors in the event of failure, remove the potential for ex post assessments, and force derivatives and repo counterparties to better assess capital and liquidity positions before entering into contracts. There are other bankruptcy amendments that should be considered as well, like new language that would enable close-out netting of repo and derivative contracts (so that all offsetting credit exposures could be combined into a single net amount).

The bottom line is that more effort should be made to update the bankruptcy code. At minimum, Senators should call on the Judiciary Committee to weigh-in. There is no question that a lot of this work should have been done in parallel with the Senate Banking Committee’s efforts over the last few months (just as Sen. Blanche Lincoln developed the derivatives title in the Agriculture Committee). But, getting this issue right (the first time) is much more important than letting committee jurisdictional issues win out. While it will be interesting to watch all the floor statements to learn which Senators actually think this measure will end bailouts (once and for all), everyone should keep an eye on Sens. Leahy and Sessions, the respective Chairman and Ranking Member on the Judiciary Committee. Will they go on record claiming that this measure ends bailouts?

The Senate bill’s insistence that large financial institutions be resolved outside of bankruptcy invites destructive speculation about how various claims holders would fare in the event of a resolution. Some believe this new extrajudicial process would institutionalize bailouts, while others believe it is a necessary step to end the current posture of ad hoc bailouts. Whoever is right on this point, we do know that bankruptcy works in nearly all settings and the track record of extrajudicial resolution mechanisms is quite poor. Rather than create the risk of perpetual bailouts, or selective losses imposed on disfavored creditor constituencies, the Senate should work to improve the bankruptcy code so that the filing of a large financial firm can be accommodated without incident.

Footnotes

1 Section 204(a)(3): the FDIC “and other agencies will take all steps necessary to assure that all parties, including management and third parties, having responsibility for the condition of the financial company bear losses consistent with their responsibility, including actions for damages, restitution, and recoupment of compensation and other grains not compatible with such responsibility.” This may sound like commonsense language, but the discretion for defining a third party is basically limitless (rating agencies, creditors, employees, the government’s regulators, etc). And, how and by what proportion will losses be handed out amongst these parties?

2 Section 210(d)(2): This section says that “all similarly situated” creditors need to be treated the same, unless disparate treatment is necessary to “maximize the value of the assets of the” company or “minimize the amount of any loss” on an asset sale. To try an assuage some concerns, the bill also sets out in Section 210(d)(2) that if the FDIC does decide to preferentially pay a creditor more, it can only do so if similarly situated creditors received at least what they would have received in a Chapter 7 bankruptcy. This may also sound reasonable, but it’s the FDIC that determines the value of any claim that would have been awarded in Chapter 7. The combination of these two provisions also means that the FDIC can “over” pay one creditor, as long as similarly situated creditors get what the FDIC says they would otherwise have received in bankruptcy. There is another subsection that authorizes (if the FDIC and Treasury Secretary agree) to make “additional payments” so long as they are deemed “necessary or appropriate to minimize losses to the FDIC as receiver.” Again, “necessary or appropriate” is anything but an airtight exception. It’s important to note here that some of these language (or “trust”) issues would not be as troubling if the Administration had handle the Chrysler and GM situations differently. Remember, politically favored unions received 3x the amount of “similarly situated” creditors. Put another way, both firms were basically used to transfer TARP money to a favored group by the government.

3 This is not to say that I am opposed to a strong derivatives title – one that pushes nearly all contracts onto an exchange and through a clearinghouse. I just think the bankruptcy code (and the resolution issue) should be the starting point of the broader debate about how to fix financial regulation generally.

Christopher Papagianis is the Managing Director of e21 and was Special Assistant for Domestic Policy to President George W. Bush.