Examination of Witnesses (Questions 20-39)
MR ROBERT
STHEEMAN, MS
JO WHELAN
AND MR
JIM JUFFS
10 JANUARY 2007
Q20 Peter Viggers: Congratulations
on your excellent record of sickness absence or non-absence for
sickness, 2.3%, I think one of the lowest levels in the Civil
Service. Have you any tips to pass on to other departments?
Mr Stheeman: If I knew how to
do that I would. I think the one point I would make is linked
to your question about staff and what motivates staff. I do think
that we are very fortunate—and I say that quite advisedly—in
having both excellent staff and staff who are genuinely very motivated
and enjoy doing what they are doing.
Q21 John Thurso: Can I ask about
borrowing costs. While the borrowing cost trend since 1988-99
has generally been downward it seems that 2006-07 will see a rise
in average issuance yields. Given that 2006-07 sees a probably
further increase in gross gilt issuance, is the increase in yields
a sign that we are reaching the upper limits of cost-effective
government borrowing?
Mr Stheeman: I would not think
so, I certainly hope not. The increase in yields effectively is
a direct result of the macro economic environment in which we
operate. In the same way as we certainly never take any credit
for the decrease in effectively coupons over the last few years,
similarly we cannot say that we have a direct impact on that.
Yields tend to be driven much more by macro economic factors,
by global markets in particular as well. Anyone who is familiar
with the gilts market is probably aware that whatever happens
in the US market will have an impact on all other markets around
the world. The same applies to the Far East or the eurozone. Very
often yields if they have been declining in the UK will certainly
be declining elsewhere, in Europe, the US and elsewhere. Similarly,
if they rise they will rise for similar reasons. I do not believe
there is a direct link to the limit of government borrowing or
that government borrowing has an impact. There has been some analysis
from time to time or attempts to analyse that also internationally
where the supply itself can have an impact but the academic literature
on that is not at all clear to what extent that happens. I would
put out there as a thought the possibility that at times when
we were effectively in a net surplus situation and there were
issues with liquidity in the gilts market, as we had perhaps five
or six years ago, at that time I think arguably the market was
potentially crying out for supply which it could not have. That
just shows the two are not directly linked.
Q22 John Thurso: It would be fair
to say, would it not, that your activities in achieving cost-effective
borrowing for the Government are a part of keeping that cost down,
ie the yields? You are not entirely at the mercy of the markets?
Mr Stheeman: In as much as we
do have this primary debt management objective of minimising costs
that is absolutely the case, yes, but I would not say we are at
the mercy of the markets, the way I would phrase it is perhaps
slightly different, we do not take a view on the level of interest
rates when we issue, we have to issue and we issue at whatever
is the prevailing rate in the market at that time. We will of
course take a view in terms of the strategic advice we give to
the Treasury about rates and about what the structure of the borrowing
in any given year should be but we are, I would say, to a large
extent, at the mercy of the market. I do not believe that we affect
the level of rates in a significant fashion
Q23 John Thurso: The NAO told us
that since you were established and Government borrowing has increased
unit borrowing costs have generally come down. I would think that
is to some degree due to your efforts at managing that?
Mr Stheeman: Again, it is something
which I would love to claim credit for.
Q24 John Thurso: But you do not want
the blame when it goes the other way; is that it?
Mr Stheeman: I do think that it
is important to stress that we are ultimately a price-taker.
Q25 John Thurso: Is there any way
of judging what impact you do have or measuring whether you have
actually got any impact on that at all?
Mr Stheeman: It goes back, I think,
to the whole discussion which we had at the beginning. It is extremely
hard, and that is why, in terms of actually saying whether we
are doing a good or a bad job, some of this has, to a certain
extent, to be intuitive. If I can just give you a specific example:
if government issues (as the current Government is issuing in
the current year) the majority of gilts in long-dated and very
long-dated bonds and in linkers and in longs, we are doing it
clearly for a reason, and the reason is that there is a demand
out there, which we know, which has impacted, or sought to impact,
on the shape the yield curve. It is cost-effective for us to do
that. If we were to have completely ignored that and issued everything
in the most expensive part of the curve, I think you could turn
round and say, "Surely you are not doing anything remotely
like trying to meet your primary debt management objective."
