Examination of Witnesses (Questions 200
TUESDAY 9 OCTOBER 2007
Q200 Mr Fallon: Why was there a five-day
gap from 9 August to 14 August before you alerted the Treasury?
Sir Callum McCarthy: Because the
identification of Northern Rock was a reasonable thing for us
to consider. The ninth to the fourteenth does not seem to me of
particular materiality in this. You may take a different view.
Mr Sants: If I can maybe amplify
a little bit. Certainly as of the ninth we were in regular discussion
with Northern Rock in monitoring their liquidity, but if you look
at their liquidity availability in terms of days, which is a way
of looking at their liquidity regime, you are not actually looking
at a significant deterioration in their profile until well after
14 August. So we properly identified that, because of their dependence
on securitisation, which required them in general to do around
£5 billion of securitisation in a quarter, with the closure
of the markets this was potentially an at-risk firm, but the actual
deterioration in the profile does not occur until well into September
and, indeed, it was not until 10 September that they reached a
conclusion that securitisation was not going to be possible. They
were actually in negotiation, as you know from our memo, with
a number of banks about the possibility of doing under-written
securitised transactions during that period. So, in terms of alerting
the Treasury to the fact that we anticipated a significant issue
with Northern Rock, which was done by myself on 15 August, I would
say that was very early to alert them to specific concerns about
a specific firm in the light of the liquidity information we had
available and not in the light of our knowledge of the business
model. So I think we very quickly identified that that business
model was at risk. Also in passing, whether we had told the Treasury
on the fifteenth or a few days earlier would not have made any
difference to the set of circumstances that transpired.
Q201 Mr Fallon: Sir Callum, you were
quoted on a BBC website in September as describing Northern Rock's
heavy reliance on short-term loans to fund its mortgage business
as "extreme". When did you first come to the conclusion
that Northern Rock's business model was extreme?
Sir Callum McCarthy: I actually
said the business model was extreme but it was in relation to
the fact that they had a heavy dependence overall on what I will
loosely describe as wholesale funding. Their overall pattern of
funding was around more 70% from securitisation, covered bonds,
long-term, and in that respect they are an outlier in terms of
most British banks, though not necessarily banks in other countries.
I would point out that that is not necessarily a source of vulnerability,
the long-term funding, because those long-term funds can match
the assets that people have.
Q202 Mr Fallon: What is the answer
to my question? When did you first realise that this model was
Sir Callum McCarthy: In terms
of its reliance on securitisation in terms of the overall balance
sheet, that was something that was well-known. I do not know what
time in the last two years, three years I became aware of it,
but it was well-known that that was a particular feature.
Q203 Mr Fallon: You were in charge
of supervising Northern Rock, you were aware that its business
model was extreme, yet over the last two years nothing was done
to prevent this particular crisis.
Sir Callum McCarthy: No, that
is not a description of events that I would recognise. I have
tried to explain that my comment on "extreme" related
to the overall balance sheet of Northern Rock. The particular
problem, as Hector has explained, related to the short-term funding,
and the short-term funding is a problem which has been acute but
has been caused by the fact that they had access to securitisation,
to covered bonds, to commercial paper and had high quality assets
to repo, and they did that in euros, in dollars, in sterling,
and all those markets, including the repo market, closed and that
is an exceptional, indeed unprecedented, set of events to have
occurred for the duration and severity that has occurred. I absolutely
accept, as Hector has said, that we did not identify the probability
of that happening. I would also say that very few people and no
regulator that I know anywhere round the world have succeeded
in identifying that.
Q204 Chairman: To clear up, Mr Sants,
you said you had three supervisors for high impact banks of which
Northern Rock was one and there are 160 institutions that are
high impact. Is that correct?
Mr Sants: Yes, I am saying that
depending on the high impact institution, the number of supervisors
tends to vary between about two and six so three was a fairly
standard number within our normal range for high impact banks.
Even the biggest UK financial institution would not have more
than six or seven supervisors.
Q205 Chairman: I have spoken to a
number of the biggest UK financial institutions in the past week
in preparation for this inquiry and one of them in particular
told me that FSA have had a line side-team dedicated to them and
over a year they could see 1,000 to 1,200 people in the FSA in
terms of supervision. The group risk director is almost in intimate
contact with the FSA almost on a daily basis and the FSA are coming
in for themed visitsfor example safety, private equityso
there is intensity there. That was described to me, that intensity.
Are you saying the same intensity was provided to Northern Rock?
Mr Sants: I am saying that certainly
in the period in question during the second half of 2006 and early
2007, yes, I was answering the very direct question, possibly
not giving a full answer in that respect of the precise number
of dedicated supervisors in our supervisory group, but the supervisory
model is like that of other investment banks, as I have indicated
earlier, namely you have coverage supervisors, you have the relationship
with the bank and then you have a series of specialist teams who
regularly visit the bank on particular issues. So, the question,
for example, of stress testing would be addressed by a specialist
team who come and visit to look at the stress test, and that was
the visits that were carried out in this case in April and May
2007 and, indeed, we also have teams looking at the securitisation
process and so forth during that period. So, if you are asking
the question about the total number of people involved in the
FSA engaged with Northern Rock, you would have a much higher number.
Q206 Chairman: The question I am
asking, before we go on, was it of the same intensity as your
relationship with say the big banks?
Mr Sants: Yes, for our high-impact
institutions we have the same coverage model, which includes all
the specialists that you refer to.
