Download Credit Risk Transfer - March 2005 - Bank for International Settlements PDF

TitleCredit Risk Transfer - March 2005 - Bank for International Settlements
File Size559.7 KB
Total Pages103
Table of Contents
                            Credit Risk Transfer
Table of contents
Report on Credit Risk Transfer: Summary
Do the transactions accomplish a clean transfer of risk?
Do participants understand the risks involved?
Are undue concentrations of risk developing?
Financial Stability Implications
Issues Arising During the Consultation
1. Background
2. Trends and Market Developments
3. Extent and Sources of Risk Transfer
	3.1 Fitch Ratings and Standard and Poor’s surveys
	3.2 Working Group interviews
	3.2.1 Banks and securities firms
		3.2.2 Insurance firms
		3.2.3 Other participants
4. Risk Management Issues Associated with CRT Activity
	4.1 Credit risk
	4.2 Counterparty credit risk
	4.3 Legal risk
	4.4 Operational risk
	4.5 Market liquidity risk
5. Outstanding issues
	5.1 Financial Stability Implications of CRT Activity
	5.2 Accounting and Disclosure Issues Associated with CRT Act
	5.3 Supervisory Approaches to CRT Activity
Annex 1: The risks of credit portfolio products
Annex 2: Disclosures on credit risk transfers
Annex 3: The supervisory approaches for credit risk transfers
Annex 4: Questions for supervisors of CRT market participants to consider
Annex 5: Members of the Working Group on Risk Assessment and Capital
Document Text Contents
Page 1

Basel Committee
on Banking Supervision

The Joint Forum

Credit Risk Transfer

March 2005

A later final paper on this topic has been published in July 2008.

Page 2

A later final paper on this topic has been published in July 2008.

Page 51

Credit Risk Transfer 45

investment bank on behalf of an asset manager who chooses the reference portfolio based
on fundamental credit analysis and has limited rights to actively manage the reference

Table 1: A stylised hypothetical CDO
(dollar amounts in millions)





Equity 0-3% 30 Not rated 1200

Mezzanine 3-10% 70 A 200

Senior 10-100% 900 AAA 10


Entire portfolio 0-100% 1,000 A 60

The three tranches are typically sold to different investors. In fact, some of the economic
value of a CDO stems from its ability to create credit risk-sensitive assets whose risk-return
characteristics are tailored to certain investor classes. The equity tranche, which in our
example pays a maximum of LIBOR + 12%, may be sold to a professional asset manager
managing money on behalf of institutional clients or to a hedge fund. The mezzanine
tranche, paying LIBOR + 2%, may be sold to a regional bank looking to diversify its credit
exposures. The senior tranche, paying LIBOR + 10 basis points, may be sold to a reinsurer
or other investor looking for low-risk, low return assets.21

In our example, all three tranches are funded; that is, when the CDO is created, investors
pay in the principal amount of their tranches and any defaults cause a writedown of principal.
Investors’ principal is put into a collateral account and invested in government securities or
AAA debt.22 CDOs with unfunded tranches are also common. Unfunded tranches are
structured as swaps with no up-front payments. Investors receive periodic spread payments
and are required to make a payment when a default affects their tranche. Because unfunded
tranches rely on the investors’ future ability and willingness to pay into the CDO, they create
counterparty credit risk that must be managed.

Suppose that in the first three months of our stylised synthetic CDO’s existence, no defaults
occur in the reference portfolio. At the first quarterly interest payment date, the CDO receives
quarterly payments on the reference portfolio’s 100 single-name credit default swaps, as well
as interest on the collateral account. It distributes the interest received according to a
“waterfall” of payments. First the senior tranche receives (LIBOR + 10 basis points)/4 on its

21 The weighted-average spread on the three tranches is 59 basis points, compared with 60 basis points on the

reference portfolio. If this difference is not sufficient to pay deal structuring and asset management fees, which
typically have first priority on the cashflows from the reference portfolio, the most subordinated tranche (the
equity tranche) would receive less than its maximum spread of 1200 basis points. Note also that the weighted-
average spread on the tranches is expected to decline over time as defaults reduce the principal of the high-
yielding equity tranche.

