Mono 2009 M As09 1 Darcy
Mono 2009 M As09 1 Darcy
Mono 2009 M As09 1 Darcy
Credit derivatives have rapidly become a key financial tool in the capital markets as a
way to accept or transfer credit risk. These instruments have had a significant effect on financial
markets, both in easing the trading of credit risk and increasing the complexity of financial
transactions. The impact of credit derivatives is so extensive that anyone involved in enterprise
risk management (ERM) must develop a basic understanding of these vehicles. Some insurers
now routinely use credit derivatives as a financial management technique, and others are likely
to do this in the future. Even if an insurer does not directly trade credit derivatives, understanding
this aspect of the financial markets, and the key credit derivative metrics, is becoming
increasingly important. This paper explains the development of the credit derivative market,
describes the most common types of credit derivatives, discusses how insurers can and do use
these instruments and explains the role they played in the financial market turmoil of 2008.
1
1. Introduction
Enterprise risk management (ERM) entails a comprehensive treatment of all the risks an
organization faces. In order to manage risks on an enterprise wide basis, decision makers need to
have a basic understanding of all significant risks and how they are related. Credit derivatives
played a major role in the financial crisis of 2008, with many banks, investment banks and
insurers incurring unexpectedly large losses from credit derivatives. In some cases, the
organizations failed; others required massive government bailouts to remain solvent. From
subsequent comments it became clear that many board members, risk officers and other
executives did not clearly understand the risks involved in the credit derivatives they traded. This
paper is intended to help those involved in ERM gain a basic understanding of these financial
instruments.
A credit derivative is simply a financial instrument that allows one to assume or cede
credit risk exposure. Credit derivatives can be based on corporate debt, government debt,
residential or commercial real estate mortgages, or other types of loans. Credit risk is transferred
with a traded insurance-like contract between two parties, neither of which need be the issuer nor
the holder of the actual bonds or loans at risk. The value of the contract is derived primarily as
a function of the default risk profile of the debtor, which leads to the term, credit derivative.
Credit derivatives are used to speculate on or insure an entitys ability to meet its debt
obligations. In some cases, credit derivatives are written on a portfolio of debt instruments, rather
than a single issue. If the participant ceding the risk is already exposed to this credit risk, then the
transaction would be considered a risk management technique; if the participant does not have
2
any other exposure to the credit risk, then this transaction would be speculation. The credit risk
element is separate from interest rate risk, or from the cost of capital for the participant
(commonly termed funding risk). Credit derivatives can be used to reduce credit risk arising
from ownership of debt instruments in a manner analogous to that used by property insurers that
purchase reinsurance in order to reduce concentrations in their accumulated risk from wind
Credit derivatives were first developed in the mid-1990s, and rapidly grew to become an
extensively traded financial instrument. The key innovation of credit derivatives is the ease with
which one may trade the credit risk separately from the underlying debt. The global credit
derivatives market grew from $180 billion notional value1 in 1996 to over $20 trillion in 2006. In
2006, the $20 trillion notional value of credit derivatives exceeded the aggregate value of
international debt securities, which was approximately $18 trillion. By June 2008, the global
credit derivative market grew to $57 trillion (BIS, 2008). However, the boom in the credit
derivatives market over the last two years was not due to a doubling in the value of international
debtin fact, international debt only grew to $24 trillion by June 2008 (Brown, 2008).
The credit crisis of 2008 drastically impacted the credit derivative market. From June
2008 to December 2008, the notional value of the market dropped by more than half, to only
$26.803 trillion Dec. 26, 2008 (Depository Trust & Clearing Corporation, 2009). According to
Robert Pickel, CEO and executive director of the International Swaps and Derivatives
1
The notional value is an index on which the performance of the contracts is calculated. Notional value need not be funded and
does not change hands, so it is not actually equivalent to the face value of bonds.
3
Association (ISDA), part of the decrease was a result of industry efforts to exit offsetting trades
Prior to 2008, the accelerated growth in credit derivatives was largely driven by increased
demand from banks and financial institutions to manage their risk accumulations to debt service.
Banks were the initial drivers of growth as they managed their capital leverage and credit risk
line. They were expected to remain important players in their roles as market makers, as well as
The increased demand was fueled in part by tighter capital management rules in the Basel
central banking accords. In the United States, loans could be allocated no capital if they were
hedged by credit derivatives. Consequently, credit derivatives were embraced as a sound risk
management tool since they allowed the separate pricing of the credit risk component of their
overall market risk, and were believed to provide an effective hedge against a position in the
credit market. However, this view overlooked counterparty risk, or the risk that the organization
that accepted the credit risk could fulfill its commitment. This risk is similar to the risk of a
catastrophe reinsurer being unable to pay claims after a major hurricane. Counterparty risk
became obvious during the credit crisis of 2008, as regulators and market participants came to
realize that credit derivatives are not guaranteed insurance. Defaults on these instruments then
spread through the market, leading credit derivatives to be blamed for helping to expedite the
4
Theoretically, hedging credit risk accumulations through various capital market options
mitigates systemic and material volatility risk to banks earnings and book values. This
highlights the very essence of risk management. A pervasive friction is eased and simultaneously
the existing market attracts capital from unrelated markets. Originators of commercial and
consumer loans were aware that they are not always the best businesses to retain the credit risk
from holding others debts over time, and were able to use credit derivatives to protect
themselves from the risk and pass it along to another player in the market. Diversification is the
key to all risk management strategies, and the business of lending and borrowing is no different.
However, it became painfully clear through the credit crisis of 2008 that the risk was not being
diversification in the credit market as many of the underlyings on the derivatives were correlated.
Furthermore, banks and other financial institutions whose business focuses on credit exposure
were not significant net buyers of credit protection (see Figure 2 below). Institutions whose
business did not focus around issuing credit and who could emerge as potential net sellers of
growth of the credit derivatives market up through 2008 was aided by the fact that credit
derivative trades could be unfunded. Investors can buy/sell arbitrarily large positions in a
particular credit for reasons of speculation, and since investors do not need to make an upfront
payment, they can leverage their positions. However, it is exactly the lack of a funding
requirement that sparked many of AIGs problems in 2008 (discussed below). As the probability
of default of the underlyings shot up, so did collateral calls on AIGs credit derivatives. Since
5
AIG was under-funded, it had to negotiate to reduce the level of collateral required. The
standardization of documentation and the emergence of newer product applications that delve
further into sub-risk components of market credit risk (i.e., volatility, recovery rate arbitrage and
correlation) have also fueled the recent expansion of the global credit markets. This expansion
has brought about an increase in the credit risk exposures assumed by financial institutions such
returns, and independent of any credit derivatives trading, some were seeking to effect asset risk
hedging strategies.
This paper will provide a general overview of credit derivatives and their applications as
they emerged primarily as a banking risk management technique, then evolved toward more
refined and complex credit risk trading schemes, and finally played a key role in the financial
crisis of 2008. It will highlight emerging risk areas that will affect actuaries and other financial
6
2. Market Participants
Although credit derivatives were developed as a method of shifting credit risk, the use of
credit derivatives is not limited to commercial banks risk management departments. As seen in
the detailed breakdown of market participants in Table 1, banks and corporations were usually
net buyers of credit risk protection, while hedge funds, mono-line insurers (financial guaranty
insurers), (re)insurance companies, pension funds and mutual funds emerged as significant net
sellers of credit risk through 2006 (and projected to 2008). Table 1 illustrates the relative role of
the different participants in the credit derivative market from 2000 through 2006 and projects a
forward forecast.
TABLE 1
Total Market Share by Notional Value Through 2006 with 2008 Projection
Year /
[Volume in 2000 / 2002 / 2004 / 2006 / 2008 (Est.) /
USD bn] [$893] [$1,952] [$5,021] [$20,207] [$33,120]
Market Buyers of Sellers of Buyers of Sellers of Buyers of Sellers of Buyers of Sellers of Buyers of Sellers of
Credit Credit Credit Credit Credit Credit Credit Credit Credit Credit
Participant Protection Protection Protection Protection Protection Protection Protection Protection Protection Protection
Banks
Trading 39% 35% 36% 33%
Activities
81% 63% 73% 55% 67% 54%
Banks
Loan 20% 9% 18% 7%
Portfolio
Hedge Funds 3% 5% 12% 5% 16% 15% 28% 32% 28% 31%
Pension 1% 3% 1% 2% 3% 4% 2% 4% 3% 5%
Funds
Corporations 6% 3% 4% 2% 3% 2% 2% 1% 3% 2%
Mono-Line
2% 10% 2% 8% 2% 8%
Insurers 3% 21%
Reinsurers 3% 7% 2% 4% 2% 4%
7% 23%
Other
Insurance 3% 12% 2% 3% 2% 5% 2% 6%
Companies
Mutual
1% 2% 2% 3% 3% 4% 2% 3% 3% 3%
Funds
Other 1% 1% 2% 0% 1% 1% 1% 1% 2% 1%
Source: British Bankers Association Credit Derivatives Report 2006.
However, as seen below in Table 2, when larger categories are examined, most large
categories of market participants were neither significant net sellers nor net buyers of credit
protection in either 2006 or 2008. Rather, most of these groups netted out their positions.
7
Reporting dealers (market makers in credit derivatives) were the largest market participants in
2006 and their market share grew to 2008, while the other categories of participants realized
share declines.
