Bankruptcy-proof finance and the supply of liquidity.
NOTE CONTENTS INTRODUCTION I. THE DEMAND FOR LIQUIDITY AND THE ROLE OF BANK INSOLVENCY LAW A. Banking as Liquidity Creation B. Implications for Bank Resolution II. THE STRUCTURE OF SHADOW BANKING A. Securitization: Bank Lending Unbundled B. Repos: Shadow Bank "Deposits" III. BANKRUPTCY-PROOFING SHADOW BANKING A. Securitization and Bankruptcy-Remoteness B. The Repo Safe Harbors IV. ASSESSING THE REPO SAFE HARBORS V. SYNTHETIC SECURITIZATION AND THE DERIVATIVE SAFE HARBORS A. Credit-Risk Transfer in Shadow Banking B. Ongoing Challenges C. The Scope of the Rationale for the Derivative Safe Harbors CONCLUSION
The tremors that shook Wall Street in 2008 radiated from a set of
novel asset markets straddling the boundary between commercial banking
and the capital markets. Mortgage-backed securities and related funding
instruments had transformed the credit landscape during the preceding
years, enabling institutional investors to supply capital for
residential mortgages and other opportunities once accessible only to
deposit-taking banks. (1) Thus a “shadow” banking system
emerged alongside the regulated banking sector, and grew to rival it in
size by its 2007 peak. (2) Only a year later this system collapsed, (3)
taking with it many of the country’s leading financial institutions
and plunging the economy into a deep recession. (4)
Despite the complexity of the transactions involved, the basic
dynamics of the 2008 crisis were distressingly familiar. A classic
banking panic had occurred, (5) although it struck outside the
traditional banking sector and the regulatory institutions protecting
it. (6) Banking crises occur when depositors demand more withdrawals
than the system’s limited cash reserves can satisfy, forcing banks
to liquidate assets or seek emergency assistance. (7) Like traditional
banks, shadow banks such as Bear Stearns and Lehman Brothers held large
portfolios linked to mortgages and other conventional bank receivables.
However, these institutions funded themselves using commercial paper and
other short-term borrowing markets that lacked the stabilizing influence
of FDIC deposit insurance. (8) This left shadow banks vulnerable to a
dramatic loss of liquidity as capital fled from mortgage-related assets
in 2007 and 2008, (9) forcing officials to rescue entities that lacked
access to backstops such as the Federal Reserve’s discount window.
Complicating matters, shadow banks faced a paradoxical legal
situation as they edged toward the precipice in 2008: although they were
regulated as nonbanks, applicable insolvency law treated these
institutions rather like traditional banks. As we shall see, this
paradox meant that shadow banks were excluded from both the regulatory
safeguards available to commercial banks under Title 12 and certain
bankruptcy protections available to nonbanks under Title 11.
Banks are not eligible debtors under the Bankruptcy Code, (11) and
the Federal Deposit Insurance Act provides neither a general stay nor
avoiding powers to protect them from their depositors’ claims. (12)
In contrast, the Bankruptcy Code gives debtors protection from their
creditors’ recovery efforts through the automatic stay, (13)
avoidance of preferential transfers, (14) and related provisions. (15)
Yet unlike most debtors, shadow banks traded heavily in contracts that
are exempt from protections ordinarily available to debtors in
bankruptcy. (16) These include repurchase agreements (repos), a
short-term borrowing instrument that Bear Stearns, for example, used to
stay afloat during its final months as an independent company; (17) and
derivative contracts, which Wall Street firms used to trade
mortgage-related risk. (18) The statutory exemptions for these contracts
allow the parties to enforce their contractual rights outside of
bankruptcy proceedings. (19) These typically include the right to
liquidate collateral from, and to terminate dealings and net mutual
obligations with, a distressed counterparty. (20)
The statutory carve-outs, or “safe harbors,” are
traditionally justified by the need to protect the financial system from
the fallout of a major market participant’s failure. (21) Yet the
exercise of these rights facilitated a kind of bank run on weak
institutions as their counterparties moved to protect themselves from
the deepening crisis in 2008.
The run on the shadow banking system was most apparent in the repo
market, which provided “the main source of funds” for the
securitization process. (22) Like bank depositors, repo lenders have the
option to withdraw credit almost immediately, as many repo loans mature
overnight and must be rolled over daily. (23) By 2008, much of the
collateral that shadow banks could offer to secure their repo borrowings
consisted of structured products tied to the mortgage market or
otherwise affected by the credit squeeze. (24) As this form of
collateral grew increasingly unacceptable to repo lenders, Bear Stearns
and other institutions struggled to raise the cash necessary to continue
Major institutions also hemorrhaged cash through their derivative
contracts, which gave parties the right to demand collateral as their
counterparties weakened and terminate contracts in events of default.
(26) For example, MG and other firms that had sold
“protection” against losses on mortgage-backed securities were
required to put up cash as these securities plunged in value and as
their own finances weakened. By mid-September 2008, AIG had posted more
than $19.5 billion in collateral on credit default swaps written by its
Financial Products subsidiary. Mounting demands from its counterparties
ultimately forced a costly rescue by the New York Federal Reserve and
the Treasury Department. (27)
These episodes have motivated a growing body of scholarship calling
for a rollback of the safe harbors for repos and derivatives, so that
distressed firms can invoke traditional bankruptcy protections against
their counterparties under these contracts. (28) Proponents of a more
protective insolvency regime find three principal defects in the safe
harbors. First, by permitting counterparties to withdraw credit and
seize collateral from weak institutions, the safe harbors may expose
weak firms to a sudden loss of liquidity that can quickly spread to
other firms. (29) Second, the race to grab the assets of an insolvent
firm can hamper its orderly resolution and destroy going-concern value.
(30) Third, counterparties whose contractual rights are unfettered by
bankruptcy procedures may lack optimal incentives to monitor their
counterparties’ risk-taking and may overuse contracts protected by
the safe harbors. (31)
Breaking with the emerging consensus, this Note argues that
repealing the safe harbors would be a misdirected response to the
fragility of nonbank financial companies. Part I lays the foundation for
an account of the role of the safe harbors in the supply of liquidity
through the shadow banking system. I proceed from the premise that
“financial instruments, markets, and institutions arise to mitigate
the effects of information and transaction costs.” (32) On this
view, bank deposits, for example, create valuable liquidity by offering
an investment contract free from the due-diligence and other transaction
costs that hamper trading in virtually every other asset class, from
home loans to technology stocks. (33) Banks accomplish this by issuing
what I call liquidation rights, or options to convert assets to cash.
For example, a bank depositor indirectly invests in the bank’s loan
portfolio but retains the right to withdraw her investment without
taking a loss. In this way, banking transforms illiquid portfolio assets
into a liquid investment contract that enlarges the supply of capital
that households are willing to invest.
The subsequent Parts argue that rights allowing repo and derivative
counterparties to liquidate these contracts outside bankruptcy play an
analogous role in attracting capital from these “depositors”
in the shadow banking system. (34) In this way, the safe harbors may not
only expand the supply of loanable capital, but they may also make that
supply more resilient by insulating investors from bankruptcy risk. If
it was a loss of repo credit that ultimately felled Bear Stearns, this
was because other funding sources with fewer privileges in bankruptcy
had long since become unavailable to the troubled investment bank. (35)
My claim that the law should enforce bank-issued liquidation rights
should not be confused with an argument that the banking system should
be allowed to fail during a crisis. (36) Critics of the safe harbors are
undoubtedly right about the consequences of allowing a large institution
to unravel under pressure from its repo and derivative counterparties.
(37) Yet if Lehman Brothers should have gained protection from its
counterparties in September 2008, this should have come in the form of
emergency liquidity, not a judicial stay imposed after the firm was
already in bankruptcy. (38) Curtailing the safe harbors would seemingly
do little to expand the options available to troubled firms ex post,
while doing much to undermine the liquidity and stability of the repo
and derivative markets ex ante.
The challenge for regulators, then, is to supply a framework that
reduces the incidence of bank runs at minimal cost to bank liquidity
creation. If this task seems daunting, it may be helpful to recall that
today’s regulated banking sector was once “an inherently
fragile, shadow banking system operating without credible public-sector
backstops and limited regulation.” (39) Yet twentieth-century
regulators enacted a mix of deposit insurance and prudential oversight
that largely ended the threat of panics in the traditional banking
sector. An insolvency regime that curtailed depositors’ rights in
order to prop up weak banks was conspicuously absent from this formula,
since it would have undermined the very liquidity services that
regulators sought to protect. (40) Seen in this light, the current
so-called “special treatment” (41) of repos and derivatives in
bankruptcy is far from anomalous: more remarkable was the absence, in
2008, of regulatory mechanisms to ensure the shadow banking
system’s resilience to crises of confidence. Instead of trying to
legislate away financial fragility through insolvency law, policymakers
should work toward a regulatory regime for shadow banking that
approaches their successes in the traditional banking sector.
I. THE DEMAND FOR LIQUIDITY AND THE ROLE OF BANK INSOLVENCY LAW
Capital traded through repos and derivative contracts has been
spotted fleeing the scene of recent history’s most prominent
financial disruptions, including the 1998 failure of Long-Term Capital
Management; the liquidity crises of Bear Stearns, Merrill Lynch, Lehman
Brothers, AIG, and other firms in 2008; and the bankruptcy of MF Global
in 2011. Accordingly, critics of the safe harbors have argued that
ordinary bankruptcy protections should be available to stanch the
outflow of liquidity through the repo and derivative books of troubled
However, this Part argues that such proposals elide important
distinctions between the functional principles of bankruptcy law and
bank regulation. Bankruptcy generally aims to protect assets from
inefficient liquidation and to return an equitable and
incentive-compatible recovery to each claimant. Most legal scholars have
analyzed repos and derivatives from this perspective, arguing that
bankruptcy-law safeguards should be used to protect firms from chaotic
unraveling by their counterparties. In contrast, modern approaches to
bank regulation recognize that the efficiency of markets for many
financial contracts would be undermined if participants could not
transact without exposing themselves to uncertainty and delay in a
counterparty’s insolvency proceeding. (42)
The choice between bankruptcy and bank-regulatory approaches
depends on the nature of the costs imposed by a firm’s failure.
Bankruptcy is adapted for the typical case where coordination problems
prevent creditors’ recovery efforts from achieving an optimal
allocation of a failed firm’s assets and default losses. Financial
institutions also encounter coordination problems among customers or
creditors racing to withdraw cash or seize collateral. (43) However, the
resolution of failed financial intermediaries raises additional
functional considerations relating to these firms’ role as
suppliers of liquidity. (44)
Liquidity is the ability to trade an asset at minimal cost or
delay. (45) Securities firms, for example, facilitate the issuance and
exchange of securities by acting as “market makers” who stand
ready to buy or sell securities on demand. (46) Commercial banks create
liquidity for their depositors by allowing them to easily invest or
divest their savings. The next Section will rehearse an account of how
bank-created liquidity generates valuable gains in market efficiency. At
this stage, however, it suffices to observe that when a financial
intermediary fails, there is a tradeoff between the typical mechanisms
of bankruptcy law-staying or clawing back creditors’ recoveries-and
the failed firm’s role in supplying customers with immediate access
to funds. For this reason, financial intermediaries have historically
been excluded from the mandatory provisions of the Bankruptcy Code. (47)
These same considerations, I will argue, justify the bankruptcy
exemptions for transactions in the repo and derivatives markets.
Ironically, the argument for excluding certain financial contracts from
conventional bankruptcy mechanisms has much in common with the argument
for bankruptcy itself. Both are rooted in the central observations of
institutional economics that information or transaction costs may limit
the efficiency of atomistic financial markets. (48)
A. Banking as Liquidity Creation
An efficient financial system should allocate society’s scarce
savings to all projects with expected returns exceeding their cost of
capital. (49) However, due to information and transaction costs, most
investments incur some amount of liquidity risk: an investor may be
unable to sell her asset without loss when she experiences a need for
cash. Relatively few assets are financed with liquid claims such as
common stock or bonds. Most, such as homes or businesses, can be sold
only after incurring substantial search costs, due diligence, and other
transaction or agency costs. (50) Thus, an investor in an illiquid asset
need not only evaluate the underlying investment opportunity: there is
also the risk that she may at some point be forced to sell the asset at
a discount to raise cash. (51) Even socially valuable projects may be
unable to attract funding for this reason. (52)
In response, banks supply liquidity by offering a deposit contract,
which allows households to invest their savings while retaining
on-demand access to funds. (53) This liquidation right performs two
liquidity-creating functions. (54) First, deposits may be withdrawn at
any time notwithstanding the long maturities of the bank loans that they
finance; this function is known as maturity transformation. (55) Second,
bank deposits are information-insensitive, meaning that the depositor
need not monitor her bank account as vigilantly as she might monitor a
risky investment. (56) This is because her deposit represents a share of
the bank’s entire portfolio, which is highly diversified across
borrowers; (57) because the bank’s shareholders absorb all
portfolio losses to the extent of their equity; (58) and because
government deposit insurance guarantees against residual default risk.
(59) The result is to allow investments that may be prohibitively
illiquid for some investors, such as homes and small businesses, to
attract financing in a form that is almost as liquid as cash. (60)
Banking thus plays an essential role in any market economy. The
value of bank-supplied liquidity can be seen, for example, in the
extremely low yields that demand deposits pay. The difference between
deposit rates and the higher interest rates paid on other debt contracts
largely reflects the liquidity premium that depositors are willing to
incur for the greater liquidity of a bank deposit. (61) But a more
important clue is the “overwhelming proportion” (62) of
capital that is intermediated through banks: “For centuries, the
vast majority of externally financed investments have been funded by
banks, for which demandable-debt instruments (bank notes and checking
accounts) have been the principal source of funds.” (63) However
elusive conceptually, the potential value created by financial contracts
that supply similar on-demand liquidity should not be ignored in
scholarship on the shadow banking system.