22 The CDO tranches are issued by a special purpose vehicle (SPV), which sources credit risk by selling
protection via credit default swaps. The government securities support the SPV’s obligation to perform on the
credit default swaps.

Page 52

46 Credit Risk Transfer

principal balance of $900 million. (Amounts are divided by 4 because payments are
quarterly.) Next, the mezzanine tranche receives (LIBOR + 2%)/4 on its principal balance of
$70 million. Then, the equity tranche receives up to a maximum of (LIBOR + 12%)/4 on its
principal balance of $30 million. Some of the equity tranche spread could be deferred to later
payment dates through structural means, such as subordination to a reserve account. If any
excess spread remains, the details of the CDO structure determine what happens to it. It
could be paid out to equity investors, paid out to the asset manager as an incentive fee, held
in a reserve account as an extra credit enhancement, or some combination of these.23

Suppose that in the fourth month, XYZ Inc., one name in the reference portfolio, defaults. If
the recovery rate on XYZ’s debt is 40%, the CDO will take a loss of $6 million on the $10
million notional single-name credit default swap referencing XYZ. This loss will cause a
writedown of the principal amount of the equity tranche by $6 million. Other tranches would
not suffer principal writedowns, but their mark-to-market value would fall because the smaller
equity tranche now provides less credit enhancement than before the default.

The default event demonstrates how CDOs transfer credit risk. Suppose the CDO’s
counterparty on the single-name credit default swap referencing XYZ, Inc. was a bank with a
loan outstanding to XYZ, Inc. That bank receives a payment of $6 million on the credit default
swap that can offset some of the loss on its loan to XYZ. Investors in the CDO, especially the
equity tranche investors, bear the loss. The bank’s exposure to XYZ’s default has been
transferred to the CDO investors.

1.2 CDO economics
How do CDOs make economic sense when they merely rearrange the payment priority of
already-existing credit-risk-sensitive bonds and loans? In a recent paper, Duffie and
Garleanu give two answers.24 First, CDOs help investors overcome market imperfections
associated with the illiquidity of bonds and loans. Few corporate bonds trade more than twice
a day.25 Loans trade even less frequently. Illiquidity makes it costly for credit investors to
assemble a portfolio that meets their diversification and risk-return targets. Interviewed banks
report that efficiently managing their credit risk portfolio is the dominant motive for their
participation in CRT markets.

The economic value of a CDO is apparent in the fact that the CDO’s spread income from the
reference portfolio can compensate investors in the CDO tranches and also pay deal
structuring and asset management fees. The rapid adoption of CDO technology by credit
investors suggests that the cost of creating a CDO is less than the cost a credit investor
would incur to assemble a portfolio of bonds and/or loans to meet the investor’s
diversification and risk-return targets. The high cost of investing directly in a portfolio of
bonds or loans is presumably driven by the high bid-ask spreads an investor would pay,
reflecting the illiquidity of bond and loan markets.

23 This description oversimplifies the waterfall – see the appendix for a detailed example of a cash CDO

24 Darrell Duffie and Nicolae Garleanu, “Risk and Valuation of Collateralized Debt Obligations,” Financial

Analysts Journal, January/February 2001, pp. 41-59.
25 Among the roughly 5,000 corporate bonds for which secondary market transaction data are disseminated on

the NASD’s trade reporting system (TRACE), only about 250, or 5%, traded fifteen times or more in the seven
days ending January 29, 2004.

A later final paper on this topic has been published in July 2008.

Page 102

96 Credit Risk Transfer

2.5 Breakdown by remaining maturity
Taken for CRT instruments by themselves, maturity breakdowns would be most relevant for
protection sellers, as such breakdowns indicate the duration of their risk and the time frame
over which the obligors’ creditworthiness could deteriorate. For protection buyers, maturity is
relevant in comparison to the maturity of the assets being hedged. It is not particularly
meaningful otherwise. For traders, the focus should be on managing any mismatch in the
maturities of protection sold and protection bought. It is preferable that maturity distributions
be reported using notional amounts. Three maturity bands (under 1 year, 1 to 5 years, and
over 5 years) are fairly standard and would be sufficient in most cases.