I am slightly rephrasing it, but that is the way I look at it.
Q26 John Thurso: Can I ask one last
question. Have the high borrowing levels in recent years placed
any operational pressures on the DMO, and what are they?
Mr Stheeman: Yes, and that is
a very valid point. If you think about it, I think there was a
four-week period in November/December, from mid November to early
mid December, where we actually conducted more auctions than we
did in the whole first year of the DMO's existence. This year,
for instance, we have got 36 auctions scheduled. If you compare
that to, I think, four in the first year of our existence, it
has risen consistently with the borrowing requirement. Each time
that happens there are operational issues, operational risks,
which we are very much aware of, and every auction contains an
element of risk. Systems have to work, people have to be there
to actually help make sure that these things occur, and the same
applies when we are doing cash management. I think the straight
answer is, yes, these are issues and they consistently occupy
our mind.
Q27 Peter Viggers: Approaching the
same issue as that raised by John Thurso but from the point of
view of the pensions industry, where I have a declared interest,
can I put to you a scenario that longevity has consistently been
underestimated and there has been some weakness in the stock exchange
and some changes in the bond sector as well, and all of this has
impacted on the pensions industry, which increasingly has gone
in for asset and liability modelling, seeking to match its liabilities
with its assets. There has, therefore, been a very heavy demand
for longer-dated gilts. You have issued increasing amounts of
long-dated gilts in response to market demand, and in 2006 we
saw historically low yields in such gilts because of the strong
demand for them. Some commentators reported the existence of a
self-perpetuating downward spiral in long-dated gilts that was
damaging to the pensions industry. Has the Government been reaping
the benefit of low borrowing costs at the expense of the pensions
industry? I can rephrase the question if you like. Have you been
very good at your job of selling at advantageous rates to Government
but has that had other damaging effects elsewhere in the city
world? Is your brief perhaps too narrow? Are you not taking as
broad a view as you should?
Mr Stheeman: I think it is a very
legitimate question. I certainly prefer the second version, or
the first part of the second version. The comment that I would
make is, first of all, yes, you are absolutely right, clearly
the decline in yields, especially at the ultra-long end, which
was very pronounced exactly a year ago and is still low, certainly
by recent historical standards, was the subject of much press
speculation: it caused significant issues, problems, arguably,
stress for the pension fund industry. We have, indeed, a debt
management objective, which is cost minimisation. We could just
happily sit back and wash our hands and say, "Well, we are
just following." We are, indeed, as you say, reaping the
benefit for the taxpayer (which I would stress) of this very low
rate environment. To suggest that we do so without due regard
to the consequences elsewhere I think would not be correct. We
are very much aware of the needs of the pension fund industry.
The pension fund industry still represents the largest part of
our investor base. We have dialogue with a number of pension funds
constantly, with representatives of investor groups and we are
very much aware of their needs, but there is an issue. You said:
are we not ignoring potentially the needs of that group? If you
were to talk about changing the debt management objective to take
in the needs, for instance, of the pension fund industry to make
us arguably more responsive towards those, you are potentially
giving an agent of the Treasury conflicting objectives, and that
is something which I think is quite hard. What I am trying to
say is that I think it is an issue for government and for wider
government, and I know that the Treasury—and I have been
part of the group—has been discussing these issues with
the Pension Regulator, with the Pension Protection Fund, also
with the Department of Work and Pensions and officials in various
meetings. It is something which we are all aware of. For us, you
could argue I am being very narrow in my perspective, but I think
I have to be and I think it is quite a good thing. If you ask
a debt manager to minimise cost, he has to stick to that.
Q28 Peter Viggers: I respect that
answer. The question might be perhaps addressed elsewhere. Your
default position in issuance is to split between short, medium
and long gilts. How valid is this default position given the high
level of demand for long-dated gilts?
