Examination of Witnesses (Questions 980
TUESDAY 13 NOVEMBER 2007
Q980 Mr Fallon: But you did not take
any action. You have written us a long submission here pointing
out how Northern Rock was over-reliant on wholesale markets, its
funding mix was historically very skewed, and so on, but you did
not change your rating of Northern Rock before the interim results,
nor between the interim results and the final crash on 14 September?
Mr Prescott: That is correct.
Mr Hancock: I would add that at
S&P we have over 2000 clients who are rated within Europe
and the fee from one of them is not going to influence a decision
that we would make on their credit standing. We would be driven
entirely by our belief on the likely probability of default.
Q981 Mr Fallon: But you are taking
fees from all of them, Mr Hancock, that is the point. You are
the traffic-lights being fixed by the speeding motorist, are you
Mr Hancock: We are driven entirely
by our reputation. If we lost our reputation, if anybody doubted
that we were driven by the fees we received from any individual
client, that would be the end of our business model.
Q982 Mr Fallon: Can you explain to
me then why you did not flash any kind of warning up about Northern
Rock between July and the middle of September?
Mr Hancock: I think if you looked
at the history of the rating of Northern Rock, we had indicated
that in the 13 years that we have been rating them they were rated
somewhat lower than many other financial institutions within the
UK. We certainly highlighted their increased, or higher, reliance
on wholesale funding. Certainly we did not expect the repercussions
of the US sub-prime market to have the impact and repercussions
on funding of all UK banks, but we have certainly highlighted
the extra risks that were involved in their reliance on wholesale
Q983 Mr Fallon: I do not think you
quite see this from the way we are seeing it. This is the first
bank run for 140 years. You are the people who are supposed to
be flashing up the warnings about the likelihood of banks getting
into trouble and then you come here and tell us you are being
paid by the very same banks and you did not give the warning.
Mr Bell: I think it is worth drawing
attention to two pieces of research or two investigations that
have taken place. As the potential conflict of interest is well-known,
we would like to draw attention to the fact that it has been researched
in the United States. A few years ago the Federal Reserve Board
asked a bunch of academics to investigate whether or not the conflict
of interest did affect the way the rating agencies assigned their
ratings. That academic research clearly came to the conclusion
that they did not and that our reputational integrity was more
important to us than the fees. In the same way, after 2003 there
was a quite legitimate investigation of the rating agencies by
the Committee of European Securities Regulators as well as the
European Parliament that was quite extensive, and they also reached
the conclusion that our conflicts of interest did not impact on
the ratings we gave. So, although we absolutely accept that these
are legitimate questions that should be asked, as there is a potential
conflict we would like to draw attention to all the evidence that
is accumulated in the case that that conflict does not impact
on the rating process.
Q984 Mr Fallon: Let us focus on the
evidence on Northern Rock. Why was it that none of you flagged
up after July the real danger facing Northern Rock from the closure
of the financial markets?
Mr Bell: I think the fact that
needs to be borne in mind is that the crisis that unfolded in
the international capital markets in August was of a very novel
kind, one never seen before, and it took the entire market by
surprise, it took the regulators by surprise, it took the Fed,
the Bank of England, the ECB, by surprise, it certainly took the
investment bankers by surprise and it took us by surprise. Our
business is to express opinions. We do so based on our best understanding
of how events are likely to unfold. Undoubtedly, if your business
is, as our business is, to predict the future and try to see into
the future, there are times and there will be things that you
do not see. We did not see the way in which the sub-prime crisis
would ripple through the capital markets of the world and impact
an English bank, like Northern Rock.
Q985 Mr Fallon: You all failed.
Mr Madelain: I would like to give
you my perspective on this situation. As has been provided to
you in various submissions, I think we made an assessment of the
liquidity situation and the risk situation, the funding strategy
of Northern Rock. I think we felt that this risk was manageable.
We had identified. I think what is important to realise
is that when you assign a rating you assign a rating to a scenario
which you think is the most plausible scenario for that institution.
You do not assign a rating to an extreme case. So, that is just
to explain what was the background for our rating. Moving into
the specifics, I think that we had obviously very early on measured
the situation on liquidity that was affecting the market, and
our view was that Northern Rock was obviously exposed. We did
engage with the bank on that situation and we moved the rating
when we were informed, effectively, of the decision of Northern
Rock to seek assistance from the Bank of England.
Mr Taylor: To answer your question
directly, the rating did not change by much, the situation changed
by a bit. Going back to your first question, which what is the
rating actually doing, the rating is addressing. If you
hold a security from Northern Rock, what is the likelihood of
you getting your money back? So, because of the strength of support
that came in there, we did not have to change the rating, the
rationale of the rating changed and was described by research
issues by us, but the risk of the piece of debt did not change
by that much if you were a bond holder.
Q986 Peter Viggers: When you were
asked whether each of you drew fees from Northern Rock, Fitch
and Standard and Poor's said, yes; Moody's was slightly less forthcoming.
Can I assume that Moody's also receive fees from Northern Rock?
Mr Drevon: We do.