TABLE 2
Total Market Share by Notional Value 2006 Through 2008
Year / [Volume in June 06/ June 07/ June 08/
USD bn] [$20,352] [$42,580] [$57,325]
Buyers of Sellers of Buyers of Sellers of Buyers of Sellers of
Market Participant Credit Credit Credit Credit Credit Credit
Protection Protection Protection Protection Protection Protection
Reporting Dealers
68% 69% 71% 71% 73% 74%
(gross)
Banks and Security
16% 16% 15% 14% 15% 15%
Firms
Other Financial
12% 12% 13% 13% 10% 10%
Institutions
Insurance and
Financial Guaranty 1% 0% 1% 0% 1% 0%
Firms
Non-financial
3% 2% 1% 1% 1% 1%
Institutions
Source: BIS Semi-annual OTC Derivatives Statistics at end-June 2008.
Figure 1
Comparison of 2006 Market Share, Buyers versus Sellers
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
s
rs
ds
rs
g
er
s
ds
er
nd
n
re
te
re
un
th
oa
ur
di
su
su
Fu
Fu
O
ra
lF
ro
-L
In
In
In
-T
p
on
ua
e
or
e-
er
e
s
g
nk
lin
i
ut
ed
s
R
ns
th
nk
M
Ba
o-
H
O
Pe
Ba
on
M
8
The two charts provide different breakdowns of who is paying for default protection and
who is receiving income from selling it. Firms that are net sellers are taking a long position on
credit risk and receiving income from their positions. Firms that are net buyers are taking a short
position on credit risk, and reducing their risk exposure. Figure 1 provides a more detailed
breakdown of market share, while Figure 2 provides market share for broader categories in both
Figure 2
Comparison of 2008 Market Share, Buyers Versus Sellers
60%
50%
Market Share
June 06/
40% Buyers
June 06/
30% Sellers
June 08/
Buyers
20%
June 08/
Sellers
10%
0%
Reporting Banks and Other financial Insurance and Non-financial
dealers (gross) security firms institutions financial institutions
guaranty
Market participant
Source: BIS Semi-annual OTC Derivatives Statistics at end-June 2008.
Banks were once the primary participants in the credit derivatives market. Following the
adoption of the Basel Accord in 1988 by the G-10 and other nations, banks have been required to
hold more capital for riskier assets. Banks looked to reduce the amount of capital required to
satisfy regulatory requirements. Banks were able to reduce capital required by securitizing loans
since securitized pools of debt require less capital under Basel than individual loans. Banks also
9
developed the credit default swap (CDS) market to decrease risk exposure accumulations within
their commercial loan operations. The banking markets will likely continue to use credit
derivatives for hedging both single-name (i.e., Exxon, for example) and broad market credit
In the past, commercial banking market makers such as Citibank and JP Morgan were
constrained in their ability to provide liquidity because of regulatory limits on the amount of
credit exposure they could have to any single company or business sector. The use of credit
derivatives allows these securities dealers to trade more efficiently and employs less capital.
Credit derivatives afford these firms the opportunity to hold onto their physical bond inventories
during the tight credit cycles by keeping their credit risk profile neutral.
Credit risk has moved from highly regulated banks (with strict capital requirements) to
other investors. Other firms or investors may be better suited or more diversified in holding the
risk, but as is evident in the positive correlation of the underwriting cycle gains and default rates,
investors may not fully understand how credit exposure fits with their portfolios. In addition,
banks are usually the cushion in times of turmoil or panic while other investors will be quick to
sell, even at fire-sale prices. Furthermore, what banks thought was a credit risk neutral position
may not have been, because they may not have correctly accounted for the counterparty risk
inherent in the purchase of credit protection. Consequently, the net effect of the Basel regulations
and the securitization and protection trends it spurred may have been to increase risk in the
markets.
10
Corporations use credit derivatives to manage credit risk exposure to key customers or
critical suppliers. The liquidity, transparency of pricing, and more refined structural options
available in the CDS market offer an efficient alternative to traditional credit insurance. Some
corporations invest in CDS indices (i.e., CDX and iTraxx) and structured credit products (i.e.,
collateralized debt obligations (CDOs) and asset-backed securities (ABS)) comprised of pools of
mortgages, auto or other consumer loans) in an effort to embellish returns on pension assets or
Hedge funds have continued to increase their presence and have been integral to
expanding the variety of trading strategies. An important interest to hedge funds is the possibility
of long-short CDS trading strategies, particularly those involving reference entities that are in the
process of merging or being acquired. Acquired companies are generally expected to have
improving credit spreads, while the purchasing company often has a diminished credit profile
over the short term. Hedge funds have been the primary users of new products that facilitate the
Due to hedge funds significant role in the credit derivatives market, trouble for hedge
funds (such as the struggles they faced as a result of exposure to the subprime market in summer
2007) has repercussions for banks and other lenders who have purchased credit protection.
Should a credit event occur for a particular debt, hedge funds that were sellers of protection for
that debt may not be able to cover their obligations, damaging purchasers of the protection who
thought they had already hedged their risk. A hedge fund would need to be in extreme duress for
this outcome to occur, but amidst the trouble in the credit markets some hedge funds have faced
11
this worst case. For example, Sowood Capital Management sold its portfolio to Citadel
Investment Group in August 2007 for fear they would not be able to meet their margin calls and
other obligations.
Hedge funds also use credit derivatives to make bets on the credit markets. Many use
borrowed money (or are exposed by default swaps that require no upfront investments or capital
in reserve) to make these bets. While some hedge funds have realized significant returns with
this strategy, players on the other sides of those bets are often hedge funds whose losses could
Asset managers have begun using the CDS market as a relative value tool, or to provide a
structural feature (i.e., particular maturity date, senior versus subordinate debt, cash flow versus
synthetic, etc.) that is not available in the bond trading market. Also, the ability to use the CDS
market to take a bearish position is readily available. For example, an asset manager might
purchase three-year protection to hedge a 10-year bond position on an entity where the credit is
currently considered to be in financial distress but is expected to perform well if it survives the
The participation of insurance companies in the CDS market can be separated into two
distinct groups: 1) life insurers and 2) mono-lines and (re)insurers. Life insurers typically use
CDSs to sell protection (long credit risk) to enhance the return of their asset portfolio. Mono-
lines (the common term for financial guaranty insurers) are in the business of assuming
investment grade debt service exposures for a variety of asset-backed securities [15].
12
3. Example of a CDS Market Transaction
Let us review the example of a three-year loan from Sterling Bank to Global Media
Corporation (GMC). Subsequent to providing the loan Sterling Bank buys protection for this
particular GMC loan from Offshore Re, a reinsurance company. Sterling Bank expects to receive
periodic interest payments from GMC and a final repayment of the principal loan balance in
three years. Sterling Bank faces the credit risk that GMC may go bankrupt and fail to meet its
The yield that Sterling Bank earns on this loan, like most loans, can be broken down
between the risk free yield and the risk premium for bearing the risk of loan default. This risk
premium is a function of both the probability of default and Sterling Banks underwriting risk
In a CDS, Sterling Bank buys protection from Offshore Re, against a default by GMC.
The aggregate contractual arrangements are summarized in the deal diagram below.
Figure 3
Credit Default Swap on a Single Corporate, Between a Bank and a Reinsurer
13
The buyer of the CDS (Sterling Bank) receives the benefit of protection for credit risk in
exchange for periodic premium payments (usually quarterly in arrears) until the contract expires
or a predetermined credit event occurs. Credit events are detailed in the credit swap insurance
contract between Sterling Re and Offshore Re and identified as: 1bankruptcy, 2obligation
restructuring. Since all credit events except restructurings (not considered a credit event since
2003) are linked to default of the reference entity (Global Media Corp), the term credit event is
replaced by default.
In the event of a default by GMC, Sterling Bank will receive a payoff from Offshore Re
equal to the difference between the face value and the market value of the underlying debt minus
the CDS premium that has accrued since the last periodic payment date.
When a GMC default occurs, there are two accepted settlement procedures or protocols
between Offshore Re and Sterling Bank: 1physical settlement (the more commonly used
protocol) and 2cash settlement. In a physical settlement, Sterling Bank provides to Offshore
Re the notional value of deliverable obligations of the reference entity (GMC). Offshore Re then
In a cash settlement, Offshore Re pays Sterling Bank the face value minus the recovery
rate of GMC loan amount; known as the loss given default. The recovery rate is calculated by
either referencing quotes from market dealers or by observing market prices over some period
14
To further illustrate, suppose that Sterling Bank purchases a three-year CDS security with
a spread (i.e., insurance risk premium) equal to 300 basis points (bps; i.e., 3.00 percent) and that
the notional value of the underlying debt is $20 million. Sterling Bank then makes quarterly
payments of 0.03 times $20 million divided by 4, which equals $150,000. In the case that GMC
compensates Sterling Bank for the loss on the face value, which is $12 million ($20 million x
.40) and Sterling Bank pays the insurance risk premium that accrued since the last payment. For
example, if the default occurs two months after the last premium was paid, the accrued premium
The bank has effectively hedged its credit risk with the CDS. It has parted with a portion
of the loan yield (to pay the premium for the insurance protection of 300 bps in the example
above) but has deemed that to be a fair trade for reduced volatility in the next credit cycle.
