Examination of Witnesses (Questions 801
- 819)
FRIDAY 27 MARCH 2009
Mr Patrick McCartan and Mr Victor Hewitt
Q801 Chairman:
Gentlemen, thank you very much indeed for coming this afternoon.
As I am sure you know, we have been asked by the House of Lords
as a Select Committee of the House to look into the operation
of the Barnett Formula. I hope you have seen our terms of reference
because you will see that they fairly focused and limited. We
are not entitled to look at tax raising powers for any of the
devolved administrations. We are not entitled to look at the division
of monies among the English regions. Our function is to look at
how Barnett operates and whether it should go on operating as
it does. It is really quite an interesting conundrum as to what
is the best way of getting money from the centre out to the three
devolved administrations, particularly in a situation in which
the three devolved administrations do not have symmetrical devolution,
it is different in the three areas. What one wants to do is devise
a system if one moves away from Barnett which produces sensible
and effective allocation of resources and resources which, by
and large, are fair. I wonder if I could ask you a general question
to start off with, which is a very simple question to ask. Do
you think the Barnett Formula treats Northern Ireland fairly and,
if so, why, and, if not, why not?
Mr Hewitt: We must remember that the Formula
is an adjustment mechanism, it is not a mechanism for setting
a baseline, that has to be done by other means. Historically it
takes the baseline and adds to it or potentially could subtract
from it depending on what is happening across the world. It is
a marginal adjustment mechanism simply based upon your relative
population proportions and the comparability of your expenditure
with expenditure essentially in England at this moment in time.
It is a relatively simple conceptual formula. What it effectively
does is to give each of the countries of the UK the same per capita
increase in expenditure as is happening in England. That is effectively
what it does each time it is applied. One of the consequences
of that is if your baseline is proportionately bigger than your
population proportion then there will be an effective squeeze
upon the amount of money coming across because you are adding
at the margins less than the proportion of the baseline. Is it
fair? Well, we would need to look at not only the Formula itself
because it only really applies to a proportion of the expenditure
which goes on in the region and part of the expenditure in the
region is outside the Formula altogether. Such things as social
security and the annually managed expenditure money is done under
an entirely separate mechanism, a sort of sale or return type
mechanism. The Barnett Formula only tackles that bit which we
call the assigned budget in the block, which is round about half
of our total budget. Has it enormously disadvantaged or advantaged
us? There has not been a sign that we have converged dramatically
over time with the situation in England. There has been some movement,
but quite often the relative figures are distorted because additional
monies have come through other mechanisms, what Professor David
Heald referred to as "bypass mechanisms". Money such
as the funding of the EU Special Programme Bodies, the Peace and
Reconciliation Funds, and in earlier times monies which were associated
with the privatisation of Harland & Wolff and Shorts, all
of these flowed into our block outside the Barnett mechanism and
tended to push up the per capita lead. It is not just a matter
of being fair but is it a mechanism which can be easily replaced
by something which is as workable, that is the question which
really needs to be addressed. It is okay in theory to devise all
sorts of sophisticated mechanisms for allocating money out, but
the practicalities are that on the day of a Spending Review the
devolved administration would want to know how much they are going
to get, and since Spending Reviews are not finished until the
last minute there has to be a relatively straightforward mechanism
for allowing them to calculate how much money is coming to them.
That is a roundabout answer to your question. It is a complicated
formula when other factors are taken into consideration. It is
very simple in itself. By and large it has served us reasonably
well. We cannot answer the issue of needs through the Formula
and whether the amount of money we are getting is proportionate
to our needs, that is a matter for needs assessment and perhaps
we might talk about that later.
Q802 Chairman:
Certainly it is true that the Formula narrowly applied only applies
to the changes up and down, but it seems to be being used and
understood and, indeed, applied now as to the block as well as
the variations. Do you think the amount Northern Ireland gets
for its baseline is a fair allocation?
Mr Hewitt: That is very difficult to say unless
one were to carry out what is known as a needs assessment but,
as in life, these things are never simple because there is not
an absolutely agreed methodology about how one should approach
a needs assessment. As you are probably aware, the only published
version of a needs assessment relates back to 1977 when devolution
was first conceived for Scotland and not carried out at that time.
