Journal of Economic Perspectives—Volume 24, Number 1—Winter 2010—Pages 73–92

Credit Default Swaps and the

Credit Crisis

René M. Stulz

C
redit default swaps are a subject of considerable ambivalence. On one side,
they seem like straightforward financial derivatives that serve standard
useful functions: making it easier for credit risks to be borne by those who
are in the best position to bear them, enabling financial institutions to make loans
they would not otherwise be able to make, and revealing useful information about
credit risk in their prices. On the other side, in trying to understand the credit
crisis, many observers have identified credit default swaps to be a prominent villain.
One segment of the “60 Minutes” television show on October 26, 2008, called credit
default swaps on subprime mortgages the “bet that blew up Wall Street.” Searching
the Internet on Google, a search under “worst Wall Street invention” came up with
credit default swaps as the first entry. George Soros, the prominent hedge fund
manager, and many others want most or all trading in credit default swaps to be
banned.
My focus in this paper is on how credit default swaps may have contributed
to the credit crisis. I first review the mechanics of credit default swaps in their
most straightforward use—providing insurance against the default of individual
companies—before turning to how they were used to take positions on subprime
mortgages. I examine the size and growth of the credit default swap market. I
then turn to arguments as to how credit default swaps may have contributed to the
crisis: that financial derivatives in general and credit default swaps in particular
enabled an unsustainable credit boom, excessive risk-taking by financial institu-
tions, and even market manipulation. I show how derivatives positions create a web

■ René M. Stulz is the Reese Chair of Banking and Monetary Economics, The Ohio State Uni-
versity, Columbus, Ohio, and Research Associate, National Bureau of Economic Research,
Cambridge, Massachusetts. His e-mail address is 〈[email protected]
[email protected]〉〉.
doi=10.1257/jep.24.1.73
74 Journal of Economic Perspectives

of linkages across financial institutions and assess the argument that this web made
the financial system significantly less safe and forced taxpayers to bail out financial
institutions. I then consider the claim that if credit default swaps were traded on an
exchange or through a clearinghouse, rather than over the counter through bilat-
eral contracts between sellers and buyers of protection, it would eliminate much
of the risk they might pose to the financial system. I conclude with some thoughts
about the difficulties of assessing the social costs and benefits of credit derivatives
in the aftermath of the credit crisis.

Credit Default Swaps on Companies

There is nothing particularly exotic about credit default swaps. They are as
easy to understand as insurance contracts. The best way to understand a plain
vanilla credit default swap is as an insurance contract against the cost of default of
a company, which is referred to as the “name” or the “reference entity.” Suppose
that you hold Ford bonds and are concerned about Ford’s default risk. You could
insure your bond holdings with a credit default swap. As with a typical insurance
contract, you pay premiums over time. If Ford does not default, you lose the premi-
ums. If Ford does default, the credit default swap allows you to exchange the Ford
bonds you hold, which are now worth little, for the principal amount of the bonds,
or alternatively, depending on the details of the contract, for a payment equal to
the principal amount of the bonds you hold minus their current value at the time
of default. Your Ford bonds could lose value even if Ford does not default—for
instance, if interest rates increase or Ford’s credit falls without a default—but you
only receive payment from a credit default swap in the case of an actual default
(and in the event of a debt restructuring for some contracts).
However, the parallel between insurance contracts and credit default swaps
does not hold in two important ways. First, you do not have to hold the bonds to
buy a credit default swap on that bond, whereas with an insurance contract, you
typically have to have a direct economic exposure to obtain insurance. Because you
don’t have to hold bonds, the amount you insure with a credit default swap is usually
called the notional amount. If you buy a credit default swap on Ford for a notional
amount of $100 million, you have insurance on $100 million of principal amount of
Ford bonds. Second, insurance contracts (mostly) are not traded; in contrast, credit
default swap contracts do trade over the counter—that is, a market where traders in
different locations communicate and make deals by phone and through electronic
messages. Dealers trade with end users as well as with other dealers.
The Depository Trust & Clearing Corporation (DTCC) keeps a record of out-
standing credit default swaps involving major dealers as counterparties.1 For the

1
Data on subprime credit default swap positions only became available late in fall 2008 through the
DTCC data warehouse. These data are incomplete. The DTCC estimated that less than 1 percent of
the credit default swaps registered with it involved the ABX as of early November 2008. It is possible
 René M. Stulz 75

week ending on May 15, 2009, for example, the DTCC had 5,387 credit default swap
contracts registered with it on Ford Motor Company, 1,583 on Ford Motor Credit
Company, and 4,649 on Ford Motor Credit Company LLC. The total notional
amount of credit default swaps on Ford Motor Company was for $36 billion. For
comparison, on December 31, 2008, the automotive sector of Ford had total debt of
$25.8 billion. It is not unusual for the total notional amount of credit default swaps
written on a name to exceed the total amount of debt issued by that name—a point
to which I will return.
There are also traded indices based on credit default swaps, which are aver-
ages of these contracts on different names. There are indices for corporates for
Europe (iTraxx Europe), the United States (CDX North America), as well as other
regions. The iTraxx Europe index, for example, represents a basket of 125 credit
default swaps. In addition to traded indices, customized indices of credit default
swaps are also available.
In principle, credit default swaps should make financial markets more efficient
and improve the allocation of capital. Historically, the investors who funded compa-
nies through debt had to bear the credit risk of these companies. Now, the investors
who provide the capital need not be those who bear the credit risk. Instead, credit
risk can reside with the investors who are best equipped to bear it. Separating
the cost of funding and the credit risk also introduces greater transparency in the
pricing of credit. Taken together, these benefits from credit default swaps should
reduce the cost of capital for firms.
For example, the credit default swap market turns out to be a better place to
assess a company’s credit risk than the market for that company’s bonds. An empiri-
cal study of how information gets incorporated by Blanco, Brennan, and Marsh
(2005) shows that information mostly flows from credit default swap prices to bond
prices. After all, abstracting from market frictions, the price of a credit default
swap is purely about the expected default loss and thus is not affected by all the con-
tractual provisions of a bond like covenants, coupon, maturity, and so on. Further,
liquidity should be less of a factor in the pricing and trading of credit default swaps
than of bonds because bonds involve funding. The credit default swap market for
a name can therefore often be more liquid than the market for the name’s bonds.
Finally, if you believe that a company’s risk of default is about to increase, it can
be difficult to sell short a company’s bonds, or certainly its loans. However, by buying
protection, you have the same economic benefit in the event of default as if you had a
short position in the bond. If you sell a bond short when it trades at 100 and it trades
at 50 after default, you earn 50 abstracting from the costs of selling short. If you buy
protection at 100, you receive 50 upon default, but have to pay the cost of protection.
Many economists argue that the existence of short-selling makes a market react more