When the loan was initially recorded as a liability on Sterling Banks balance sheet, the
risk of default, and the expected rate of recovery given default, were assessed by Sterlings loan
officers based upon a review of private data from its long banking relationship with GMC about
developments in the sector and changes to the companys structure and management. This
assessment was largely separate from daily movements in GMCs stock price or bond ratings.
Offshore Re does not have a long-term relationship with GMC. Its primary interest is in
the premium it can earn on GMCs default swap. When the CDS is on Offshore Res balance
sheet, it must be booked in accordance with fair value accounting (FAS 157), which is subject to
15
daily fluctuations. Daily market movements in GMCs equity and bond prices feed directly into
the estimation of default and expected loss given default, thereby affecting the recorded asset
value on its books, via a booking process known as the mark-to-market (MTM) adjustments. As
the bond and equity prices fall, the probability of loss increases, and the MTM adjustment yields
a diminished position (i.e., a decrease in the quarterly income and also a reduction in Offshore
Res book value). As described subsequently in the case of AIG, companies that assumed large
positions of credit protection experienced huge accounting losses from the write-down of the
When asset values of the credit derivatives positions fall, the reinsurer may look for an
effective hedge for its own credit derivative retentions. Offshore Re could hedge its position by
shorting bonds if a liquid market exists, although if such a market existed, there would be little
need for CDSs on bank loans in the first place. The equity market of the borrower (e.g., GMC) is
much more liquid, and as a result, insurers will often short the GMC stock in order to hedge their
credit risk.
(Re)insurers often sell protection (i.e., taking a long position on credit risk) as a source of
These insurance markets need to be careful to avoid the short squeeze, whereby they
may seek to hedge their credit risk positions by shorting equities. As markets fall, insurers
typically respond by selling more, which in turn may lead to an accelerated downturn and
16
Insurers have often resorted to a strategy of selling when the market is falling and buying
when the market is rising. When falling markets are met not by speculators but by further selling,
a liquidity vacuum develops and volatility can spike suddenly. For insurers this situation is called
a short squeeze. This would be very similar to the experience of portfolio insurance in
October 1987, when institutions following this strategy sold extensively on the initial market
decline, precipitating further selling and further price declines [11]. It is not yet known if this
17
4. Credit Derivative Product Types and Key Terms
Single name CDSs are still the most widely used product and have been the cornerstone
of the credit derivatives market. However, full index trades and tranched index trades have taken
TABLE 3
Credit Derivative Volumes by Product Type
Product Type 2004 2006
Single-name credit default swaps (CDSs) 51.0% 32.9%
Full index trades 9.0% 30.1%
Synthetic collateralized debt obligations (CDOs) 16.0% 16.3%
Tranched index trades 2.0% 7.6%
Credit linked notes 6.0% 3.1%
Others 16.0% 10.0%
Source: British Bankers Association Credit Derivatives Report 2006.
Table 4
Credit Default Swaps Volumes by Product Type
Product Type June 06 June 07 June 08
Single-name CDSs 68% 57% 58%
Multi-name CDSs 32% 43% 42%
Single name CDSs are just that. The reference entity is a single corporation (i.e.,
Microsoft) or a sovereign (i.e., Chile). CDSs on sovereigns (see Table 5) increased from 2006 to
2008, along with swaps on entities with ratings BB or below. The notable increase in share of
swaps on BB or below rated entities is likely due to ratings changes as companies troubles were
revealed through the unfolding of the credit crisis in 2007 and 2008. The trade turns on whether
or not the reference entity encounters a credit event during the term of the contract. Unlike loss
events with traditional insurance, the credit derivative guarantee is irrevocable and unconditional
(and waives all defenses, including fraud) and results in the guarantor stepping into the shoes of
the issuer. This is a significant departure from traditional insurance whereby a claim is made and
18
negotiations begin as to what extent the claim is deemed valid. In the credit derivatives market,
the protection seller pays the protection buyer at the time of credit event and argues with the
issuer later. Absent that type of claims paying protocol (known as a capital market standard),
TABLE 5
Single Name Credit Derivative Volumes by Type 20062008
Type June 06 June 07 June 08
Sovereigns 5% 6% 7%
Non-sovereigns 95% 94% 93%
Investment grade
(AAA-BBB) 67% 65% 66%
BB and below 14% 13% 20%
Non-rated 18% 22% 13%
Synthetic CDOs are among the most complex instruments in the market today. Synthetic
CDOs are typically "structured" transactions in which a special purpose entity (SPE) is
established to sell credit protection on a range of underlying assets via individual CDSs. The SPE
in turn issues several prioritized tranches of notes to investors, with note proceeds typically
payments, while note holders (in order of seniority) receive both cash flows on the underlying
collateral and the premiums on the special purpose vehicle (SPV) default swaps. Synthetic CDOs
provided an attractive way for banks and other financial institutions to transfer credit risk on
pools of loans or other assets without selling the assets. Examples of the type of underlying
collateral that can be securitized include mortgages, auto loans and other consumer debt.
Lending companies that specialize in subprime mortgages will periodically bundle up these loans
and sell off the credit risk of the riskier lower tranches of the pool to investors for roughly two-
thirds of the collective spreads on the loans. In 2007 and 2008, the housing market crash sparked
defaults of the riskier debt in these CDOs (especially mortgage-based CDOs). The fall in value
19
of the CDOs, plus the outright payout obligations as more and more debt defaulted, caused
massive losses for investors in CDOs. Furthermore, the crisis was exacerbated as key cogs
investment banks and commercial banksto the financial system also experienced losses since
they kept some CDOs on their books when they became harder to sell. Many insurers had
invested in highly rated synthetic CDOs as an alternative to bonds. As the ratings on these
instruments are lowered, insurers may be forced to sell them to meet regulatory requirements on
the quality of debt held. These forced sales can further depress the market and generate even
Figure 4
Structure of Synthetic CDOs
Loan 1
Loan 2
Asset Special Cash
Purpose Investors
Loan 60 Cash Vehicle Bond
Credit derivative index trades are usually comprised of a generic basket of single name
swaps with standardized terms. Recently, tradable CDS indexes have also been introduced that
allow investors quick and easy ways to buy and sell market-wide or sector credit risk. In June
2004, the two main CDS indexes, iboxx and Trac-x, were merged into the Dow Jones iTraxx
index that since has set a new standard when it comes to liquidity, transparency and
diversification [7]. Large exposure (negative or positive) to a diversified pool of credit risks is
now much easier to gain, and the liquidity of the iTraxx market has attracted hedge funds. Unlike
other multi-name CDSs, such as first-to-default baskets, index swaps provide exposure to
20
unnamed reference entities. These blind pools were the fastest growing products, allowing
investors to buy and sell a customized cross section of the credit market much more efficiently
than they could if they were dealing in individual credit derivatives. However, such blind pool
trades raise a modicum of moral hazard as the buyer and seller do not always have the same
Credit-linked notes, or CLNs, are known as funded credit derivatives, because the
protection seller pays the entire notional value of the contract up front. In contrast, CDSs pay
only in case of default and are therefore referred to as unfunded. CLNs are often used by
borrowers to hedge against credit risk and by investors to enhance their holdings yields.
Other product types include credit spread options, swaptions, variance swaps, constant
21
5. Credit Derivative Contract Standards
Credit derivatives are similar to options in the form of bond insurance, and have been
available since interest rate volatility became a key concern for financial markets. Banks seeking
to hedge credit risk in their loan portfolios led the initial growth in the early 1990s. The size and
liquidity of the credit derivatives market has recently developed in response to a broader range of
participants (i.e., private equity firms, hedge funds, asset managers, insurers, etc.) seeking to
purchase and sell credit risk protection. By the late 1990s it was apparent that standardized
wordings and the uniform application of capital market standards were needed to enable these
dissimilar markets to trade credit risk exposure more efficiently and effectively.
The International Swap and Derivatives Association (ISDA) set and defined standard
credit events (i.e., bankruptcy, failure to pay and restructuring) and standardized terms of CDS
contracts in 1999, substantially aiding development of this market [JP Morgan]. CDS contracts
now commonly include a standard confirmation letter, incorporating the 2003 ISDA Credit
Derivative Definitions, and follow the ISDA Master Agreement. These forms cover the
D. What events would define a credit event that would trigger coverage
22
Standardization reduces the costs and increases liquidity. The only unique information
required for a particular transaction is the elements that differ for each trade, including the
reference entity, maturity date, spread over Libor2 and notional value. These conventions have
One concern over any new credit instrument is how it would fare in adverse
circumstances. Before the most recent and severe test of these instruments during the credit crisis
of 2008, the corporate defaults of the early 2000s (e.g., WorldCom, Parmalat, Delphi, Calpine,
Northwest and Delta) tested this market and the strength of the standard conventions. At least in
the first instance, markets did function effectively, with over $50 billion of contracts involving
thousands of separate contracts settled without litigation or disputes that adversely impacted the
market [JP Morgan]. Thus far, settlements in the credit derivatives market have been orderly
through 2008. The most notable bankruptcy of Lehman did not lead to any disruptions in the
market caused by the settlement process. The settlement process should be enhanced once central
clearinghouses are established for these derivatives. Despite the clarification the contract
standards provided, valuation of credit derivatives remains an issue for market players. Sellers of
credit protection dont have as much information about loans and their issuers as the original
lenders do, making it hard for sellers to adequately price credit insurance. Information
2
Libor is an acronym for London InterBank Offered Rate, and is the most widely used benchmark for short-term
interest rates. Libor is the rate that credit-worthy international banks generally charge each other for large
overnight Eurodollar loans. It is effectively the AA bank loan rate in the capital markets and is compiled and
released each business day. As banks are primary participants in the credit markets, Libor is viewed as the funding
rate, or what it costs an institution to raise money to engage in the credit markets.