Q207 Mr Todd: Can I get some procedural
stuff straight. In your annual report you refer to stress testing:
"We reviewed the stress testing practices in ten large firms
in the banking, building society and investment bank sectors.
Was Northern Rock one of those ten?
Mr Sants: No, it was reviewed
Q208 Mr Todd: So it was reviewed
after this report was concluded?
Mr Sants: Yes.
Q209 Mr Todd: So you did have a stress
test under this model that is referred to in your annual report
Mr Sants: We reviewed their stress
test in the context of their Basle application, and it was that
review which led to the conclusion being reached in July, which
I referred to earlier, that their stress testing could take into
account more extreme scenarios than they were and, as I have already
acknowledged in the earlier statement, I think, in terms of our
lessons-learnt exercise, we do need to return to our supervisory
engagement with the stress test.
Q210 Mr Todd: You have said what
is later said in your annual report in the same paragraph in which
you say (and this was after seeing the ten firms which did not
include Northern Rock, and so presumably you had already worked
out some of the inadequacies in your stress testing then) that
further improvements were needed, particularly where firms were
not fully taking into account severe but plausible scenarios when
making strategic all-risk management decisions. So you already
had some intelligence from the stress testing models that you
had applied elsewhere than Northern Rock on the perhaps limited
compass of that exercise. Did that not give you any hints as to
the insight you ought to apply to a business model which I think
Sir Callum was not alone in regarding as an outlier in this market
Mr Sants: I agree with you. As
I said before, if you look at the type of stress test Northern
Rock was using, they were not anticipating closure of the securitisation
market and the repo market. The only set of circumstances actually
which they had in which those type of closures occurred were operational
failures rather than market failure and, as I said earlier, I
think that type of scenario should be in a stress test; and we
would like to see more extreme stress tests and we were making
those points, as you kindly point out, in the document in question,
and it is incumbent on us to make sure that we carry that through
with all the major firms that we regulate, and I think that lessons
learnt point, as I have said before, needs to be picked up in
our supervisory practices and we will be returning to
Q211 Mr Todd: My point was a slightly
different one, which was that from what one can understand from
your annual report, which refers to a period before this, you
were already learning some of those points about the lack of testing
of severe but plausible scenarios. Presumably this particular
scenario did not fit into your category of severe but plausible.
You regarded it as wholly implausible, did you?
Sir Callum McCarthy: I think it
was unprecedented, and I would say that we have for some time
been emphasising the importance of severe but plausible. I would
also point out that that is a matter of judgment, it is particular
to any institution and it is quite difficult to decide what level
of stress to test against.
Q212 Mr Todd: Can I follow this line
of argument a little further? We have got again in your Annual
Report, if you turn to 66, 67, the role of the Risk Committee
of the FSAand you have already referred to one of the august
members we met earlier. In the list of risks that the Committee
does consider one can see some resemblances to some of the issues
that have occurred in this particular case. Is it perhaps the
case that this Risk Committee treated this as a rather academic
exercise of running through risks in a routine way or did not
actually consider this in the depth that one might expect? What
was this risk committee actually up to?
Sir Callum McCarthy: Perhaps I
could describe the Risk Committee of the Board. It is chaired
by Hugh Stevenson; its members are Deidre Hutton, Peter Fisher,
who is ex New York-fed, New York based, both an ex-central banker
and an investment banker nowadays, David Miles, whose proper title,
I think, is the European economist for Morgan Stanley, and John
Q213 Mr Todd: They are not lightweights.
Sir Callum McCarthy: They are
absolutely not lightweights and were carefully chosen for that
Q214 Mr Todd: They have big reputations
Sir Callum McCarthy: If I may
say so, the idea that they were looking just at what I think you
described as academic points is absolutely not the case. They
looked at a variety of issues and those included the credit risk,
derivative risk, the sub-prime risk in the US; so they were examining
Q215 Mr Todd: I have got that. We
are tight for time. I just want to explore the linkage between
your Risk Committee and the stress testing models that you have
used. Is there some linkage? You have just touched on one issue,
sub-prime markets and its possible implications. Is there any
linkage in which the risk committee communicates to those who
deliver this process on the ground?
Sir Callum McCarthy: Yes, indeed.
One of the responsibilities of the Risk Committee is to examine
the way in which the FSA mitigates against the risks that have
Q216 Mr Todd: You prepare fact books.
Was there a fact book on Northern Rock?
Mr Sants: A fact book is a very
particular statement about a particular set of data which we are
preparing for the Tripartite, and that is an electronic set of
data which is held on a particular IT system and is not yet completed
for any institution currently in terms of a central IT depository.
We carry data of the same nature on all major high-impact institutions.
So the answer in that sense is yes.
Q217 Mr Todd: I am sorry, the answer
is yes but you sound as if you are at an early stage. It sounds
as if the answer is no actually.
Mr Sants: I am slightly confused
by your question. If you are asking me is the FSA holding the
same set of data on Northern Rock as it does on the other major
UK institutions, the answer is yes. If you are asking a very particular
question about an IT system, the answer is no because it is not
generally ready, it is in development.
Q218 Mr Todd: Even though actually
this was referred to as to something that was needed back in October
Mr Sants: Yes, I agree. It would
be a useful tool for speedy decision-making.
Q219 Mr Todd: It is one of these
projects which is rolling away gently in the background.
Mr Sants: Yes. It would be a useful
tool for speedy decision-making, but in the context of this issue,
which arose over a long period of time, I do not think it is a
relevant gauge of our handling of it.