that the size of the subprime credit default swap market was much larger because not all contracts are
registered with the DTCC, because many contracts may have been unwound by the time the DTCC
started reporting contracts, and perhaps because there were many contracts on subprime that did not
use the ABX indices. For example, AIG contracts involving subprime collateral not indexed to the ABX
would not be registered.
76 Journal of Economic Perspectives

quickly to new information (for a model, see Diamond and Verrecchia, 1987). But as
we will discuss later, some have argued that the ability to take short positions through
the credit default swap market has made this market destabilizing.
The greater efficiency of the credit default swap market in incorporating infor-
mation benefits the pricing of all securities of a firm. However, the separation of risk
bearing and funding made possible by credit default swaps can create problems as
well in incentives for monitoring and for working out situations of financial distress.
As an example of a problem with incentives for monitoring, consider a bank
that made a large loan to a firm and then buys credit default swap protection
against a default of that loan. If the bank has protected itself by buying protection,
its incentives to monitor the loan may have become less powerful. Of course, the
seller of protection cannot monitor the firm in the same way as the bank would
because it has no contractual relationship with the firm. As a result, there may be
too little monitoring of the firm.
However, in practice, banks have many reasons to monitor their borrowers so
that the effect of hedging on their monitoring incentives need not be material. More-
over, the ability of banks to hedge loans that they make also has benefits. For example,
banks can keep lending to firms with which they have close relationships, even when
they have already lent large amounts, because they can limit their risk exposure to
such firms through the use of credit default swaps. As a result, firms can get more
credit than they would otherwise receive and on better terms. In the past, the use
of credit default swaps by banks has been surprisingly limited. In Minton, Stulz, and
Williamson (2009), my coauthors and I show that only 23 U.S. bank holding compa-
nies had credit default swaps positions in 2005 and, using a proxy for hedging, that
they hedged on average 2 percent of their loans with these instruments. A possible
reason why banks’ use of credit default swaps to hedge is limited is that, while the
credit default swap market is typically quite liquid for large companies, it is usually
not liquid for the smaller companies to which banks make a lot of loans.
The availability of credit default swap contracts can change the incentives of
investors, too. Consider an investor who holds bonds of a company in financial dis-
tress. This company may approach the investor to suggest a restructuring of its debt.
The attitude of the investor towards the company’s proposal will depend on whether
the investor hedged his position through a credit default swap, as Yavorsky (2009) dis-
cusses in detail. Some credit default swap contracts treat a restructuring of debt as an
event that causes a payout; others do not. An exchange of new bonds for old bonds;
for example, will not trigger payment under a credit default swap. An investor in this
situation might prefer to drive the firm into bankruptcy, and thus trigger payments
under the credit default swap, rather than work out a refinancing plan.

Credit Default Swaps on Subprime Mortgage-Backed Securities

As with other mortgages, subprime mortgages are securitized: that is, the
mortgages are placed in a pool (typically set up as a trust) and notes are issued
 Credit Default Swaps and the Credit Crisis 77

against that pool. These notes, often called tranches, differ in their priority in
receiving payments. The most senior tranche has a first claim on interest payments
and mortgage payoffs. The super-senior notes always have a AAA rating. If and
when mortgages default, the lowest-rated tranches suffer first from the default
losses. As default losses mount, it becomes possible for the highly rated securities
to suffer from default losses as well. Ashcraft and Schuermann (2008) offer a more
detailed description of the securitization of subprime mortgages and the problems
that can arise.
Consider now super-senior AAA-rated debt issued against a pool of mortgages.
A financial institution holding that debt, who wishes to insure it, could do so by pur-
chasing protection through a credit default swap. However, a problem arises here.
The default of a debt-holder like Ford is a well-defined event, typically leading to
bankruptcy or restructuring. But when holding a tranche of subprime securitized
debt, a rising level of defaults on the underlying mortgages leads to a reduction in
debt payments, but the lower payments do not lead directly to a bankruptcy filing
and the debt keeps making payments.
Because of this difference, credit default swaps written on securitized debt
work differently from those written on corporate debt. Suppose that an inves-
tor holds a AAA tranche with a principal amount of $100 million and the other
tranches of the securitization have been wiped out; further, suppose that during
a month $1 million of mortgages default so that the principal balance falls from
$100 million to $99 million. At that time, the investor would be paid $1 million
from the credit default swap. Moreover, the credit default swap would still exist
after that payment and would make payments as further mortgages default until
maturity of the contract.
In 2006, the ABX indices on subprime securitizations were introduced,
representing a basket of credit default swap contracts on securitized subprime
mortgages. An index would be based on an average of credit default swaps for
same seniority securitization tranches. For instance, the AAA index for 2007-1 was
based on an average of individual credit default swaps on the largest AAA-rated
securitization tranches issued in the second half of 2006. In 2007, these indices fell
sharply, reflecting a loss in value of subprime securities. These indices introduced
greater transparency in the market for subprime debt as their trading facilitated
price discovery for that debt. The ABX indices made it possible for investors to
take views on the subprime market without owning subprime mortgages directly or
indirectly as well as to obtain insurance for subprime exposures. The indices also
made it possible for investors to take more exposure to subprime mortgages than
there were such mortgages.
Though credit default swaps based on subprime mortgages provided investors
with several valuable benefits, including improved price discovery and an ability
to hedge the risks of subprime mortgages, many questions have been raised about
whether the market for these instruments was efficient. For instance, the Bank of
England (2008) argues that the ABX indices overreacted to the troubles of the
subprime market. Future academic research will eventually show whether such
78 Journal of Economic Perspectives

