Examination of witnesses (Questions 20 - 39)
WEDNESDAY 15 JULY
MR C REEVES,
MR A BARTON
and MR R DOUGLAS
20. It might be helpful to the
Committee as to what the sort of ballpark risk figure for a basket
of PFI schemes is as opposed to your actual experience for similar
projects of a similar size and similar number under the previous
arrangements. Would that be possible?
(Mr Reeves) Yes. What we do have is information on
exchequer schemes over the past in that we use three-year moving
averages in terms of time and cost overruns. Traditionally there
have been substantial time and cost overruns in the 1960s and
1970s. That has diminished in recent years. On average now we
are saying that a time overrun would be approximately nine per
cent and a cost overrun would be about five per cent. Of course
we are not in a position to compare that with PFI schemes at present
because none of the larger schemes has reached three years.
21. Except that under the PFI schemes you are
allowing up to eight per cent of risk and you said you produced
that figure through some calculation or some assessment. Presumably
these assessments are made on historical data. It would be useful
to see that.
(Mr Douglas) The Committee has asked for copies of
the business cases for these and the risk assessments. The business
cases and risk assessments will show very clearly the assumptions
we have made about the cost of risk on each individual scheme.
What I can do is make sure that elements like the construction
risk are pulled out very clearly.
22. Could you give us a very basic figure on
what your actual experience is over capital funding and extra
cost, not anticipated but obviously factored in somewhere, and
what you are formally factoring in under PFI?
(Mr Douglas) Yes.
23. That would be very helpful. I cannot find
anywhere in the tables whether an assessment has been done over
the lifetime of the scheme, say 30 years, because that tends to
be the break clause, of the services element, the total package
of capital plus services, comparing that cost under the PFI against
that cost if it were provided through the traditional publicly
funded route.
(Mr Douglas) That is in there. The assessment we do,
where we compare the public sector comparator with the privately
financed scheme, does not just bring in the capital construction
element, it brings in the whole cost that we will be paying over
the lifetime of that scheme.
24. Can you point me to the relevant figures
in our papers?
(Mr Douglas) The tables we have in there in 4.8f give
a breakdown of the net present value calculations. There is a
run of tables starting with Carlisle Hospital, North Durham. These
bring in the discounted costs over 30 years or 25 years, depending
on the length of the scheme, of all the elements of that scheme.
25. It would be lovely to see how that is actually
worked out. What are your risks? I thought that the main risk
was in your design not being right or your construction not being
right or unexpected construction costs. If you are going to factor
in risks over 30 years, you are a jolly clever person if you can
assess that as being £21 million out of a £152 million
contract.
(Mr Douglas) We have a basket of risks for each scheme,
30 or 40 different risks, which will cover design, construction,
the operation of a facility, availability, performance, a whole
range. If we provide the Committee with the full business case
and analysis of those risks that will actually break it down into
that level of detail.
Chairman
26. May I pursue one point, having looked at
the illustrations you give us here from Carlisle Hospital onwards?
All the schemes you have described in your answer indicate that
the PFI option costs less than the public sector comparator, taking
account of risk. Do you have any examples of schemes where the
PFI option has been tested against the public sector comparator
and found to be wanting, where it has gone the other way?
(Mr Douglas) We would not approve a scheme and let
it go ahead if that were the case.
27. Yes, but have you found any examples where
that has been the case?
(Mr Douglas) I am trying to think whether we have
or not. I think the main reason for going ahead at Royal Berks
and Battle as a publicly funded scheme was that we could not get
a value for money PFI scheme at the time.
(Mr Reeves) I was going to confirm exactly that. That
is the biggest hospital to date which is now being funded by exchequer
capital, where we tried on a number of occasions to do recalculations
in terms of the relative merits of PFI and exchequer capital and
in terms of net present value the PFI option was not a favoured
one.
Audrey Wise
28. It all depends on the risks does it not?
(Mr Reeves) Very much so.
29. Completely. If you look at North Durham,
you have it working out there exactly the same.
(Mr Douglas) Yes.
30. I should be really interested to see more
detailed examples because I cannot see how you can work this out.
My question really is: where do you buy your crystal ball?