For firms active both as dealers and as either hedgers or investors, it could be helpful to
distinguish trading and non-trading portfolios in making maturity breakdowns. By showing the
trading book separately, it would be easier to assess the risk arising from maturity
mismatches in that book. It should also be noted that maturity mismatches in the two
portfolios are often managed differently (e.g., by VaR or other asset/liability management

For protection sellers, it could be helpful to have cross-tabulations of credit rating and

2.6 Effect on income
In some jurisdictions, accounting rules require that derivatives be marked-to-market. When
the hedged instrument is not similarly valued, income volatility results. While the income
results are included in the firm’s total reported income, supervisors might want to obtain
detailed information for CRT activities specifically.

• With respect to hedgers, supervisors may recognise that the asymmetrical
accounting treatment, when marking-to-market of derivatives is required, may
obscure the firm’s true underlying earnings quality. They could make adjustments
when evaluating earnings quality as considered appropriate.

• With respect to investors, supervisors might want to compare the results of
investment via CRT to that made directly (e.g., loans or other assets) or to the risks
inherent in the activity.

• With respect to dealers, supervisors might wish to evaluate the profitability of the
firm’s CRT activities in relation to the risks undertaken.

2.7 Losses
In order to help supervisors and the industry evaluate the risks in CRT transactions, it would
be helpful if firms could report the following data.

• The number and notional value of defaults that the firm has actually experienced on
CRT transactions. In order to be meaningful, the data could be given for both the
latest year and an extended (e.g., 5-year) period.

• The number of external credit rating downgrades experienced in the past year and
the resulting change in market value.

A later final paper on this topic has been published in July 2008.

Page 103

Credit Risk Transfer 97

Annex 5

Members of the Working Group on Risk Assessment and Capital

Co Chairmen: Darryll Hendricks, Federal Reserve Bank of New York
Roger Cole, Board of Governors of the Federal Reserve System

Belgium Mr David Guillaume Commission Bancaire et Financière

Canada Ms Jane Lamb Office of the Superintendant of Financial

France Ms Nadège Jassaud Commission Bancaire

Mr Roland Moquet Ministère de l’Economie, des Finances
et de l’Industrie

Mr Benoît Sellam Commission de contrôle des assurances, des
mutuelles et des instituts de prévoyance

Germany Mrs Jana Klein
Mr Reinhard Köning

Bundesanstalt für Finanzdienstleistungsaufsicht

Italy Ms Laura Pinzani Banca d’Italia

Japan Mr Yoshinori Nakata The Bank of Japan

Netherlands Mr Klaas Knot De Nederlandsche Bank NV

Spain Ms Marta Estavillo Banco de España

Ms Maribel Herrero Comisión Nacional del Mercado de Valores

Switzerland Mr Roland Goetschmann Eidgenoessische Bankenkommission

United Kingdom Mr John Carroll
Mr Piers Haben

Financial Services Authority

Mr Ronald Johannes Bank of England

United States Mr Michael Gibson
Ms Anna Lee Hewko

Board of Governors of the Federal
Reserve System

Mr John DiClemente Federal Deposit Insurance Corporation

Ms Elise Liebers
Mr Richard Mead

Federal Reserve Bank of New York

Mr Jack Buchmiller New York Insurance State Department

Ms Teresa Rutledge
Mr Kurt Wilhelm

Office of the Comptroller of the Currency

Mr Michael Macchiaroli
Mr George Lavdas

Securities and Exchange Commission

EU Commission Mr Peter Smith

FSF Ms Kristel Grace Poh

IAIS Mr Alex Lee

IMF Mr Todd Groome

Secretariat Mr Laurent Le Mouël Secretariat of the Basel Committee on Banking

Similer Documents