overreactions occurred. For now, it’s clear that credit default swaps on complicated
debt instruments such as securitized subprime mortgages can be hard to price.
In principle, the hedging benefit of credit default swaps should have made it
possible for subprime risk to be located with those investors and institutions for
which bearing such risk was most efficient. However, there are two problems with
this simple view. First, the sellers of these credit default swaps, including some
specialty “monoline” insurance companies2 that had historically mostly insured
municipal bonds, as well as the well-known case of AIG, ultimately did not have
the ability to bear the risks they took on, so some of the hedging benefit of credit
default swaps turned out to be illusory (or would have turned out that way without
taxpayer support). Second, because of their built-in leverage, credit default swaps
may make it possible for investors to take riskier positions than they could other-
wise. To the extent that the most optimistic and least risk-averse investors may be
those whose investment opportunities are expanded by the availability of these
instruments, these instruments may lead to price distortions where risk is under-
priced. Before the credit crisis, the compensation required by investors to bear the
risk of high yield debt (so-called “ junk bonds”) was at historic lows. Future research
will hopefully help us understand whether the price of credit risk was ultimately
too low and whether credit default swaps played a role in making it too low if it
was. If these instruments contributed to a false sense of safety of investors through
hedges that were more imperfect than they thought and led to prices that under-
estimated risk, they may have led to an excessive build-up of subprime exposures.
Ultimately, however, such arguments rely not on the properties of credit default
swaps, but on market inefficiency brought about by limits of arbitrage—because
otherwise arbitrageurs could exploit any mispricing of risks.

The Size of the Credit Default Swap Market

Back in the mid-1990s, one of the first credit default swaps provided protection
on Exxon by the European Bank for Reconstruction and Development to JP Mor-
gan (Tett, 2009). It took months to negotiate. By 1998, the total size of the credit
default swap market was a relatively small $180 billion (Acharya, Engle, Figlewski,
Lynch, and Subrahmanyam, 2009). The credit default swap market has grown
enormously since then, although there is no definitive measure of how much.
Based on survey data from the Bank for International Settlements (BIS) at
⟨http://www.bis.org/statistics/derstats.htm
http://www.bis.org/statistics/derstats.htm〉〉, the total notional amount of the
credit default swap market was $6 trillion in 2004, $57 trillion by June 2008, and
$41 trillion by the end of 2008. Credit-default swap contracts that insure default
risk of a single firm are called single-name contracts; in contrast, contracts that pro-
vide protection against the default of many firms are called multi-name contracts.

2
They are so named because they provide only one type of insurance contract, that is they have only
one line of business.
 René M. Stulz 79

In 2004, single-name contracts were 80 percent of the credit default swap market;
at the end of June 2008, these single-name contracts were only 58 percent of the
market.3 Based on data from the DTCC, the size of the credit default swap market
was $29 trillion on May 22, 2009. It’s possible that the survey measure from the BIS
inflates the size of the market somewhat by leading to some double counting; it’s
also likely that because not all contracts are registered with the DTCC, the DTCC
underestimates the size of the market to some extent. Of the $29 trillion of credit
default swaps registered with the DTCC on May 22, 2009, $15 trillion were single-
name swaps.
As with all derivatives, the total notional amount outstanding of credit default
swaps and the market value of such contracts differ considerably. In the credit
default swap market, for each buyer of protection, there is a corresponding seller
of protection. From that perspective, the total market value of outstanding credit
default swaps is zero.
Moreover, while the notional value of credit default swaps is established at the
time they are created, the market value of the protection bought through credit
default swaps varies with market conditions. At inception, a credit default swap’s
value is zero for the protection buyer because the value of the protection obtained
is equal to the present value of the payments the protection buyer will have to
make. The value of the credit default swap subsequently falls if default becomes less
likely and increases if default becomes more likely. The value of the credit default
swap depends on many factors, though. For instance, the protection buyer could
make a profit even though the probability of default stays unchanged because the
amount expected to be recovered in the event of default falls so that protection
becomes more valuable.
Table 1 shows the evolution of the market value of credit default swap contracts
from 2004 to the end of 2008. In 2008, the market value of credit default swaps fell
when measured using the total notional amount of the contracts, but it almost tripled
when measured using the market value of the outstanding swaps. Such an evolution
is not surprising because default risks increased for many companies in 2008.

Counterparty Risks and the Financial Crisis

In many ways, the credit default swap market worked remarkably well during
much of the credit crisis. Despite huge and unexpected losses in underlying mort-
gage securities and near chaos in the financial sector at times, the credit default
swap market remained fairly liquid for long periods over the last two years. Further,
the market handled extremely large defaults efficiently. A good example is how well
it processed the default of Lehman.

3
The International Swaps and Derivatives Dealers Association (ISDA) conducts a survey of the credit
default swaps market as well. Its estimate for mid-year in 2008 is slightly lower and shows a decrease
from the end of 2007.
80 Journal of Economic Perspectives

Table 1
Credit Default Swaps (CDS) Notional and Market Values Outstanding

Notional amount outstanding Gross market values

(in billions of dollars) (in billions of dollars)

Single-name Multi-name Single-name Multi-name

Period CDS CDS All CDS CDS CDS All CDS

12/31/04 5,117 1,279 6,396 112 22 133

12/31/05 10,432 3,476 13,908 171 71 243
12/31/06 17,879 10,771 28,650 278 192 470
12/31/07 32,246 25,648 57,894 1,143 859 2,002
06/30/08 33,334 23,991 57,325 1,889 1,283 3,172
12/31/08 25,730 16,138 41,868 3,695 1,957 5,652

Source: Bank for International Settlements.

The notional amount of protection bought on Lehman was unclear at the

time of the bankruptcy. Estimates for the total notional amount of credit default
swaps written on Lehman ranged from $72 billion to $400 billion.4 The lower-end
estimate is firm: the DTCC had contracts on Lehman for a notional amount of
$72 billion registered in its warehouse. Protection sellers had to pay 91.375 cents on
the dollar to settle the contracts. The settlement for these contracts went smoothly.
The net exchanges of cash for Lehman, despite wild rumors in the markets, were
rather small: $5.2 billion were exchanged through the DTCC. Many institutions
were both buyers and sellers of protection on Lehman, which contributed to keep-
ing the net positions small. As mentioned before, not all contracts are registered
through the DTCC, and surely some additional contracts referencing Lehman
existed, but there is no evidence that the additional contracts were especially dif-
ficult to settle either or that parties defaulted on these contracts.
If the credit default swap market worked well, why is it considered to have been
so dangerous? Credit default swaps were clearly part of the story of how banks and
other financial institutions ended up holding mortgage securities on which they
made large unexpected losses. Because of the way capital requirements are deter-
mined, financial institutions generally were able to hold less regulatory capital if
they packaged loans in securities and held them on their balance sheet than if
they just kept the loans on their balance sheet (for examples, see Goldman Sachs,
2009). Further, some financial institutions apparently believed that it was advanta-
geous for them to hold super-senior tranches of securitizations on their books if
they insured them with credit default swaps. Regulators across countries allowed
financial institutions to set aside less capital because these institutions had bought
protection through credit default swaps. There was therefore a large demand for