(Mr Douglas) It is clearly not an exact science. There
are methodologies which are endorsed by the Treasury as to how
to do this based on assessing the probability of events happening,
what the consequence of that event happening is and then coming
to a figure, testing the sensitivity of that. The whole basis
of PFI is that we do transfer risk and I would not expect any
scheme to be cheaper before the risk adjustment. That would be
very, very unusual. It is the risk and the transfer of risk and
the management of risk by someone else which actually gives you
the benefit from PFI.
31. If there is a benefit from PFI.
(Mr Douglas) On our assessment we would only go ahead
with these schemes if that showed us there was a benefit from
PFI.
32. It is just about impossible for us to take
any sensible look at it without more detailed examples of what
these risks are, how they are calculated, when they might occur
and all that. Looking at it, if I were uncharitable, I would think
with the slightest skewing, if it were done by somebody with my
predilection, I could readily get it going just a tiny bit the
other way. It is just a tiny bit your way, is it not? These are
not huge sums, are they?
(Mr Douglas) No, these are not. It is quite clear
to us that this first round of PFI schemes, although they do offer
us value for money, do not offer us a very, very big improvement
on the public scheme. We would hope, as PFI develops, that we
would actually start to get more value for money out of these
schemes. You are right, in most cases, there is a relatively marginal
difference on these figures.
33. It would be well worth interested people
looking at exactly what is given up for this very small assumed
benefit.
(Mr Douglas) We have moved to making the business
cases publicly available to people. These will be available to
anyone to test and examine. I should add that they do go through
a very rigorous approval process. First of all the trust has to
satisfy its board. They then have to satisfy the regional office,
the regional office then has to satisfy us at headquarters and
we then have to satisfy the Treasury. At all stages one of the
things being looked at is this risk assessment and the valuation
of that risk.
Dr Brand
34. Might it be useful, when we have some exemplar
business cases, to have a session on PFI and take evidence on
it? I was interested to hear you say that you look at formal schemes
and first consider them for PFI and if that fails you go along
other routes. That is actually rather a change from using PFI
for very pressing cases where they cannot think of anything else
which might finance it. Clearly that is something the Committee
would benefit from having some more views on.
(Mr Reeves) There has been a sea change in terms of
PFI and certainly, going back to the early 1990s, I recall it
was very much a provider orientated approach that trusts would
look in isolation at what its requirements were. We have done
one major change in terms of making the whole issue more strategic
and more purchaser orientated. The whole idea of a strategic outline
case, which must be submitted by each health authority and trust
with the agreement of the regional office and that strategic outline
case being submitted to the CPAG group, is a very important breakthrough.
The other point to make is that although we are focusing more
on PFI that is not to say we do not put a great deal of importance
on exchequer capital and, if nothing else, yesterday's statement
on the comprehensive spending review suggests much more focus
will be placed than in the past on exchequer capital. It is back
to Dr Brand's point about time and cost overruns. Certainly the
Secretary of State is very keen to change the approach to exchequer
capital. May I name three ways he is thinking about doing that?
First, making contractors responsible not only for the construction
but also the subsequent maintenance of the scheme to put more
pressure on them to reduce time and cost overruns. Second, we
are thinking about the possibility of fixed price contracts instead
of variable price contracts in future. Third, we are looking at
the idea of possible turnkey contracts where the payment is made
to the contractor after the construction of the scheme is completed.
They are all three variations on the basic theme of trying to
reduce time and cost overruns. The final point we would make as
well is that we are actually thinking of a hybrid approach between
exchequer capital and PFI. We use the acronym DBFO, which is design,
build, finance and operate, and under a PFI scheme the private
sector would be involved in all those four elements. They would
design the scheme, build it, finance it and operate it. Under
a hybrid approach which we are looking at, there is a possibility
of the private consortium designing, building and financing the
scheme but the operation would lie solely with the NHS. We feel
that because the private sector is putting up its own money in
terms of that, that again will have an implication in terms of
reducing future time and cost overruns. To summarise: two major
changes. One is the focus on the purchaser much more than in the
past. Second, the importance of thinking not just in terms of
PFI for the future but also the importance of exchequer capital.