B. Implications for Bank Resolution
Banking creates value by shifting liquidity risk from depositors to
the bank itself.” banks offer a kind of liquidity insurance to
their depositors. (64) However, the underlying assets on the bank’s
balance sheet remain illiquid–that is, they can be sold only at a
discount. For this reason, a bank encountering a surge in withdrawals
may itself run short of cash, because it cannot repay its depositors at
par by liquidating portfolio assets. (65) Moreover, individual bank
failures tend to spread to other banks, (66) triggering a deleveraging
of the banking system and a contraction in economic activity. (67)
Banks’ inherent fragility and their susceptibility to
contagion effects generate a compelling need to protect them from the
demands of panicked depositors. However, if the value of bank deposits
is closely related to their liquidity–specifically, the
depositor’s right to withdraw or transfer her account balance on
demand–then the ordinary bankruptcy protections built around automatic
stays have, at best, problematic application to the resolution of failed
Ex ante, the risk that an investor’s savings will be frozen in
a failed bank makes the deposit contract less attractive, (68) and would
likely reduce the supply of capital made available to the banking system
generally. Perhaps more importantly, this risk would destroy the
information-insensitive quality of bank deposits by forcing depositors
to monitor the solvency of their bank. Credit conditions would then tend
to tighten in response to rising risk perceptions. (69) In extreme
cases, a panic can ensue when
a shock occurs that is large enough for bank debt to become information sensitive. It loses its important feature, and so agents do not want it anymore; they want an asset which is surely information insensitive--cash. When that happens, the banking system cannot honor the demands and is insolvent. (70)
Historically, swings in confidence in the banking system have had
sharply pro-cyclical effects, as the resulting pullback in bank lending
worsens underlying economic conditions. (71) Ex post, a resolution
mechanism that freezes a failed bank’s contracts may badly disrupt
the activities of the households and investors that depend on them. (72)
These factors explain an important aspect of modern approaches to
bank resolution: rather than stymie depositors’ recovery efforts,
the law has historically sought to shift losses away from banks’
depositors and noteholders to other, less risk-averse stakeholders. As
Bray Hammond wrote in his landmark history, in the antebellum period the
view emerged “that the obligations of banks were not ordinary debts
but money; and that a public interest was at stake in them which
overrode that of any particular debtor and creditor.” (73) Thus,
historical bank-insolvency regimes have looked to four categories of
stakeholders to bear losses in lieu of banks’ customers: (1) bank
shareholders or insiders; (74) (2) other banks; (75) (3) taxpayers; (76)
and (4) the central bank. (77) Under present-day FDIC receivership
procedures, the experience of even uninsured depositors “is far
better than could be expected under general corporate bankruptcy where
most payments to creditors are usually delayed until final
resolution.” (78) The current approach, which is highly protective
of depositors, has enjoyed substantial success in reducing systemic
risk: following the Great Depression, the banking system enjoyed “a
panic-free period of 75 years–considerably longer than any such period
since the founding of our republic.” (79)
The foregoing suggests two salient observations. First, the
liquidation rights built into the deposit contract supply the vector for
bank runs-withdrawal of deposits–but also the liquidity that makes
banking valuable, (80) Imposing bankruptcy-style restrictions on
depositors’ accounts would render the banking system less efficient
and more prone to panics. Second, bank insolvency law should prevent
runs on weak banks without impairing depositors’ rights, and can do
so by shifting losses to other stakeholders. It should be stressed that
such approaches do not eliminate risk, but shift it, so that one set of
claims on the bank (the equity) is risky and information-sensitive,
while another set of claims (the deposits) is information-insensitive
and can function as a form of money. (81)
These design principles explain important and surprisingly enduring
features of American insolvency law. In contrast to the pro-debtor
bankruptcy model advocated by critics of the safe harbors, the law
governing the banking sector has traditionally protected the enforcement
of liquidation rights against their issuers, while employing other
methods, including deposit insurance, to maintain system stability. This
approach reflects the valuable role that banks’ creation of safe,
money-like instruments plays when real assets are imperfectly liquid.
II. THE STRUCTURE OF SHADOW BANKING
While the banking system has exhibited remarkable stability since
the advent of federal deposit insurance, the cap on insured balances
(currently $250,000) has forced asset managers, governments, and
corporations to look elsewhere for “safe, interest-earning,
short-term investments” akin to an insured deposit. (82) This
demand has been met by an array of nonbank financial companies and
markets that comprise a kind of parallel banking system for
institutional investors. (83) While a comprehensive discussion is well
beyond the scope of this Note, this Part will describe two transactions,
securitization and repo, that lay at the center of the shadow banking
system and its 2008 crisis.
A. Securitization: Bank Lending Unbundled
Part I argued that banks create value by transforming illiquid
assets into liquid deposits. Thus “the existence of financial
intermediaries implies the creation of bank loans that banks should be
unable to sell [to outside investors].” (84) However, the advent of
securitization–the packaging of bank loans into tradable debt
securities–overcame this constraint by allowing banks to market their
assets to a wide array of institutional investors beyond a bank’s
depositor base. Traditionally, banks have intermediated credit by
originating loans and taking deposits, which appear on their balance
sheets as assets and liabilities, respectively. (85) In a
securitization, by contrast, the bank sells pools of bank loans (such as
mortgages) to a special-purpose vehicle, which finances the purchase by
selling investors claims on the pool; these claims are known as
asset-backed securities (ABS). (86) In this way, securitization blurs
the line between banking and the capital markets by allowing large
outside investors to finance loans originated by a bank.
The origins of securitization date to 1968, when the Government
National Mortgage Association (Ginnie Mae) began issuing securities
backed by federally guaranteed mortgages. The Federal National Mortgage
Association (Fannie Mae) began offering mortgage-backed securities (MBS)
in 1981. (87) Because the loans underlying these deals (known as
“agency” MBS) are guaranteed by the government, investors in
these securities enjoy a backstop analogous to FDIC deposit insurance.
However, the following decades saw securitization spread quickly to
non-agency-backed (“private-label”) mortgages, and to other
types of consumer and business debt. (88) In lieu of a government
guarantee, these transactions employ other methods to protect investors
from credit risk in the loan pool. For example, ABS are divided into a
series of “tranches” that absorb losses in reverse order of
seniority. The most junior tranche is the first to take losses, while
the senior tranche is paid first and is consequently the safest. (89)
The use of tranches gives ABS a capital structure analogous to that of
traditional banks, in which stockholders absorb losses to the extent of
their equity before any uninsured depositors are impaired.
Asset-backed securities, including subprime MBS, “became
subject to explosive demand from investors around the world” in the
years before the crisis. (90) By 2005, issuance of securitized bonds
exceeded corporate bond issuance in the United States, “even
excluding mortgage-related securitization.” (91) According to the
Bank of England, the global ABS market reached $10.7 trillion at the end
of 2006. (92) By 2008, hedge funds and investment banks had greater
combined exposure to subprime mortgages than the commercial banks that
had originated these loans. (93)
Securitization thus provided a way for institutional investors to
finance bank loans that had previously been funded by bank deposits.
(94) But rather than buy securitized bonds outright, many investors
“deposited” funds in short-term loans, including asset-backed
commercial paper (ABCP) and repos, which used securitized assets as
collateral. While the ABCP market reached $1.2 trillion at its peak, our
focus is on the repo market, which may have reached $10 trillion,
roughly the same size as the total assets of the U.S. commercial banking
sector. (95) Safe and highly liquid, repos furnished the
“deposits” of the shadow banking system.
B. Repos: Shadow Bank “Deposits”
Like bank deposits, repos are short-term debt instruments that are
designed to be highly liquid and insulated from credit risk. (96) In
substance, a repurchase agreement is a short-term secured loan,
typically made by a cash-rich investor, which takes securities such as
Treasuries, MBS, or other debt securities as collateral. (97) The lender
extends credit by “purchasing” the collateral from the
borrower, which agrees to “repurchase” the collateral (perhaps
the next day (98)) at a small premium over the purchase price. (99) Gary
Repos are like demand deposits. One party deposits (lends) money in a bank, usually overnight, and will receive interest. To make the deposit safe, the depositor is provided with collateral in the form of a bond.... If the bank fails, then the institutional investor can sell the bond, without going into a bankruptcy procedure.... The institutional investor can always withdraw the money, so to speak, by not rolling the repo. (100)
Consequently, the lender should incur only negligible liquidity
risk to the extent that the borrower is solvent or the collateral is
relatively liquid. Repos are not just functionally analogous to
deposits: the law also treats repos as money-like reserve assets rather
than as risky debt securities. (101)
Repos’ predominant use for much of the twentieth century was
to finance Treasury securities, (102) but since the 1980s they have
become a key source of day-to-day funding for financial institutions and
an important vehicle for idle cash held by corporations, governments,
and asset managers. (103) Repos ultimately became an important method of
financing the securitization process prior to the 2008 crisis. (104)
Shadow banks such as Goldman Sachs and Bear Stearns borrowed extensively
in the repo market, offering their structured-finance portfolios as
collateral. (105) Thus investors “depositing” funds in the
repo market financed asset-backed securities based on bank loans. In
this way, shadow banking “open[ed] up potentially new sources of
funding for the banking system by tapping new creditors.” (106)
III. BANKRUPTCY-PROOFING SHADOW BANKING
As Part I showed, the liquidity of a traditional bank deposit
depends on two basic features that have changed little since the 1930s:
the withdrawal right and deposit insurance. Together, these devices
render bank deposits highly liquid and information-insensitive: a
depositor is guaranteed substantially uninterrupted access to her
savings, even if her bank fails. Not only does this protection increase
the attractiveness of banking to prospective depositors, but it also
appears to have eliminated the threat of panics in the traditional
banking sector, which occurred with alarming frequency before the advent
of the FDIC. (107)
The development of similarly liquid, information-insensitive
instruments in the shadow banking system has proven to be a far more
intricate–and in many ways unfinished–problem. First, as the previous
Parts have shown, the marriage of asset securitization and repo-based
financing brings together a complex chain of financial companies and
off-balance-sheet vehicles, each of which introduces an increment of
counterparty risk. Second, unlike insured deposits, the markets for
private-label ABS and repos lack an explicit government guarantee. That
is, the shadow banking system must mitigate risks affecting not only the
ultimate borrowers (such as subprime homebuyers), but also the various
entities involved in the intermediation process. Thus the development of
low-risk shadow bank “deposits” has required extensive
contractual innovation and legal accommodation, most importantly within
applicable insolvency law.
A. Securitization and Bankruptcy-Remoteness
To transform illiquid bank loans into highly rated securities, a
securitization must ensure that an investor has exposure only to risks
affecting the loans themselves–not the creditworthiness of the sponsor.
To accomplish this, a sponsor moves the loan pool off its balance sheet
by “selling” it to a special-purpose vehicle, which finances
the purchase by issuing ABS. This structure ensures that the securitized
loans remain segregated from the prospective bankruptcy estate of the
sponsor. For example, MBS backed by mortgages originated by Countrywide
Financial would have been unimpaired even if Countrywide had gone
bankrupt in early 2008. Additionally, the transaction documents in a
securitization are written to exclude events of default that would place
an ABS issuer itself in bankruptcy. For example, a missed coupon payment
or other significant breach triggers an accelerated repayment schedule
(“early amortization”) rather than a court-supervised
Together, these contractual devices ensure that ABS deals are
“bankruptcy-remote”– that is, potential buyers can be assured
that amounts owed under the transaction will not be tied up in the
unanticipated bankruptcy of the sponsoring bank or the issuer. In so
doing, they ensure that investors need not monitor the financial
condition of the other participants in the securitization process; they
also allow ABS to carry a higher credit rating than the other entities
involved in the deal. Thomas Plank explains that “[s]ecuritization
reduces the bankruptcy tax … and therefore has reduced the bankruptcy
premiums charged to the obligors of mortgage loans and other
B. The Repo Safe Harbors
The relevance of bankruptcy is not limited to the securitization
process. Like bank depositors, repo investors require protection from
the risk that the borrower will fail. In particular, they must be able
to sell the collateral in their possession if the borrower defaults on
its repurchase obligation. The investor’s rights in the collateral
are akin to the benefit of deposit insurance; they should guarantee
uninterrupted access to cash even if the repo borrower becomes
insolvent. (110) This is why repos, like securitizations, are structured
as “true sale” transactions rather than as secured debt. Just
as securitized loans are “sold” to an off-balance-sheet
vehicle to separate them from the sponsor’s bankruptcy estate, repo
collateral is “sold” to the repo investor so that the investor
may sell the assets if the borrower breaches its obligations under the
A repo investor’s liquidation rights in the collateral are
thus central to the deposit-like characteristics of the transaction.
Realizing these rights in practice, however, has involved a decades-long
evolution of repo-market institutions, (111) which, as the 2008
financial crisis showed, remains far from complete. (112) To appreciate
the importance of these rights, it is helpful to recall the tumultuous
path of the repo market during the early 1980s, when repos lacked
protection under the Bankruptcy Code and other institutional problems
limited the security of repo collateral. As a result of these issues, a
string of failures by government-securities dealers left many repo
investors holding illiquid, impaired claims on transactions that were
supposed to be liquid and relatively riskless. These ordeals underscored
the need to ensure the effectiveness of each party’s liquidation
rights in the event of its counterparty’s failure.
An early, dramatic example was the 1982 failure of Drysdale
Government Securities, which had lost money shorting Treasury bonds via
reverse repos. (A reverse repo is simply a repo that the cash lender
enters with the motive of borrowing securities. (113)) Consistent with
existing custom, Drysdale’s counterparties had taken cash equal to
the market value of the securities Drysdale had borrowed, with no
provision for the interest accruing during the term of the contract.
(114) This meant that when Drysdale failed, its counterparties held too
little cash to replace the securities they had lent. Concern arose that
Drysdale’s “failure to pay interest on these borrowings could
leave the dealers short of funds to meet their own obligations” and
could cause other dealers to fail as well. (115)
The Drysdale affair sparked a crisis of confidence in the repo
market. (116) Fearing the fallout, (117) “the Fed sharply expanded
the volume of government securities it lends temporarily to dealers who
need them to complete transactions” (118) and made an extraordinary
announcement that it “‘stood ready as lender of last
resort’ to help commercial banks meet ‘unusual credit demands
related to market problems.'” (119) The ensuing flight from
the repo market depressed demand for Treasury securities as market
participants reassessed the stability of their counterparties and firms
shifted away from short-maturity financing. (120)
The failure of E.S.M. Government Securities several years later
would deal a similar blow to market confidence. Like Drysdale, E.S.M.
had borrowed securities worth far more than the cash given to its
securities lenders. (121) Thus, E.S.M.’s collapse cast a pall over
the solvency of its creditors, which included numerous municipalities.
(122) And it precipitated a run on Ohio’s thrift industry when it
became clear that the Cincinnati-based Home State Savings Bank had some
$600 million in repo exposure to the bankrupt dealer–an amount
exceeding the entire assets of Ohio’s deposit guarantee fund. (123)
The resulting insolvency of that fund forced the Governor to shutter
Ohio’s seventy-one state-chartered thrifts. (124) News of the chain
of failures rocked international markets, causing the dollar to suffer
its biggest one-day plunge in fifteen years. (125)
The Drysdale and E.S.M. failures highlighted the need to
effectively collateralize dealings in the repo market. Yet the uncertain
treatment of repos under the Bankruptcy Code prior to 1984 raised a
parallel concern about liquidity risk: even adequately collateralized
creditors could face problems if they became caught in the bankruptcy
proceeding of a failed borrower. This concern gripped the markets after
the 1982 failure of Lombard-Wall, a small government securities dealer,
(126) and lingered for months afterward. (127)
Following its bankruptcy filing, Lombard-Wall had argued that the
automatic stay barred moves by its counterparties to exit their repo
positions, (128) and the court temporarily froze hundreds of millions of
dollars in repo collateral. (129) Lombard-Wall’s failure did not
pose credit exposure concerns comparable to those raised by Drysdale
because its repo obligations were adequately collateralized, (130) but
the automatic stay left Lombard’s creditors unable to sell the
collateral and badly short of liquidity. (131) Government agencies
affected by Lombard’s bankruptcy threatened to default on their
bonds, (132) and a major money-market mutual fund warned “that its
holders might panic and sell their shares.” (133) Hoping “to
keep the wheels moving,” the bankruptcy judge later granted partial
relief from the stay for many repo lenders, (134) but he later
recharacterized some of Lombard’s repos as secured loans subject to
the automatic stay. (135) Thus, the Lombard-Wall bankruptcy frustrated
the repo market’s expectation that participants would enjoy
immediate recourse to the cash (or securities) in their possession
following a counterparty’s default. The episode “severely
dislocated [the] financial markets,” (136) and accelerated
investors’ post-Drysdale flight from the repo market. (137)
In reaction to the Lombard-Wall bankruptcy, the Federal Reserve
urged Congress to exempt certain repurchase agreements from the
operation of the Bankruptcy Code (138) Congress responded in 1984 with
Code amendments exempting many repo transactions from the automatic stay
and the trustee’s avoiding powers. (139) Additionally, the law
established a new, open-ended exemption for parties’ contractual
rights to unwind a repurchase agreement, modeled after earlier
carve-outs for securities and commodity contracts (140) Forming a direct
rebuke to the Lombard-Wall ruling, the 1984 amendments sought to
eliminate the Bankruptcy Code as an obstacle to repo participants’
access to liquidity in the event of a future dealer failure.