23
As illustrated by the ratings for the subprime mortgage market, sellers of protection cant
count on ratings agencies to be reliable. In 2000, when mortgages with piggyback loans (a
mortgage and a second loan for the down payment) were first being issued, ratings agencies
determined piggyback loans to have risk similar to that of typical mortgages. Consequently,
more of these mortgages were marketed than might have been had they been given lower ratings.
In 2001, Standard & Poors did recognize an increased level of risk associated with these loans,
but still allowed mortgage pools to be comprised of up to 20 percent subprime mortgages before
receiving a ratings penalty. In 2006, Standard & Poors found through a proprietary study that
subprime loans were 43 percent more likely to default than conventional loans. However, they
did not change ratings on existing securities. It was not until spring of 2007 that agencies slashed
their ratings for subprime loans following high default and foreclosure rates as the housing
market deteriorated. Congress and regulators are scrutinizing the ratings process in an attempt to
24
6. ISDA Credit Events and Settlements
A credit event is the frequency trigger for a loss given default and the 1999 ISDA
definitions appeared to provide fairly straightforward and concise default triggers (delayed
payment, restructuring, bankruptcy, etc.). However, the market went through some growing
This can be illustrated by the real market event that follows. In August 2000, Conseco's
bank debt was restructured. While the restructuring included a deferral of the loan's maturity by
three months, it also included an increased coupon, a new corporate guarantee and additional
covenants in favor of the lenders. Thus, because lenders were compensated for the maturity
extension, Moody's did not consider the restructured debt to be a "diminished financial
obligation," and thus not a "distressed exchange" default. However, because of the maturity
extension, the restructuring was considered a "credit event" under the ISDA "restructuring"
definition and triggered loss payments under the CDSs written on Conseco. This is a perfect
example of a loss event under the ISDA "restructuring" definition that Moody's did not consider
to be a default.
Thus, any "restructuring" definition should look at the totality of the circumstances (e.g.,
whether the lenders/investors have been compensated for the reduction or deferral) to determine
25
The 2003 ISDA revisions eliminated restructuring as a default event and also removed
any material gray areas. However, in the capital markets all terms and conditions are subject to
Based on the 2003 ISDA conventions, there is a three-step procedure for physical
settlement:
Either the buyer or the seller of credit protection can provide notice to the
provisions allow notice to be given any time up to 14 days after the contract
matures, which could be long after the event. This delay is important for valuing a
CDS, as the seller of the protection, once the credit event has occurred, will be
should be booked.
of physical settlement within 30 days. This notice must specify the particular
3. Delivery of bonds
Following the notice of physical settlement, the credit protection buyer generally
delivers the debt instruments to the seller within three days. The buyer should
26
deliver debt instruments with a face value equal to the notional value of the CDS
in order to get the full protection purchased. However, if fewer bonds are
More recently, an alternative settlement mechanism known as the CDS protocol has been
developed by ISDA in conjunction with the derivatives and securities dealers. The new
settlement protocol allows investors to cash or physically settle contracts at a recovery rate
27
7. Bonds Versus Credit Default Swaps
A typical fixed rate coupon bond for a single corporation is designed to compensate the
bond holder for three components: credit risk, funding risk and the risk free rate.
Credit risk compensates the investor for the chance that the particular corporation may
default on its debt. Funding risk represents the cost of capital for the investor. As the credit
market originated with major banks, the spread of Libor (the rate at which banks can obtain
short-term funding) over the risk free rate (as measured by U.S. Treasury securities) is
commonly used to measure this component. In general, the swap spread (which is the swap yield
risk free rate) is used to reflect funding risk. This can be viewed as the hurdle rate for an
Bonds and CDSs on the same individual credit will usually trade closely with each other,
since both reflect the markets view of its default risk. The pure credit risk element of the loan at
issue is encapsulated by the CDS. CDSs are not measured by a spread over a benchmark; rather,
the spread is the annual premium (coupon) the credit protection buyer is willing to pay and the
protection seller is willing to receive. Any credit that is considered to have a relatively higher
default probability will in turn carry a relatively higher spread. In order to compare bonds with
CDSs it is requisite to split out the spread of the bond that compensates the bondholder for
28
Figure 5
Comparing Bond Yields and the Spread on CDS
Credit
Risk
Credit
Credit
Risk
Risk
Funding
Risk
Funding
Risk
Risk
Free
Rate
Source: JPMorgan, Credit Derivatives Handbook Detailing Credit Default Swaps Products, Markets and
Trading Strategies, page 45.
For example, assume a bond that is paying a yield of U.S. Treasury rate plus 120 bps. In
order to remove the interest rate risk from owning this bond, an investor can swap the fixed
interest payments received from owning the bond for a series of floating rate payments through
an agreement with a third party known as an asset swap. In order to place a value on the credit
risk inherent in a fixed rate bond, a two-step process must be followed. First, the cost of
exchanging the fixed rate bond for a floating rate bond needs to be determined. In financial
floating swap. The investor holding the fixed income security agrees to swap the fixed rate
payments in return for payments from a counterparty based on a floating rate that is usually
defined as Libor (a rate that can change each day) plus a spread. The spread is calculated to set
the present value of the fixed rate payments to equal the present value of the floating rate
payments, and depends on the term structure of interest rates that exists when the contract is
established.
29
In the example illustrated on Figure 5, the floating rate equivalent turns out to be Libor
plus 80 bps. The fixed rate payer in this transaction now pays the fixed interest rate on the
notional value to the floating rate payer, and receives a payment each quarter (or whatever timing
was included in this swap) equal to Libor plus 80 bps. Now the fixed rate payer is no longer
concerned about interest rate changes as it receives payments that fluctuate with interest rates
The second step involved in estimating the credit risk requires the investor to determine
the cost of capital for funds to purchase the underlying bond. As the market initially developed
around banks, it is commonly assumed that an investor can borrow at Libor. Thus, if a bank
purchased this bond, its spread would be the yield on the bond less its borrowing costs, or (Libor
For example, consider a three period bond. Given that the price is determined by:
We solve for z in the above equation which yields the bonds implicit spread.
If a bond has an implicit spread that is less than the CDS spread for the same credit and
same maturity trades, the market has a negative (bearish) view of this credit over the term of the
loan. Conversely, if the spread on the bond is higher than the CDS spread, the market has a
30
8. Credit Spreads, Option Volatility and Equity Prices
The markets for equities, options on equities and credit derivatives are all large, liquid
markets, where news about companies, macro-economic events and market conditions are
quickly reflected in prices and spreads. When an unexpected piece of adverse information is
reported, for example, equity prices tend to fall, credit spreads tend to rise, and the implied
volatility of options on equity usually increases. Thus, these markets are closely related, and
Based upon analyses and investor presentations by credit experts at JP Morgan, the
In general, these markets would be inversely related. If market expectations regarding the
financial conditions of a company are improving, then equity prices would normally increase and
the CDS spreads would decrease as the companys chances of defaulting on debt would be
reducing. Occasionally this relationship will not hold, such as when a leveraged buyout increases
the equity price, but as the likelihood of a default on the debt of the restructured company would
The price of an option on a companys equity is based on five factors: the current price
of the stock; the exercise price of the option; the time to maturity of the option; the current
interest rate; and the volatility of the underlying stock price. All of these factors except the
31
volatility are readily determined. Using the current price of an option and applying an
appropriate option pricing model allows investors to calculate the implied volatility of the equity
of that company (the value that, when used for volatility, sets the model price equal to the actual
price). The price of a companys equity and the implied volatility of its options tend to be
inversely related based on the leverage effect. The debt-to-equity ratio declines as a stocks
price increases, which reduces the leverage of the firm. The stock price is then expected to be
Both the CDS spread and the implied volatility measure the risk of a company, although
on different components. The CDS spread reflects the likelihood of a default on its debt; implied
volatility reflects the stability of its equity price. A very risky firm would be expected to have
high values for both of these measures, whereas a low risk firm would have a lower CDS spread,
32
9. Role of Insurers in the Credit Derivative Market
companies as a whole have become the third largest sellers of credit protection with a combined
market share of 17 percent. Insurance companies are also significant purchasers of credit
Insurance companies sell credit protection to gain quick exposure to the credit market, sell
protection to gain a higher return than direct investment in underlying credit, and buy protection
derivatives can provide a more liquid market than bonds. An insurance company might sell
protection for a single bond with a CDS or for a credit default index and purchase Treasuries or
AAA asset-backed bonds rather than purchasing a bond. The company would look for a credit
derivative with risks in line with that of the desired security. The purchase of a safe security
provides a venue for the funds with a low return, funds the credit derivative trade and provides
inflows whose duration corresponds to that of the underwriting portfolio. When bonds of the
appropriate maturity and risk become available, the insurance company can sell its safe
The purchase of an AAA bond or Treasury along with the sale of credit protection can
also be used to earn a higher return than the purchase of a credit obligation. Sometimes an
33
insurance company will be able to earn a higher return from the yield on the safe security and
the premiums from the sale of credit protection than on a corporate bond.