Q987 Peter Viggers: In assessing
these fees (and we would like you to let us know what the fees
were) can you, please, also tell us whether these fees were proportionate
compared with other institutions comparable to Northern Rock?
In other words, were you remunerated in a manner which was perhaps
in excess of that which one would normally see from such an institution?
I would you like to put that question to you, please, and I hope
you will be able to respond. You have been criticised for the
speed with which ratings adjusted to the sub-prime crisis. Previously
you were criticised for the speed with which you adjusted to problems
with Enron. Were you satisfied with the speed that you responded
and what plans have you to improve the speed with which you intend
Mr Madelain: I think actually
we feel we did effectively respond to the situation in a timely
manner. I would like to make two comments. The first comment relates
to what we have done. All over the summer we had very intense
activity around the assessment of the situation and the communication
to the investors of the impact of the situation on the bank we
rated. We had weekly conference calls for global investors, we
published more than 25 research reports through the summer on
that very topic, so I think that the level of response and the
expectation of the investors at that stage were effectively met.
I guess the other question that one would have is what effectively
we were able to communicate. I think it is very important that
you understand that the opinions we produce are based on the public
information that is available to us in the form of financial reporting
as well as information that is made available by the banks that
we rate; and there is obviously a limit in what we can do and
that limit is affected by the amount of information that is available.
Q988 Peter Viggers: But you are experts,
otherwise it is just rubbish in, rubbish out.
Mr Madelain: But the expert is
forming an opinion using information that is made available to
you. We cannot create information that is not available to us.
Q989 Peter Viggers: Switching to
another point, to Fitch, Fimalac claim that you warned investors
about the dangers of a sub-prime mortgage market in 2005. Is that
Mr Taylor: It is correct, yes.
Q990 Peter Viggers: What did you
do about it?
Mr Taylor: What did we do about
it? One of the challenges we have had as an industry over the
last couple of years is when the markets have been so buoyant,
as we have put out comments about maybe credit deterioration or
concerns in increasing leverage, for example, in transactions,
many, many players in the investor community simply were not viewing
credit perhaps at the forefront of their investing decision. So,
even though we can draw attention to commentaries about deteriorating
credit risk, it was not being reflected in investor behaviour.
A good way of describing that is the average ratings we have been
applying, for example, in leveraged corporate transactions in
Europe have been coming down on average for the last two years.
Until recently the price was coming down as well. The increase
in risk as we reviewed it has not been reflected in market pricing;
so there was an imbalance between what we were saying, the direction
of credit and the way the market was responding to deals being
Q991 Peter Viggers: Were you content
that the information you were putting out reflected the reality
that investors would probably want?
Mr Taylor: It is very difficult
to say that, because what actually happened in the US sub-prime
market was very much unprecedented. We had not seen anything like
it before. You can argue about the magnitude of what happened
versus the magnitude of what we were talking about, but I think
the direction was clear, but it was not just us, many commentators
in the market were saying similar things.
Q992 Peter Viggers: Do you at Standard
and Poor's and Moody's accept that Fitch was ahead of you in warning
of the dangers of sub-prime?
Mr Bell: I am not exactly sure
when their warning came out. We had warnings at roughly the same
time. They may well have been ahead. I think one has to remember
two things. First of all, we did, as did the other agencies, warn
investors about the deterioration in sub-prime, but also one has
to bear in mind how you rate structured finance transactions,
which is that the criteria require a credit cushion. In order
to have a triple-A you have to have a credit cushion so the pool
of mortgages can absorb a certain amount of losses before there
are any losses at the triple-A level. These cushions were very
substantial in the case of sub-prime. They reflect our analysis
of past data. Those cushions are not simply reflective of what
happened in the past. For triple-A ratings we simply do not say:
"we assume the future will be the same as the past; what
we do is we stress the past". We look at how bad things got
in the past and we say: "we will multiply that level of rates
to create a cushion that should survive not just credit problems
in the past but a much worse credit scenario". What happened
in the case of US sub-prime, however, is that the future was much
worse than even our worst case assumptionsthe deterioration
in credit, the deterioration in underwriting, the deterioration
on the fraud side, the amount of fraud involved, and also a series
of very unusual and completely novel patterns of behaviour by
sub-prime borrowers, things that had never been seen before in
any other market. All of these combined to create a situation
that was worse than our worst case assumptions and also completely
novel, not just to us but to the entire markets. As a result,
with hindsight, if we had known exactly what was going to unfold,
we would have rated these transactions differently. However, we
maintain that we rated them with all our knowledge and all our
skill to the best of our ability.
Q993 Peter Viggers: But you are experts
and presumably well paid. Did you study the initial sub-prime
market, notice the way they were collateralised and then draw
conclusions to enable you to warn people?
Mr Bell: Absolutely. I think also
one needs to remember that we have been rating sub-prime transactions
since the early 1990s. This was not a new market for us. In 2001
there was a crisis in the then sub-prime markets, so we drew all
the lessons that we could from this. What happened is that the
crisis in 2007 was of a speed and amplitude much greater, not
just than in the past, but much greater than anything we had seen
as a worst case scenario in the future.