IV. ASSESSING THE REPO SAFE HARBORS
Together, asset securitization and improvements in wholesale
funding techniques allowed institutional investors to finance bank
lending through a structure with liquidity and flexibility approaching
that of a traditional bank deposit. However, the resulting drastic
expansion of credit availability seemingly spurred on the improvident
“search for yield” that brought mortgage lending to
increasingly risky borrowers in the last years of the housing bubble.
Hyun Song Shin invokes an image
of an inflating balloon which fills up with new assets. As the balloon expands, the banks ... look for borrowers they can lend to. However, once they have exhausted all the good borrowers, they need to scour for other borrowers--even subprime ones. The seeds of the subsequent downturn in the credit cycle are thus sown. (141)
Moreover, the collateral-based safeguards created to protect shadow
bank “depositors” could not ensure the stability of the entire
system. (142) As Jeremy Stein explains, “one of the most damaging
aspects of the crisis was not just the problems of … big firms”
such as Bear Stearns, Lehman Brothers, and AIG, “but also the
collapse of an entire market, namely the market for asset-backed
The collapse of Bear Stearns in March 2008 provides an instructive
stress test of the repo market and the applicable bankruptcy safe
harbors. With mounting mortgage-related losses and a limited capital
cushion, by the end of 2007 Bear Stearns was clearly the shakiest of the
large investment banks. (144) Increasingly unable to borrow on an
unsecured basis, Bear Stearns turned to the repo market to replace its
commercial paper funding. (145) In early 2008, Bear Stearns’s
counterparties began to demand additional collateral or to refuse
exposure to the ailing investment bank. By March 13, it was clear that
the firm could not raise enough cash the following day to continue
operating, and Bear Stearns avoided bankruptcy only by a Federal
Reserve-backed merger with J.P. Morgan. (146)
For critics of the safe harbors, the collapse of Bear Stearns and
other firms symbolized the dangers of leaving a systemically important
financial institution without access to bankruptcy protection against
its repo and derivative counterparties. Mark Roe argues that the safe
harbors allowed counterparties to disregard Bear Stearns’s
weakening risk profile and encouraged the firm to become overly reliant
on a funding method that could dry up virtually overnight. (147) David
Skeel and Thomas Jackson add that the bailout of Bear Stearns suggested
that regulators were concerned that the disposal of Bear Stearns’s
collateral outside the bankruptcy process “could drive down the
values of mortgage-related securities and further destabilize the
markets.” (148) This fear, they argue, “suggests that the very
[bankruptcy] exclusions that were justified as reducing systemic
risk–allowing counterparties to terminate (and sell collateral)
notwithstanding the automatic stay–can actually exacerbate it through
the very sale of that collateral.” (149)
These criticisms have motivated a number of proposals to curtail
the bankruptcy safe harbors for repurchase agreements. Roe has proposed
that the filing of a bankruptcy petition should stay repo creditors from
demanding and liquidating collateral. (150) Skeel and Jackson’s
gentler call for “transaction consistency” would allow repo
creditors to immediately liquidate the most cash-like collateral, such
as Treasury securities or agency debt, without court approval, but would
impose an automatic stay in the case of “other, more opaque, forms
of collateral.” (151) Yet these proposals misdiagnose the source of
the problems affecting the repo market in 2008, and they risk
undermining the efficiency and stability of the financial system
The view that the safe harbors created attractive
“incentives” (152) for financial institutions to rely on
overnight funding misconstrues the financing choices facing Bear Stearns
and other repo borrowers at the height of the financial crisis. It was
not by choice that Bear Stearns was dependent on $102 billion in repo
borrowings by the end of 2007. Instead, the commercial paper
market’s growing aversion to Bear Stearns forced the bank to fall
back on repos as the safest debt structure it could offer its creditors.
(153) Put differently, the problem was not that Bear Stearns was
excessively dependent on repos at the market’s peak, but that Bear
Stearns’s other funding sources (principally short-term commercial
paper) became unavailable at the height of the firm’s troubles and
could not be replaced except by repos. Far from a source of volatility,
then, repos were in fact a lifeline to a firm suffering from high
leverage, a toxic balance sheet, and growing problems raising funds in
the commercial paper market.
On this analysis, a world without the repo safe harbors would
likely have left Bear Stearns in an even more precarious position.
Without the ability to pledge collateral to its lenders free of
bankruptcy risk, the firm’s options would have been limited to (1)
issuing commercial paper, which it was unable to do by late 2007, or (2)
pledging even more collateral, to compensate its repo lenders for the
growing risk that they would be caught in a Lombard-Wall-style
bankruptcy. Referring to traditional banks, economist Franklin Allen
Raising new capital is problematic when a bank is beset with difficulties. The bank is effectively suffering from a debt overhang. Suppliers of capital will know that in the event of default their money will go to the depositors and other creditors and so will be unwilling to supply it. (154)
This problem came to the fore during the financial turmoil of the
early 1980s, when investors not yet protected by the safe harbors chose
to withdraw from the repo market rather than risk tying up funds in
costly bankruptcy proceedings. (155) As a Wall Street professional told
the Wall Street Journal in the wake of one securities dealer’s
failure in the early 1980s, “There are hundreds of [repo]
transactions out there that look safe until one participant goes
These risks were also realized during the 2008 crisis when the
market for commercial paper–an analogous market without safe-harbor
protection–seized up in response to losses taken in the Lehman Brothers
bankruptcy. Nearly $2 trillion in commercial paper was outstanding at
the beginning of 2007, the vast majority of which was issued by
financial services firms. (157) Like repo, commercial paper is a form of
short-term debt that must be rolled over frequently. When Lehman’s
bankruptcy filing dealt losses to the Reserve Primary Fund, a large
money-market fund, in September 2008, the news “triggered the
modern-day equivalent of a bank run.” (158) Money funds faced $172
billion worth of redemptions over the next three days, and the flight of
capital abated only when the federal government announced that it would
guarantee all money fund shares. Still, the industry massively reduced
its holdings of commercial paper, whose total outstanding value fell by
15 percent within a month. To stem the decline, the Federal Reserve made
the extraordinary decision to begin purchasing commercial paper
directly; by early January 2009, it held on its balance sheet 22.4
percent of the commercial paper market. (159)
Like the episodes that gripped the repo market before the enactment
of the safe harbors in 1984, the panic in the commercial paper market
illustrated the dangers of injecting bankruptcy risk into the
instruments used as deposits by the shadow banking system. (160) As
collateralized lending instruments, repos mitigate this problem by
allowing distressed firms to issue a form of debt with a risk profile
tied primarily to the collateral rather than to the bankruptcy risk of
the borrower. (161)
Critics of the safe harbors correctly note that the mass
liquidation of repo collateral following the failure of a major
institution might toil asset markets, as regulators apparently feared
when they decided to bail out Bear Stearns. Even this insight, however,
risks confusing cause and effect. If the financial markets were
incapable of absorbing a massive sell-off of Bear Stearns’s assets
in March 2008, it was because liquidity had vanished from every segment
of the credit markets: “[M]oney, corporate debt, securitization,
[and] collateralized debt obligations (CDOs) … ground to a halt.”
(162) Even solvent firms were conducting fire sales in every asset class
in a desperate bid to hoard cash and retire debt. (163)
Selling off the assets of a major institution could easily
exacerbate these extreme conditions. However, the critics’ proposal
to stay the liquidation of counterparty collateral could hardly improve
matters. Instead, it would merely replace one form of contagion mediated
by asset markets with another form, mediated by the impact of an
automatic stay applicable to billions of dollars in financial contracts
on a distressed and highly interconnected market. It could also make
crises more likely by encouraging counterparties to rush for the exits
at the first signs of bankruptcy risk.
Like withdrawing bank deposits, liquidating collateral may be
benign under normal conditions yet damaging in the midst of a crisis; as
Gorton argues, “Debt during crises is not the debt of noncrisis
times.” (164) But the distress facing asset markets during the 2008
crisis suggests a need for regulators to respond to the crisis, not for
Congress to overturn, after the fact, the legal foundations of the
credit markets. Indeed, timely interventions such as the Term Securities
Lending Facility were “uniquely effective” in removing toxic
assets from shadow banks’ balance sheets and allowing them to
return to the debt markets with high-quality collateral. (165) And
forward-looking changes, such as minimum liquidity standards
contemplated by the Dodd-Frank Act, (166) offer regulators more targeted
responses to funding-market fragility than do proposals to subject the
repo market to bankruptcy risk.
Skeel and Jackson suggest that using a bankruptcy stay to prevent
repo lenders from selling their collateral could give regulators
breathing room to achieve an orderly disposition of these assets. (167)
Yet this is precisely the approach already taken by the Federal Deposit
Insurance Act (FDIA) and the Dodd-Frank Act’s Orderly Liquidation
Authority, which provide for a one-day stay on the closeout of
derivatives, repos, and other safe-harbored contracts pending their
transfer to a third-party acquirer or bridge institution. (168) This
gives regulators a window to rescue a troubled firm without
overburdening its counterparties with an indefinite freeze of their
rights in the collateral. Proposals for a lengthier stay, in contrast,
risk undermining the goal of creating a deposit-like instrument that
offers continuous liquidity to investors and a lifeline for troubled
Skeel and Jackson also suggest limiting the safe harbors to the
most liquid (“cash-like”) collateral, such as Treasury
securities, since the liquidation of these assets is least likely to
generate fire-sale conditions. (169) However, this proposal would deny
the benefits of the safe harbors to the firms and asset classes that
most need them. Firms depleted of high-quality assets would be unable to
offer their remaining assets as bankruptcy-proof collateral to repo
creditors, and would likely fail. At the same time, illiquid assets
could see their value plunge during crises, since without a bankruptcy
safe harbor they would be useless to repo borrowers. Thus, even a
partial rollback of the safe harbors could limit the efficiency of
shadow banking without meaningfully ameliorating fire-sale risks.
The repo safe harbors allow contracting parties to protect their
liquidation rights from the hazards of a formal bankruptcy process. To
be sure, this self-help regime cannot maintain the stability of the
financial system as a whole: when liquidity is scarce, one party’s
efforts to seize and sell collateral can negatively affect the condition
of other market participants. However, repealing the safe harbors would
have the perverse effect of exposing entire markets to instability by
leaving parties without any means of contracting around bankruptcy risk.
Moreover, if shadow banks could not offer a deposit-like liability to
investors, then institutional investors would be forced to look to
other, likely less efficient financing methods. (170)
V. SYNTHETIC SECURITIZATION AND THE DERIVATIVE SAFE HARBORS
Repurchase agreements are not the only instruments with liquidation
rights exempted from formal bankruptcy procedures. Derivatives–a much
larger, (171) more diverse, and more complex class of contracts–also
frequently permit parties to unilaterally terminate and liquidate their
dealings outside of bankruptcy, ahead of other creditors. These
instruments, which allow parties to take positions on a
“reference” asset (such as a corporate bond) or indicator
(such as the London Interbank Offered Rate, or LIBOR) without investing
in the reference entity itself, have an enormous array of applications
that are well beyond the scope of this Note.
However, prior to the 2008 mortgage crisis, these contracts found
prominent use in constructing “synthetic” mortgage-related
products that grew to represent a large share of the assets traded in
the shadow banking system. The bankruptcy safe harbors for derivatives
were thus important to ensure that investors who had purchased these
assets (or taken them as collateral) were unimpaired by the failure of
the party with payment obligations under the contracts. In this limited
but important context, this Part argues that the rationale for the repo
safe harbors also applies to the safe harbors for derivatives used to
construct synthetic financial products. Whether the derivative safe
harbors are justified with respect to the far wider universe of uses to
which derivative contracts can be applied is a more challenging issue,
but this Part will conclude by suggesting that they are.
A. Credit-Risk Transfer in Shadow Banking
Part II of this Note described traditional “cash” or
“cash-flow” securitizations, in which loans are sold to an
off-balance-sheet vehicle that issues securities representing claims on
the pool. Frequently, the ABS are themselves pooled and structured into
CDOs. These financial products were often financed by repo lenders,
which took these assets as collateral for a form of short-term lending
analogous to a bank deposit. In this way, “the banking system …
developed a method by which it could focus on generating assets while at
the same time getting these funded by the capital markets.” (172)
Synthetic securitization recapitulates this process but with an
important difference: the sponsor need not transfer the underlying
assets to the issuer. Instead, the sponsor executes a credit derivative,
such as a credit default swap or a credit-linked note, that transfers
only the credit risk affecting the asset pool to the investors in the
structure. (173) Synthetic transactions proved attractive as a way for
“financial institutions to pass their unwanted credit risks on to
the capital markets” in a cheaper, more flexible manner than could
be achieved by securitizing whole loans. (174) The synthetic CDO market
grew rapidly following its emergence in 1997, accounting for a large
share of total CDO issuance and becoming a major component of the credit
derivative market. (175)
Because cash flows in a synthetic CDO are based on an agreement
between the parties, rather than derived from the reference portfolio
itself, parties to a synthetic CDO incur counterparty risk: the risk
that another party will default on payment obligations. (176) Thus, as
in traditional “cash” securitizations, there is a need to make
the structure bankruptcy-remote, so that the resulting securities are
linked solely to the performance of the reference portfolio, rather than
to the credit risk of the sponsor. (177)
Accordingly, the derivatives market and the applicable law have
responded in a manner similar to the repo market: the trend has been
toward use of collateral to eliminate counterparty risk, (178)
accommodated by the enactment of safe harbors to ensure the
effectiveness of the parties’ liquidation rights. As with repos,
“in most cases collateral posted against derivatives positions is
under the control of the counterparty and may be liquidated immediately
upon a covered ‘event of default.'” (179) Additional
liquidation rights embedded in derivative contracts allow a party to
terminate and net out payment obligations under outstanding transactions
if the counterparty defaults. This is known as “closeout
A simple example will illustrate the importance of these rights to
the parties to a derivative contract. Consider a typical interest rate
swap, which might require Party A to periodically pay Party B $5 million
based on a floating interest rate tied to LIBOR, in return for $4
million according to a fixed rate. These obligations, which sum to a
periodic $1 million payment from A, represent a valuable asset for B;
thus the swap might have a mark-to-market value of, say, $10 million.
Yet B may also have countervailing obligations to A under other
contracts that reduce A’s net liability to B. In this context, the
sudden insolvency of B poses potential problems for A that may be
ameliorated by appropriate liquidation rights.
First, B’s insolvency could leave A locked into a position
with no potential upside. While the swap is currently “out of the
money” for A, which owes B a payment stream valued at $10 million,
an insolvent B would be unable to perform any future obligations to A
that would arise if interest rates were to swing in A’s favor.
A’s right to terminate the transaction protects it from this risk.
Second, B’s insolvency could substantially increase A’s
total exposure to B by impairing the value of B’s countervailing
obligations to A. Assume that A and B have other outstanding derivative
contracts, collectively worth $10 million, which are “in the
money” for A and thus perfectly offset A’s liability under the
interest rate swap. IrA and B terminated all their dealings, the safe
harbor for closeout netting would allow them to net their offsetting
exposures to zero, so that no termination payment would change hands.
(182) Without the safe harbor, A could be required to seek relief from
the automatic stay or file a bankruptcy claim for the value of B’s
offsetting liability. (183)
Third, if we assume instead that A’s net position is “in
the money,” any collateral available to A would limit its losses in
the event of B’s default. (184) Crucially, A’s ability to
promptly dispose of the collateral and replace the terminated contract
(instead of being hamstrung by the automatic stay) minimizes any
disruption to its hedging strategy arising from B’s default.