4
The $72 billion figure is the amount reported by the DTCC for swaps that settled through the DTCC.
The $400 billion figure was reported by the Financial Times on October 6, 2008, quoting a Citi analyst
stating that “there could be $400bn of credit derivatives referenced to Lehman”(FT.com, 2008).
 Credit Default Swaps and the Credit Crisis 81

insurance of super-senior tranches that was partly met by credit default swaps
from AIG. However, the losses on credit default swaps referencing subprime mort-
gage securitizations came about because of defaults on subprime mortgages and
because of disappearing liquidity for such securitizations. The credit default swap
market caused neither the mortgage defaults nor the disappearance of liquidity.
Though some market participants were surely too optimistic about the prospects of
the subprime market, credit default swaps on subprime securitizations cannot be
blamed for that excessive optimism. In fact, it is more likely that the ABX indices
made it harder for investors to remain excessively optimistic.
Many observers have focused on problems caused by counterparty risk in argu-
ing that derivatives and especially credit default swaps made the credit crisis worse.
The argument has two parts. First, derivatives lead to a huge web of exposures
across financial institutions. If an institution fails in this web of exposures, it can
lead other institutions to fail as they make losses on their exposures. As a result,
this web of exposures could lead to a collapse of the financial system and to con-
siderable uncertainty about the solvency of financial institutions in the event of the
failure of a major financial institution. Second, credit default swaps heighten this
concern because their value jumps, and often by large amounts, when a default
occurs. I examine these arguments in turn.
When Lehman failed, it had close to one million derivatives contracts on
its books with hundreds of fi nancial fi rms. Some of these fi rms expected to
receive payments from Lehman on their derivatives. Suddenly, Lehman was no
longer in a position to make these payments because it had fi led for bankruptcy.
One might therefore be concerned that these fi rms became fi nancially weaker,
leading to contagion of Lehman’s problems through losses on derivatives con-
tracts because of the failure of a counterparty. However, the typical derivatives
transaction uses protections against the risks of a counterparty not meeting its
obligations. The biggest protection is generally the use of collateral, and usually
the amount of collateral insuring a counterparty’s performance on a contract
changes with the value of the contract. Consider bank Y that had derivatives
positions with Lehman which cost $100 million to bank Y to replace. If bank
Y had $110 million of collateral from Lehman when it failed, bank Y could use
the collateral and make no loss from Lehman’s failure. Alternatively, had it had
collateral for $90 million, it would have made a $10 million loss if it had no recov-
ery from the bankruptcy estate. While collateral arrangements were frequent,
they were not universal. According to a survey by the International Swaps and
Derivatives Association, 63 percent of derivatives contracts were subject to such
agreements in 2007, compared to 30 percent in 2003. Consequently, there is still
a possibility of contagion through derivatives exposures. However, that possi-
bility is limited by the incentives of counterparties to manage their exposures
actively as counterparty risk changes and by the fact that parties not subject to
collateral arrangements are often very highly rated counterparties. At the same
time, however, a failure of a fi nancial institution can lead to large changes in
derivatives prices as well as in derivatives liquidity, so that the collateral amounts
82 Journal of Economic Perspectives

held immediately before the failure may not be sufficient to cover possible losses
if other counterparties default.
Another issue with credit default swaps is that because a default is a discrete
event, it can lead to large jumps in the value of these contracts. To see this, sup-
pose that the market expects that there is a 20 percent chance a dealer will default
and the recovery is expected to be 40 percent. In the event of default, the value
of the bonds falls to 40 percent, the recovery value, so that the bondholder loses
60 percent. The credit default swap pays the 60 percent. Ignoring the time value of
money and risk premia, the value of a $10 million notional credit default swap for
the protection buyer would be $1,200,000 (there is a 20 percent chance of receiving
a payout of (1.00 – .40) × $10 million). At default, the value of the credit default
swap would be $6 million. The protection seller would lose $4.8 million on the day
of default. Such losses could possibly lead to default by some other party that has
a large net exposure as a protection seller. For example, on the last working day
before Lehman’s bankruptcy filing, it cost roughly $700,000 to insure $10 million
of Lehman debt for a year, so that a buyer of protection against Lehman on that
day would have earned a huge gain since the swap paid off more than $9 million
on settlement. With such jumps to default, collateral will not be enough to protect
buyers of protection in the event of a counterparty default, which could then lead
to additional failures of financial institutions.
Another reason for concerns about the credit default swaps market is
the sheer size of gross exposures of dealers. In 2008, the credit default swap
contracts outstanding of JPMorgan Chase had a notional amount of almost
$8 trillion, and those of Citibank almost $3 trillion. Investment banks did not
provide as much information about their derivatives exposures, but the credit
default swaps of Bear Stearns may have amounted to a total notional amount of
$2.25 trillion (Madigan, 2008). Under normal circumstances, these gross expo-
sures are not much of a problem. The market value of the credit default swaps
of JPMorgan Chase, for instance, was estimated to be just $44 billion—and
even that amount substantially overstated the exposure of JPMorgan because it
ignores netting agreements and that the bank has collateral for a majority of its
contracts. In contrast, JPMorgan Chase’s shareholder equity at the end of 2008
was $166 billion.
However, even if a dealer’s net derivatives receivables are zero, the dealer
might still pose significant risks to the fi nancial system. Consider a dealer who
has $1 trillion notional of protection bought and $1 trillion notional of protec-
tion sold. Thus, this dealer has $2 trillion of gross exposure, but the net amount
is $0. Moreover, suppose that all the dealer’s contracts have collateral agree-
ments where the collateral changes daily as the market value of the contracts
changes (a feature called mark-to-market), so that those who are on track to
lose from the trade must post collateral as these losses accumulate. Even in
this case, a default of this hypothetical dealer still has the potential to create
havoc in the fi nancial markets. If a major dealer defaults, counterparties to the
dealer have to replace the credit default swaps. This process can take time and
 René M. Stulz 83

