Examination of Witnesses (Questions 80
- 99)
WEDNESDAY 4 FEBRUARY 2009
Professor Iain McLean
Q80 Chairman:
Perhaps I can sum up by looking at the development of the formula
as it seems to have developed in the 1970s. What do you see as
the purpose of the Barnett Formula? What do you think the purpose
of the Barnett Formula was when it was first introduced and has
the purpose changed over time?
Professor McLean: My understanding from talking
to civil servants involved in the early years and from listening
to and reading the evidence of Lord Barnett himself, notably to
the Treasury Committee some years ago, is that there were two
purposes at the beginning: the first was to get precisely a single
block of expenditure so that the Chief Secretary would not be
arguing with the three territorial departments programme by programme;
and the second was the convergence purpose of Barnett. As members
likely know, at the same time as developing what we now know as
the Barnett Formula the Treasury was working with the territorial
departments on a needs assessment. I say "with the
territorial departments" rather tentatively because it is
now well established that some of the territorial departments
were more resistant to this than others. That was intended to
assess the relative need for expenditure on the services that
would have been devolved under the Scotland and Wales Acts 1978
which, as members know, were never brought into operation. The
Treasury calculations from this needs assessment were that Scotland
and Northern Ireland were receiving relatively more expenditure
than their relative need seemed to indicate and Wales was not;
Wales was getting less. However, the second purpose of Barnett,
as I understand it, was to bring about convergence until such
time as the expenditure in each of the three territories was brought
down to the level of their relative needs, whereupon, as Lord
Barnett told the Commons Treasury Committee some years ago, he
envisaged that a needs formula would replace it; instead, as members
know, that has never happened and Barnett is still in place.
Q81 Lord Lawson of Blaby:
What evidence do you have that convergence was the purpose? So
far as I am aware, and I was only responsible for the Treasury
for six and a half years, convergence was never a purpose and
when Lord Barnett gave evidence to us he said that in his opinion
that was not the purpose. What evidence do you have to contradict
all that?
Professor McLean: I have two sources of evidence:
one is the mathematics of the formula which bring about convergence
and as the formula was devised in the Treasury, and the Treasury
is populated by very clever people, I am sure that it was not
an accident that its mathematical effect is to bring about convergence.
Q82 Chairman:
It was to Lord Barnett. That is what he told us anyway. Clever
people in the Treasury may have thought it was going to happen
but I do not think he did.
Professor McLean: I can only repeat that it
is the mathematical effect. The second source for my evidence
is talking to successive officials in the devolved countries and
regions' team and other public spending teams at the Treasury
who have put on record that this was the purpose. I say two pieces
of evidence but I should say a third, which is that the records
of the needs assessment 1979 were released to me under a Freedom
of Information request in 2005 and further records have been since
put into the public domain of discussions between the Treasury
and the territorial departments in 1984. These are my three pieces
of evidence.
Q83 Baroness Hollis of Heigham:
Are you confident that you are not aligning or running together
what appears from Lord Barnett two completely parallel but unconnected
pieces work, which is the work going into the needs assessment
which you referred to which you had released and was available,
in which there clearly was some consideration of needs resources
and so on, and what went into Lord Barnett's Formula which was
entirely temporary? As far as he was concerned he told us it was
entirely and simply based on population without even any obvious
way of adjusting for population pluses and minuses because he
did not expect it to last for 12 months. As far as he was concerned
he seemed to suggest to the Committee that convergence, if it
happened, was an abstract appliquéd onto it by academics
looking for elegance which was not there or intended. I wonder
whether the evidence you are getting from your civil servants,
from the devolved territories and so on, are related to the other
piece of work which was the needs assessment which came out in
December 1979 as opposed to the Barnett Formula itself. Again,
Lord Barnett seemed to indicate to us very clearly that he knew
virtually nothing about that piece of work which was going on
and that the one, as far as he was concerned, did not inform the
other.
Professor McLean: You are right that the needs
assessment was an entirely separate exercise although done at
the Treasury end by, as I understand, the same people in the Treasury,
but the needs assessment gave a static picture of the relative
needs as they were judged to stand at the time of the 1978 Acts.
The convergence properties of the formula are dynamic. If you
start from a constant base line and you add a population proportion
to that base line each year then its effects are convergent. This
is not the first time I have been put on the spot of trying to
give a full explanation which would involve a white board and
showing you the differential equation system which lies at the
root of it. I would imagine that is not what you want but I could
submit it in written evidence subsequently.