Following the template used for repos and other financial products,
Congress has written safe harbors into the Bankruptcy Code and the FDIA
in order to protect the enforceability of many liquidation rights
commonly embedded in derivatives. (185) In 1990, Congress provided a
safe harbor to closeout netting of swaps in a new Code section 560,
(186) and in provisions relating to the automatic stay and the
trustee’s avoiding powers. (187) Additionally, these protections
were given expansive applicability: swaps were defined as a laundry list
of agreements or “any other similar agreement,” (188) and
eligible “swap participant[s]” included any “entity that
… has an outstanding swap agreement with the debtor.” (189) The
Bankruptcy Abuse Prevention and Consumer Protection Act of 2005
represented a major further expansion of the safe harbors for derivative
In these ways, rights to terminate, net out, and liquidate
derivative contracts protect each party from the risk of counterparty
default by allowing each to exit the transaction unimpaired.
Significantly, these liquidation rights also allow synthetic
transactions to replicate the bankruptcy-remoteness of a traditional
“cash” securitization by limiting the impact of a
counterparty’s bankruptcy on the derivative contracts backing the
CDO. Just as the “true sale” structure of a cash
securitization ensures that the sponsor’s bankruptcy will not
affect an investor’s rights in the securitized assets, so the use
of collateral and the application of the safe harbors ensure that
amounts owed under a synthetic transaction are insulated from the
counterparty’s bankruptcy risk. For example, in a typical
“funded” transaction, the investors’ principal is
invested in safe, liquid collateral that secures any amounts owed to the
sponsor or the investors under the terms of the transaction. (191)
“Unfunded” transactions, in which no collateral is purchased
at the outset, may give investors the right to demand security based on
subsequent declines in the sponsor’s credit rating. (192)
These precautions enable the creation of highly liquid synthetic
financial products that can circulate through the shadow banking system
unimpaired by bankruptcy risk. For institutional investors, these
synthetic assets expand the collateral available for the creation of
shadow bank “deposits,” while for borrowers, they may expand
credit availability by allowing lenders to efficiently offload credit
risk to outside investors. (193)
B. Ongoing Challenges
While the expansive wording of the derivative exemptions aimed to
provide legal certainty to market participants, (194) the boundaries of
the safe harbors remain uncertain and have been extensively litigated in
the Lehman Brothers bankruptcy, indicating that legal risk related to
liquidation rights in the derivatives market remains far from resolved.
(195) This litigation provides a helpful context for investigating the
impact of a limitation on the safe harbors on the liquidity and
efficiency of derivatives used in the shadow banking system.
An important problem concerns the enforceability of provisions
giving priority over a collateral pool to the non-defaulting party in a
synthetic transaction. In a common structure, the investors purchase
notes linked to credit risk in the reference portfolio. The issuer of
the notes, a special-purpose vehicle, enters into a credit default swap
with the deal’s sponsor (e.g., Lehman Brothers), under which the
issuer incurs obligations based on credit losses in the reference
portfolio (“sells protection” to the sponsor). The proceeds of
the issue are used to purchase high-quality collateral that is held in
trust for the benefit of the sponsor to secure its exposure to the
issuer’s obligations under the swap.
Such transactions include an important liquidation right for the
investors to protect them from counterparty risk. (196) To prevent the
collateral from being funneled into the bankruptcy estate of the
sponsor, a “flip” clause reverses the parties’ priority
over the collateral, allowing the investors to recover their remaining
principal once the deal is terminated. (197) However, in two recent
rulings, the judge presiding over the Lehman Brothers bankruptcy ruled
that such clauses are unenforceable under the ipso facto provisions of
the Code, (198) holding that they do not fall within the safe harbor
protecting a party’s contractual right to liquidate, terminate, or
accelerate a swap agreement. (199) A potentially damaging consequence of
these rulings has been to reintroduce counterparty risk into the credit
ratings of synthetic transactions. (200) If investors in a synthetic CDO
lack protection from the sponsor’s bankruptcy, then the credit
rating applicable to a particular synthetic product can be no higher
than the rating of the sponsor. Investors in such a transaction are thus
in a position comparable to a depositor in an uninsured bank. The likely
impact will be to reduce the attractiveness of the synthetic CDO market
while making these instruments less resilient to volatile market
C. The Scope of the Rationale for the Derivative Safe Harbors
Thus far, this Part has focused on the limited use of credit
derivatives to create synthetic CDOs that may be used as collateral for
repos. However, because the bankruptcy safe harbors protect a far wider
universe of applications of derivative contracts, this Section suggests
a broader rationale for these provisions. To be sure, the heterogeneity
of the derivatives market resists general statements about the social
value of these contracts. Nevertheless, a set of general functional
considerations suggests a plausible justification for the exemption of
the broader derivatives market from the bankruptcy process.
The value of the derivatives market reflects the basic insight that
most financial instruments involve a wide variety of risks, not all of
which are efficiently borne by a given investor. It might be efficient,
for example, for a loan originator to lay off credit risk to outside
investors with greater credit-risk appetites in the form of a synthetic
CDO. Similarly, an investor in Tokyo real estate may find a willing
buyer for its exposure to the Japanese yen through a currency swap; a
municipality financing new infrastructure may seek protection from
volatile borrowing costs through an interest rate swap; a hedge fund may
use derivatives to gain greater exposure to risks unwanted by other
market participants. By allowing the distinct risks of a given
investment strategy to be decomposed and shifted to their most efficient
bearer, the derivatives market may facilitate the more efficient supply
of capital to valuable investment projects. (201) Part II argued that
depository banking improves the supply of capital by allowing households
to insure their investments against liquidity risk. The derivatives
market may provide an analogous facility to investors facing a more
diverse array of unwanted risks.
The next step in making a more general case for the derivative safe
harbors is to show that these forms of risk transfer could not be
conducted efficiently if market participants were exposed to the risk of
a counterparty’s bankruptcy. For example, a yen hedge with a risky
counterparty may be no more palatable to our Tokyo real estate investor
than the underlying yen risk. As a general matter, derivative contracts
that are not insulated from counterparty risk are less liquid and, to
that extent, less efficient than transactions that are liquid. For
example, an “unfunded” synthetic instrument (in which no
collateral is purchased at the outset) cannot be freely traded across
the capital markets because each counterparty must be freshly evaluated
for default risk. (202) These considerations suggest that counterparty
risk might frequently prove prohibitive in the context of an otherwise
efficient derivatives transaction.
To be sure, contractual liquidation rights protected by statutory
safe harbors are not the only strategy for managing counterparty risk. A
potentially important innovation in the Dodd-Frank Act is the imposition
of mandatory clearing on many swap agreements, (203) which would
interpose a central clearinghouse between the parties so that each party
faces the clearinghouse, rather than the other party, as its
counterparty. (204) By alleviating market participants’ need to
monitor and guard against the risk of default by their counterparties,
mandatory clearing may improve liquidity in the swaps market and
mitigate some of the core concerns that the derivative safe harbors were
designed to address. (205) However, traditional counterparty-risk
protections, including collateral and the safe harbors, will have
ongoing relevance for the vast segments of the derivatives market that
are not subjected to mandatory clearing under the Dodd-Frank Act.
Critics of the safe harbors have correctly noted that the use of
bankruptcy-proof liquidation rights to manage counterparty risk is not
without substantial costs. Moreover, these costs are largely borne not
by the contracting parties themselves, but by junior claimants (such as
a firm’s bondholders) who likely have little visibility into the
size of the firm’s liability to its derivative counterparties.
(206) Thus, for example, whether the benefits of the safe harbors for
Lehman Brothers’s heterogeneous derivative counterparties and the
wider derivatives market outweighed the costs imposed on its commercial
paper holders is inevitably an empirical question and a fruitful area
for research. Perhaps the most promising task for future scholarship
would be to specify more clearly the social costs and benefits of
derivatives trading, so that regulators can achieve the most compelling
benefits of bankruptcy-proof finance while aiming to minimize its most
The safe harbors have faced withering scholarly criticism following
the 2008 financial crisis. Critics argue that the safe harbors left
troubled firms without recourse to bankruptcy protection as their
counterparties exercised contractual rights to withdraw credit and seize
collateral. In addition to draining liquidity from weak institutions,
these self-help efforts seemingly roiled the broader financial markets
as parties liquidated collateral and replaced terminated contracts in
distressed markets. Scholars also claim that the safe harbors fueled the
pre-crisis expansion of the repo and derivative markets by according
them advantages vis-a-vis other contracts in bankruptcy. Accordingly,
there has been overwhelming agreement in the recent literature that the
safe harbors should be rolled back, leaving repo and derivative
counterparties on more equal footing with other bankruptcy claimants.
In contrast, this Note has argued that scholarly emphasis on the
actions of failed firms’ repo and derivative counterparties is akin
to blaming bank runs on depositors’ right to withdraw funds. That
is, it identifies the contractual vector that brought down Lehman
Brothers and other firms, but does not illuminate a promising path for
reform. Imposing post-petition limits on repo and derivative
counterparties’ liquidation rights could not have kept a troubled
firm like Lehman Brothers on its feet. (207) Worse, such proposals would
promote self reinforcing panics in these markets by exposing risk-averse
capital suppliers to impairment in a counterparty’s bankruptcy.
Critics of the safe harbors have overlooked the extent to which
bankruptcy-proof repos offered a lifeline to Bear Stearns, for example,
which had lost access to unsecured funding sources amid the
market’s flight from bankruptcy risk.
To be sure, the safe harbors alone do not furnish a comprehensive
framework for resolving large financial institutions, let alone for
ensuring their stability. The rescues of Bear Stearns and AIG in 2008,
like that of Long-Term Capital Management ten years earlier, put
regulators in the ironic position of bailing out institutions precisely
to prevent the disorderly unwinding of these firms through the recovery
efforts of their counterparties. (208) Yet in each case, regulators
chose not to prop up the bailed-out firm through a stay of its
obligations, but rather to supply emergency funds to ensure these
obligations were met. Thus these episodes are consonant with the basic
structure of modern bank regulation, which insures depositors rather
than staying their claims. (209) Criticism of the safe harbors’
role in these episodes is, at best, a distraction from more pressing
questions about how to regulate and structure emergency assistance to
More broadly, I have argued that the safe harbors play a valuable
role in enabling institutional investors to finance traditional bank
lending through the structured-finance and repo markets. If
deposit-taking banks have historically supplied the overwhelming share
of the economy’s external investment capital, (210) then scholars
should not underestimate the potential value of analogous contracts
issued by shadow banks. By aiding the flow of institutional investment
into traditional bank assets, liquidation rights embedded in repos and
derivative contracts help to transform vast reserves of “dead
capital” into “live capital” available to homebuyers,
entrepreneurs, and other borrowers. (211) To be sure, the mortgage
bubble and subsequent credit crisis underscore the potential
consequences of sudden transformations in the financial system. (212)
These opportunities and dangers should challenge lawmakers to manage the
flow of this capital based on principles drawn from our successful
experiments in bank regulation during the twentieth century.
(1.) Shadow banks financed hundreds of billions of subprime
mortgages in 2005 and 2006 alone. See Gary Gorton & Andrew Metrick,
Securitized Banking and the Run on Repo, 104 J. FIN. ECON. 425, 430
(2012); see also Douglas W. Diamond & Raghuram G. Rajan, The Credit
Crisis: Conjectures About Causes and Remedies, 99 AM. ECON. REV. 606,
606 (2009) (explaining how securitization attracted foreign investment
to the U.S. housing market).
(2.) Zoltan Pozsar et al., Shadow Banking, FED. RES. BANK OF N.Y. 5
& fig.1 (July 2010), http
(3.) For example, quarterly issuance of consumer asset-backed
securities (ABS) had fallen from an average of $50 billion to $70
billion in pre-crisis years to just $2 billion at the end of 2008.
Jeremy C. Stein, Securitization, Shadow Banking & Financial
Fragility, 139 DAEDALUS 41, 41 (2010).
(4.) Federal Reserve Chairman Ben Bernanke told the Financial
Crisis Inquiry Commission that of “the 13 … most important
financial institutions in the United States, 12 were at risk of failure
within a period of a week or two” in September 2008. FIN. CRISIS
INQUIRY COMM’N, THE FINANCIAL CRISIS INQUIRY REPORT: FINAL REPORT
OF THE NATIONAL COMMISSION ON THE CAUSES OF THE FINANCIAL AND ECONOMIC
CRISIS IN THE UNITED STATES 441, 479 (2011).
(5.) See GARY B. GORTON, MISUNDERSTANDING FINANCIAL CRISES: WHY WE
DON’T SEE THEM COMING 182 (2012) (“The global financial crisis
… was triggered by a bank run, just like those of 1837, 1857, 1873,
1893, 1907, and 1933.”). For a useful summary of this thesis, see
Bryan J. Noeth & Rajdeep Sengupta, Is Shadow Banking Really
Banking?, REGIONAL ECONOMIST, Oct. 2011, at 8,
(6.) “The shadow banking system has existed outside the
explicit banking safety net and, in most cases, with minimal regulatory
constraints.” Morgan Ricks, Regulating Money Creation After the
Crisis, 1 HARV. BUS. L. REV. 75, 87 (2011). In contrast, traditional
banks generally expanded their balance sheets even as the nonbank
financial sector deleveraged sharply during the crisis. See Zhiguo He et
al., Balance Sheet Adjustments During the 2008 Crisis, 58 IMF ECON. REV.
118, 120-21 (2010).
(7.) See GORTON, supra note 5, at 5 (“A financial crisis in
its pure form is an exit from bank debt…. Financial intermediaries
cannot possibly honor these short-term debt obligations if they are
withdrawn or not renewed.”); Ricks, supra note 6, at 84
(“[R]uns … occur when large numbers of funding providers with
near-term maturities decline to renew their contracts upon
(8.) See Tobias Adrian & Hyun Song Shin, Money, Liquidity, and
Monetary Policy, 99 AM. ECON. REV. 600, 600 (2009) (“At the margin,
all financial intermediaries (including commercial banks) have to borrow
in capital markets, since deposits are insufficiently responsive to
funding needs. But for a commercial bank, its large balance sheet masks
the effects of operating at the margin.”).
(9.) See Markus K. Brunnermeier, Deciphering the Liquidity and
Credit Crunch 2007-2008, 23 J. ECON. PERSV. 77, 78 (2009) (“[Blanks
increasingly financed their asset holdings with shorter maturity
instruments. This change left banks particularly exposed to a dry-up in
funding liquidity.”); Paul A. McCulley, Teton Reflections, GLOBAL
CENTRAL BANK FOCUS (PIMCO), Aug./Sept. 2007, at 2,
(“[U]nregulated shadow banks fund themselves with un-insured
commercial paper, which may or may not be backstopped by liquidity lines
from real banks. Thus, the shadow banking system is particularly
vulnerable to runs–commercial paper investors refusing to re-up when
their paper matures, leaving the shadow banks with a liquidity
(10.) This situation proved untenable. By September 2008, “all
the five of the largest independent investment banks had either closed
down (Lehman Brothers), merged into other entities (Bear Stearns and
Merrill Lynch), or converted to bank holding companies to be supervised
by the Federal Reserve (Goldman Sachs and Morgan Stanley).” FIN.
CRISIS INQUIRY COMM’N, supra note 4, at 154.
(11.) 11 U.S.C. [section] 109(b)(2) (2006).
(12.) See Robert R. Bliss & George G. Kaufman, U.S. Corporate
and Bank Insolvency Regimes: A Comparison and Evaluation, 2 VA. L. &
BUS. REV. 143, 157-58, 164 (2007).
(13.) 11 U.S.C. [section] 362(a) (2006).
(14.) 11 U.S.C. [section] 547(b) (2006).