Insurance companies may also purchase protection as a hedge for bonds the company
owns. Insurance companies might do this rather than sell the bond to avoid a loss on the bond or
to speculate about the credit curve of the issuer. The insurance company would execute the latter
strategy by, for example, purchasing five-year credit protection on a bond with a 15-year
maturity.
When entering into credit derivative trades, insurance companies should carefully consider
their expected return along with their maximum loss compared with the purchase of a low-risk
security such as a Ginnie Mae. Consider this strategy for investment in a Ginnie Mae along with
the sale of protection through a CDX. The investor has $10 million to invest and purchases $10
million in Ginnie Maes for a return of 5.63 percent. The investor also sells protection for Dow
Jones CDX.NA.IG with a notional value of $10 million for an inflow of 0.70 percent from the
spreads. The credit default index has a default range of 0100 percent and an expected value of
0.30 percent.
Min return = -100% (all the bonds in the portfolio default and nothing
can be recovered)
34
The investor has gained an additional expected return of 0.30 percent; however, the
investor has assumed exposure to a 100 percent loss if all the credit obligations in the index
default.
Figure 6
Probability Distribution
There are two significant risks inherent in this strategy that go beyond predicting the next
credit cycle. These include MTM booking effects and the correlation of credit defaults with the
insurance cycle.
The MTM is necessitated by financial reporting as well as the need to monetize existing
contracts. It is largely a function of the probability of default (and survival) over time. The
CDSW model is available on Bloomberg and is a market standard tool for calculating the MTM
adjustment on CDS contracts. To access this model simply type CDSW <Go> on Bloomberg.
In concept the MTM on a CDS contract is roughly equal to the notional amount of the contract
35
multiplied by the difference between the contract spread and the market spread (in basis points
Consequently, for insurance companies that are increasingly becoming net sellers of
credit risk protection, if the spread widens (i.e., credit outlook deteriorates) in a given accounting
quarter, this results in a material reduction in current period income. This income loss is
unrealized but impairs book value and is difficult to explain or reconcile to market analysts
unfamiliar with these types of fluctuations in the valuation of an insurance company. Likewise,
when credit spreads tighten (i.e., credit outlook improves), there are apparent windfall impacts
that can cause irrational exuberance that will not be repeated in future accounting periods.
Because the notional value of credit derivative trade is so very large, small changes in quarter
end market spreads can cause material volatility to the earned income bookings and distort
A significant portion of the growth by the insurance industry in selling net credit
protection is based on the notion that introducing a modicum of credit risk into the existing
portfolio of life and/or property & casualty risks will further diversify its book. The expectation
is that the credit cycle is separate and distinct and therefore uncorrelated with the traditional
insurance underwriting cycles. By introducing a slice of credit risk into the mix, these insurers
expect the cycles to partially offset and thereby reduce income volatility and further stabilize
36
Figure 7
Correlation of Combined Ratio and Corporate Default Rate
Cor r el at i on of Combi ned Rat i o and Cor por at e Def aul t Rat e
140 9
8
Combined Ration
120
7
Default Rates
100
6
80 5 Combi ned Rat i o
60 4 Y
40
3 Cor por at e Def aul t
2 Rat es
20 1
0 0
1930 1933 1936 1939 1942 1945 1948 1951 1954 1957 1960 1963 1966 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999
Ti me
Source: Moodys Investors Service Inc., Default and Recovery Rates of Corporate Bond Issuers, 1920-2005; A.M. Best
Company, Inc. Best's Aggregates & AveragesProperty/Casualty, 2000, 2001; Alfred M. Best Company Inc., Bests Fire
and Casualty Aggregates and Averages, 1940, 1941.
The correlation ratio of combined ratio and corporate default rate is 0.258, so it clearly
shows the positive correlation of bond default and insurance company combined ratio.
The positive correlation between the insurance underwriting cycle and default frequency
in the graph is a bit surprising, and although the R 2 is relatively low there are two key
observations to be made. The correlation is neither zero nor is it negative. As such, any positive
correlation between the credit cycle and the insurance underwriting cycle will act as a drag on
earnings. Additionally, positive correlations (albeit even a small one) will increase the capital
requirements for those insurers that have assumed significant net credit risk exposure. This is a
bit of a double whammy as the pro forma planning expects some diversification benefit from
37
10. CDSW Model
Bloomberg uses the credit default swaps function to evaluate the price of CDSs. It bases
the function on the credit default swaps model that was modified by Hull and White in 2000. The
following will present the price model and how to use the Bloomberg software to evaluate the
To evaluate a $1 notional principal plain vanilla CDS, Hull and White assume that
default events, Treasury interest rates and recovery rates are mutually independent. This
assumption should be carefully assessed. In practice some of the above conditions are unlikely to
be met. Some research has been done on the relationship of recovery rate and default rate. Hu
and Perraudin (2002) used Moodys data from January 1971 to January 2000 to find the
correlation between quarterly aggregate recoveries and default rates; the typical correlation for
post-1982 quarter is 22 percent, while from 19712000 the correlation is 19 percent. Moody's
(19302005) and other researchers corroborated the apparent negative correlation [17].
At the same time, the assumption overlooks the effect of economic conditions. For
example, in a severe recession (as seen in the recession of 2008), interest rates could decline,
default rates increase and recovery rates decline as well if there is a glut of fire sale assets on the
market simultaneously. John Frye concluded in an economic downturn, recovery of bond will
decline 2025 percent from normal year average [9]. The current experience in the subprime
mortgage market shows this interaction. Subprime lenders offer mortgages to borrowers with
weaker credit scores at interest rates of higher than regular mortgage market. Consequently,
38
subprime borrowers are more likely to be susceptible to the interest increasing, since their
interest payments are already so significant. In this case, rising interest rates increased mortgage
defaults on floating rate mortgages while housing prices (which impact recovery rates) declined,
in part because of the higher interest rates and tighter mortgage lending rules.
They also assume that the claim in the event of default is the face value plus accrued
interest. Given these assumptions, the Hull and White model is as follows:
T: Life of CDS
R : Expected recovery rate on the reference obligation in a risk-neutral world. This is assumed to
be independent of the time of the default and the same as the recovery rate on the bonds used to
calculate q (t ) .
u (t ) : Present value of payments at the rate of $1 per year on payment dates between time zero
and time t
e(t ) : Present value of an accrual payment at time t equal to t t * when t * is the payment date
: The risk-neutral probability of no credit event during the life of the swap
A(t ) : Accrued interest on the reference obligation at time t as a percent of face value
39
The value of is one minus the probability that a credit event will occur by time T .It can be
calculated from q (t ) :
= 1 q(t )dt .
T
When a default occurs at time t (t T ) , the present value of the payments is w[u (t ) + e(t )] . If
there is no default prior to time T , the present value of the payments is wu (T ) . The expected
Given our assumption about the claim amount, the risk-neutral expected payoff from the
CDS is
0
T
[1 R A(t )R ]q(t )v(t )dt
and the value of the CDS to the buyer is the present value of the expected payoff minus the
0
T
[1 R A(t ) R ]q(t )v(t )dt w 0
T
q(t )[u (t ) + e(t )]dt wu (T ) .
The CDS spread, s is the value of w that makes this expression zero:
s= 0
.
( )[ ( ) ( )] ( )
T
q t
0
u t + e t dt + u T
The CDSW model uses the function above to price the CDS spread. There are three areas
in the CDSW screen of Bloomberg: deal information, spreads and calculator. In Figure 8 we
40
Figure 8
Bloomberg CDSW Function Using Hull-White Pricing ModelIBM Five-Year Spread
Evaluated on March 16, 2007
In the deal section investors can input the company name of underlying asset; notional
value; and time payment frequency. B is show buyer of default protection. In the calculator
field, investors can choose the swap valuation model, i.e., the Hull-White model, and type <GO>
to get the price that sellers received. The term Sprd DV01 means the change in market value
when the par spread curve is increased by 1 bp (0.01 percent) and IR DV01 means the change in
market value when the swap curve is increased by 1 bp (0.01 percent). In the spread section
Benchmark indicates the swap curve selected as the riskless curve. By default the swap curve is
determined by deal currency (e.g., curve 23 for USD, 45 for EUR, 13 for JPY). These spreads
(bps) represent the swap spread added to the benchmark curve to credit risky CDS deals. For
41
Figure 8, the reference is International Business Machine Corp with a notional of 10 million. The
output shows deal spread is 9.63 bps, the amount the protection buyer pays to the seller each
year. For example, a deal spread of 9.63 bps to be paid quarterly implies four installments of
Subsequent to March 2007, the spread for protection on IBM debt increased. A protection
buyer would have to pay a spread of 29.151 bps for protection on Aug. 30, 2007 for five years of
credit protection compared to the indicated 9.630 bps in March for a five-year swap. Spreads for
credit derivatives continued to rise throughout the economy through 2009 as the credit crisis
deepened. In the case of IBM, credit default spreads on five-year bonds (with the same reference
identity) were up from 29.151 bps in August 2007 to 99.400 on Jan. 24, 2009 (Figure 9).
