Supplementary memorandum from Standard
This memorandum reponds to the Committee's requests
at the evidence session on 13 November 2007. These responses contain
information which is available to Ratings Services in the time
The Committee has asked for details of fees
received from Northern Rock plc. We have always been transparent
that we are paid by the issuers that we rate and have always been
transparent as to the means by which we ensure that potential
conflicts of interest are managed. That transparency deos not
extend, however, to our disclosing confidential and commercially
sensitive details of the fees received from particular issuers.
For that reason we are not in a position to disclose that information.
By way of response to your subsequent question, however, we can
confirm that the fees received from Northern Rock plc were set
in accordance with our normal commercial arrangements.
The Committee has asked for information on our
ratings by rating category, by type of issuers, by number of issuers
and by volume. Please find annexed to this letter, tables showing:
(1) the total volume of global issuance rated
by Ratins Services by rating category in US $ millions;
(2) the total number of global issuers by
department and rating category; and
(3) the percentage of global issuers by department
and rating category.
Please be aware that the table showing the total
volume of rated issuance by rating category is based on initial
maturities. For example, if a transaction has three classes and
all three are active and currently rated AAA, the count shows
three AAAs and not on transaction. For the table showing the total
number of global issuers by rating category, regardless of issuance
volume or number, each individual institution was counted once
Please note that the term "financial institution"
is quite broad. It ranges from commercial banks to development
agencies to issuers of covered bonds. Some financial institutions
benefit from national, state or regional guarantees which elevate
their ratings. Due to varying regulations, some issuers set up
funding entities in multiple locations.
The data has been calculated as at 28 November
2007 by reference to long term issuer ratings, but excluding national
scale ratings and US public finance ratings in order to be responsive
to your request. This is a point in time data set and no trends
can be deducted from it. Please also not that bond insured transactions,
mostly rated AAA, are included in the universe.
By way of response to Mr Dunne's specific question,
using the above definitions approximately 43% of the debt at issue
rated by Ratings Services globally is rated AAA.
You referred to the suggestion, in the Bank
of England's October 2007 Financial Stability Report, that some
investors may not have "fully appreciated that rating agency
assessments are currently intended to cover only credit risk".
While we cannot speculate as to the extent to which investors
utilised structured finance credit ratings to address issues other
than credit risk, we have always been very clear that our credit
ratings speak only to the default risk associated with an issuer
or an issue and that they do not speak to other factors such as
their likely market performance or the amount that may be recovered
in a post-default scenario. We seek continually to enhance our
communications on this point.
In relation to the Bank's specific "suggestions
for improvement", at page 57 of its Financial Stability Report,
we would make the following comments in relation to each of the
suggestions, which are included below for ease of reference. Please
note that in light of the fact that these suggestions were only
published last month, we have not yet had an opportunity to discuss
them in detail with the Bank of England.
"Agencies could publish the expected loss
distributions of structured products, to illustrate the tail risks
around them. Agencies have made significant efforts over the past
few years to increase the transparency of their rating methodology
for structured finance products, through publication of research
reports describing their modelling methodoloty and their assumptions
on correlations and recovery rates. But published distributions
could provide a visual reminder of the fatter tails embedded in
the loss distribution in structured products".
A Standard & Poor's rating assigned to a
structured finance security reflects our opinion of the likelihood
that the security will pay timely interest over its remaining
lifetime and/or full principal no later than its final maturity
date. In other words, it is an opinion on the security's default
risk. Our ratings currently do not seek to quantify default risk
explicitly, for example, by associating a given rating level with
a certain "probability of default" estimate. Rather,
the performance of our ratings over time is demonstrated by the
ex post probability of default of the assigned ratings
as shown in our published default and transition studies. Similarly,
our current rating methodologies for most structured finance asset
classes do not model the probability of default for a given security
or involve deriving a probability distribution for possible losses
incurred by holders of the security. However, as noted above,
we are still reviewing this proposal and may have additional views
at a later time.
Our existing CDO rating analysis does generate
modelled loss distributions for the asset portfolio backing the
transaction, though not for the rated securities themselves. This
information is already available as an output from our freely-distributed
credit model, CDO Evaluator.
"Agencies could provide a summary of the
information provided by originators of strcutured products. Information
on the extent of originators' and arrangers' retained economic
interest in a product's performance could also be included. Such
a summary may satisfy investors that incentives are well aligned
or encourage investors to perform more thorough risk assessments".
Our rating analysts publish pre-sale and post-sale
reports for the majority of transactions that we rate publicly.
These reports include summary information relating to the transaction's
characteristics, which may include information provided by originators,
as well as details of our rating analysis. To the extent the originator
is provided explicit support to the transaction, these aspects
of the originator's risk will be disclosed. However, to the extent
the originator is holding some of the securities, and these securities
are saleable in the market, we are aware of no current way to
monitor which market participants hold of sell any given security
at any give time during the life of those securities.
Much of the information we receive when rating
a transaction is also disseminated to market participants by the
issuer in various forms, eg in the transaction's offering circular.