Q84 Lord Lang of Monkton:
I am surprised that Professor McLean has not been put on the spot
before and memorised a simple answer. I do not profess to know
precisely how convergence works but when I was at the Scottish
Office the CFO, perceiving my mathematical shortcomings, made
it easy for me. What she wanted pointed out was that Goschen had
applied a per capita percentage increase to the block each year
at the same percentage level as the increase in England but as
the Scottish block was larger than the English expenditure base
line divergence took place. The Barnett Formula allegedly, although
I have had no confirmation of this from any other source other
than the CFO of the Scottish Office at the time, changed from
a percentage increase to a pounds and pence increase per head.
In other words, if England got a 4 per cent increase, Scotland
would get the cash product of that 4 per cent increase which,
because of the higher Scottish base line, would produce a lower
percentage increase figure thus creating vary gradual convergence.
Professor McLean: That is exactly correct and
is clearer than what I have just said so thank you very much.
Q85 Chairman:
There is one problem about your theory, if I may call it that,
that convergence was at the heart of the Barnett Formula when
people brought it up and that is by the time the report on needs
was finished he had ceased to be in the government because the
government had changed. We find no evidence at all to show that
convergence was in Lord Barnett's mind nor anybody else's mind
at the time it was done.
Professor McLean: Firstly, as to the name members
are probably aware it was an entirely unofficial nickname conferred
by David Heald, the public finance economist, after a Commons
hearing in 1980. Secondly, you are quite right that Lord Barnett's
accounts in the early days and going up to when he gave evidence
to the Commons Committee some years ago concentrated on the first
of the two rationales for Barnett. As he has frequently said,
it made his life as Chief Secretary easier to be dealing with
a single block rather than programme by programme amounts, but
the convergence property is exactly the one which Lord Lang has
just expressed more clearly than I have and that was inbuilt from
the start.
Q86 Lord Lawson of Blaby:
May I put it to you that the convergence property which you think
the Treasury officials put in from the start is only of significance
if the formula is going to be adhered to over a considerable number
of years and that was not the intention at the time. Therefore,
since it was not intended to last for a long period of time convergence
was not a purpose of the formula.
Professor McLean: My understanding from Treasury
officials and from the two lots of FOI releases of 1979 and 1984
was that convergence was indeed in the minds of Treasury officials
from the start. I think that was also perceived by the officials
of the devolved administration, certainly by those in Scotland.
I have talked to civil servants in Northern Ireland also who have
given me a similar story. They understood it as being intended
from the outset as a convergent formula. I do not know what Lord
Barnett said to you at the hearing last week but I have seen him
say in other places that he understood that in so far as it was
a convergent formula it would run until such time as it would
be replaced by a needs assessment but of course he was out of
government by the time that was said.
Q87 Chairman:
Can I read to you, if I may, the last question that he answered
when he came to us last week. I put to him "You devised a
mechanism which you hoped would last for a few years. You did
not expect it to last for as long as it has lasted. You are not
sure now whether it is based on the right criteria and you lean
towards having, among other things, a needs-based assessment.
Is that fair?" and he said "That is fair." It seems
to me fairly clear on the evidence we have seen that convergence
was not part of the exercise.
Professor McLean: I do not follow the last bit
because it seems to me that what he has said to you implies that
he understands that it was intended to converge until such time
as a needs formula replaced it.
Q88 Baroness Hollis of Heigham:
It is exactly the opposite. There may have been an accidental
and, in the Treasury's eyes, fortuitously happy consequence if
this formula were to stay in place over time but it was never
intended. It seems to me we are running together consequences
and intentions here.
Professor McLean: On that matter I am with Lord
Lang's CFO plus my interviews with Treasury officials. I believe
it was entirely intended by Treasury officials to have a convergent
effect.
Lord Lang of Monkton: Even as a short-term
deal it had the merit of having a convergent component because
it enabled the Chief Secretary, under considerable pressure to
cut expenditure, to claim to Cabinet colleagues that he had a
formula in place that would gradually create convergence and it
gave the Secretary of State for Scotland, Bruce Millan, the opportunity
to say that he had protected for the long term expenditure levels
in Scotland.
Chairman: I do not think we will wring
any more out of this particular dishcloth.