Furthermore, one of the main indicators of spreads in the credit markets, the spread for the North
America Investment grade CDX, reached its high point near 280 bps in November 2008, up from
42
Figure 9
Bloomberg CDSW Function Using Hull-White Pricing ModelIBM Five-Year Spread
Evaluated on Jan. 24, 2009
43
11. Credit Derivatives and the Financial Crisis of 2008
While credit derivatives experienced explosive growth through 2007 as investments and
risk management instruments, credit derivatives themselves introduced the potential for
disruption in the financial system and losses at entities with large exposures to the instruments.
The potential hazards of credit derivatives came to fruition with the credit crisis of 2008; several
major financial institutions were driven to bankruptcy, sale or bailout, due to or exacerbated by
exposure to the credit derivatives market. Of those firms, credit derivatives caused the most
significant problems for AIG. Part of the difficulties at the hardest hit institutions stemmed from
their reliance on short-term debt and subsequent inability to roll over or pay off debt the next
period.
Credit derivatives, the instruments once hailed as innovative and sound risk management
tools were blamed for many of the problems in the financial markets in 2008. In response,
derivatives traders and market participants are working on risk-reduction measures for the
market including working with regulators to expedite approval of clearinghouses for credit
derivatives trading. Interested market participants are also pointing to the liquidity and
information provided by the derivatives market as the credit market dried up in the crisis.
The systemic risk resulting from counterparty exposure in the credit derivatives market
first came to light with the near collapse of Bear Stearns in March 2008. Bear was overleveraged
and faced losses related to the mortgage and debt markets. The Fed worked with Bear and JP
Morgan (the main potential buyer for Bear) to negotiate a deal to save Bear Stearns. The Fed
44
pushed the deal in order to avoid the shock to the financial system associated with the failure of
JP Morgan agreed to buy Bear for $2 per share on March 16, 2008a total of $236.2
million, a huge discount for equity that was trading at $70 per share the week before (USAToday,
2008). The government helped push JP Morgan to acquire Bear by providing $30 billion to
provide for some of Bears illiquid assets (The Economist, 2009). However, Bears shareholders
threatened to veto the deal at that price, and JP Morgan had not yet agreed to guarantee Bears
positions. Without JP Morgans Fed-backed and proprietary capital to back Bears positions,
Bear would have to file for bankruptcy. As long as doubt remained as to whether JP Morgan
would back Bear, Bears customers refused to continue to do business. Bear needed the deal for
survival and was able to renegotiate with JP Morgan to a sale price of $10 per share, enough to
prevent Bears investors from wanting to block the sale. JP Morgan then bought enough Bear
While the Fed helped reduce some of JP Morgans risk in the acquisition, the deal did
increase risk for JP Morgan as it assumed Bears bad assets. For example, JP Morgan already
had $77.2 billion in notional value of credit derivatives at the end of 2007 and added $13.4
billion in additional exposure with its acquisition of Bear. However, overall notional exposure
may have been less than this number as some of Bears and JP Morgans may have netted out
offsetting derivatives. Furthermore, JP Morgan was better positioned than Bear to handle losses
45
The settlement mechanism in the CDS market was tested when Fannie Mae and Freddie
Mac were placed into conservatorship by the government, which counted as a credit event. The
recovery rates on Fannie and Freddies debt were high because of their government backing,
reducing the effects of the credit event on the market. However, the event did cause ripples in the
markets as buyers of the $1 trillion in protection of Fannie and Freddie worked to try to recover
One way to settle the CDSs was to deliver a Fannie or Freddie bond and receive close to
100 percent of face value (because the government backing made the recovery rate almost 100
percent). However, this method of settling had two adverse effects. First, smaller players who
wrote credit default protection on Fannie or Freddie might not be able to deliver the face value to
the protection buyers. Furthermore, credit protection buyers who did not actually possess Fannie
and Freddie bonds were driving up the prices of these bonds as they sought to use them to settle
their credit default purchase. Other investors decided to wait for the cash settlement auction to
Lehman was the next significant financial institution facing bankruptcy during the credit
crisis. In September 2008, Lehman experienced losses and write-downs stemming from real
estate holdings (mostly on subprime mortgage assets Lehman retained on its books) and turmoil
in the credit markets. The credit freeze left Lehman short of new capital, and Lehman Brothers
declared bankruptcy in September despite efforts to convince the government and others in the
industry of the systemic risks associated with its bankruptcy. The Wall Street firms that met in a
New York Fed meeting the weekend before the bankruptcy decided that they would not
46
guarantee a purchase of Lehman by a bank or another firm. Instead, they decided to pool their
resources and achieve an orderly wind-down of Lehmans assets by purchasing Lehmans assets
ranging from leveraged loans to real estate (Craig, Lucchetti, Mollenkamp and Ng, 2008).
One of the main worries associated with the Lehman Brothers bankruptcy resulted from
Lehmans positions in the CDS market. Lehman had both bought and sold billions in CDSs, and
many investors bought credit protection on Lehman itself. A Lehman bankruptcy would force
Lehmans counterparties to re-hedge their credit exposure, and would force those institutions that
sold insurance on Lehmans debt to pay out large sums to their counterparties. In order to reduce
some of the systemic effects, the International Swaps and Derivatives Association (ISDA) held a
session over the weekend before Lehmans bankruptcy to net out positions between
counterparties with Lehman. However, traders involved said that little progress was made due to
heterogeneity in the contracts terms. (Craig, Lucchetti, Mollenkamp and Ng, 2008). As CDSs
with Lehman disappeared following the official bankruptcy, the turmoil for Lehmans
counterparties of re-hedging their positions and for entities that wrote insurance on Lehman to
pay out claims spilled over to other financial institutions. Worry increased amongst investors
who sought to hedge positions with other hedge funds, brokerages and the two remaining
investment banks, sending swap spreads soaring. For example, on Sept. 18, 2008, it cost
$900,000 to insure $1 million of Morgan Stanley debt for five years (Ng, 2008).
As Lehman neared bankruptcy and frantically looked for a buyer or bailout, Merrill
Lynch recognized it would have to take quick action to prevent ending up alongside Lehman.
Bank of America opted to buy the relatively stronger Merrill as Lehman moved to declare
47
bankruptcy. Merrill had taken aggressive steps to shore up capital and to reduce holdings of risky
assets, including the sale of $30 billion of CDOs in July at 22 cents on the dollar. Despite these
efforts, Merrill had billions in write-downs in 2008 from exposure to mortgage-backed securities
and other risky assets, combined with a highly levered balance sheet (Fitzpatrick, Karnitschnig
At the same time that Lehman was forced into bankruptcy and Merrill rushed to find a
buyer, credit derivatives were causing even more problems for AIG, who wrote derivatives on
billions of dollars of debt (from corporate debt to multi-sector CDOs) through its subsidiary,
AIG Financial Products Corp. Exposure was especially problematic for AIG because it wrote
billions in swaps, meaning it was required to post collateral as default risk on the underlying
increased and subject to the full payout if the underlying debts defaulted.
The risks caused by these assets at AIG came to the fore as the mortgage-backed
securities (for which AIG had written insurance in the form of credit derivatives) risk of
default shot up. As the credit crisis deepened, default risk on corporate debt and other loans rose
as well. AIG was required to post additional collateral on the credit protection it had sold. Some
of this collateral was required in the form of cash payments to AIGs counterparties
(Mollenkamp, Ng and Siconolfi, 2008). One of AIGs main failings in valuing its swaps when
they were originally sold was their failure to recognize their risk from collateral calls if many of
the swaps required significant collateral simultaneously and the effect of write-downs on the
value of the swaps (even if none of the underlying debts actually defaulted) (Mollenkamp, Ng,
48
Counterparties began to ask AIG for billions in collateral beginning in August 2007.
However, AIG was able to negotiate and post less than asked because its valuation of the swaps
was still higher than the valuations of many of AIGs counterparties. As the value of the debt on
which the swaps were written decreased even further and AIGs exposure to the swaps became
more obvious in 2008 (especially when AIGs credit rating was cut in September), AIG was
forced to post more collateral and write-down the swaps, leaving them near bankruptcy
AIGs demise was especially threatening at the height of the credit crisis. Many of the
entities who purchased credit protection from AIG did so as a hedging measure. These entities
may have purchased protection to reduce capital requirements for the credit positions those
institutions held or to hedge their own positions in the credit derivatives market. For example,
Merrill Lynch bought significant protection from AIG on the multi-sector CDOs (credit default
obligations comprised of hundreds of securities, each of which was in turn backed by multiple
loans) it sold. When AIG stopped writing new protection on multi-sector CDOS in 2005, its
exposure was already near $80 billion (Mollenkamp, Ng, Pleven and Smith, 2008).
Many large commercial banks had large total exposure to CDS, although their net
exposure was much smaller because they had hedged their positions with counterparties such as
AIG. If AIG were to go bankrupt, rendering those contracts void, AIGs counterparties would
need to re-hedge their positions. Then protection buyers would be short capital in the eyes of
creditors and regulators. These counterparties would rush to buy protection in order to hedge
their positions again. Given the massive amount of credit protection sold by AIG, pressure would
49
have mounted in the credit derivatives market as AIGs counterparties attempted to find new
Furthermore, companies using positions with AIG to reduce capital requirements might
still be able to meet requirements at the lower protection level, but they may face difficulty in the
future if the underlying were actually to default. If the loss of protection left them short of
capital, they could raise capital instead of purchasing protection on the underlying debt from
another party. Capital was in short supply and thus expensive or impossible to come by at the
counterparties themselves could be forced into bankruptcy. A ripple effect could ensue if these
The U.S. government stepped in to prevent the systemic turmoil that might have resulted
from an AIG bankruptcy. The government issued an $85 billion loan in September 2008 to AIG,
which came along with a 79.9 percent share in AIG (preferred, dividend shares). After AIG used
much of the original loan to fund collateral calls, the government altered its intervention plan,
first raising the total value of the package to $123 billion, then to $150 billion on Nov. 10, 2008.