Most of the information we receive that is not disclosed in this
way is confidential. Our analysts are bound by our Confidential
Information Policy and the relevant provisions of the IOSCO Code
(incorporated in our own Code of Conduct). For these reasons,
we believe the extent to which data is made available to investors
in structured finance securities is ultimately a decision for
issuers or originators taking into account applicable regulations,
which are a matter for government and regulators. However we will
continue to consider whether we could provide the market with
a summary of information provided by originators in a way that
would not call into question our confidentiality undertaking and
still provide meaningful information.
Originators or arrangers often have an ongoing
involvement in structured finance transactions: for example, the
arranger may also act as a swap counterparty. In these caseswhere
the originator's or arranger's credit standing may have implications
for the tranches' default riskwe analyse their involvement
and consider their creditworthiness in the same way that we would
analyse any other entity performing the same role. In general,
the identity of investors in the tranches of a structured finance
transaction does not have a bearing on those tranches' default
risk. Similarly, the level of originators' or arrangers' retained
economic interest in this sense (eg in cases where the originator
retains some of the rated securities) does not generally have
a bearing on our rating analysis.
"Agencies could produce explicit probability
ranges for their scores on probability of default. Probability
ranges would provide a measure of the uncertainty surrounding
their ratings. Although such figures are already available retrospectively
as transition matrices, an explicit probability range would allow
investors to monitor agencies' performance when rating different
As noted above, our ratings currently do not
correlate to an explicit probability of default range. However,
we are still reviewing this possibility.
An effective way to monitor rating agencies'
performance when rating different asset classes is via historic
default statistics. These are published as part of our regular
transition and default studies, which are available on our public
website at www.standardandpoors.com. This historic data shows
that structured finance ratings have been successful diffentiators
of default risk. For example, between 1981 and 2006, the average
three-year default rate for "AAA" strcutured finance
securities was around 2 basis points (bps), compared to 168 bps
for "BBB" rated securities. Regarding comparability
of ratings across asset classes, our intention is that ratings
should provide a reasonably consistent framework for assessing
default risk across different sectors. This is also borne out
by historic data. For example, the equivalent average three-year
default rates for "AAA" and "BBB" corporate
issuers have been 9 bps and 132 bps respectively, ie broadly comparable
with the structured finance data mentioned above.
While such long-term average default rates are
useful proxies for rating performance, default rates over a shorter
timeframe must be interpreted with caution. Significant variations
in default rates can occur over time, even if ratings are assigned
with great precision. This is because default rates are derived
by observing the co-movement of multiple ratings, whose behaviour
may be more or less correlated, depending on the sample used.
"Agencies could adopt the same scoring definitions.
Currently, some use probability of default, some loss given default
and others a combination. Converging on a single measure would
reduce the risk of misinterpretation by investors".
We believe the presence of multiple rating agencies
in the structured finance market affords a valuable diversity
of opinions and choice for the users of our ratings. For example,
an investor may decide they most value the opinion of one agency
when making investment decisions in CNBS, whereas they most value
the opinion of a different agency when making similar decisions
in CDOs. On balance we would see little merit in enforcing a homogeneous
analytical framework on multiple institutions working in parallel.
Indeed, such homogeneity would deprive the market of the variety
of opinions that exist today. We believe investors are capable
of understanding what the ratings of different agencies mean,
what they can infer from them, and how they incorporate them into
their investment decision-making process. It is also important
for agencies to work with markets so participants understand their
rating definitions and methodologies. Ratings Services continues
to make efforts in this regard via multiple media, including publication
of research articles, teleconferences, and investor seminars and
"Finally, rating agencies could score instruments
on dimensions other than credit risk. Possible additional categories
include market liquidity, rating stability over time or certainty
with which a rating is made. Clear scores on there dimensions
could encourage more sophisticated investment mandates and easier
monitoring of non-credit risks in a portfolio. It would clearly
take some time and money for agencies to develop the necessary
expertise on these other risks, but some agencies have already
proposed these additions".
As discussed above, we believe it is positive
for the structured finance market for investors to be offered
a wide range of risk assessment tools and opinions. We believe
it is important that credit ratings continue to meet market expectations
by addressing credit risk. We continue to consider ways in which
we would further serve the needs of structured finance market
participants, including in the areas of liquidity and rating stability.
We note that rating agencies (or any other organisations)
already have an incentive to bring such new products to market
if they are commercially viable. The fact that no publicly available
third party assessments of different sectors' liquidity risks,
for example, are currently in widespread use, may be a reflection
of the practical challenges in forming such assessments with a
sufficient level of confidence, even though there may be demand
for such a product.
Nevertheless, as part of our ongoing assessment
of ways in which we could potentially further serve the needs
of structured finance market participants, we continue to consider
possible additions to our current product range.
|GLOBAL AMOUNTS AS ISSUE (US $ MILLIONS)
|Financial Institutions Ratings||3,869,608
|Sovereigns and International Public Finance
|Structured Finance Ratings||9,307,585
|NUMBER OF GLOBAL ISSUERS
|Financial Institutions Ratings||82
|Sovereigns and International Public Finance
|Structured Finance Ratings||9,418
|NUMBER OF GLOBAL ISSUERS AS A %
|Financial Institutions Ratings||5.25
|Sovereigns and International Public Finance
|Structured Finance Ratings||60.35