can be costly, especially if the dealer’s collapse renders the market less liquid or
even dysfunctional. As a result, counterparties to the defaulting dealer can be
exposed to risks over some period of time, which could lead to further defaults
and instability.
Though Lehman was a big dealer in credit default swaps, these contracts
were not the cause of Lehman’s failure. Neither were they the direct cause of Bear
Stearns’s demise. Lehman and Bear Stearns were dealers, and in credit default
swaps, their books were largely balanced and collateral arrangements were in
place. Both Bear Stearns and Lehman Brothers failed because market participants,
rightly or wrongly at the time, believed that there was a high probability that the
assets of these institutions were worth less than their liabilities. It is undoubtedly
true that without derivatives, their assets and liabilities would have been quite
different, but derivatives were not the proximate cause of their collapse.
AIG is a different and more complex story. Exposure to credit default swaps
did play a big role in AIG’s failure, but it’s worth noting that AIG did not behave like
a dealer. It did not run a matched book. It did not appear to hedge significantly.
What AIG did was provide credit default swaps on AAA tranches in securitizations
on an extremely large scale. As of June 30, 2008, it had written a net amount of
$411 billion notional of credit derivatives on super senior tranches of securitizations.
Included among these were derivatives on super-senior tranches with subprime col-
lateral for a notional amount of $55.1 billion. At the time that AIG wrote the credit
protection, all the tranches were rated AAA. The probability of a default on an
AAA-rated obligation is in principle extremely small, less than 0.1 percent per year.
However, with the major downturn in the U.S. housing market, these tranches lost
substantial value and the credit default swap liability of AIG became very large.
As losses mounted and the company’s credit rating dropped, AIG needed to post
ever more collateral until it did not have the cash to post the collateral amounts its
agreements required. Importantly, AIG could not meet its obligations not because
of realized losses on its credit-default swaps (that is, not because of payouts on the
contracts because of defaults) but because of collateral arrangements that required
posting of collateral because its credit rating was downgraded.
But even in the case of AIG, credit default swaps were not the only or even
the primary reason for its problems—nor were its credit default swaps the only or
even the primary reason why the firm was bailed out. AIG didn’t just write pro-
tection on subprime securitizations, it also borrowed heavily to purchase these
securities on its own. In fact, AIG made even larger losses on its portfolio of mort-
gage-related securities than on its credit default swaps. It’s true that the danger of
an AIG default on its credit default swaps was of concern to many financial insti-
tutions, which as noted before had been encouraged by regulators to purchase
such protection. But many financial institutions would also have been largely
protected by collateral agreements and by purchases of protection on AIG. An
additional danger of an AIG default was that AIG would have defaulted on its
debt and commercial paper at a time in September 2008 when there already was
a run on money markets.
84 Journal of Economic Perspectives

Transparency, Instability, and Manipulation

One of the main arguments for financial derivatives like credit default swaps
is that by enabling the trading of specific risks, they help make financial markets
more efficient and transparent in price discovery and increase liquidity. However,
in the fall of 2008, many executives were complaining that the market for credit
default swaps was being manipulated. Of course, it is very difficult to manipulate
profitably a highly liquid market through trades, but in the fall of 2008, many
financial markets were not always liquid. Thus, the accusation was that few well-
placed trades in the name’s credit default swaps could give the impression that
the name was in trouble, which would drive down the name’s stock price and debt
prices. The manipulator could then benefit by having established short stock and
debt positions. Financial institutions could be especially vulnerable to such actions,
because they are susceptible to runs.
There were extreme movements in credit default swap premiums in the fall
of 2008. The peak cost of insuring Morgan Stanley’s debt was roughly 1,500 basis
points per year (in other words, to insure $100 principal amount of debt, you would
have to pay $15 per year). Even firms like Berkshire Hathaway experienced sharp
increases in the cost of protection—from early September to mid-November 2008,
the cost of insuring Berkshire Hathaway’s debt increased from 140 basis points a
year to 415 basis points a year, apparently on rumors that a particular derivatives
bet that the company had made could turn out to be hugely expensive.
However, despite all the talk of manipulation, the Securities and Exchange
Commission has as of yet filed no action. This may mean either that no manipu-
lation occurred or that in an over-the-counter market it is it too difficult to find
evidence of manipulation. After all, regulators can only investigate manipulation
if they can find who traded what and when.5 The fact that traders in credit default
swaps exchange a lot of information during the day is not evidence of manipu-
lation. Dealer markets work through traders talking to each other! It was clear
in 2008 that regulators had an insufficient understanding of dealers’ derivatives
exposures and that a better understanding would have been beneficial.
It is conceptually important here to separate transparency for market partici-
pants from transparency for regulators. It would not be in the interests of financial
institutions to be too transparent about their derivatives positions. Such trans-
parency could make it difficult for an institution to trade, to provide liquidity to
clients, or to take advantage of its views on the market. In addition, there have
been calls for limiting or banning “naked” positions in these securities—that is,
situations where an investor buys protection without owning the underlying bonds
or securities. Naked positions are controversial because they enable investors effec-
tively to sell debt short. Financial economists generally believe that short-selling

5
Trade reporting could also help in identifying potential insider trading, as the literature suggests
that investors at times can use the credit default swap market to exploit insider information (Acharya
and Johnson, 2007).
 Credit Default Swaps and the Credit Crisis 85

helps efficiency, but neither the theoretical case for this belief nor the empirical
evidence are unambiguous; for example, Khanna and Matthews (2009) show con-
ditions under which manipulation through short-sales can succeed and make the
market inefficient. However, prohibiting naked positions in credit default swaps
would essentially destroy this market. If the credit default swaps market is reduced
to having only hedgers, with speculators banned, hedgers will not find counterpar-
ties because the market will have no liquidity. Speculators have to be able to trade
on either side of a market for there to be trading in that market. Prices cannot
be efficient if investors who see profit opportunities cannot exploit them. Dealers
have to be able to offset their positions to manage their risks. The credit default
swap market would stop being a source of credit information and a means of credit
hedging if buyers of protection could only do so if they owned the underlying
bond. There is no evidence I know of which suggests that removing naked buying
of credit protection—which is equivalent to selling short a name’s bonds—would
help the economy any more than attempts to reduce stock short-sales did during
the crisis. Most likely, once the evidence is in, we will find out that the attempts to
reduce short-sales of stocks hurt the stock market and the economy and worsened
the credit crisis.