Q89 Lord Sewel:
Can we have a look at formula by-pass and its effect on the period
up to devolution. How regular was formula by-pass? What was the
cumulative effect of formula by-pass on the distribution of expenditure?
Has the whole concept and idea of formula by-pass changed or indeed
disappeared since the devolved settlements came into place?
Professor McLean: The easiest part of that question
to answer is the last because the answer is one word: yes, as
I understand it. The details of formula by-pass are very difficult
and, if I may say so, somewhat embarrassing for me to attempt
to assess from outside with at least three or four former players
in the system on your side of the table. I understand that it
went in several phases. The first phase ran until, I believe,
financial year 1982-83. In that first period the public expenditure
was planned on volume terms year to year and, therefore, there
was in effect no convergence and the issue of formula by-pass
did not arise. From then until 1992 it ran on a cash basis, as
it has done ever since, and therefore the potential for convergence
was there. It is said in the press, and I am not in a position
to confirm or deny it, that there was in the later Conservative
years, or the years up to 1992 at any rate, formula by-pass sometimes
for pay settlements but I am repeating hearsay when I say that
because I do not know. The next phase began in 1992 when Chief
Secretary Portillo re-based the population bases which had drifted
away in a way which was relatively favourable especially to Scotland,
and it may have been relatively unfavourable to Northern Ireland
whose relative population was increasing. My impression is that
there has been considerably less formula by-pass since 1992 than
there may have been before then. I can only say that to the academic
community this is hearsay as we have no inside information on
this.
Q90 Lord Trimble:
When you say that there was no formula by-pass since devolution
I defer to your knowledge of the position in Scotland and Wales,
but I had the very distinct impression in my time there was formula
by-pass.
Professor McLean: You are right and I cut my
story off a little bit too early. Formula by-pass in more recent
years has been in relation to EU programmes including the peace
programme in Northern Ireland. My impression is since 1997 there
have been relatively modest amounts of formula by-pass in Northern
Ireland, modest in relation to public expenditure. There has been
modest formula by-pass in Wales because the politics of that were
the attempt by First Minister Alun Michael not to concede that
which led to his deposition by the National Assembly and therefore
it was necessary to concede, from the Treasury's point of view,
some formula by-pass in Wales. I believe there has been no formula
by-pass in Scotland since 1997 because the Treasury, I believe,
has held the line that any EU public expenditure which comes under
Objective 1 is to be treated in Scotland as being within the block.
That is my understanding although I would hope that you will take
evidence from the Treasury on this point.
Q91 Lord Trimble:
My impression on the matter is that with regard to the European
programmes only the peace programme was genuinely additional in
Northern Ireland and that was at the insistence of the European
Union itself. Quite apart from that I think there was formula
by-pass in some of the Comprehensive Spending Reviews although
I cannot give you hard evidence on that.
Professor McLean: You, Lord Trimble, are better
placed than I am to say. I am not aware of any formula by-pass
since the present CSR regime began.
Q92 Lord Sewel:
Let us assume that formula by-pass did take place in the period
up to devolution and when it was used it was used predominantly
to fund public sector wage settlements, nurses and schoolteachers,
because if you did not then it had a disproportionate impact on
the rest of Scottish expenditure. Post-devolution, as you say,
it has not occurred. How do you think the devolved governments
and the UK government would respond to the sorts of financial
and political situations that caused formula by-pass in the first
place?
Professor McLean: I think the response of the
UK government, or at any rate of UK Treasury civil servants, is
easy to say. They would say, "You have a block, you get on
with it and it is your problem". The response of devolved
administration ministers now I suppose would be if they had a
complaint they would be invited to take it to the Joint Ministerial
Committee. This mechanism which was set up at devolution and has
been barely been used. It was, I think, never used during the
period in which Wales and Scotland had labour-led governments
and for a lot of the same period Northern Ireland was under direct
rule. We can expect that we will see more of it in the next few
years.
Q93 Baroness Hollis of Heigham:
What I want to follow up is in your very interesting paper when
you explain despite this why convergence did not occur. You gave
a couple of reasons why you felt that convergence despite what
you regard as the intent behind the original Barnett Formula nonetheless
did not occur in Scotland. I was wondering whether you could enlarge
on that.