(15.) “Thus, while most contracts … are automatically stayed
by courts in the event of a corporate bankruptcy, the opposite situation
obtains in the event of a bank’s insolvency.” Bliss &
Kaufman, supra note 12, at 158-59.
(16.) E.g., 11 U.S.C. [section] 362(b) (2006) (automatic stay); id.
[section] 546(e)-(g) (avoiding powers); id. [section] 548(d)(2)(B)-(E)
(fraudulent transfers); id. [section] 555 (general exemption for
securities contracts); id. [section] 556 (commodities or forward
contracts); id. [section] 559 (repos); id. [section] 560 (swap
agreements); id. [section] 561 (cross-product netting). These carve-outs
are reflected in analogous provisions of the Dodd-Frank Wall Street
Reform and Consumer Protection Act, [section] 210(c), Pub. L. No.
111-203, 124 Stat. 1376, 1477 (2010) (codified at 12 U.S.C.A. [section]
5390(c) (West 2012)), and the Federal Deposit Insurance Act, which uses
the term “qualified financial contract” to describe “any
securities contract, commodity contract, forward contract, repurchase
agreement, swap agreement, and any similar agreement that the [FDIC]
determines … to be a qualified financial contract” eligible for
certain exemptions from mandatory resolution procedures, 12 U.S.C.
[section] 1821(e) (8) (D) (i) (2006).
(17.) A repo is essentially a secured loan characterized as a
“sale” of collateral (usually securities) coupled with a
promise to buy back (“repurchase”) the collateral at the
transaction’s maturity for a small premium. In this way, a cash
lender receives both security and a promised rate of return on a
short-term loan. FIN. CRISIS INQUIRY COMM’N, supra note 4, at 31.
Bear Stearns’s repo borrowings rose from $69 billion to $102
billion during 2007 as it found itself locked out of the unsecured
commercial paper market. Id. at 283.
(18.) See OFFICE OF THE SPECIAL INSPECTOR GEN. FOR THE TROUBLED
ASSET RELIEF PROGRAM, FACTORS AFFECTING EFFORTS TO LIMIT PAYMENTS TO AIG
COUNTERPARTIES 3 (2009).
(19.) See supra note 16.
(20.) See Robert R. Bliss & George G. Kaufman, Derivatives and
Systemic Risk: Netting, Collateral, and Closeout 1 (Fed. Reserve Bank of
Chi., Working Paper No. 2005-03, 2005), http://ssrn.com/abstract=730648
(stressing “the ability of these contracts to net or setoff
offsetting positions between counterparties, to access collateral
promptly, and to close-out or terminate positions quickly without being
subject to prolonged legal stays”).
(21.) See, e.g., Shmuel Vasser, Derivatives in Bankruptcy, 60 Bus.
LAW. 1507, 1510 (2005) (locating the public rationale of the safe
harbors in the need “to protect American financial markets and
institutions from the ripple effects resulting from a bankruptcy filing
by a major participant in the financial markets”). “Since its
adoption in 1978, the Bankruptcy Code has been amended several times to
afford different treatment for certain financial transactions upon the
bankruptcy of a debtor … to further the policy goal of minimizing the
systemic risk potentially arising from certain interrelated financial
activities and markets.” H.R. REP. No. 105-688, pt. 1, at 2 (1998).
(22.) Gorton & Metrick, supra note 1, at 425.
(23.) See Tri-Party Repo Infrastructure Reform, FED. RES. BANK OF
N.Y. 13 (2010), http://www.newyorkfed.org/banking/nyfrb_triparty_whitepaper.pdf.
(24.) See, e.g., FIN. CRISIS INQUIRY COMM’N, supra note 4, at
284 (“Often, backing Bear’s borrowing were mortgage-related
securities and of these, $17.2 billion–more than Bear’s
equity–were Level 3 assets,” meaning that they lacked observable
prices.). In the case of nonsubprime asset-backed securities, “the
problem was that if a large bank failed or had to dump assets for other
reasons … then prices of these asset classes would fall.” GARY
GORTON, SLAPPED BY THE INVISIBLE HAND: THE PANIC OF 2007, at 134 (2010).
(25.) See Martin N. Baily et al., Improving Resolution Options for
Systemically Relevant Financial Institutions 7-8 (Oct. 2009) (Squam Lake
Working Group on Financial Regulation),
http://dspace.cigilibrary.org/jspui/bitstream/123456789/27243/1/Improving%20resolution %20options%20for%20systematically%20relevant%20financial%20institutions.pdf. From August 2007 to January 2009, repo haircuts rose
from near 0% to 45%, signaling a massive drop in the amount a financial
institution could borrow against a given portfolio. See Gorton &
Metrick, supra note 1, at 429 fig.4.
(26.) For example, Lehman Brothers’s counterparties “had
the right under U.S. bankruptcy law to terminate their derivative
contracts with Lehman upon its bankruptcy, and to the extent that Lehman
owed them money on the contracts they could seize any Lehman collateral
they held.” FIN. CRISIS INQUIRY COMM’N, supra note 4, at 354.
(27.) See id. at 344-50. See generally GORTON, supra note 24, at
128 (“Collateral calls … were massive, creating liquidity
problems for some and windfall funding for others.”).
(28.) See, e.g., David A. Price, The Dodd-Frank Act and Insolvency
2.o, REGION FOCUS, 3d Quarter 2011, at 8,
/federal reserve.pdf; Darrell Duffle & David A. Skeel, Jr., A
Dialogue on the Costs and Benefits of Automatic Stays for Derivatives
and Repurchase Agreements (Scholarship at Penn Law, Working Paper No.
397, 2012), http://lsr.nellco.org/upenn_wps/397; Baily et al., supra
note 25, at 3 (“The priority treatment currently given to these
contracts should be reevaluated to determine if it unnecessarily adds to
systemic risk.”); Stephen J. Lubben, A Consensus Begins To Emerge
on Derivatives in Bankruptcy, N.Y. TIMES: DEALBOOK (Apr. 27, 2012, 11:59
AM), http ://dealbook.nytimes.com/2012/04/27/a-consensus-begins-to-emerge-on-derivatives-in -bankruptcy.
(29.) See, e.g., Viral V. Acharya et al., Resolution Authority, in
REGULATING WALL STREET: THE DODD-FRANK ACT AND THE NEW ARCHITECTURE OF
GLOBAL FINANCE 213, 229 (Viral V. Acharya et al. eds., 2011); Franklin
R. Edwards & Edward R. Morrison, Derivatives and the Bankruptcy
Code: Why the Special Treatment?, 22 YALE J. ON REG. 91, 106 (2005);
Randall S. Kroszner, Making Markets More Robust, in REFORMING U.S.
FINANCIAL MARKETS: REFLECTIONS BEFORE AND BEYOND DODD-FRANK 51, 77
(Benjamin M. Friedman ed., 2011); Edward R. Morrison, Is the Bankruptcy
Code an Adequate Mechanism for Resolving the Distress of Systemically
Important Institutions?, 82 TEMP. L. REV. 449, 451-52 (2009) ; Mark J.
Roe, The Derivatives Market’s Payment Priorities as Financial
Crisis Accelerator, 63 STAN. L. REV. 539, 564-69 (2011); David A. Skeel,
Jr., Bankruptcy Boundary Games, 4 BROOK. J. CORP. FIN. & COM. L. 1,
10-13 (2009); David A. Skeel, Jr. & Thomas H. Jackson, Transaction
Consistency and the New Finance in Bankruptcy, 112 COLUM. L. REV. 152,
166-68 (2012); Bryan G. Faubus, Note, Narrowing the Bankruptcy Safe
Harbor for Derivatives To Combat Systemic Risk, 59 DUKE L.J. 801, 827
(2010); Bliss & Kaufman, supra note 20, at 19; Baily et al., supra
note 25, at 7.
(30.) See, e.g., Too Big To Fail: The Role for Bankruptcy and
Antitrust Law in Financial Regulation Reform (Part I): Hearing Before
the Subcomm. on Commercial & Admin. Law of the H. Comm. on the
Judiciary, 111th Cong. 72 (2009) (statement of Harvey R. Miller, Senior
Partner, Well, Gotshal & Manges LLP) (arguing that the exercise of
close-out rights by 733,000 counterparties “caused a massive
destruction of value for Lehman” following the investment
bank’s bankruptcy filing); DAVID SKEEL, THE NEW FINANCIAL DEAL:
UNDERSTANDING THE DODD-FRANK ACT AND ITS (UNINTENDED) CONSEQUENCES 162
(2011) (“The simple expedient of giving managers the benefit of the
stay would make bankruptcy a much more viable option for a systemically
important firm.”); Onnig H. Dombalagian, Requiem for the Bulge
Bracket?: Revisiting Investment Bank Regulation, 85 IND. L.J. 777, 811
(2010); Stephen J. Lubben, Repeal the Safe Harbors, 18 AM. BANKR. INST.
L. REV. 319, 320 (2010); Robert R. Bliss & George G. Kaufman,
Resolving Large Complex Financial Institutions: The Case for
Reorganization 10-11 (Apr. 11, 2011) (unpublished manuscript),
(31.) See, e.g., SKEEL, supra note 30; Thomas H. Jackson, Chapter
11F: A Proposal for the Use of Bankruptcy To Resolve Financial
Institutions, in ENDING GOVERNMENT BAILOUTS AS WE KNOW THEM 217, 236
(Kenneth E. Scott et al. eds., 2009); Roe, supra note 29, at 555
(“The Bankruptcy Code’s core negative consequence from
favoring derivatives contracts and repurchase agreements is to slacken
the contracting parties’ efforts to contain the risk of
counterparty failure.”); see also Acharya et al., supra note 29, at
230 (“The effective outcome is tremendous liquidity in repo markets
for these products in good times, with systemic stress and fragility
when the products are anticipated to experience losses. The expansion of
safe harbor to repo transactions with underlying mortgage-backed assets
… has been cited as one of the reasons….”); Patrick Bolton
& Martin Oehmke, Should Derivatives Be Privileged in Bankruptcy?
(July 3, 2012) (uupublished manuscript), http://www.gsb.columbia-edu
/faculty/moehmke/papers/BoltonOehmkeDerivatives.pdf (arguing that risks
shifted to firms’ general creditors are more efficiently borne by
their derivative counterparties).
(32.) Ross Levine, Financial Development and Economic Growth: Views
and Agenda, 35 J. ECON. LITERATURE 688, 689 (1997). See generally
Douglass C. North, Economic Performance Through Time, 84 AM. ECON. REV.
359, 360 (1994) (“When it is costly to transact, then institutions
(33.) See Douglas W. Diamond & Philip H. Dybvig, Bank Runs,
Deposit Insurance, and Liquidity, 92 J. POL. ECON. 401, 409 (1983)
(“It is precisely the ‘transformation’ of illiquid assets
into liquid assets that is responsible both for the liquidity service
provided by banks and for their susceptibility to runs.”). See
generally Michael Aitken & Carole Comerton-Forde, How Should
Liquidity Be Measured?, 11 PAC.-BASIN FIN. J. 45, 46 (2003) (“A
perfectly liquid market is one where any amount of a given security can
be instantaneously converted to cash and back to securities at no
(34.) See GORTON, supra note 24, at 7 (arguing that “demand
deposits [and] repo with collateral” play analogous roles in the
supply of “[t]ransactions (or ‘liquidity’)”).
(35.) See FIN. CRISIS INQUIRY COMM’N, supra note 4, at 283. To
be sure, eventually all “[s]hort-term near money market instruments
with a risk of loss–uninsured deposits, commercial paper, and
repos–respond to increases in risk primarily through [a contraction in]
quantity.” Charles W. Calomiris, The Subprime Turmoil: What’s
Old, What’s New, and What’s Next, 15 J. STRUCTURED FIN. 6, 24
(36.) As Gorton has written, “banks should be liquidated if
they cannot honor their debt in noncrisis periods, but not during a
crisis.” GORTON, supra note 5, at 149.
(37.) See, e.g., Acharya et al., supra note 29, at 229 (describing
the attendant “form of systemic risk involving fire sales … and
liquidity funding spirals”).
(38.) As Morgan Ricks argues, a bankruptcy stay on
counterparties’ recovery efforts would offer a counterproductive
method for arresting a bank run because, rather than preserving the
bank’s liquidity, “imposing a legal stay on money-claims would
instantly turn them into non-money, which is exactly [the outcome to be
avoided].” Ricks, supra note 6, at 112; see also Bliss &
Kaufman, supra note 30, at 10 (“Adding [automatic] stays … will
not be suffic[ient] to solve this dilemma…. Stays can suspend
collection of debts but they cannot force continued rolling over of
funding or provision of services.”).
(39.) Pozsar et al., supra note 2, at 1; see also GORTON, supra
note 5, at 28 (“[T]he private sector’s attempts at money
creation–first private banknotes and then demand deposits–were plagued
by difficulties rooted in the inability of the private sector to create
(40.) See generally Charles W. Calomiris, Is Deposit Insurance
Necessary? A Historical Perspective, 50 J. ECON. HIST. 283, 284 n.4
(1990) (identifying “the desire to preserve liquidity” as the
core motivation of deposit insurance legislation).
(41.) See Edwards & Morrison, supra note 29.
(42.) See Bliss & Kaufman, supra note 12, at 154-55.
(43.) See Diamond & Dybvig, supra note 33, at 401; cf. Sudipto
Bhattacharya & Douglas Gale, Preference Shocks, Liquidity, and
Central Bank Policy, in LIQUIDITY AND CRISES 35 (Franklin Allen et al.
eds., 2011) (characterizing the central bank as a solution to the
flee-rider problem in interbank lending).
(44.) See, e.g., supra note 40 and accompanying text.
(45.) See Aitken & Comerton-Forde, supra note 33, at 46.
(46.) See generally Sanford J. Grossman & Merton H. Miller,
Liquidity and Market Structure, 43 J. FIN. 617, 618 (1988) (describing
how “market makers” satisfy “the demand for
immediacy” among market participants by “maintaining a
continuous presence in the market”).
(47.) See 11 U.S.C. [section] 109(b)-(d) (2006) (excluding
insurance companies, banks, stockbrokers, and commodity brokers from
relief under portions of the Bankruptcy Code).
(48.) See Levine, supra note 32, at 690 (“The costs of
acquiring information and making transactions create incentives for the
emergence of financial markets and institutions.”). See generally
Sudipto Bhattacharya & Anjan V. Thakor, Contemporary Banking Theory,
3 J. FIN. INTERMEDIATION 2 (1993) (surveying theories of banking).
(49.) See, e.g., Robert G. King & Ross Levine, Finance,
Entrepreneurship, and Growth: Theory and Evidence, 32 J. MONETARY ECON.
(50.) See GORTON, supra note 5, at 46 (“You could not realize
the value of [a Van Gogh] painting at short notice unless you were
willing to take a great loss.”). For further discussion, see
Levine, supra note 32, at 690-96.
(51.) Imagine an investor who has lent $1 million to a
small-business owner and now seeks to sell the loan to an
arm’s-length buyer. If the loan matures in five years, each buyer
would have to incur costs evaluating the borrower’s business
prospects over a five-year time horizon. Moreover, buyers might rightly
interpret the investor’s attempt to sell the loan as a discouraging
signal about the borrower’s credit risk. Thus the investor may find
the loan impossible to sell.
(52.) See Bengt Holmstrom & Jean Tirole, Private and Public
Supply of Liquidity, 106 J. POL. ECON. 1, 2 (1998) (“The wedge
between the full value of the firm and the external value of the firm
prevents it from financing all projects that have a positive net present
value.”); Jean Tirole, Illiquidity and All Its Friends, 49 J. ECON.