The second restructuring reduced the loan amount to $60 billion, but cut the interest rate and
extended the loan term from two to five years as well. Additionally, the government also gave
AIG $40 billion in capital in exchange for preferred, dividend paying shares (the governments
stake in AIG remained the same with the restructuring) (Pleven, 2008). The government lent a
total of $52.5 billion to two LLCs newly established to buy CDOs that AIG had insured and
residential mortgage-backed securities that were part of AIGs securities lending collateral
50
portfolio (Federal Reserve Press Release, 2008). The government would first acquire any
payments from these CDOs to repay their loan, and once it was repaid, to recoup AIGs capital.
Despite the relief provided to AIG by the November restructuring, the company is still in
need of cash. Much of the bailout funds from the government had been used as collateral against
AIGs swaps, which continued to fall in value (Mollenkamp, Ng, Pleven and Smith, 2008). AIG
has been scrambling to divest businesses and assets to raise capital, a feat only made more
difficult by the credit crisis, which leaves potential buyers with fewer options for funding (Wall
Street Journal, 2008). AIG has had difficulty finding buyers of even profitable businesses that
were not directly dealing in the investments that caused AIGs near-bankruptcy. For example,
AIG sold Hartford Steam Boiler to Munich Re for $742 million in January 2009, far short of the
$1.2 billion AIG paid for the specialty reinsurer in 2000 (Cimilluca, Karnitschnig and Pleven,
2008). AIG is especially interested in quickly raising capital to pay back the government in order
to attempt to renegotiate the terms of the bailout (Eckblad and Pleven, 2008).
AIGs troubles (and now the taxpayers troubles, who have an 80 percent share in AIG)
also continue to be reflected in its share price, which dropped from $21.96 on Sept. 2, 2008 to
$3.95 by Oct. 1, 2008, and has continued to fall to $1.69 on Jan. 2, 2009 (Yahoo! Finance, 2009).
Part of the reason for the sustained low share value has been the recent exposition of losses from
speculative gambles on the part of AIG on the credit derivatives market, which are not covered
51
Aside from the potential for far-reaching systemic risks, credit derivatives also pose
idiosyncratic risks to investors far removed from large investment banks or financial institutions.
Spreading idiosyncratic risks across countries and investors was exactly one of the benefits of
credit derivatives hailed by academics and industry advocates, and is now leaving many entities
decimated by losses from credit derivative exposure. Public sector and non-profit organizations
from municipalities to charities invested heavily in synthetic CDOs. For example, Australian
town councils invested $A600 million in CDOs (U.S. $423.85 million), while five school
districts in Wisconsin invested $200 million in CDOs. These investors essentially sold insurance
on the bond defaults of hundreds of companies through the CDOs. Thus, the investors gained
However, the CDOs were riskier than a pool of the bonds themselves, because investors
would be wiped out if only a small percentage of the companies covered (usually around 36
percent) defaulted. Investors in these CDOs are writing down the CDOs and taking losses as the
risk of default increases and, in some cases, as some of the underlying bonds default. Many of
these entities facing losses claim that they did not understand the risks of CDOs when they
invested. Many CDOs were designed to meet the minimum requirement for the highest credit
ratings (and received high credit ratings while still promising high returns), and most of these are
In addition to the causing losses at the level of the individual entity, systemic risk from
synthetic CDOs is exacerbating the current credit crisis. Investors in these CDOs are seeking to
hedge these instruments by purchasing CDSs on the underlying companies, which pushed swap
52
spreads to record highs in November 2008. In addition to pushing up the price for others seeking
to hedge credit risk, the increased swap spreads are making it hard for corporations to borrow,
because potential lenders look to the CDS markets to determine how costly the loans would be to
Credit derivatives are privately traded contracts on the over-the-counter (OTC) market,
through inter-dealer brokers, and thus lacked the pervasive oversight of many other financial
instruments. Calls for increased regulation and oversight were heightened by the credit crisis of
2007 and 2008. As a response, three exchanges pushed for the establishment of credit derivatives
clearinghouses, and regulators in the United States moved to expedite their creation. Players
expect that moving credit derivatives trading to central clearing will increase information about
buyers, sellers and trade volumes and reduce counterparty-risk by ensuring funds are available
before clearing trades and effectively acting as a counterparty to any trade. However, critics are
worried about concentrating counterparty risk with the clearinghouse (Fairless, 2008).
The OTC market poses an impediment to the transparency required to reduce or at least
allow market participants to fully understand their counterparty risk. In OTC markets for stocks
and bonds, the Financial Industry Regulatory Authority prohibits brokers from sharing or using
customer information in a way that distorts trades. Knowledge of another market participants
intentions before trading can allow certain parties to distort prices in their favor, and thus is
usually viewed as unethical. However, given the potential counterparty risk associated with the
OTC swaps trades, participants argue information about the identity of the counterparty for credit
derivatives trades may actually help reduce participants risk, and potentially systemic risk.
53
Central clearing could potentially help to eliminate this ethical problem by reducing counterparty
risk. Since CDSs arent technically defined as securities and thus are not officially regulated by
the Financial Industry Regulatory Authority, the rules are unclear as to how brokers can use
information about credit derivatives trades. Aside from this gray area, regulators such as the New
York State Attorney General and the SEC are investigating manipulation in the swaps trading
market; they are looking for market players who may have spread false information to
manipulate prices. Manipulation of swap prices has repercussions for more than just the
derivatives market; especially as the credit market has thinned, prices in the CDS market have
greatly influenced prices for the underlyings stocks and bonds for some firms (Ng and
Rappaport, 2008b).
In addition to these investigations into the swaps market, there are calls to mandate
Harkin is proposing a bill that requires all OTC derivatives to be traded on an exchange. His bill,
the Derivatives Trading Integrity Act, would mandate regulation of all derivatives by the
Commodity Futures Trading Commission (CFTC), including derivatives such as CDSs that are
currently excluded from regulation. Regulation by the CFTC would mean credit derivatives
could be subject to speculative position limits and federal reporting requirements (Lynch, 2008).
NYSE Euronext (New York Stock Exchange parent) partnered with LCH Clearnet Group
Ltd., a U.K. trade-clearing house, to set up a CDS clearing house in Europe. Following its
European launch, the NYSE Euronext clearinghouse was the first to receive a temporary
exemption by the SEC on Dec. 23, 2008 for CDS trades to be cleared in the United States. The
54
SECs move finally moved the United States closer to creating a clearinghouse for credit
derivatives tradinga system that regulators and some industry professionals had been calling
for since the Bear Stearns sale in March 2008 (Cameron and Scannell, 2008). The CME Group
(Chicago Mercantile Exchange parent) and partner Citadel Investment Group received approval
for the Credit Market Derivatives Exchange from the New York Fed and the CFTC and is
awaiting the SEC exemption (Cameron and Scannell, 2008). They are also in discussions with
six potential equity stakeholders as part of an effort to raise a $7 billion fund to back credit
derivatives trades made on its clearing system. The third exchange attempting to establish a
credit derivatives clearinghouse, Intercontinental Exchange Inc., is still awaiting all approvals,
but has the backing of nine of the credit derivatives markets largest participants (Bunge, 2008).
competition in the credit derivatives clearing industry (Cameron and Scannell, 2008).
Brokers in the OTC market are also working to reduce risk in the credit derivatives
market through measures to increase transparency. They are pushing for more electronic trades
(where prices are more transparent)no credit derivatives were traded electronically in early
2006, but by late 2008, 80 percent of index trades and 50 percent of single name trades were
executed electronically. Brokers are also stepping up efforts to check trade details and cancel
55
12. Lessons for Enterprise Risk Management
The rapid growth in credit derivatives and the recent financial crisis caused, at least in
part, by these instruments provide several valuable lessons for those involved in ERM. These
include:
1. Manage for risk, not for regulation. Although banks were able to reduce or
regulations, the banks still faced counterparty risk in the event the seller of the
CDS could not perform. Risk does not disappear just because regulations do not
recognize it.
2. Understand all the significant risks the organization is assuming. This admonition
applies to the chief risk officer, to other executives and to board members. If a
risk is too complex for its board and officers to understand, the organization
3. Link incentives to risk. Compensation must reflect the risk assumed to achieve
particular outcomes, not just the outcome. Employees should not be compensated
simply for volume or growth if those factors significantly increase risk for the
firm. If employees are allowed to bet the fortunes of the company and receive
high compensation when the results are favorable, they will naturally have an
incentive to do so. During the financial crisis of 2008, the high compensation
were not providing the proper incentives to employees from the standpoint of their
56
13. Summary
An attractive feature of credit derivatives is their ability to express more granular credit
views than the cash bond markets. These more refined positions increase the financial
complexity of insurance valuation, which necessitates more informed monitoring of the impacts
Macro strategies, i.e., investment grade versus high yield portfolio trading using
index products
Once the upheaval of the current financial crisis is over, it is likely that the credit
derivatives market will resume its growth and continue to proliferate newer and more exotic
credit risk products. It is also reasonable to expect that management of the credit risk retentions
will increasingly become the purview of financial mathematicians and actuaries. Developing a
working knowledge of the uses and applications of credit derivatives may soon be requisite of
57
References
1. AIG Unloads TMV Stake. The Wall Street Journal. Dec. 3, 2008: http://wsj.com.
2. Alfred M. Best Company Inc. 1940, 1941. Bests Fire and Casualty Aggregates and
Averages.