A system was devised to estimate our relative need and that of
Scotland and Wales relative to England, and I can go into the
detail of that if you like. That has actually been updated annually
but never publicly. The Treasury updates the needs assessment
formula annually, or certainly it did well into the 1990s. They
are less in favour of that methodology these days because they
think the underlying structures for the UK have changed so much
and different parts of the UK tend to have experiences different
from London. For example, they will say that parts of England
have a high influx of immigrants, and that is not quite the case
for, say, Scotland or Wales, so that creates a particular need
in England which is difficult to roll out to the others. The things
which go into a needs assessment formula are supposed to be objective
factors—population, the structure of your population, how
many young people, how many old people and so forth—but
inevitably there will be subjective things put into the mix as
well. There are always arguments around the subjective things.
What is the best indicator of ill-health, for example, is a classic
one. Is it the standardised mortality ratio by the number of people
who die, or is it some other measure of ill-health? That has always
been a bone of contention when these things are looked at. Needs
assessment is something which a lot of people talk about but very
few people have ever experienced what a needs assessment is really
like.
Q803 Chairman:
But you say it cannot be done.
Mr Hewitt: It can be done.
Q804 Chairman:
It does not have to be as complicated as it was in 1979.
Mr Hewitt: Once you start on these things it
tends to become complicated.
Q805 Chairman:
Only if you make it so surely. If you were to cut down the number
of variables that you were going to take account of to four, five
or six, I do not know, you would probably get 95 per cent fairness
although you would not get 100 per cent.
Mr Hewitt: Possibly so. You will always have
arguments from the other side that you have left out a very important
dimension to the problem and there is the problem of what weight
do you attach to these various things and there will be different
views about the weight that should be attached to one factor as
opposed to another. Once you get into those sorts of arguments
you rapidly begin to think whether it was worth going down this
route in the first place because you are not going to get agreement
at the end of it.
Q806 Chairman:
Have you looked at the Australian system?
Mr Hewitt: Yes, the Commonwealth Grants Commission.
It is a pretty good example of the complications which arise in
these circumstances but it is there for a different purpose than
the allocation of public expenditure in the UK. It is there to
essentially allocate out the proceeds of certain taxation among
the states. It operates on a federal system. It is about horizontal
equalisation, ie putting the various states on the same basis
so that they can deliver services, but taking account of the amount
of taxes which they actually raise. It does focus on a slightly
different issue. We are a purely expenditure based system, taxation
has no role other than through local government.
Q807 Chairman:
But the needs assessment bit would be the same, would it not?
Mr Hewitt: Yes.
Q808 Chairman:
In other words, if they assess need in Australia we can assess
need here.
Mr Hewitt: Yes, indeed. Needs assessment is
very simple in concept. The policies are broadly similar throughout
the United Kingdom, so if it costs £100 to deliver some sort
of service in England how much does it cost to deliver the same
service in Scotland, Wales and Northern Ireland. That is the starting
point for the whole thing. It then generates an index and typically
Northern Ireland would come out at something like £125 through
to £130 as opposed to an English spend of £100 in that
context. The full-blown thing is an extremely data heavy exercise.
The original one took three years essentially.
Q809 Chairman:
Yes.
Mr Hewitt: Updates on these usually take at
least 18 months.
Q810 Chairman:
Do you know how long the Northern Ireland Government took to do
its one in 2001?
Mr Hewitt: Since I was doing it, yes, 18 months.
Of course, that was a unilateral thing, it was not done with the
co-operation of the Treasury.
Q811 Chairman:
But it was 18 months?
Mr Hewitt: Yes, about 18 months.
Q812 Chairman:
If needs assessment could be done by an objective commission like
the Australian Grants Commission, why should it not be applied
to the UK?
Mr Hewitt: There is no logical reason why it
should not. It would have to be done by something outside the
Treasury.
Q813 Chairman:
I think we would all agree with that.
Mr Hewitt: Because the devolved administrations
will not accept that the Treasury is an honest broker in these
matters, and rightly so. It cannot be done that often because
we cannot really rely upon this to be an annual mechanism.
Q814 Chairman:
They do it annually in Australia.
