15:57 07Jan11 RTRS-IFR-ABS-Raters ignored internal feedback during CDO frenzy
by Adam Tempkin
NEW YORK, Jan 7 (IFR) - During the peak of the CDO boom, analysts
at one rating agency may have ignored warnings from their very own
surveillance department in order to continue handing out high ratings
on complex securities, says a research paper to be presented this
Sunday at the annual meeting of the American Economic Association and
featured in the 2011 edition of the American Economic Review.
The report, titled "Did Credit Rating Agencies Make Unbiased
Assumptions on CDOs?", claims to be one of the first to provide
concrete empirical evidence, as rating agencies remain a prime target
of academics trying to understand causes of the financial crisis.
The paper asserts that in the run-up to the credit crisis, the
group responsible for rating new CDOs within a major rating agency
used far less stringent assumptions to rate the deals than those used
by the firm's own structured-finance surveillance department, which
is a separate group responsible for monitoring the transactions after
issuance.
The evidence supports that this difference can only be explained
by the agency's desire to protect high ratings, the paper says.
"We find systematic discrepancies between the groups -
assumptions made by the ratings division are more lenient than those
by the surveillance department," wrote John Griffin, a finance
professor at The University of Texas at Austin, and Dragon Tang, a
finance professor at the University of Hong Kong.
Moreover, the paper alleges that updated estimates by the
surveillance group "were more accurate but seemingly ignored" by the
new-ratings group, which continued to rate new CDOs with looser
correlation and collateral credit-quality assumptions, despite
signals from the surveillance department that the transactions should
have been rated lower.
The authors say that their data comes from "one of the two
leading rating agencies" but do not divulge which one. However, the
academics discuss in detail a certain model output they analyzed
called "scenario default rate", or SDR - a data point and acronym
only used by Standard & Poor's, a unit of McGraw Hill Cos Inc
<MHP.N>.
The sample CDOs were issued between 2002 and 2007.
S&P said that they did not have a comment on the research paper.
Tom Lemmon, a spokesman for Moody's <MCO.N>, gave the following
statement: "While we have not fully reviewed the paper in question,
the CDO rating methodology it describes does not appear to be a
Moody's methodology. Moreover, the surveillance practices it
describes are inconsistent with Moody's approach to CDO surveillance,
which employs the same methodology when assigning new ratings and
conducting surveillance of existing ratings."
OTHER EXPLANATIONS REJECTED
Significantly, the researchers' dataset is restricted to the 355
CDOs with surveillance reports dated within 180 days of the initial
rating assignment, meaning that the surveillance group's adjusted
assumptions were made known shortly after issuance.
According to the authors, despite the surveillance group's more
negative outlook, there were no downgrades issued at the time of the
first surveillance report for each of the CDOs, and the tightened
assumptions were not immediately incorporated into the new-issue
ratings group's methodology.
The professors also present statistical evidence that the
surveillance reports were more accurate than the ratings group's
initial "pre-sale" or "new issue" reports, which are typically
published around the time the rating is issued to facilitate the
closing of the CDO.
The data indicates that the surveillance group calculated much
more pessimistic collateral credit quality and higher asset-default
correlation for the CDOs compared to the assumptions made by the
ratings team.
"If the rating agency had new information from the surveillance
group and acted on it then it would indicate that the rating agency
was learning from the surveillance team and trying to correct
mistakes made by the ratings group," the paper says.
"However, if the ratings agency did not act on information coming
from the surveillance group, then this would indicate that the firm
was compromising its standards."
Professors Griffin and Tang maintain that the discrepancies do
not appear to be explained by changes in collateral composition, the
length of time between reports, or the collapse of the subprime
mortgage market.
"Since the breakdown in CDO credit ratings was at the heart of
the credit crisis of 2007-2009, our findings suggest that conflicts
of interest may be much more economically important than previously
surmised," the authors wrote.
SURVEILLANCE OFTEN AT ODDS WITH NEW-RATINGS GROUP
According to the paper, common perception is that prior to the
financial crisis, ratings analysts received higher compensation and
more staffing than surveillance teams.
The authors imply that there may have been a natural tension
between the two groups, as surveillance analysts are less influenced
by conflicts of interest, and hence could make more objective
assumptions, the professors said.
"We expect the ratings committee to have more discretion than
the surveillance group - the surveillance group is more reminiscent
of a compliance or risk management division," the paper notes. "While
a surveillance department may be forced to corroborate the ratings
department, they have some autonomy and may not fully communicate
with the ratings group."
As the credit crisis deepened and new issues in RMBS and CDOs all
but dried up, the agencies put more focus on and expanded their
surveillance teams in order to re-assess billions of dollars of
complex securities that needed to be downgraded due to souring
mortgage collateral.
[Adam Tempkin is a senior IFR analyst]
[Adam Tempkin was communications manager of
S&P's structured finance group from 2002 through the end of 2009.]
((adam.tempkin@thomsonreuters.com; Reuters Messaging;
adam.tempkin.thomsonreuters.com@reuters.net))