LITERATURE 287, 291 (2011) (“Financial market imperfections, which
encompass moral hazard, adverse selection (asymmetries of information
about assets in place and projects), and mere transaction costs, make it
hard for cash-strapped corporations to raise financing even for positive
(53.) John Bryant, A Model of Reserves, Bank Runs, and Deposit
Insurance, 4 J. BANKING & FIN. 335, 338-39 (1980); Diamond &
Dybvig, supra note 33, at 405. In contrast, Bengt Holmstrom and Jean
Tirole stress the problem of unforeseeable liquidity shocks to firms.
They argue that credit lines (contracted for ex ante) provide borrowers
with a source of liquidity that is incentive-compatible because it is
supplied at a lower rate than ex post refinancing. Holmstrom &
Tirole, supra note 52, at 12-14. Integrating these perspectives, Anil
Kashyap and others argue that banks exploit synergies between the
provision of on-demand liquidity to borrowers and depositors. Anil K.
Kashyap, Raghuram Rajan & Jeremy C. Stein, Banks as Liquidity
Providers: An Explanation for the Coexistence of Lending and
Deposit-Taking, 57 J. FIN. 33 (2002). For further discussion, see XAVIER
FREIXAS & JEAN-CHARLES ROCHET, MICROECONOMICS OF BANKING 46-49 (2d
(54.) Cf. Ricks, supra note 6, at 93 (citing “liquidity and
price-protection” as the crucial characteristics of
“transaction reserves,” or money-like instruments).
(55.) See id. at 82; see also FREIXAS & ROCHET, supra note 53,
at 4 (“[M]odern banks can be seen as transforming securities with
short maturities, offered to depositors, into securities with long
maturities, which borrowers desire.”).
(56.) This function of bank deposits is discussed in Gary Gorton
& George Pennacchi, Financial Intermediaries and Liquidity Creation,
45 J. FIN. 49, 50 (1990). See generally GORTON, supra note 5, at 219
(“[D]ebt is ‘least information-sensitive,’ meaning that
it minimizes the incentives for agents to produce private information,
creating adverse selection, and that debt maintains the most value in
the presence of aggregate shock.”).
(57.) See Gary Gorton & Andrew Winton, Financial
Intermediation, in 1 HANDBOOK OF THE ECONOMICS OF FINANCE 432, 455
(George M. Constantinides et al. eds., 2003) (“Financial
intermediaries are the natural entities to create such securities, as
they hold diversified portfolios of assets. Consequently, their debt
should be used for transactions purposes.”).
(58.) See GORTON, supra note 5, at 48 (“Bank debt is a senior
claim on the collateral: debt holders are paid first and stock or equity
is paid last.”); Gorton & Pennacchi, supra note 56, at 50.
(59.) See GORTON, supra note 24, at 20 (“The need for
information-insensitive debt is the logic behind deposit
(60.) In fact, deposits are tracked in the Federal Reserve’s
official monetary aggregates. See H.6 Money Stock Measures: About the
Release, FED. RES. STAT. RELEASE, http://www.federalreserve.gov
/releases/h6/about.htm (last updated Nov. 3, 2001).
(61.) Cf. Ricks, supra note 6, at 96 (making an analogous argument
about the most liquid money-market instruments).
(62.) Gorton & Winton, supra note 57, at 433.
(63.) Charles W. Calomiris & Charles M. Kahn, The Role of
Demandable Debt in Structuring Optimal Banking Arrangements, 81 AM.
ECON. REV. 497, 497 (1991).
(64.) Diamond & Dybvig, supra note 33.
(65.) FREIXAS & ROCHET, supra note 53, at 4 (“This
maturity transformation function necessarily implies a risk, since the
banks’ assets will be illiquid, given the depositors’
(66.) During the era of uninsured banking, bank failures tended to
produce knock-on effects that spread to generate system-wide banking
panics. By one count, at least ten major banking panics occurred in the
United States prior to the Great Depression. See Charles W. Calomiris
& Gary Gorton, The Origins of Banking Panics: Models, Facts, and
Bank Regulation, in FINANCIAL MARKETS AND FINANCIAL CRISES 109, 113-15
(R. Glenn Hubbard ed., 1991).
(67.) See Ricks, supra note 6, at 78 (“By extension, large
numbers of near-simultaneous bank failures can lead to a sudden and
severe reduction in the money supply–with correspondingly severe
(68.) See Bliss & Kaufman, supra note 12, at 149
(“Liquidity losses occur when depositors are denied immediate
access to the insured par value or, in the case of uninsured depositors,
the recovery value of their accounts.”).
(69.) See Calomiris, supra note 35, at 26 (“The risk
intolerance of money market instruments has been visible historically
and in recent times, both in response to idiosyncratic events at
particular banks and firms, and in response to aggregate shocks.”).
(70.) See GORTON, supra note 24, at 32-33.
(71.) See generally Ben Bernanke et al., The Financial Accelerator
and the Flight to Quality, 78 REV. ECON. & STAT. 1, 1 (1996)
(“[C]hanges in credit-market conditions amplify and propagate the
effects of initial real or monetary shocks.”). The result may be
“a sharp, brief, ultracyclical deterioration of all or most of a
group of financial indicators–short-term interest rates, asset …
prices, commercial insolvencies, and failures of financial
institutions.” Raymond W. Goldsmith, Comment on Hyman P. Minsky,
The Financial-Instability Hypothesis: Capitalist Processes and the
Behavior of the Economy, in FINANCIAL CRISES: THEORY, HISTORY, AND
POLICY 41, 42 (Charles P. Kindleberger & Jean-Pierre Laffargue eds.,
(72.) See, e.g., Robert R. Bliss, Bankruptcy Law and Large Complex
Financial Organizations: A Primer, 27 ECON. PERSP., 1st Quarter 2003, at
48, 48 (“[F]inancial institutions provide capital and other
financial services to all sectors of the economy and they form the
backbone of the financial markets, markets that rely to a great extent
on trust. Thus, the failure of a financial intermediary calls into
question a multitude of business relations.”); Bliss & Kaufman,
supra note 12, at 149 (stating that disruptions in depositors’
access to their savings “reduces the ‘moneyness’ of
demand and other short-term deposits by effectively transforming a
short-term liquid deposit into a time deposit of uncertain
maturity,” and “may produce substantial negative externalities
in the markets served by the bank” which depend on its liquidity);
Ricks, supra note 6, at 108 (“[A] sudden inability to meet
transactional needs may lead to consequential losses–opportunity costs,
operational disruption, reputational damage, or even default.”).
But see Marvin Goodfriend & Robert G. King, Financial Deregulation,
Monetary Policy, and Central Banking, ECON. REV., May/June 1988, at 3,
16 (“[B]ank failures … even at their worst … were roughly of
the same order of magnitude as nonbank business failures. Their
aggregate effects appear to have been reasonably well
(73.) BRAY HAMMOND, BANKS AND POLITICS IN AMERICA: FROM THE
REVOLUTION TO THE CML WAR 180 (1957).
(74.) One early approach was to impose superadded liability on
banks’ shareholders, abrogating the common law rule of limited
liability. For a discussion of this approach, see Jonathan R. Macey
& Geoffrey P. Miller, Double Liability of Bank Shareholders: History
and Implications, 27 WAKE FOREST L. REV. 31 (1992); and Joseph M.
Leonard, Note, Superadded Liability of Bank Stockholders, 14 TEMPE U.
L.Q. 522, 522 (1940). Congress ultimately imposed double liability on
bank shareholders in the National Banking Act of 1864, ch. 106,
[section] 12, 13 Stat. 99, 102-03. By the late 1920s, federal law and
the laws of thirty-nine states provided for some form of superadded
(usually double) liability. Leonard, supra, at 523. More recent law
“singles out those with some insider connection to the failed bank
and attempts to shift the costs of failure from the [deposit] insurance
fund to the insiders.” Peter P. Swire, Bank Insolvency Law Now that
It Matters Again, 42 DUKE L.J. 469, 485 (1992).
Ex ante capital and liquidity standards also insulate banks’
customers at shareholders’ expense. For discussion of early reserve
and capital requirements, see, for example, Arthur J. Rolnick &
Warren E. Weber, The Causes of Free Bank Failures: A Detailed
Examination, 14 J. MONETARY ECON. 267, 270-71 (1984), discussing the
Free Banking Era; and David M. Gische, The New York City Banks and the
Development of the National Banking System 1860-1870, 23 AM. J. LEGAL
HIST. 21, 25 (1979), discussing the National Banking era. Of particular
significance was a New York statute, enacted in 1838 and widely
imitated, that required bank notes to be backed by deposited collateral.
See Michael D. Harter, American Banking and the Money Supply of the
Future, 3 ANNALS AM. ACAD. POL. & SOC. SCI. 559, 562 (1893).
(75.) A number of early statutes created government insurance funds
or imposed mutual liability among banks to cover individual
institutions’ shortfalls. New York established the nation’s
first bank-obligation insurance scheme in 1829, which combined a
member-funded insurance fund with government supervision of member
hanks. For comparative discussion, see Calomiris, supra note 40, at
286-88; and Carter H. Golemhe, The Deposit Insurance Legislation of
1933: An Examination of Its Antecedents and Its Purposes, 75 POL. SCI.
Q:.. 181, 182-86 (1960). Indiana enacted a very different plan in 1834,
later emulated in Ohio and Iowa, which relied on industry
self-regulation and unlimited mutual liability among member banks. One
commentator argues that the Indiana scheme proved more successful than
the New York fund because it “aligned the incentive and authority
to regulate and made insurance protection credible through unlimited
mutual liability among banks.” Calomiris, supra note 40, at 288.
Member-funded deposit insurance through the FDIC forms a crucial
feature of the current regulatory architecture. For a useful overview of
the FDIC’s operations, see Resolutions Handbook, FED. DEPOSIT INS.
CORP. (2003), http://www.fdic.gov /bank/historical/reshandbook. In
addition to straight deposit payoffs, a notable 1935 amendment to the
Glass-Steagall Act expanded the FDIC’s toolkit by allowing it to
facilitate mergers among insured banks in order to eliminate weak links
from the system. Banking Act of 1935, Pub. L. No. 74-305, ch. 614,
[section] 101, 49 Stat. 684, 684-703 (codified as amended at 12 U.S.C.
[section] 1823(c)(2) (2006)). These “purchase and assumption”
transactions came to form the cornerstone of the FDIC’s postwar
approach to resolving failed financial institutions. See William M.
Isaac, The Role of Deposit Insurance in the Emerging Financial Services
Industry, 1 YALE J. ON REG. 195, 202 (1984) (“During the past 30
years, the majority of bank failures, and practically all large bank
failures, have been handled through [purchase and assumption]
transactions.”); Resolutions Handbook, supra, at 19-40 (describing
purchase and assumption transactions).
(76.) Congress established the Reconstruction Finance Corporation
in 1932, but many banks failed to borrow from it, since doing so
signaled weakness and often provoked bank runs. The First Fifty Years: A
History of the FDIC 1933-1983, FED. DEPOSIT INS. CORP. 36-37 (1984),
(77.) The Federal Reserve, as the lender of last resort, supplies
backup liquidity to otherwise solvent institutions that might fail due
to a liquidity shortage, “when no other lender is either capable of
lending or willing to lend in sufficient volume to prevent or end a
financial panic.” Allan H. Meltzer, Financial Failures and
Financial Policies, in DEREGULATING FINANCIAL SERVICES: PUBLIC POLICY IN
FLUX 79, 83 (George G. Kaufman & Roger C. Kormendi eds., 1986).
Prior to the establishment of the Federal Reserve System, the New York
Clearinghouse Association coordinated a private form of central banking
that allowed depositors to replace claims on individual banks with
certificates issued against the clearinghouse, backed by all of its
member banks. See Gary Gorton, Clearinghouses and the Origin of Central
Banking in the United States, 45 J. ECON. HIST. 277, 282 (1985); T.B.
Paton, The New York Clearing House and the Associated Banks, 12 BANKING
L.J. 593, 607-09 (1895) (detailing the program).
(78.) Bliss & Kaufman, supra note 12, at 175; see also id. at
172 (“The prompt and full payment of insured-depositor claims at
legally closed institutions before the FDIC may have collected the
proceeds from selling the assets has gone a long way to reducing the
liquidity losses of most depositors.”). By comparison, bankruptcy
creditors are subject, for example, to the Code’s provisions for
the automatic stay, 11 U.S.C. [section] 362 (2006); the trustee’s
power to assume or reject executory contracts and unexpired leases, id.
[section] 365; statutory priorities, id. [section] 507; the effects of
discharge, id. [section] 524; the trustee’s avoiding powers, id.
[section][section] 547- 549; and limitations on setoff rights, id.
[section] 553; notwithstanding contractual terms providing otherwise.
(79.) Gary Gorton & Andrew Metrick, Regulating the Shadow
Banking System, BROOKINGS PAPERS ON ECON. ACTIVITY, Fall 2010, at 261,
(80.) See Diamond & Dybvig, supra note 33, at 403
(“Illiquidity of assets provides the rationale both for the
existence of banks and for their vulnerability to runs.”).
(81.) See generally Gorton & Pennacchi, supra note 56, at 50
(“By issuing debt and equity securities against their risky
portfolios, intermediaries can attract informed agents to hold equity
and uninformed agents to hold debt which they can use for
(82.) Gorton & Metrick, supra note 79, at 263; see also GORTON,
supra note 24, at 15 (“These depositors are not willing to deposit,
say, $500 million in a bank because it cannot be insured.”).
(83.) By one estimate, the total assets under management by U.S.
institutional investors alone exceed two-hundred percent of U.S. gross
domestic product. Gorton & Metrick, supra note 79, at 276 fig. 7.
(84.) Gary B. Gorton & George G. Pennacchi, Banks and Loan
Sales: Marketing Nonmarketable Assets, 35 J. MONETARY ECON. 389, 390
(1995); accord Lawrence H. Summers, Macroeconomic Consequences of
Financial Crises: Planning for the Next Financial Crisis, in THE RISK OF
ECONOMIC CRISIS 135, 147 (Martin Feldstein ed., 1991) (“[B]ank
assets are illiquid. If all bank assets could readily be traded on a
secondary market, the need for banks would be greatly reduced.”).
(85.) See Gorton & Metrick, supra note 1, at 426 fig.1.
(86.) See Gary B. Gorton & Nicholas S. Souleles, Special
Purpose Vehicles and Securitization, in THE RISKS OF FINANCIAL
INSTITUTIONS 549, 550 (Mark Carey & Rene M. Stulz eds., 2007); John
H. Langbein, The Secret Life of the Trust: The Trust as an Instrument of
Commerce, 107 YALE L.J. 165, 172-73 (1997); Paul Mizen, The Credit
Crunch of 2007-2008: A Discussion of the Background, Market Reactions,
and Policy Responses, 90 FED. RES. BANK ST. LOUIS REV. 531, 537 (2008);
Steven L. Schwarcz, The Alchemy of Asset Securitization, 1 STAN. J.L.
Bus. & FIN. 133, 135-36 (1994). These securities, in turn, may be
packaged into collateralized debt obligations (CDOs) backed by the
mortgage pools underlying each asset-backed security in the deal. See
FIN. CRISIS INQUIRY COMM’N, supra note 4, at 128 fig.8.2. For
useful overviews of the securitization process see Gorton & Metrick,
supra note 1, at 427 fig.2; and Pozsar et al., supra note 2, at 11-13.
(87.) Mizen, supra note 86, at 536-37.
(88.) Id.; Lynn M. LoPucki, The Death of Liability, 106 YALE L.J.
1, 24 (1996).
