CHAPTER 5: IS REFORM NECESSARY?
49. There was general agreement among most of
our witnesses that the Barnett Formula is no longer an appropriate
mechanism to allocate funds to the devolved administrations. Only
a minority of witnesses argued for the Formula to be retained.
The Treasury told us that they considered the Barnett Formula
to be effective and appropriate and they had no plans to consider
any alternatives (QQ 413-423). The Chief Secretary to the
Treasury described the Formula as "fair enough" (QQ 979,
982, 984, 988, 992-4). John Swinney MSP, Cabinet Secretary for
Finance and Sustainable Growth in the Scottish Executive, argued
for retaining the current Formula but only unless or until there
was further reform of the devolution settlement to include fiscal
autonomy (Q 226).[41]
Professor Arthur Midwinter, in a submission to the Calman
Commission, highlighted the importance of stability, predictability,
political acceptability and simplicity.[42]
50. The Barnett Formula has several important
advantages: it is simple, stable and not pre-committed; allowing
freedom of choice to the devolved authorities as to how the grant
is spent. There have been improvements in the transparency of
the Formula and its application, prompted in part by devolution
in the late 1990s,[43]
but the price of that has been reduced flexibility in funding
the devolved administrations. We have found that the way in which
the Formula has been administered in the past is opaque and subjective.[44]
51. If the existing Formula is to be replaced
or reformed, any alternative must represent an improvement on
the current system. The advantages of the Barnett Formula—simplicity,
stability and the absence of ring-fencing—are important
and should be maintained whatever the future method of allocating
funds to the devolved administrations.
52. Although the advantages of the Formula are
clear, the disadvantages call into question whether the Formula
is an adequate mechanism to allocate funding to the devolved administrations.
Whilst most witnesses conceded that the present system was relatively
quick and easy for the Treasury to administer, only a minority
thought that it could be justified in terms of fairness (Q 979).
THE BASELINE
53. The most serious criticism of the current
basis of funding is that, whilst the core allocation (the baseline)
has been built upon since the Formula was first applied, it has
never been reviewed. "The Formula does not determine the
overall sizes of budgets (these are based on past allocations
and decisions) … any perceived inequity is due to the historical
levels of funding since these are by their very nature 'locked-in'
the system" (p 351). "Baseline expenditures to
which 'consequential' increases were cumulatively added were based
on expenditure patterns prior to 1978: indeed in the case of Scotland
allocations were partly related to the Goschen formula established
in the late nineteenth century" (Q 568, pp 134, 353).
According to The Economic Research Institute of Northern Ireland,
the core funding allocated through the Barnett Formula reflects
and preserves "a myriad of past changes" (p 227).
Therefore it cannot be said that the baseline reflects the needs
of each devolved administration as they are today.
POPULATION EFFECTS
54. After many years during which the population
statistics were not kept up to date, population statistics are
now updated every three years with the reissue of the Statement
of Funding Policy. But they are applied only to the annual
increment of the Formula. The underlying baseline grant (the most
significant proportion of the funds allocated through the application
of the Barnett Formula) is not re-examined to take account of
population changes. If the populations of the four countries and
regions had followed similar paths over the years, this would
not have mattered very much. But as Figure 1 shows, the populations
have diverged. Over the period since 1981, the populations in
Wales, England and Northern Ireland have grown by some six per
cent, nine per cent and 14 per cent respectively. By contrast,
the Scottish population, after a period of absolute decline, is
now only back at its 1981 level.
FIGURE 1
Actual and projected populations 1971 to 2031, as a percentage
of the 1981 level[45]
55. What matters, as far as the Formula is concerned,
is not whether populations grow or decline in absolute terms but
how they change relative to one another. As well as showing that
change of this sort has been going on for several decades, Figure
1 also shows (using projections made by the Government Actuary)
that they are expected to continue to do so at least up until
the 2030s. Whilst some changes took place to annual increments
to reflect population statistics, the failure of the Barnett Formula
to adjust the baseline to reflect this population effect is a
serious weakness. Over the short time for which it was originally
expected to operate, this would not have mattered. But left to
run unchecked for decades, the absence of any correction mechanism
has led a pattern of grant per head which is now arbitrary.
56. The changing populations of the devolved
administrations and the failure of the Formula to take account
of population changes over time within the baseline create a significant
problem for the Barnett Formula today. In our view, the resulting
per capita allocations are arbitrary and unfair. In essence the
baseline of the grant provides funds for a level of population
that has changed.