3. Ali, A.M., and Choudhry, M. Join the Game. 2003. GARP Risk Review, May/June,
Issue 12.
4. Altman, E.I., and Arman, P. 2002. Defaults and Returns in the High Yield Bond Market:
Analysis through 2001. Working Paper, Salomon Center, New York University.
5. A.M. Best Company, Inc. 2000, 2001. Best's Aggregates & Averages
Property/Casualty.
6. Barr, A. 2008. Measure of Credit Market Woes Hits another Record; Costs of Protection
Against Investment-Grade Bond Defaults at All-Time High. MarketWatch, November
20: www.marketwatch.com.
7. Barris, M., and Cowley, M. 2008. Fund Buys More Debt from AIG. The Wall Street
Journal, December 26: http://www.wsj.com.
8. Bears Pits; JP Morgan Takes Over. 2008. The Economist. March 17:
www.economist.com.
9. Berd, A., and A. Silva. 2004. Credit Portfolio Management in the Turning Rates Cycle,
U.S. Credit Strategies Commentary, June 19, Lehman Brothers.
10. BIS Semi-annual OTC Derivatives Statistics at end-June 2008. 2008. Bank for
International Settlements, December 29: http://www.bis.org/statistics/derstats.htm.
11. Boyd, R. Bear Stearns Second Brush with Bankruptcy. 2008. CNNMoney, May 2:
http://money.cnn.com.
13. Brown, M. 2008. Volume of Swaps Contracts Declines. The Wall Street Journal,
November 13: http://www.wsj.com.
14. Bunge, J. 2008. CME Lines Up CDS Backers. The Wall Street Journal, December 22:
http://www.wsj.com.
15. Bystrom, H. Credit Default Swaps and Equity Prices: the iTraxx CDS Index Market.
Department of Economics, Lund University, Sweden.
16. Cameron, D., and Scannell, K. 2008. Regulators Back System to Clear Credit Swaps.
The Wall Street Journal, December 24: http://www.wsj.com.
58
17. Chen, R.-R., Cheng, X., and Wu, L. 2005. Dynamic Interactions between Interest Rate,
Credit, and Liquidity Risks: Theory and Evidence from the TermStructure of Credit
Default Swap Spreads. Working Paper, Rutgers University.
18. Cimilluca, D., Karnitschnig, M., and Pleven, L. 2008. AIG Sells Hartford Steam Boiler
for $742 Million; Munich Re Deal Fetches Less Than It Cost to Buy Unit in 2000. The
Wall Street Journal, December 23: http://www.wsj.com.
19. Cole, M. Bear Churns. 2008. Financial Week. March 24: www.financialweek.com.
20. Craig, S., Lucchetti, A., Mollenkamp, C., and Ng, S. 2008. Crisis on Wall Street as
Lehman Totters, Merrill Is Sold, AIG Seeks to Raise Cash; Fed Will Expand Its Lending
Arsenal in a Bid to Calm Markets; Moves Cap a Momentous Weekend for American
Finance. The Wall Street Journal, September 15: http://www.wsj.com.
21. DArcy, S.P., and Gorvett, R.W. 2000. Measuring the Interest Rate Sensitivity of Loss
Reserves, Proceedings of the Casualty Actuarial Society LXXXVII 365400. Casualty
Actuarial Society, Arlington, Va.
22. Eckblad, M., and Pleven, L. 2008. Under Burden of U.S. Rescue, AIG Chiefs Aim is a
Redo. The Wall Street Journal, December 3: http://www.wsj.com.
23. Fairless, T. 2008. Derivatives Market Allows in More Light; Fearing Regulation, Some
OTC Players Boost Transparency. The Wall Street Journal, December 24:
http://www.wsj.com.
24. Federal Reserve Press Release. 2008. Federal Reserve Statement on Restructuring of
AIG Support Plan. The Wall Street Journal, November 10: http://www.wsj.com.
25. Fitzpatrick, D., Karnitschnig, M., and Mollenkamp, C.. 2008. Bank of America to Buy
Merrill. The Wall Street Journal, September 15: http://www.wsj.com.
27. Frye, J. 2000b. Collateral Damage Detected. Working Paper, Emerging Issues Series,
Federal Reserve Bank of Chicago, October.
28. Harding, P.C. 2004. A Practical Guide to the 2003 ISDA Credit Derivatives
Definitions, Euromoney Institutional Investor, August.
30. Hu, Y., and Perraudin, W. 2002. The Dependence of Recovery Rates and Defaults.
CEPR Working Paper.
31. Hull, J.C., and White, A. 2000. Valuing Credit Default Swaps I: No Counterparty
Default Risk. The Journal of Derivatives, Fall.
59
32. ISDA. 2008. News Release; ISDA Mid-Year 2008 Market Survey Shows Credit
Derivatives at $54.6 Trillion. ISDA, September 24:
http://www.isda.org/press/press092508.html.
33. Jacklin, C.J., Kleidon, A.W., and Pfleiderer, P. 1992. Underestimation of Portfolio
Insurance and the Crash of October 1987. Review of Financial Studies, pp. 3563.
34. JP Morgan. n.d. Credit Derivatives Handbook, Detailing Credit Default Swap Products,
Markets and Trading Strategies: www.morganmarkets.com.
35. Kumar, P., and Mithal, S. 2001. Relative Value between Cash and Default Swaps in
Emerging Markets, Global Credit Derivatives Strategy. Salomon Smith Barney,
November 8.
36. Lynch, S. 2008. Harkin Seeks to Force All Derivatives Onto Exchanges. The Wall
Street Journal, November 20: http://www.wsj.com.
38. McKnight, M.B. 2001. Reserving for Financial Guaranty Products. CAS Forum, Fall;
and McNichols, Monoline Insurance & Financial Guaranty Reserving, CAS Forum, Fall
2003.
39. Mollenkamp, C., Ng, S., Pleven, L., and Smith, R. 2008. Behind AIGs Fall, Risk
Models Failed to Pass Real World Tests. Wall Street Journal, October 31:
http://www.wsj.com.
40. Mollenkamp, C., Ng, S., and Siconolfi, M. 2008. AIG Faces $10 Billion in Losses on
Bad Bets. The Wall Street Journal, December 10: http://www.wsj.com.
41. Moodys Investor Service. 2006. Default and Recovery Rates of Corporate Bond Issuers,
19202005, January.
42. Ng, S. 2008. Credit-Default Market Freezes as Risk Grows. The Wall Street Journal,
September 19: http://www.wsj.com.
43. Ng, S., and Rappaport, L. 2008a. Credit-Swap Players Puzzle Over Fan-Fred Fallout;
Lehman Situation Adds to Urgency to Settle Questions. The Wall Street Journal,
September 15: http://www.wsj.com.
44. Ng, S., and Rappaport, L. 2008b. Spotlight Shines on Swap Brokers; Regulators Probe
Possible Manipulation by Wall Street's Middlemen. The Wall Street Journal, November
13: http://www.wsj.com.
45. Persaud, A. 2004. Credit Derivatives, Insurance Companies and Liquidity Black Holes.
The Geneva Papers on Risk and Insurance 29(2): 300312.
60
46. Pleven, L. 2008. Newly Bailed-Out AIG Reports Huge Loss. The Wall Street Journal,
November 11: http://www.wsj.com.
47. Trade Information Warehouse Data: All Credit Products by Customer Type and Breakout
by Product Type. Dec. 26, 2008. 2009. Depository Trust & Clearing Corporation, January
1: http://www.dtcc.com/products/derivserv/data_table_i.php.
48. Weitz, M. 2002. After Enron, Credit Default Swaps have Taken on New Urgency,
Valuing Credit Swaps. Bloomberg Markets, April.
49. Whitehouse, M. 2008. Insurance Deals Spreads Pain of US Defaults Worldwide. The
Wall Street Journal, December 23: http://wsj.com.
50. Zhu, H. 2004, An Empirical Comparison of Credit Spreads Between the Bond Market
and the Credit Default Swap Market, BIS Working Paper No. 160 (Basel, August).
51. Zingales, L. 2008. Testimony of Luigi Zingales on Causes and Effects of the Lehman
Brothers Bankruptcy. Before the Committee on Oversight and Government Reform,
United States House of Representatives, October 6: http://research.chicagogsb.edu/igm.
Stephen P. DArcy, FCAS, MAAA, Ph.D., is professor emeritus of finance at the University of
Illinois. He can be reached at s-darcy@illinois.edu.
James McNichols, ACAS, is consulting actuary with Aon Global. He can be reached at
james_mcnichols@aon.com.
Xinyan Zhao, Ph.D. is lecturer in the Finance Department of Tianjin University of Finance and
Economics in China. He can be reached at xinyanzhao@vip.163.com.
61