(89.) See Mizen, supra note 86, at 537-38. For example, a structure
may have a “first-loss” or “equity” tranche, which
absorbs losses up to a par value of 3% of the pool; a
“mezzanine” tranche that absorbs losses impairing the next 7%
of the pool; and a “senior” tranche claiming the other 90% of
the pool. This example is borrowed from Credit Risk Transfer, BASEL
COMMITTEE ON BANKING SUPERVISION 45 (Mar. 2005), http://www.bis.org
/publ/joint13.pdf. Thus, for example, if the loan pool experienced a
default rate of 6.5%, the equity tranche would be wiped out, and the
remaining 3.5 percentage points of losses would deal a loss of 50% to
the mezzanine tranche, while the senior tranche would be unimpaired.
Note that the sponsor typically retains first-loss exposure in ABS
deals. By giving the sponsor a junior position in the payment waterfall,
this risk retention should, in principle, align the sponsor’s
incentives with those of the other investors, like bank capital in a
commercial bank. See Gorton & Pennacchi, supra note 84, at 409
(“If the selling bank retained a fraction of the loan or it gave
loan buyers an implicit guarantee against default, this could explain
why market participants would buy loans….”); Steven L. Schwarcz,
Enron and the Use and Abuse of Special Purpose Entities in Corporate
Structures, 70 U. CIN. L. REv. 1309, 1316 n.38 (2002).
(90.) The Financial Crisis and the Role of Federal Regulators:
Hearing Before the H. Comm. of Gov’t Oversight & Reform, 110th
Cong. 2 (2008) (statement of Alan Greenspan, Former Chairman of the
Board of Governors of the Federal Reserve System),
(91.) GORTON, supra note 5, at 50.
(92.) Mizen, supra note 86, at 538.
(93.) David Greenlaw et al., Leveraged Losses: Lessons from the
Mortgage Market Meltdown 25 ex. 3.8 (2008) (U.S. Monetary Policy Forum
Conference Draft), http://www.chicagobooth
(94.) See Stuart I. Greenbaum & Anjan V. Thakor, Bank Funding
Modes: Securitization Versus Deposits, 11 J. BANKING & FIN. 379, 379
(1987) (describing “the transformation of illiquid financial
claims.., held by depository financial intermediaries, into tradeable
ones” through securitization).
(95.) See GORTON, supra note 5, at 190-91.
(96.) See Kenneth D. Garbade, The Evolution of Repo Contracting
Conventions in the 1980s, FRBNY ECON. POL’Y REV., May 2006 at 27
Tri-Party Repo Infrastructure Reform, supra note 23, at 6 (“Cash
lenders use tri-party repos as investments that offer liquidity
maximization, principal protection, and a small positive return, while
cash borrowers rely on them as a major source of short-term
funding.”). Note that while many market participants enter repo
transactions to borrow cash, an identical transaction (dubbed
“reverse repo”) may be used to borrow securities rather than
cash. Garbade, supra, at 31-32.
(97.) Tri-Parry Repo Infrastructure Reform, supra note 23, at 3, 7
tbl.1, 19 app. ii.
(98.) Safe-harbor-eligible repos may have maturities of no longer
than one year. 11 U.S.C. [section] 101(47)(A)(i) (2006); 12 U.S.C.
[section][section] 1787(c)(8)(D)(v)(I), 1821(e)(8)(D)(v)(I) (2006); see
also Tri-Party Repo Infrastructure Reform, supra note 23, at 11 fig.4
(illustrating an hour-by-hour breakdown of an overnight repo
(99.) “A market participant might, for example, sell
securities for $10 million and simultaneously agree to repurchase them
ten days later for $10,005,555…. [T]his is comparable to borrowing $10
million for ten days at an interest rate of 2 percent per annum.”
Garbade, supra note 96, at 27.
(100.) See GORTON, supra note 5, at 38. In addition, “[r]epo
collateral can be rehypothecated; that is, the collateral received in a
repo deposit can be freely reused in another transaction with an
unrelated third party,” so that the repo lender enjoys continuous
liquidity even before the contract matures. Gorton & Metrick, supra
note 79, at 277.
(101.) Ricks, supra note 6, at 90 (“In area after area [of
law], these instruments are treated like deposits–a classic form of
‘money’–rather than ordinary debt securities.”).
(102.) The original bankruptcy safe harbor recognized only repos
backed by certificates of deposit, eligible bankers’ acceptances,
and government or government-guaranteed securities. Bankruptcy
Amendments and Federal Judgeship Act of 1984, Pub. L. No. 98-353,
[section] 391, 98 Stat. 333, 365 (codified as amended at 11 U.S.C.
[section] 101(47)(A)(i) (2006)). However, mortgage-related assets were
recognized in the Bankruptcy Abuse Prevention and Consumer Protection
Act of 2005, Pub. L. No. 109-8, [section][section] 901(e), 907, 119
Stat. 23, 152-55, 171 (codified at 11 U.S.C. [section] 101(47)(A)(i); 12
U.S.C. [section][section] 1787(c)(8)(D)(v)(I), 1821(e)(8)(D)(v)(I)
(103.) See S. REP. NO. 98-65, at 45-46 (1983); Garbade, supra note
96, at 29; Gorton & Metrick, supra note 1, at 432. As the Wall
Street Journal reported in 1979, “Repurchase agreements are
attractive to corporate treasurers for a variety of reasons: they make
money, they involve little if any risk and they provide the liquidity
that can’t be found in other short-term investments.” Lawrence
Rout, More Firms Use Repurchase Agreements as a Way To Earn Interest on
Idle Funds, WALL ST. J., Apr. 16, 1979, at 15.
(104.) See Gorton & Metrick, supra note 1, at 425
(“Securitized banking is the business of packaging and reselling
loans, with repo agreements as the main source of funds.”).
(105.) See GORTON, supra note 5, at 191.
(106.) Hyun Song Shin, Securitisation and Financial Stability, 119
ECON. J. 309, 310 (2009).
(107.) See supra note 66.
(108.) For further discussion, see Gorton & Souleles, supra
note 86, at 549; and Schwarcz, supra note 86, at 135-36.
(109.) Thomas E. Plank, Toward a More Efficient Bankruptcy Law:
Mortgage Financing Under the 2005 Bankruptcy Amendments, 31 S. ILL. U.
L.J. 641, 654 (2007).
(110.) See Gorton & Metrick, supra note 79, at 263 (describing
collateral in repos as offering “protection similar to that
provided by deposit insurance”).
(111.) See Garbade, supra note 96 (detailing consequential changes
in repo-market institutions during the 1980s).
(112.) For discussion, see Tri-Party Repo Infrastructure Reform,
supra note 23.
(113.) See supra note 96.
(114.) Mfrs. Hanover Trust Co. v. Drysdale Sec. Corp., 801 F.2d 13,
16 (2d Cir. 1986).
(115.) Michael Quint, Lessons in Drysdale’s Default, N.Y.
TIMES, May 20, 1982, http://
(116.) See, e.g., John Andrew, Some Expect Shakeout in U.S.
Securities, WALL ST. J., July 6, 1982, at 27 (“Now government
securities dealers and others on Wall Street are wondering who’s
next.”); Robert A. Bennett, Less Risk, More Worry for the Banks,
N.Y. TIMES, Oct. 10, 1982,
http://www.nytimes.com/1982/10/10/business/less-risk-more-worry-for-the-banks.html (“[I]f the retreat from risk goes too far, there are
grave dangers. For as the banks pull back, borrowers are left short of
cash, making it tough–or even imposslble–for them to repay their
(117.) See Bd. of Governors of the Fed. Reserve Sys., Record of
Policy Actions of the Federal Open Market Committee, 68 FED. RES. BULL.
417, 418-19 (1982) (noting the Federal Open Market Committee’s
attention on May 18, 1982, to the impact of Drysdale’s failure).
Referring to brokerage houses, a Federal Reserve official warned the
Securities Investor Protection Corporation that “some of these
people might be going under.” Richard L. Hudson & Kenneth H.
Bacon, How Agencies Helped Avert Drysdale Panic, WALL ST. J., June 1,
1982, at 29, 36.
(118.) Hudson & Bacon, supra note 117, at 36.
(119.) Robert J. Cole, A Dealer in Bonds Defaults on Debt, N.Y.
TIMES, May 19, 1982,
(120.) Michael Quint, Weakness in Treasury Issues: Post-Drysdale
Caution Seen, N.Y. TIMES, June 7, 1982,
(121.) Warner v. Zent, 997 F.2d 116, 120-21 (6th Cir. 1993); State
v. Warner, 564 N.E.2d 18, 23 (Ohio 1990).
(122.) Martha Brannigan, ESM Collapse Prompts S&P to Add 4 More
Municipalities to Credit Watch, WALL ST. J., Mar. 8, 1985, at 10.
(123.) John Bussey, Gregory Stricharchuk & Martha Brannigan,
Thrift’s Ex-Owner, Stung by Failure of ESM, Says He’s One of
the Victims, WALL ST. J., Mar. 14, 1985, at 20; Todd S. Purdum,
Cincinnati Thrift Unit Seeks Sale: E.S.M. Failure Caused a Run at Home
State, N.Y. TIMES, Mar. 11, 1985,
(124.) Thrift Crisis: Closing of Ohio S&Ls After Run on
Deposits Is One for the Books, WALL ST. J., Mar. 18, 1985, at 1.
(125.) Al Swift, Fed, Not Treasury, Should Take the Lead, N.Y.
TIMES, Aug. 4, 1985,
(126.) Robert J. Cole, Wall St. Securities Firm Files for
Bankruptcy, N.Y. TIMES, Aug. 13, 1982,
http://www.nytimes.com/1982/08/13/business/wall-st-securities-firm-files-for-bankruptcy .html; Repo Market Remains Weak as Legal Issues Trouble
Many Dealers, WALL ST. J., Jan. 30, 1984, at 43.
(127.) Repo Market Remains Weak as Legal Issues Trouble Many
Dealers, supra note 126, at 43.
(128.) Robert J. Cole, Forced Sale of Securities Is Opposed by
Lombard, N.Y. TIMES, Aug. 14, 1982, at 37; Gary Putka & George
Anders, Two U.S.-Securities Firms File Plea of Bankruptcy, Jarring Wall
Street, WALL ST. J., Aug. 13, 1982, at 3.
(129.) Tim Carrington, Securities in Lombard-Wall Case Termed Loan
Collateral by a Bankruptcy Judge, WALL ST. J., Sept. 20, 1982, at 10;
Lombard Securities with Buy-Back Plan Are Frozen by Court, WALL ST. J.,
Aug. 18, 1982, at 7 (“In its ruling, the court agreed with Lombard
that securities held in connection with repurchase agreements should be
considered loans rather than purchases.”).
The same year, the FDIC made a parallel move following the failure
of Mount Pleasant Bank and Trust, deciding that the bank’s repo
creditors “would have to wait along with other creditors for their
share of the bank’s assets remaining after liquidation.”
Michael Quint, Repo Backing Is Under Cloud: Lombard and Iowa Rulings
Spur Review, N.Y. TIMES, Sept. 29, 1982,
(130.) Cole, supra note 128, at 37 (“Unlike last May’s
default of Drysdale Government securities, which nearly touched off a
financial crisis, Wall Street took Lombard’s collapse with
comparatively little reaction.”).
(131.) Daniel Hertzberg, Lombard-Wall Failure May Cause Losses for
Dozens of New York State Institutions, WALL ST. J., Aug. 17, 1982, at 4
(“[A] more immediate threat is a cash squeeze for a handful of the
institutions whose unspent construction funds are 100% invested at
Lombard-Wall. It is possible that the bankruptcy proceedings could tie
up the money for months.”).
(132.) Id. Moody’s, the credit rating agency, suspended the
ratings on thirty-eight bond issues affected by Lombard-Wall’s
collapse. See Robert Metz, Lombard Fall and Ratings, N.Y. TIMES, Aug.
25, 1982, http://www.nytimes.conff1982/08/25/business/market-place
(133.) George Anders & Daniel Hertzberg, Lombard-Wall Felt the
Effect of Other Crisis, WALL ST. J., Aug. 16, 1982, at 15, 18; see
Martin Baron, Money-Market Funds Are Cash-Rich but Face Some Problems,
L.A. TIMES, Sept. 19, 1982, at E1 (noting that “[m]oney-fund
managers believe one of the most serious close-calls occurred” when
Lombard-Wall’s bankruptcy threatened the liquidity of the Reserve
Fund’s repo positions).
This episode repeated itself on a catastrophic scale when the
Reserve Primary Fund “broke the buck” in September 2008 after
Lehman Brothers’s collapse left the Fund holding $785 million in
impaired commercial paper and triggered a run on the money markets. See
Marcin Kacperczyk & Philipp Schnabl, When Safe Proved Risky:
Commercial Paper During the Financial Crisis of 2007-2009, at 2
(Nat’l Bureau of Econ. Research Working Paper, No. 15538, Nov.
(134.) Judge Approves Lombard-Wall Creditor Pacts, WALL ST. J.,
Aug. 19, 1982, at 5 (quoting Edward Ryan, Judge, U.S. Bankruptcy Court
for the Southern District of New York); see More Securities Sales by
Lombard Creditors Are Cleared by Judge, WALL ST. J., Aug. 26, 1982, at
(135.) Stephen A. Lumpkin, Repurchase and Reverse Repurchase
Agreements, in INSTRUMENTS OF THE MONEY MARKET 59, 63-64 (Timothy Q.
Cook & Robert K. Laroche eds., 1998).
(136.) Putka & Anders, supra note 128, at 3.
(137.) Repurchase Accord Issue, N.Y. TIMES, Jan. 25, 1983,
(138.) Lumpkin, supra note 135, at 64; Michael Quint, Securities
Dealers in Reforms, N.Y. TIMES, Jan. 31, 1983,
.html; see, e.g., H.R. CONF. REP. NO. 98-882, at 8 (1984) (statement of
Rep. Robert W. Kastenmeier).
(139.) Bankruptcy Amendments and Federal Judgeship Act of 1984,
Pub. L. No. 98-353, [section] 392, 98 Stat. 333, 365 (1984) (codified as
amended at 11 U.S.C. [section] 362(b)(7)(2006)) (automatic stay
exemption); id. [section] 393, 98 Stat. at 365 (codified as amended at
11 U.S.C. [section] 546(f) (avoiding powers exemption).
(140.) As amended, 11 U.S.C. [section] 559 provides in pertinent
[T]he exercise of a contractual right ... to cause the liquidation, termination, or acceleration of a repurchase agreement because of [the financial condition of the debtor] shall not be stayed, avoided, or otherwise limited by operation of any provision of [the Bankruptcy Code] or by order of a court or administrative agency in any proceeding under the [Code]....
(141.) Shin, supra note 106, at 331.
(142.) Cf. Holmstrom & Tirole, supra note 52, at 5 (explaining
that bank-created liquidity will be insufficient when there is an
aggregate shortage of liquidity).
(143.) Stein, supra note 3, at 41.
(144.) FIN. CRISIS INQUIRY COMM’N, supra note 4 at 256.
(145.) Id. at 283.
(146.) Id. at 286-90.
(147.) Roe, supra note 29, at 552-53.
(148.) Skeel & Jackson, supra note 29, at 163.
(150.) Roe, supra note 2% at 572-73.
(151.) Skeel & Jackson, supra note 29, at 179. The
debtor’s post-petition obligations would be limited to adequate
protection of the collateral’s value; the debtor would have no
further obligation to post collateral. Id. at 176-77.
(152.) Id. at 168.
(153.) FIN. CRISIS INQUIRY COMM’N, supra note 4, at 283
(“Throughout 2007, Bear Stearns reduced its unsecured commercial
paper … and replaced it with secured repo borrowing (which rose from
$69 billion to $102 billion).”).