Mr Hewitt: Yes, they do, and it is a very large
effort indeed. I was having a look at their website the other
day and a very large effort goes into that. It is full of methodological
pitfalls as you go along to calculate things on net or gross terms,
you get different results on those sorts of things. How much of
the block do you actually take into account? When it was done
in the 1970s the law and order and protective services were effectively
outside of it because of the extraordinary circumstances in Northern
Ireland, so there will be judgments about things like that. It
is certainly an interesting exercise but it is not a magic bullet
in terms of guaranteeing that you are going to get more money
from the Treasury at the end of the day.
Q815 Chairman:
If there is a finite amount some will get more and some will get
less. What we have really been concerned about is whether or not
it could be feasible to have a needs assessment procedure along
the lines of the Australian system which we could apply to the
allocation of resources here in the UK. I think with all the qualifications
you made about it you do think that is possible.
Mr Hewitt: Yes. The methodologies are relatively
straight forward. The needs assessment system which we operated
did not involve a lot of statistical work in the sense of doing
estimations and stuff like that. It was certainly less complicated
than the old standard spending assessments which operated in local
government, which you might recall. It was never designed to be
that, it was to take into account people's experience of how programmes
worked, what was important in driving expenditure in a programme
and so on. It is not quite as scientific as some of the work which
is done in Australia. I think you will find the Australian system
still produces results which are not universally welcomed by the
participants.
Q816 Chairman:
What we heard about it was it produces its report and then for
about two or three days everybody says how unfair it is, but it
gradually goes to another sleep for another 360 days until it
produces another report and there is the same eruption. It is
not a serious business but it is loud while it is going on.
Mr Hewitt: Even if you did this, what would
you do in the intervening years? You would have adjusted the baseline
and how are you going to continue to adjust the baseline outside
that period. That is what the Formula is there to do, it is not
there to set it but to adjust it.
Q817 Chairman:
The Formula is not there to adjust the baseline.
Mr Hewitt: No.
Q818 Chairman:
You would have to adjust the baseline
Mr Hewitt: Are you proposing to run the exercise
every year?
Chairman: I should think so, yes. The
Australians seem to do it and I do not see why we should not.
I do not have violently strong views as to whether you do it every
year, 18 months or two years but it has got to be done within
a fairly narrow band of time. It has got to be able to stick to
the length of time it is meant to be done for.
Q819 Lord Sewel:
Can we look at convergence. I have seen some figures that indicate
over recent years there has been some convergence in Northern
Ireland. Obviously the weight of the past to prevent convergence
was bypass and really bypass was not easy but at least it was
a route that was accessible prior to devolution when we basically
had the territorial ministers going to the Chief Secretary, and
if that did not work going to the Chancellor and if that did not
work trying the Prime Minister, and they were all ministers of
the one government. Some were more successful than others, it
has to be said. Then with devolution that system breaks down so
the opportunity to bypass reduces. If you then have, if you like,
more pure Barnett convergence is likely to kick in much more heavily.
If your population is going up, which I understand the population
in Northern Ireland is, then because your base is a product of
a lower population you are also going to get a squeeze, yet people
seem to be relatively content with Barnett whereas because of
those factors if I was in Northern Ireland I would be pretty worried
about that.
Mr Hewitt: The "Barnett squeeze" is
primarily a mathematical phenomenon. The faster expenditure goes
up in England, the faster the squeeze will be applied. In recent
years we have been living through very rapid increases in public
expenditure in the UK, certainly since 2000, therefore it is not
surprising to see there has been some convergence. Of course,
the absolute amounts of money which have been transferred through
Barnett are very, very substantial. The second thing, as I emphasised
before, is it only applies to part of the totality of public expenditure
in the Province, the other half at least is coming through annually
managed expenditure and the benefits system. There is also expenditure
by UK departments actually in the Province, so the Ministry of
Defence will be spending money in Northern Ireland which is not
part of the block. Should we be worried about it? It has not really
been a problem. I think David Heald put it quite well, that one
of the great fears at the beginning of devolution was there was
going to be too little money whereas in reality it appears that
there has been rather too much money available to the devolved
administrations. What is the evidence for that because that is
a fairly harsh thing to say? If you actually look at the under-spends
from the devolved administrations over the years, these have been
very substantial and have not been diminishing very rapidly, so
we are carrying forward from year-to-year substantial sums of
money under-spent from the previous year. That does not suggest
there is a vast shortage of money, but it may suggest there has
been not very good estimating by departments.
| 