How Over-the-Counter Markets and Exchange Trading Work

A popular proposal for reforming the market for credit default swaps is to
move trading away from the over-the-counter market, and instead use exchange-
based trading. Proponents of such a move argue that exchange trading could
greatly reduce the problem of counterparty risk, as well as create greater trans-
parency and order in the market. To get a handle on this argument, this section
describes in some detail just how an over-the-counter deal for a credit default swap
works, and contrasts it with how exchange trading would work. The next section
then evaluates the benefits of over-the-counter markets and exchange trading. It
turns out that these are not mutually exclusive options, but rather both choices
exist in a number of markets. In addition, there is a third choice—greater use of
clearinghouses without exchange trading—which has benefits of its own.
To understand the benefits and costs of over-the-counter trading for credit
default swaps, let’s start with an example of how it works. Suppose that you are
a hedge fund manager who wants to purchase a five-year credit default swap on
bank X. You call up dealers like Goldman Sachs, Deutsche Bank, and Natixis to
obtain quotes. (There are also some fully automated platforms on which you could
trade.) Say Natixis offers the best deal. You agree to make regular payments, say
quarterly, at an annual rate of 100 basis points on a notional amount of $10 million.
Until March 2009, credit default swaps were priced so that the market’s assessment
of the present value of the payments the buyer of protection expected to make
roughly equaled the present value of the regular premium payments he expected
to receive in the event of a default of bank X. There has been a push by regulators
86 Journal of Economic Perspectives

and by market participants to standardize credit default swap contracts to reduce

the operational risks associated with trading these contracts. In particular, since
March 2009, credit default swaps originated in North America have a fi xed pay-
ment of either 100 basis points or 500 basis points per year.6 If the net present
value of the protection you receive on company X is positive when you pay 100 basis
points per year, then you have to make an initial payment to the seller of protection
corresponding to that net present value.
Immediately after you enter the swap, Natixis has an exposure to bank X.
Natixis would therefore immediately look for ways to hedge or transfer its expo-
sure. If Natixis doesn’t already have some risk exposure that your swap offsets, most
likely Natixis would buy protection on bank X from some other dealer, say UBS,
who might also turn around and buy protection, say from Goldman Sachs. Natixis
might also implement hedges for its whole portfolio of credit default swaps, instead
of attempting to match each individual credit default swap with one that is offset-
ting. Eventually, however, somebody has to bear the risk from the credit default
swap you entered.
Continuing this example, suppose that another hedge fund, Contrarian, now
wants to sell protection on bank X and agrees to sell protection on bank X to
Goldman Sachs. When this happens, in a sense, you, the hedge fund manager, are
buying your protection on bank X from Contrarian. However, three intermediar-
ies make this happen—namely Natixis, UBS, and Goldman Sachs—and you never
know that Contrarian is the ultimate seller of protection. More specifically, in this
example, there are four credit default swaps of $10 million notional each, which is
a way in which the over-the-counter market makes the ultimate size of the market
appear larger than in some sense it truly is.
Now the trade has to clear, which means that both you and Natixis have to
ratify the agreement formally. The trader at Natixis will be off to other trades
immediately after talking to you. Back offices now get involved. However, when the
back office from Natixis contacts your back office, due to some misunderstanding
they may believe they have a contract on $20 million notional instead of $10 million
notional. At that time, this disagreement would have to be resolved. The “clear-
ing process” for derivatives consists of all the steps that take place from the trade
completion to the settlement. The “settlement” occurs when a party receives cash
from the counterparty to fulfill the obligation agreed to through the trade.
It should be immediately clear that a disorganized clearing process can cre-
ate substantial risk. Suppose that for whatever reason confirmation does not take
place. It could be that the back office of Natixis was too busy and forgot or that the
trader forgot to inform the back office properly. When the market for credit default
swaps expanded rapidly in the mid-2000s, a number of banks and dealers skimped
on their back offices, and there was a pervasive if usually low level of uncertainty

6
A number of changes took place in North America for credit default swap contracts in April 2009.
These changes are generally described under the name of the “CDS Big Bang” (Markit, 2009). The
objective of these changes was to standardize the single-name credit default swap contracts.
 René M. Stulz 87

as to the status of credit default swaps. The New York Fed worked hard to get the
industry to solve these problems and had significant success. That success had
limits: after Bear Stearns was acquired by JPMorgan Chase, its new owners discov-
ered a large amount of unconfirmed credit default swaps (Tett, 2009, p. 224).
The run-of-the-mill outcome here is that after you enter the contract, you hold
on to the contract until either bank X defaults or the credit default swap matures.
But two other possibilities are interesting: 1) you might decide you want to exit
the contract or 2) your counterparty, in this case Natixis, might become unable to
honor its contract.
If you, the hedge fund manager, wish to exit your credit default swap position,
there are at least three ways to proceed: 1) go to Natixis and negotiate terms for ter-
mination, which may involve payments depending on how the market has evolved
since the agreement was made; 2) enter into a contract to sell protection in a way
that exactly offsets your original contract; or 3) enter an agreement with a dealer,
with appropriate payments, that this dealer will take on your obligation to Natixis.
Such an agreement is called a novation, and Natixis would have to agree to such a
change. A sign that Bear Stearns’ situation was desperate was when counterparties
to Bear Stearns wanted to novate their trades and eventually they could not find
dealers willing to take on Bear Stearns as a counterparty.7
Now look at the case of counterparty risk. If Natixis fails, the swap is termi-
nated. If bank X’s credit worsened, you would have gained on your swap (you have
a claim on the bankruptcy estate). If bank X’s credit improved, on the other hand,
most likely you will owe to the bankruptcy estate of Natixis. Irrespective of whether
you gained or lost, the precise quantification of the gain or loss can be complicated.
To be in the same situation as you were in before the termination, you would have
to replace the swap, which would involve costs that you would want to be compen-
sated for by Natixis. A well-established solution to address the issue of counterparty
risk is an agreement to post collateral as the market value of the swap evolves. If you
suffered a loss in excess of the collateral you had in your possession, you would have
a claim against the bankruptcy estate of Natixis.
How would your trade in the credit default swap have worked differently had
the credit default swap traded on an exchange? Derivatives are standardized on
exchanges. Consequently, you would have had to choose a contract that is available
on an exchange. You would have placed an order with a broker to open a credit
default swap position. On the exchange, your trade would have taken place when
somebody else would have been willing to take the opposite position. In contrast
to the over-the-counter market, your counterparty would not be a dealer, but it
would be the clearinghouse of the exchange; in other words, the promise of your
credit default swap contract would be honored as long as the clearinghouse has the
resources to do so.