(154.) Franklin Allen, Financial Structure and Financial Crisis, 2
INT’L REV. FIN. 1, 9 (2001) (internal citations omitted).
(155.) See supra Section III.B.
(156.) Tim Carrington & George Anders, Drysdale’s Default
Shows Dangers of Intricate Financing Arrangements, WALL ST. J., May 20,
1982, at 29.
(157.) Kacperczyk & Schnabl, supra note 133, at 1.
(158.) Id. at 2.
(160.) See Calomiris, supra note 35, at 24 (“Short-term near
money market instruments with a risk of loss–uninsured deposits,
commercial paper, and repos–respond to increases in risk primarily
through [a contraction in] quantity.”).
(161.) See generally Rene M. Stulz & Herb Johnson, An Analysis
of Secured Debt, 14 J. FIN. ECON. 501, 519 (1985) (“[I]f the
existing debt of the firm is risky enough and there is a significant
underinvestment problem one would expect secured debt to be
(162.) Tirole, supra note 52, at 287.
(163.) See Gary B. Gorton, Questions and Answers About the
Financial Crisis 14 (Nat’l Bureau of Econ. Research, Working Paper
No. 15787, 2010), http://www.nber.org/papers/w15787.pdf.
(164.) GORTON, supra note 5, at 99.
(165.) Id. at 58. Other interventions were conducted through
programs such as the Federal Reserve’s Discount Window, Primary
Dealer Credit Facility, “Maiden Lane” programs, Term
Asset-Backed Securities Loan Facility, and, later, successive rounds of
quantitative easing. See FIN. CRISIS INQUIRY COMM’N, supra note 4,
at 290, 294-95, 376, 396. The Treasury Department’s involvement
included, for example, the Troubled Asset Relief Program (TARP) and the
temporary guarantee for money-market mutual funds. See id. at 359,
(166.) For example, Dodd-Frank directs the Federal Reserve Board of
Governors to establish prudential liquidity requirements for banks and
nonhank financial companies under its supervision. Dodd-Frank Act
[section] 165(b)(1)(A)(ii), Pub. L. No. 111-203, 124 Stat. 1376, 1424
(to be codified at 15 U.S.C. [section] 8305(b)(x)(A)(ii)).
(167.) Skeel & Jackson, supra note 29, at 179 n. 118.
(168.) See 12 U.S.C. [section] 1821(e)(10)(B) (2006); Dodd-Frank
Act [section] 210(C)(10)(B), Pub. L. No. 111-203, 124 Stat. 1376, 1491
(to be codified at 12 U.S.C. [section] 5390(c)(10)(B)).
(169.) Skeel & Jackson, supra note 29, at 179; accord Acharya
et al., supra note 29, at 231 (“[Assets] that are liquid should
keep the exemption…. [Assets] that are illiquid–or potentially
illiquid … would be subject to the ordinary rules of bankruptcy,
including the automatic stay.”).
(170.) See Gorton & Metrick, supra note 79, at 267 (“[I]f
repo had not been granted this [bankruptcy-proof] status, the private
sector would have sought a substitute, which likely would have been even
(171.) The notional value referenced in the credit default swap
market alone reached $62.2 trillion at its peak. Summaries of Market
Survey Results, INT’L SWAPS & DERIVATIVES ASS’N,
http://www.isda.org/statistics/recent.html (last visited Oct. 5, 2012).
“Since 1996, the credit default swap market has seen almost 100
percent annual growth…. It is by far the largest part of the overall
credit derivatives market….” GEORGE CHACKO ET AL., CREDIT
DERIVATIVES: A PRIMER ON CREDIT RISK, MODELING, AND INSTRUMENTS 186
(172.) Ian Bell & Petrina Dawson, Synthetic Securitization: Use
of Derivative Technology for Credit Transfer, 12 DUKE J. COMP. &
INT’L L. 541, 550 (2002).
(173.) CHACKO ET AL., supra note 171, at 200-01. A cash-based deal,
in contrast, “brings in addition to credit risk all the normal
risks associated with owning an asset, such [as] interest rate,
prepayment, and currency risk.” Id. at 221.
(174.) Bell & Dawson, supra note 172, at 550.
(175.) See CHACKO ET AL., supra note 171, at 220; YURI YOSHIZAWA,
MOODY’S INVESTORS SERV., MOODY’S APPROACH TO RATING SYNTHETIC
CDOS 1 (2003).
(176.) See EUR. CENT. BANK, CREDIT DEFAULT SWAPS AND COUNTERPARTY
RISK 20 (2009); see also JAN JOB DE VRIES ROBBE, STRUCTURED FINANCE: ON
FROM THE CREDIT CRUNCH-THE ROAD TO RECOVERY 21-22 (2009); Bell &
Dawson, supra note 172, at 555-56; James Bullard, Christopher J. Neely
& David C. Wheelock, Systemic Risk and the Financial Crisis: A
Primer, 91 FED. RES. BANK ST. LOUIS REV. 403, 407 (2009).
(177.) ERIK BANKS, THE CREDIT RISK OF COMPLEX DERIVATIVES 40 (3d
ed. 2004); FRANK J. FABOZZI & VINOD KOTHARI, INTRODUCTION TO
SECURITIZATION 5 (2008) (describing the need to “de-link … the
credit risk of the collateral asset pool from the credit risk of the
originator”). The International Swaps and Derivatives Association
(ISDA) estimates that some ninety-seven percent of over-the-counter
(OTC) credit derivatives are collateralized. Market Review of OTC
Derivative Bilateral Collateralization Practices, INT’L SWAPS &
DERIVATIVES ASS’N 6 (Mar. 1 2010),
(178.) See, e.g., EUR. CENT. BANK, supra note 176, at 48 (noting
“a sharp increase in the use of collateral in the last ten years:
according to the ISDA’s findings, two-thirds of the net credit
exposures derived from OTC credit derivatives were collateralised at the
end of 2008”).
(179.) Bliss & Kaufman, supra note 20, at 7.
(180.) See BANKS, supra note 177, at 22.
(181.) See id. at 395. Termination rights are enforceable under 11
U.S.C. [section] 560 (2006). Absent this safe harbor, the bankruptcy
trustee would have the right to “assume or reject any executory
contract” notwithstanding the debtor’s default. 11 U.S.C.
(182.) Bliss & Kaufman, supra note 20, at 6; see also Adam R.
Waldman, OTC Derivatives & Systemic Risk: Innovative Finance or the
Dance into the Abyss?, 43 AM. U. L. REV. 1023, 1058-60 (1994)
(describing the mechanics of netting). 11 U.S.C. [section] 560 (2006)
protects “[t]he exercise of any contractual right … to offset or
net out any termination values or payment amounts arising under or in
connection with the termination, liquidation, or acceleration of one or
more swap agreements.”
(183.) See 11 U.S.C. [section] 362(a)(7) (2006) (providing for a
stay of “the setoff of any debt owing to the debtor that arose
before the commencement of the [bankruptcy] case … against any claim
against the debtor”).
(184.) See Bliss & Kaufman, supra note 20, at 11-12.
(185.) Derivatives enjoy safe harbor under the Bankruptcy Code
“in three main areas: (i) allowing the enforceability of bankruptcy
termination or ipso facto clauses, (ii) exempting close-outs, setoffs
and foreclosure on collateral from the automatic stay, and (iii)
exempting payments made under them from preference and constructive (but
not actual) fraudulent transfer causes of action.” Vasser, supra
note 21, at 1509.
(186.) Act of June 25, 1990, Pub. L. No. 101-311, [section] 106,
104 Stat. 267, 268 (1990) (codified as amended at 21 U.S.C. [section]
560 (2006)); see H.R. REP. No. 101-484, pt. IV, at 5 (1990).
(187.) Act of June 25, 1990, [section] [section] 102-103, 104 Stat.
at 267-68 (codified as amended at 11 U.S.C. [section] [section]
(188.) Id. [section] 101, 104 Stat. at 267 (codified as amended at
12 U.S.C. [section] 101(53B)); see H.R. REP. NO. 109-32, pt. I, at 121
(189.) Act of June 25, 1990, [section] 101, 104 Stat. at 267
(codified as amended at 22 U.S.C. [section] 101(53C)).
(190.) Edward R. Morrison & Joerg Riegel, Financial Contracts
and the New Bankruptcy Code: Insulating Markets from Bankrupt Debtors
and Bankruptcy Judges, 23 AM. BANKR. INST. L. REV. 641, 642 (2005)
(“The reforms of 2005 … might appear only to eliminate
longstanding uncertainty surrounding the protections available to
financial contract counterparties…. But the ambit of the reforms is
(191.) See YOSHIZAWA, supra note 175, at 4.
(192.) See Bell & Dawson, supra note 172, at 556.
(193.) But see Beverly Hirtle, Credit Derivatives and Bank Credit
Supply, 18 J. FIN. INTERMEDIATION 125 (2009) (finding limited evidence
that derivatives expand credit supply).
(194.) See, e.g., H.R. REP. NO. 109-31, pt. I, at 125 (2005)
(“Express recognition of the enforceability of such cross-product
master agreements furthers the policy of increasing legal certainty and
reducing systemic risks in the case of an insolvency of a large
financial participant.”); Morrison & Riegel, supra note 190, at
644 (“By relying on broad market definitions, the  Act gets
judges out of the (largely futile) business of second-guessing financial
(195.) See generally Lehman Brothers, Sharper Image,
Bennigan’s, and Beyond: Is Chapter 11 Bankruptcy Working?: Hearing
Before the Subcomm. on Commercial & Admin. Law of the H. Comm. on
the Judiciary, 110th Cong. 8 (2008) (statement of Jay Westbrook,
Professor, University of Texas School of Law) (“Unfortunately, the
2005 amendments not only expanded the scope of the exemptions but it
made them much fuzzier, and much more ambiguous than they had been
before, so that now it is not clear exactly what a swap agreement is for
(196.) Aline van Duyn & Nicole Bullock, Lehman Ruling Creates
New Doubts for CDOs, FIN. TIMES, Feb. 9, 2010,
(“It had long been assumed that investors in structured deals-the
ones owning the notes–will get paid before swap counterparties
(197.) INT’L SWAPS & DERIVATIVES ASS’N & SEC.
INDUS. & FIN. MKTS. ASS’N, COMMENTS ON THE SECURITIES AND
FUTURES COMMISSION CONSULTATION PAPERS [paragraph] 6.10, at 15 (2009)
(“[I]t is not uncommon for documentation to provide for the
priority between the swap counterparty and the investors to be reversed
in the event that the swap counterparty is in default or
(198.) These render unenforceable any agreements purporting to
modify the debtor’s interests under an executory contract or in
property in the event of the debtor’s bankruptcy. See 11 U.S.C.
[section] 365(e)(1) (2006) (“[A]ny right or obligation under [an
executory contract] … may not be terminated or modified, at any time
after the commencement of the [bankruptcy] case solely because of a
provision in such contract … that is conditioned on … (B) the
commencement of a case under this title…. “); id. [section]
541(c)(1) (A debtor’s interest in property “becomes property
of the estate … notwithstanding any provision in an agreement … (B)
that is conditioned on … the commencement of a case under this title
… and that effects or gives an option to effect a forfeiture,
modification, or termination of the debtor’s interest in
(199.) Lehman Bros. Special Fin. Inc. v. Ballyrock ABS CDO 2007-1
Ltd. (In re Lehman Bros. Holdings Inc.), 452 B.R. 31, 40 (Bankr.
S.D.N.Y. 2011); Lehman Bros. Special Fin. Inc. v. BNY Corporate Tr.
Servs. Ltd. (In re Lehman Bros. Holdings Inc.), 422 B.R. 407, 421
(Bankr. S.D.N.Y. 2010). Relying on Judge Peck’s ruling in BNY,
Lehman Brothers’s trustee has filed suits claiming billions of
dollars in additional losses based on liquidation rights embedded in
synthetic transactions. See, e.g., Complaint, Lehman Bros. Special Fin.
Inc. v. Bank of Am. Nat’l Ass’n (In re Lehman Bros. Holdings
Inc.), No. 08-13555, Adv. No. 1003547 (Bankr. S.D.N.Y. Sept. 14, 2010) ;
Chelsea Emery & Jonathan Stempel, Lehman Seeks $3 Billion from CIBC,
Others in Lawsuits, REUTERS (Sept. 15, 2010, 10:25 AM),
(200.) See, e.g., Evan Jones et al., Lehman Bankruptcy Judge
Prevents Trigger of CDO Subordination Provision Based on Credit Support
Provider and Swap Counterparty/Bankruptcy Filings, 127 BANKING L.J.
(201.) See Alan Greenspan, Chairman, Bd. of Governors of the Fed.
Reserve Sys., Remarks Before the Futures Industry Association (Mar. 19,
/boarddocs/speeches/1999/19990319.htm (arguing that “It]his
unbundling improves the ability of the market to engender a set of
product and asset prices far more calibrated to the value preferences of
consumers than was possible before derivative markets were
developed”). But see Hirtle, supra note 193.
(202.) See Bell & Dawson, supra note 172, at 556-57 (“[T]o
the extent that the originator/protection buyer relies on the investor
to make a payment if a credit event occurs, the former needs some degree
of comfort that the latter is going to be good for the money. This
cannot be achieved with a traded instrument that may change hands at the
whim of its present holder.”).
(203.) Dodd-Frank Act [section] 713, Pub. L. No. 111-203, 124 Stat.
(204.) See Duffle & Skeel, supra note 28, at 13 (“By
‘clearing’ a derivatives contract, a [central clearing
counterparty] … becomes the counterparty to each of the two original
participants to the contract. That is, the [central clearing
counterparty] becomes the seller to each buyer, and the buyer to each
seller. The main purpose of clearing is to insulate the original
counterparties from counterparty default risk.”).
(205.) For a discussion of the interaction between mandatory
clearing and the derivative safe harbors, see id. at 13-17.
(206.) See Roe, supra note 29, at 555-60.
(207.) See Bliss & Kaufman, supra note 30, at 10 (“The
major problem stems from the fact that … the firm is in
liquidation…. Stays can suspend collection of debts but they cannot
force continued rolling over of funding or provision of
(208.) See Edwards & Morrison, supra note 29, at 94 (noting
this irony in the context of Long-Term Capital Management).
(209.) Critics have argued that curtailing the safe harbors could
enhance market discipline by forcing parties to manage their exposure to
potential losses stemming from their counterparties’ risk taking.
See supra note 31. Yet as Ricks observes, “market discipline by
money-claimants is incompatible with financial stability: runs and
panics are the very manifestations of market discipline by short-term
creditors.” Ricks, supra note 6, at 139.
(210.) See supra notes 62-63 and accompanying text.
(211.) Tirole, supra note 52, at 299 (citing HERNANDO DE SOTO, THE
MYSTERY OF CAPITAL: WHY CAPITALISM TRIUMPHS IN THE WEST AND FAILS
EVERYWHERE ELSE (2003)).
(212.) Historically, sudden credit expansions frequently fuel
credit bubbles followed by downturns in the credit cycle. See Allen,
supra note 154, at 7.
AUTHOR. Yale Law School, J.D. 2012. In preparing this Note, I have
benefited from invaluable discussions with Andrew Metrick, Roberta
Romano, Gary Gorton, Jonathan Macey, and Daniel Hemel, and from
challenging feedback offered at the Yale Law Journal Student Scholarship
Workshop. I owe special thanks to Glenn Bridgman and Amanda Lee of the
Journal’s Notes Committee for their thoughtful and tireless
commitment to this project. All errors are undoubtedly my own.