Mr Stheeman: You are right. We
have talked about this default position, and I would stress it
is only that. It is a starting point for the analysis. Clearly,
if you see what we are doing this year, the fact that we have
skewed as much as we have in literally record amounts towards
longs and linkers suggests that we are willing to deviate from
this default position where we believe it does meet these objectives.
I stress that the default position does not necessarily mean that
this is a specific preference of what government wants its portfolio
to look like. It is more a default position which is used as a
starting point for analysis and for working out what the issuance
strategy should actually be given the environment, given the yield
curve, given all these things which we face each year. It is based
on things such as risk minimisation in terms of maturity profile,
all these sorts of things, but it is perhaps worth mentioning
that the UK already has a longer average maturity in its debt
portfolio than any other major OECD country by a very long way.
Q29 Mr Newmark: May I just ask you
a pension question. I think you are absolutely right; I think
your remit should be focused on cash management and debt issuance
and I would be saying, "Yippee", if there was all this
demand from the pension industry, but I will get back to that
in a minute. According to the NAO's briefing,[3]
overseas holdings of gilts have increased significantly since
2002. How do you account for this trend and how dependent is the
successive government borrowing programme on the demand from overseas
investors?
Mr Stheeman: You are absolutely
right. It has increased significantly. You have seen the chart
in the report, but just to put that in numbers, I think since
I have joined the amount of overseas holdings have doubled, effectively,
from about 60 to roughly 120 billion, which is clearly significant.
There are a couple of reasons for that. One is quite simply because
we have increased issuance, the gilts market has become that much
more liquid. It is therefore, arguably, as a more liquid mature
market, more attractive, but I think the key point is that it
is a result of investor diversification in particular by overseas
investors, who, for instance, in the case of, let us say, central
banks, are looking to divest holdings out of purely US Treasuries
or Euro government bonds, potentially also into gilts. It is also
arguably a consequence of—
Q30 Mr Newmark: How have our rates
fared versus the dollar? What has been driving that? Why are we
a more attractive market to go into than dollars? It has to be
other than simply diversifying currency risk.
Mr Stheeman: It is largely to
diversify currencies. Put it this way, absolute yield levels in
some, but not all, maturities are at times higher here than elsewhere,
but in terms of what drives, for instance, reserves managers,
I think that does tend to be very much diversification trades.
Q31 Mr Newmark: I am curious as to
where on the yield curve this demand has been happening. Is it
throughout, is it mid term, is it the far end? Where have overseas
people really been demanding our gilts?
Mr Stheeman: To a large extent,
but not exclusively, interestingly enough, intuitively most people
are familiar with what reserves central banks invest in, what
reserves managers do, and would expect that demand to be very
much in the shorter and medium level, three to five years to 10
years.
Q32 Mr Newmark: So very different
from the pension industry demand, for example?
Mr Stheeman: Exactly. Having said
that, I should say that I have specifically come across some investment
management firms, for instance in Japan, which buy UK 30-year
bonds, in one case 50-year bonds.
Q33 Mr Newmark: I am curious. Has
the fact that the price of oil went from $12 a barrel to $70 a
barrel meant that effectively the suppliers of that oil now are
flush with cash? Has that driven the demand for gilts as well,
or is that not the case?
Mr Stheeman: Quite possibly. Some
of this is only conjecture on our part exactly where all this
is going. There is a lot of talk about the far east and central
banks, but it is fair to say that, for instance, the Russian Stabilization
Fund—you have heard of that—which invests effectively
oil revenues in Russia, announced publicly that they were diversifying
last year their holdings, which were then effectively only in
US dollars and Euros, to 45% each in those and 10% in gilts. So
that is arguably a direct result of capital flows.
Q34 Mr Newmark: Would the lower cost
of borrowing happen without this foreign interest, or would it
have been happening anyway or would you say it would only affect
the shorter end of the yield curve and not the longer end of the
yield curve?