It actually comes back very much to the opening question you put:
one of the big concerns I have at the moment is that the amount
of maintenance in terms of exchequer capital tends to diminish
when there are pressures on the revenue account. 1996-97 was a
classic example of that, which is why the capital expenditure
fell that year. The quid pro quo of that, the corollary of that,
is that backlog maintenance increases as a result of this capital
virement to revenue. We have backlog maintenance now of £2.7
billion. An important focus on exchequer capital in the future
will be to try to ensure we can reduce this backlog maintenance.
Mr Gunnell: It sounds as though there is a case
for the Committee having a session on PFI, particularly to know
what difference it would make to the schemes and to the overall
financing of the schemes if the operation of the scheme once complete
were an operation carried out by NHS personnel.
Audrey Wise
35. When you supply the extra information, would
you make sure that you include the Norfolk and Norwich Trust?
I should be really interested to see the case made there, because
I know there was controversy in that area about that particular
thing, indeed it ran to demonstrations outside Richmond House.
I know that there was removal from a hospital which was on a centre
site to a very much less convenient outskirts site, probably with
release of value of land. I should like to make sure that when
you supply the information, it takes in all those sorts of factors.
(Mr Douglas) I could supply the four business cases
for all the schemes we have now completed, and if it would be
helpful to the Committee I could supply someone to talk to your
advisers and take them through it. The pile of documents will
be about that big.
Chairman
36. We should be very grateful. I am sure our
advisers would be very excited to do that.
(Mr Reeves) It is a very complex area. I remember
four or five years ago coming along when capitation was a very
big issue in 1995-96 and doing a presentation to the Select Committee
to try to increase the understanding of what is a very complex
issue. The same applies in terms of this area.
Mr Walter
37. I wanted to look more at the global spending
and I am looking at Table 2.1.1 in your submission. Looking at
the tables as presented, three tables into 2.1.1, total spending
on the NHS £38,199 million in 1998-99 compared to the forecast
outturn the previous year of £36,443 million, which is a
2.3 per cent increase, the question I want to ask is whether or
not we are still looking at the right figures in light of what
was announced yesterday? Obviously yesterday's comprehensive spending
review was announced but we only have very brief outlines of figures;
we may get more detail tomorrow. Are we still looking at the right
figures and do you or do you not think that kind of increase will
prove sufficient to meet the changing demographic and technological
needs we have talked about before in this Committee and also those
new policies which were in the NHS White Paper?
(Mr Reeves) In terms of Table 2.1.1 the figures for
1998-99 show that the planned increase in total cash spending
will be 2.3 per cent in real terms. Obviously in terms of the
announcement yesterday, the comprehensive spending review, that
will affect the years after 1998-99 and indeed there has been
a substantial increase. May I quote the actual increases in terms
of yesterday's announcement? In terms of the year 1999-2000, which
is the first of the three years of the comprehensive spending
review, we are talking about a real terms increase there of 5.7
per cent over the 1998-99 figure and then in terms of 2000-2001
compared with 1999-2000 it is an increase of 4.5 per cent. Finally,
in the third year of the comprehensive spending review, an increase
of 3.9 per cent. In terms of the average over those three years,
we are talking about an increase of 4.7 per cent based on an assumed
GDP deflator of 2.5 per cent. Obviously in juxtaposition the comprehensive
spending review is putting substantially more monies into the
NHS, as is indicated from the figures in Table 2.1.1. Having said
that, everything is relative. If you compare the planned figure
for 1998-99 of 2.3 per cent that compares with the similar figure
in terms of the forecast outturn for 1997-98 and indeed is substantially
in excess of the figure for 1996-97. At this stage, yes, we are
anticipating in the comprehensive spending review—I am sure
the Secretary of State will say this next week—more real
term growth in the NHS. May I come back to the figure you quoted
of 2.3 per cent? This is a planned increase in total cash spending
and the important thing there is what the additional cash can
buy, in other words what its purchasing power is. That is affected
by a number of issues. You quite rightly raise demography as one
issue. Having said that, the increasing costs of the elderly are
projected to reduce in the future and for 1998-99 we are talking
about an assumption about additional costs of 0.3 per cent. This
is substantially less than the costs in the early 1990s and up
to 1995-96 we were talking about potential increases of 0.8 per
cent. The cost of demography is certainly less than it was in
previous years. You mentioned technology. That is a very difficult
area. Sometimes increasing technology is more expensive even if
it is more cost effective. Sometimes of course new treatments
are less expensive. It is very difficult to make a judgement on
this. I do recall some research being done by the King's Fund
some years ago which suggested that technology could increase
cost by 0.5 per cent per annum, though I have to say I think that
was some years ago and I think the figure was extremely tentative.