Professor McLean: The first reason is the one
that you have just mentioned in passing, which is that social
security is outside the Barnett block entirely and until recently
agriculture was. Therefore, as these are entitlement programmes,
the spending on them is a straightforward function of the relative
proportions of claimants in the four territories. The second reason
I think is that the population re-basing introduced by Chief Secretary
Portillo operates in arrears. If the relative population of Scotland
is continuously declining then Scotland does well until the next
re-basing at which point there is a step downward change which
is painfully noticed in Scotland. If the population proportions
continue to diverge, so that Scotland's relative and sometimes
absolute population is declining, the effect is that expenditure
per head does not converge as fast as the formula would have it
do.
Q94 Lord Lawson of Blaby:
May I ask a supplementary to that? What you have said about the
failure to have a population adjustment when the trends of population
in England and Scotland were in different directions is obviously
of central importance. Was there not a greater mischief or a greater
inequity than you have implied in the sense that although the
population was adjusted in 1992 after more than ten years of the
Barnett Formula there was no retrospective adjustment? In other
words, if the expenditure per head had been increasing relatively
in Scotland during that period because of the failure of the population
adjustment and all that happened with the population adjustment,
leaving aside the point you make correctly that adjustments are
belated and there is always this catch up, the new base line in
1992 was inevitably a better base line from Scotland's point of
view than the one in 1979?
Professor McLean: I am almost sure that is correct.
People outside government are not in a position to confirm that
because the relevant papers are not yet publicly available but
I have a strong hunch that that is correct.
Q95 Baroness Hollis of Heigham:
Could you help me on your first point, the growth in social security
expenditure, of which you say the expenditure in Scotland is higher
has served to protect Scotland from any effect of convergence
by going from percentage to cash? Actually I do not think I really
understand that because social security is cash but, secondly,
all of the stats I have seen on what is happening to the pattern
of social security expenditure, given that pension expenditure
accounts for 40 per cent if not 50 per cent of social security
expenditure and given the difference in mortality rates in Scotland,
this seems to suggest to me that actually Scottish expenditure
on social security while possibly increasing should not have been
increasing at the same rate certainly as in England. I do not
actually understand how this works. It may be that if I saw your
stats I could have a go at them but from reading it again it actually
puzzles me because it should have, at best, a neutral effect and
possibly even, given what we know has been happening to social
security expenditure in Scotland, added to the convergence effect
not the other way around.
Professor McLean: I am afraid our stats do not
help on this matter because throughout the IPPR report, which
I think was the one you were looking at, we have excluded social
security and agriculture for the reason I gave in an earlier answer.
We do not have a table of social protection expenditure per head
in the four countries or the 12 territories of the United Kingdom,
however the Treasury does produce this every year in PESA. If
it was helpful I could supply a note to the Committee attempting
to answer the question about the per head trends in social protection
expenditure.
Q96 Baroness Hollis of Heigham:
The only reason I am pressing it is a large part of your argument
as to why convergence was built into the formula has not happened
is because of the way the social security stats have gone and
you have excluded them from your workings and I find that counterintuitive
or against what I would assume to be the case.
Professor McLean: We simply have no information
from outside government. The only information that we have is
the information which is provided in the PESA, about which I am
willing to speak if members wish me to. I am afraid I cannot improve
on my previous answer.
Q97 Chairman:
If you could let us have a note on that.
Professor McLean: I will do my best.
Q98 Lord Moser:
I have another supplementary on convergence. I have been trying
to understand not so much what the original intentions were, you
have discussed that with my colleagues, to what extent convergence
was the intention or was not the intention, what I am interested
in is are you surprised to what extent convergence has or has
not worked out? To put it statistically, could you answer an examination
question, not now, on just what was expected and what has actually
happened? I still do not understand that.
Professor McLean: I hope that I never have to
answer an examination question on that subject because the basic
information which will be required to give a statistically acceptable
answer to that question is missing, and that basic information
is planned expenditure in England on the programmes which form
the base line.
Q99 Lord Moser:
When you say it is missing, which is what all the papers say,
it is not actually missing but it is not being made available,
is that not right? Surely the data exists on what was spent by
whom for whom?
Professor McLean: The data is collected retrospectively
in the PESA exercise. The planning data is considerably harder
to come by for people outside government. The only public documents
that people like me have are the annual PESA numbers and the statement
of funding policy which comes out with each spending review. I
am sorry, that made me forget the first part of your question
if you would kindly ask it again.
|