Q994 Peter Viggers: Can I turn to
Moody's. Brian Clarkson from Moody's said in the Financial Times
on 18 September, "A more uniform method of valuation is essential
for efficient and rational price discovery and to address future
liquidity issues." Could you expand on that statement?
Mr Drevon: I think it appears
right now that one of the main reasons why we are in these very
troubled times is not necessarily linked to the credit risk we
are seeing but more to liquidity-related issues and liquidity-related
issues due to the fall in values we are seeing on a number of
securities. Why are we are seeing this level of falls? First of
all, it is very clear that these instruments are complex and require
significant amounts of time to fully understand and, therefore,
to try to find a tool for valuation. There is no standardised
solution to that. Different banks will come up with different
solutions to value these securities and, by definition, this will
result in different methodologies across different institutions
across different markets. This creates generally concerns that
institutions may have more risk than they have been disclosing
because they are using different methodologies. So there certainly
is, I think, a clear understanding in the market place that there
needs to be more done in terms of agreeing on solutions to have
a more standardised approach for valuations. From Moody's point
of view, this is not an area that we are involved in at this stage.
We provide credit ratings, but we think it is an area which may
benefit in the market in the future. We plan on developing tools
to provide fundamental valuations to help investors with valuing
Q995 Peter Viggers: You have a unique
status under US security laws. You profit from your privileged
status and it was Mr Bell, I think, who referred to your reputational
integrity. Do you think you deserve this unique status in the
Mr Drevon: Moody's has generally
been in a position to say that we are not in favour of using ratings
for regulations. We have been very clear about that in all our
observations. We think that ratings are used for many things already,
including by investors, by issuers. They are used by investors
who buy whole issues, investors who trade, by investors who will
short sell. We think that adding regulations to that as an instrument
that will be using ratings is not something that we recommend,
Mr Bell: I would go further to
say that, although undoubtedly we do commercially benefit from
that status, we have always, at S&P, as well as Moody's, been
opposed to being part of regulation. We did not ask for this status,
we did not lobby for this status, we disagreed with the SEC when
they created this status and we have said ever since that it is
a bad idea. We welcome the changes in the law that provide more
clarity about how this status is going to be provided in the future.
Q996 Chairman: If I can go back to
another question, roughly speaking you all rated Northern Rock
Mr Hancock: Not dissimilar. There
were minor differences.
Chairman: Not dissimilar, and you all
had business with Northern Rock. I will be writing to you formally
on that regarding your relationship with Northern Rock and the
income you have, and if you decide to write back to me and say
it is confidential, then it is going to be a matter of public
record, but you will be getting a letter from me on that.
Q997 Ms Keeble: I must say what it
reminds me of is the children's game, pass the parcel. You have
got some risk bundled up, it gets passed from pillar to post and,
when the music stops, people open it and find that there is nothing
there, or next to nothing there. It makes me wonder how hard you
actually looked at the securities that you were supposed to be
rating: because once the information had come out about the sub-prime
market, it is hard to imagine how anyone could have regarded them
as sound investments at all. I just wondered, if you listen to
what the professors said as well, how far you look and scrutinised
what you were rating?
Mr Bell: We have 30 years of experience
of rating structured finance and the first structured finance
transactions, both in Europe and in the United States, were residential
mortgages. The advantage of residential mortgages is that because
the pools are quite large, thousands and thousands of mortgages,
they do respond fairly well traditionally to statistical analysis.
We have used that statistical type of analysis, taken quite a
lot of information from the pools, and we do a lot of due diligence
in that sense, over the 30 years. Our record shows that actually
it has been extremely successful in predicting the probabilities
Q998 Ms Keeble: In this particular
case it is not just that we had the first run on the bank and
all that; a lot of people, a lot of my constituents, could have
lost a lot of money, all their money, so it is desperately serious.
How far down through the structure did you actually prod to test
how viable these loans were and what this was built on? Not just
general mortgages, specifically these risky loans, obviously risky
because they are sub-prime, and it is very clear that there was
basically pretty much nothing there?
Mr Bell: I think that is not actually
correct. What you are seeing is a large number of down grades.
Right now we do not know what the ultimate default will be, but
right now the defaults on these pools are very, very small; they
are less than 1% of actual defaults. Those will almost certainly
rise, but to say of something that was rated triple-A, for example,
was downgraded to double-A or double-A minus that there is nothing
there, in all likelihood those bonds will be paid out in full.
Q999 Ms Keeble: If you bundle things
up and pass them on and keep on doing that, it is just like pass
the parcel: it keeps on going until somebody blows the whistle
and it stops. What I want to know is how far down did you go,
not just saying they have not defaulted, the record is good, we
have got 30 years experience. What analysis did you actually do
of what these securities were actually based on?
Mr Drevon: Again, our analysis
is really statistically based because we are looking at very large
pools, in the case of US sub-prime we are looking it more than
40 different pieces of data on each loan to come to a conclusion,
and that helped us to understand the level of risk in each individual
loan and then, based on that data, make assumptions about the
credit risk of those pools. It is a very serious amount of work,
extremely detailed, and we make available to the market place
the models we use to analyse that type of risk. We are very transparent
about the process.
4 See Ev 258, 272, 280 Back