So a little bit uncertain in terms of the implications of technology.
Two other points I could make. One is that this figure of 2.3
per cent is based on the current GDP deflator based on March 1998
of 2.9 per cent. If that is wrong of course the 2.3 per cent will
change as well. The one important point I could make is not reflected
in these figures and that is that we are thinking about efficiency
to a great extent in 1998-99. We did ask our health authorities
to increase efficiency by 2.3 per cent and we do not think that
is unrealistic, bearing in mind the increases in recent years
and certainly some of our current work suggests the differences
in unit cost between hospitals is quite substantial. If all hospitals
could reduce their costs to the level of the most efficient 25
per cent of hospitals in the country, the potential reduction
in unit costs would be about 12 per cent. We believe there is
a lot of potential for increasing efficiency in the future. It
is not just related to acute services, but also related to FHS
services which form part of this increase of 2.3 per cent. We
are doing a lot to try to increase efficiency in primary care.
I can give numerous examples of the way we are doing that. Yes,
the 2.3 per cent seems in relative terms compared with previous
years to be a good figure. There are several caveats obviously
in terms of demography and technology. On the other side there
is increasing efficiency which needs to be taken into account
as well. The very final point I should like to make relates to
the purchasing power of the cash which is very much affected by
the income expenditure position of health bodies around the country.
I can certainly state categorically that there is no intention
to improve the I&E deficit by not paying creditors which has
an implication in terms of cash. People possibly do not know,
but over the last 12 months there has been quite a focus on reducing
the financial deficit in the NHS and focus on financial stability.
We have actually reduced the I&E deficits of health bodies
from a figure of £460 million to around about £100 million
and we are also improving our payment of bills, public sector
payment policy, which is an indicator of the timescale for paying
bills. It now stands that we pay 83 per cent of bills on time
which is very much better than in the previous two or three years.
It would be easy to say we can actually increase the 2.3 per cent
by increasing our I&E deficit and playing around with creative
accountancy in terms of paying bills, but Ministers and certainly
myself, are very, very keen to ensure that is not the position.
It was not in 1997-98 and it is not the intention in 1998-99 either.
38. May I go to the efficiency point you raised
and you said that if all health authorities operated like the
upper quartile then you could get all these efficiency savings?
Are you demanding of the health authorities in the upper quartile
that they improve in efficiency by 2.3 per cent, because they
might say they have done everything, it is the guys lower down
you should be squeezing?
(Mr Reeves) Yes. In terms of reducing unit costs in
previous years I remember I have come to this Committee and talked
about what we have done to reduce lengths of stay, what we have
done to increase day cases. Although a lot has been done, and
that is shown very clearly in our reduction in unit costs, we
believe if you compare the average hospital with the most efficient
hospital that there is still an awful lot of scope for the poorer
hospitals to improve their efficiency. That is our intention and
that indeed is reflected in the requirements of health authorities
in their purchaser efficiency index. In other words, the requirement
set for health authorities by their regional offices will very
much take into account the unit costs of the hospitals which serve
those health authorities.
39. I am sorry, you may have missed my point.
What I am trying to ask you is whether you are asking those health
authorities and those trusts which are already efficient to be
equally as more efficient as the ones who are inefficient?
(Mr Reeves) The balance is that the health authorities
purchaser efficiency index is a reflection of the relative efficiencies
of both the health authorities and the trusts within their area.
The corollary of that would be that a very efficient health authority,
a very efficient set of hospitals within the area of that health
authority, is likely to be asked to increase its purchaser efficiency
index by a lesser extent than a more inefficient health authority
and separate trust.
|