7
See Cohan (2009, pp. 27–30) for a description of discussions at Goldman Sachs about whether to
novate a trade in March 2008 for a hedge fund that had Bear Stearns as a counterparty.
88 Journal of Economic Perspectives

The ability of the clearinghouse of the exchange to honor the promises of

the contracts it clears depends on its resources, including its capital, and on its
risk management. Clearinghouses use margin agreements to reduce their risk.
With these agreements, each contract is marked to market each day, and the
gains and losses are settled. Through judicious use of these margin levels, a
clearinghouse can make its risk of default low. There would be no chain of con-
tracts if the contracts were traded on exchanges. Trade prices on exchanges are
publicly available, so that there is transparency. In addition, regulators could
identify the counterparties to trades through the clearinghouse should they
need to do so.

Tradeoffs between Over-the-Counter Trading, Exchange Trading,

and Clearinghouses

When the first credit default swaps were introduced in the 1990s, or the first
interest-rate swap agreement was introduced in the early 1980s, each new deal
took much time and effort. Eventually a modest degree of standardization came to
exist, even in over-the-counter trading. For example, as interest rate swaps became
better known, the industry formed the International Swaps Dealers Association
(ISDA), which devised standardized agreements. 8 When parties trade derivatives,
they enter a so-called ISDA Master Agreement. The Master Agreement has many
options, each with standard forms. However, the Master Agreement still makes it
possible to have infinite variations of amounts and maturities counterparties can
choose. They can also choose new forms of derivatives and combine derivatives as
they see fit.
In contrast, exchanges do not typically let derivatives traders set the terms of
the contracts. Instead, derivatives traders have a choice of contract terms and can-
not depart from these terms.
Why are exchanges inflexible while over-the-counter markets are so flexible?
Exchanges create pools of liquidity by standardization—they have few contract
types trading. With this standardization, investors and firms give up the opportu-
nity to obtain a contract that exactly fits their needs for the benefit of trading in a
liquid contract.
Exchanges offer an efficient solution to matching buyers and sellers when
they succeed in drawing large pools of liquidity. However, creating such pools for
derivatives can be difficult, because there is often a demand for terms that meet
specific hedging needs. Consider a manufacturing firm that wants to sell forward
its anticipated euro receipts from exports to Germany. On exchanges, the con-
tracts for future delivery of currencies (futures contracts) mature at specific dates,
which might not match the dates when the actual payments are expected. How-
ever, an over-the-counter dealer could offer a contract that matures on the day that

8
ISDA eventually renamed itself the International Swaps and Derivatives Association.
 Credit Default Swaps and the Credit Crisis 89

the exporter expects to receive the euros and for the exact number of euros the
exporter expects to receive. The manufacturing firm can choose to pay extra for a
contract that exactly meets its needs or use the exchange.
One benefit of exchange trading is the existence of a clearinghouse that
becomes counterparty to all trades. Clearinghouses are also used for some
derivatives trades in the over-the-counter market. Indeed, regulators have been
pushing hard for the use of clearinghouses as counterparties for credit default
swaps both in the United States and in Europe. Ice Trust, part of the Interconti-
nental Exchange, started clearing credit default swap index contracts in March
2009. By August 2009, the open interest on contracts cleared through Ice Trust
was in excess of $180 billion and Ice Trust had cleared more than $1 trillion
notional amount of contracts.
Use of clearinghouses for over-the-counter derivatives trading could decrease
the risks posed by derivatives exposures for the financial system for several reasons.
First, a clearinghouse can diversify and manage risks associated with the failure
of individual counterparties, so that counterparty risk is reduced. Second, if a
dealer uses a single clearinghouse, that clearinghouse can net out all of a dealer’s
exposures, which also reduces counterparty exposure. Third, a clearinghouse can
monitor the exposures of its counterparties and can prevent counterparties from
taking additional exposures.
The use of clearinghouses is not a panacea to eliminate systemic risk associ-
ated with over-the-counter trading of derivatives (Pirrong, 2009). A dealer who
trades hundreds of times a day with another dealer most likely will have a better
assessment of the credit of that dealer than would a clearinghouse that interacts
with the dealer for only one type of derivative trade. In addition, a clearinghouse
is inefficient at dealing with products that are not very liquid—which means most
new financial products and customized derivatives. The resources of clearing-
houses are limited. For instance, CME Clearing, the largest futures clearinghouse
in the United States, can draw on resources of $64 billion to cope with failures,
which might not have been enough to deal with the problems of a huge firm like
AIG. Finally, if the economy evolves toward multiple clearinghouses and clearing-
houses specialized to derivatives types, it is even possible that the netting that takes
place through clearinghouses is less than the netting that would take place without
clearinghouses (Duffie and Zhu, 2009).
In summary, the over-the-counter market is better at enabling innovation, at
addressing specific derivatives requirements from end-users, and at finding coun-
terparties when liquidity for a derivative on an exchange would be low. In contrast,
exchanges are more efficient when there is a large volume of trading for standard-
ized contracts. The over-the-counter market can compete well with exchanges
precisely because of this customization. For example, there is a huge over-the-
counter forward currency market—a market for the purchase of foreign currencies
for future delivery—and at the same time a large parallel currency futures market
on exchanges. The forward and futures currency markets have co-existed for more
than 30 years.
90 Journal of Economic Perspectives

Conclusion

Economists have generally believed that financial derivatives increase economic

welfare by facilitating risk-sharing among investors, by improving price discovery,
and by making the allocation of capital more efficient. These arguments certainly
apply to credit default swaps. However, as we have seen repeatedly in this paper,
there are legitimate reasons to be concerned about potential problems that can be
created because of exposures to derivatives and because of the trading of deriva-
tives. In the aftermath of the financial crisis, credit default swaps and other financial
derivatives have clearly lost any presumption of innocence that they once enjoyed
among economists—and they probably never had such a presumption with the gen-
eral public. But it would be premature and quite misguided to turn 180 degrees
from a presumption of innocence to a presumption of guilt. There is a dearth of
serious empirical studies on the social benefits and costs of credit default swaps and
other derivatives—not just in the last two years, but in the last several decades.
My own sense is that the deep dramatic problems of the financial credit crisis
were not caused by credit default swaps, nor by other financial derivatives. Neither
Bear Stearns nor Lehman failed because of derivatives. AIG lost money by selling
unhedged credit default swaps, but it also lost money in all kinds of other ways,
including by borrowing money to buy super-senior AAA-rated tranches of sub-
prime-mortgage-backed securities. The common denominator of the large losses
of AIG was that they occurred on subprime exposures and hence were brought
about by a dramatic unexpected fall in house prices.
In my view, the financial crisis was primarily driven by two factors. First, inves-
tors and financial institutions generally did not expect that real estate prices would
fall dramatically. This dramatic fall in real estate prices led to large defaults on sub-
prime mortgages and large falls in value in securitizations of subprime mortgages.
The fall in value was especially dramatic for AAA-rated tranches of subprime secu-
ritizations even though few of these tranches have suffered from defaults so far
(AAA-rated tranches of securitizations of securitizations, called collateralized debt
obligations, have suffered default losses). The second factor is that many financial
institutions were operating with extremely high levels of leverage and held large
investments in subprime securitizations, so that significant unexpected losses on
these investments could quickly lead market participants to question their solvency,
which led to cash hoarding by these institutions, to fire sales of assets to bring
about decreases in leverage, and to a contraction in their willingness to lend.
As these events unfolded, financial derivatives like credit default swaps were
associated with losses and uncertainty at some institutions, but also enabled other
institutions to hedge and hence to reduce the impact of the fall in subprime mort-
gage and other securities. Rather than blaming derivatives markets such as the
credit default swap market for being too large, it might make as much sense to
regret that derivatives markets were not larger. For instance, it may well be that
more robust derivatives markets in housing would have produced useful informa-
tion for investors that would have changed the evolution of housing markets and
 René M. Stulz 91