Mr Stheeman: I think probably
it would have happened almost regardless. A benefit for us is
that our investor bases become more diverse. That is good. It
does not mean, as I say, that we ignore the needs of the pension
funds or we do not realise how important they are overall as part
of our investor base, but it is also quite nice to have a somewhat
more diverse portfolio.
Q35 Mr Newmark: I was fascinated
by your strategic debt analysis model, so I just have a couple
of questions on that, and I might have a supplementary depending
on what your answers are. The NAO's briefing on debt management
includes details about the DMO's work to develop a strategic debt
analysis model that analyses quantitatively the expected cost
and risk of various gilt issuance strategies.[4]
How do you think this particular model of yours will be and what
practical benefits will it provide? For example, will it enhance
public accountability?
Mr Stheeman: In answer specifically
to that last question, yes, I hope it will, is the straight answer.
Q36 Mr Newmark: Through what mechanism?
Mr Stheeman: By effectively providing—
Q37 Mr Newmark: Is it, effectively,
a get-out-of-jail clause for the Treasury? You have said, "This
is what our model says" and, therefore, they can say, when
Gordon Brown, or whoever makes the decisions that he makes, "This
is what the model has told me to do"?
Mr Stheeman: This is why I should
be quite explicit. The model is not there to tell us exactly what
we are going to do, and that is quite important. If I may, can
I just quote you something which Charlie Bean wrote in one of
the Bank's quarterly reports which I think is worth quoting. He
said that, "all economic models are highly imperfect reflections
of the complex reality of the UK economy and, at best, they represent
an aid to thinking about the forces affecting economic activity
and inflation". I picked on that for a reason, which is that
this is a tool which will illustrate, effectively, the results
of potentially certain strategies in cost and risk terms, but
it is based purely on assumptions and it is as good as any assumption
or as bad as any assumption you put into it.
Q38 Mr Newmark: It has no practical
benefit to the decision-making you at the debt office make. It
is an interesting intellectual exercise you and I can have a discussion
about but there is no practical benefit, it sounds to me?
Mr Stheeman: I would not go that
far. I think, to pick up your words before, it is an aid to accountability.
It allows us to show what the results are, what the trade-offs
are, if we follow various strategies, to the extent that, if you
can illustrate something by figures, it is useful and, if we can
show the scale by which something might be cheaper or more expensive,
it aids accountability. The article read probably very much to
you (which it is) like a piece of academic research. It is the
sort of work that one or two other major OECD debt managers are
also interested in trying to pursue, and we have discussed this
with them, because everybody has the Holy Grail, if you like,
of trying to develop this issue of performance measurements, of
trying to find out how can we be more accountable, and that is
the real reason. The model is not a black box which will spew
out an easy answer and tell us this is what we have got to do:
because if we happen to put in assumptions that turn out to be
wrong in—
Q39 Mr Newmark: So there is no cost-benefit
analysis to the output of the model? I guess what particularly
perplexes me is that it is intellectually fascinating but I can
pull out my HP12C or I can actually read the newspapers and decide
the yield curve at the far end is at the lowest it has ever been,
let us refinance. I do not need a model to tell me this. Shorter
term gilts, middle term gilts. Let us do a favour for future generations,
for my grandchildren and your grandchildren, and issue as many
of these 50-year long-dated gilts as possible because we are doing
our grandchildren a great benefit. Why go on with the fiction
of paying a higher cost of capital for middle-term and shorter-term
gilts today? I do not see the logic to it and I do not need a
model to tell me that.
Mr Stheeman: The model will not
tell you that, that is absolutely correct, but (and I will move
away from the model to try and answer that specific question)
there is a reason why we have specifically, even in this environment,
even with the shape of the yield curve that we have in the current
remit, not ignored the short or the medium end of the market,
and we have issued effectively the minimum that we can in order
to create a single benchmark in both short to mediums, five or
10 years, 10 billion each, and everything else we put into longs
and linkers.
3 http://www.nao.org.uk/publications/nao_reports/06-07/debt_management_office.pdf Back
4
http://www.nao.org.uk/publications/nao_reports/06-07/debt_management_office.pdf Back
|