averted or minimized the effects of a crash by enabling investors to hedge against

drops in house prices.
However, until a significant empirical literature develops on the costs and ben-
efits of credit default swaps and other financial derivatives in the last few decades,
it will be difficult for financial economists who view derivatives to be valuable to
convince their colleagues—and a large share of the public—that such derivatives
have contributed significantly to social welfare and played a positive role in the
robust economic growth of the last 30 years.

■ I am grateful for comments from Viral Acharya, David Autor, Harry DeAngelo, Bernadette
Minton, Til Shuermann, Andrei Shleifer, and Timothy Taylor. I thank Mike Anderson and
Jérôme Taillard for assistance.

References

Acharya, Viral V., Robert F. Engle, Stephen Fi- Econometrica, 73(6): 1815–47.
glewski, Anthony W. Lynch, and Marti G. Subrah- Duffie, Darrell, and Haoxiang Zhu. 2009.
manyam. 2009. “Centralized Clearing for Credit “When Does a Central Clearing Counterparty
Derivatives.” Chapter 11 in Restoring Financial Stabil- Reduce Counterparty Risk?” Stanford University
ity: How to Repair a Failed System, ed. Viral V. Acharya Graduate School of Business Research Paper 2022.
and Matthew Richardson. New York, NY: Wiley. Goldman Sachs. 2009. “Effective Regulation–
Acharya, Viral V., and Timothy C. Johnson. Part I: Avoiding another Meltdown.” http://www2
2007. “Insider Trading in Credit Derivatives.” Jour- .goldmansachs.com/ideas/global-markets-institute
nal of Financial Economics, 84(1): 110–141. /featured-research/effective-reg-part-1.pdf.
Aggarwal, Rajesh, and Guojan Wu. 2006. Goldstein, Michael A., Edith S. Hotchkiss,
“Stock Market Manipulations.” Journal of Business, Erik R. Sirri. 2007. “Transparency and Liquidity:
79(4): 1915–53. A Controlled Experiment on Corporate Bonds.”
Ashcraft, Adam B., and Till Schuerman. Review of Financial Studies, 20(2): 235–73.
2008. “Understanding the Securitization of Sub- Gorton, Gary B. Forthcoming. “The Subprime
prime Mortgage Credit.” FRB of New York Staff Panic.” European Financial Management.
Report 318, Federal Reserve Bank of New York. Khanna, Naveen, and Richmond D. Matthews.
Bank of England. 2008. Financial Stability 2009. “Bear Raids and Short Sale Bans: Is Govern-
Report, Issue 23. ment Intervention Justifiable.” Available at SSRN:
Blanco, Roberto, Simon Brennan, and Ian W. http://papers.ssrn.com/sol3/papers.cfm?abstract
Marsh. 2005. “An Empirical Analysis of the Dy- _id=1434387.
namic Relation between Investment-grade Bonds Madigan, Peter. 2008. “JP Morgan CDS Expo-
and Credit Default Swaps.” Journal of Finance, sure May Top $10trn Notional.” Risk, April. http://
60(5): 2255–81. www.risk.net/risk/news/1506245/jp-morgan-cds
Cohan, William D. 2009. House of Cards. New -exposure-usd10trn-notional.
York, NY: Doubleday. Markit. 2009. “The CDS Big Bang: Under-
Diamond, Douglas W., and Robert E. Ver- standing the Changes to the Global CDS Con-
recchia. 1987. “Constraints on Short-selling and tract and North American Conventions.” http://
Asset Price Adjustment to Private Information.” www.markit.com/cds/announcements/resource
Journal of Financial Economics, 18(2): 277–312. /cds_big_bang.pdf.
Duffie, Darrell, Nicolae Garleanu, and Lasse Minton, Bernadette, René M. Stulz, and Ro-
Pedersen. 2005. “Over-the-counter Markets.” han Williamson. 2009. “How Much Do Banks
92 Journal of Economic Perspectives

Use Credit Derivatives to Hedge Loans?” Journal of “The Limits of Arbitrage.” Journal of Finance, 52(1):
Financial Services Research, 35(1): 1–31. 25–55.
Pirrong, Craig. 2009. “The Economics of Clear- Tett, Gillian. 2009. Fool’s Gold. New York, NY:
ing in Derivatives Markets: Netting, Asymmetric Free Press.
Information, and the Sharing of Default Risks van Duyn, Aline, and Hal Weitzman. 2008.
through a Central Counterparty.” Available at SSRN: “Fed to Hold CDS Clearance Talks.” FT.com,
http://papers.ssrn.com/sol3/papers.cfm?abstract October 6.
_id=1340660. Yavorsky, Alexander. 2009. “Analyzing the
Salmon, Felix. 2008. “In Defense of the CDS Potential Impact of Credit Default Swaps in
Market.” Portfolio.com, December 9. Workout Situations.” Moody’s Investors Services,
Shleifer, Andrei, and Robert W. Vishny. 1997. New York, NY.
This article has been cited by:

1. Xiaoling Pu, Junbo Wang, Chunchi Wu. 2011. Are Liquidity and Counterparty Risk Priced in the Credit
Default Swap Market?. The Journal of Fixed Income 20:4, 59-79. [CrossRef]