The administration of the present
system
OPAQUE DECISION MAKING
57. Treasury officials argued that their decisions
on the categorisation of expenditure as "England" only
or "UK-wide" spend were relatively clear because they
simply flowed from legislation that specified which areas were
devolved and which were reserved (Q 471). However the issue
is not as simple as they suggest. Under the Scotland Act 1998,
all matters are devolved save those expressly excepted or reserved
to Westminster (set out in Schedules 4 and 5 to the Act). For
Northern Ireland, all matters are similarly devolved save those
excepted or reserved (though the lists of such matters are different
there, and the meaning of 'excepted' and 'reserved' is also different
under the Northern Ireland Act 1998). For Wales, under the Government
of Wales Act 2006, only matters expressly devolved are within
the competence of the Welsh Assembly Government and National Assembly
for Wales. While, in substance, many matters are devolved in each
case, there are important variations in both the topic devolved
and the form in which they are devolved. While some functions—such
as education and health care—are devolved in each case,
policing and criminal justice are devolved in Scotland but not
Wales (and have yet to be devolved to Northern Ireland). These
variations mean that the system is inherently complex and asymmetric.
58. There was general agreement from all our
witnesses, except the United Kingdom Government, that the application
of the Formula is unclear. Some decisions, especially those relating
to Formula by-pass, are taken within the Treasury with little
or no consultation. Phil Jarrold of the Wales Council for Voluntary
Action summed this up for us: "Our members are regularly
confused, I think, by the way the Formula applies, or maybe does
not apply, to individual announcements. Spending announcements
are frequently silent on whether they apply to the devolved nations
and whether or not there is any consequential. I do think that
that lack of transparency creates problems for organisations that
are trying to track these issues" (Q 624).[46]
59. Treasury Officials argued that the dispute
resolution process set out in the Statement of Funding Policy
was adequate to resolve any issues arising from Treasury decisions
(QQ 476-78). However, they admitted that the dispute resolution
process has never been used (Q 478). It remains untested.
Moreover, the fact that the dispute resolution process has never
been used is not to be taken as indicative of satisfaction across
the United Kingdom with the decisions made by the Treasury on
grant allocations, bypass and 'consequentials'. The Committee
has heard considerable disquiet in the nations and regions over
the decisions taken by the Treasury (paragraphs 42-48 above).
Whilst devolution legislation provides a broad framework within
which administrative decisions are taken, a published set of detailed
criteria by which the Treasury make decisions on how to categorise
expenditure would improve transparency.
60. On every funding decision the Treasury
is judge in its own cause, including whether to bypass or include
any expenditure within the application of the Barnett Formula.
We recommend that before decisions are made on whether the system
is bypassed or create a 'consequential' there is a clear process
and open consultation with the devolved administrations.
INADEQUATE AND INACCESSIBLE DATA
61. A further concern relates to the data available
on the operation of the Formula. Despite its importance, the Treasury
only publish limited data about devolved public spending, and
the published official data appear in a number of places—in
the Statement of Funding Policy, the Public Expenditure
Statistical Estimates, and the annual reports of the Scotland
and Wales Offices. Older published data do not distinguish clearly
which level of government is responsible—United Kingdom
or devolved—for particular spending in the breakdowns published
in the Public Expenditure Statistical Estimates. There
is no time series showing how expenditure has changed as a result
of spending decisions made in previous years or spending reviews.
It is difficult to establish comparable levels of spending in
England for devolved functions as they are different in each part
of the United Kingdom. At the start of our inquiry we asked the
Treasury to provide data to show how the Formula has been applied
to shape the grant allocations to the devolved administrations
since 2003-04. We now publish this information in order to make
sure that it is placed in the public domain (see pp 103-120).
The Chief Secretary to the Treasury acknowledged that Treasury
data could be improved and he suggested that the Committee had
a role in making recommendations to that end (Q 1030).
62. Although we acknowledge that the data
on public spending have improved since 1999, we continue to be
concerned that clear, thorough and readily accessible data on
public spending across the United Kingdom are not yet being provided.
63. Rather than having a multiplicity of official
documents providing some data on devolved finance, we recommend
that the Treasury publish their statistics of the workings of
the Barnett Formula, or its successor, in a single, coherent and
consistent publication. This annual publication should contain
all material data on devolved finance, showing the allocations
of grant to the devolved administrations, changes from previous
years and explanations for any changes made. We recommend that
the statistics be monitored by the UK Statistics Authority.
41 We are prevented from considering fiscal autonomy
by our orders of reference. Back
42
A Midwinter written submission to the Commission on Scottish Devolution
18 March 2009 pp 30-34 available at: http://www.commissiononscottishdevolution.org.uk/engage/submissions-received.php. Back
43
See Chapter 4. Back
44
See paras 57-63. Back
45
Source: ONS population trends dataset, table 1.2. In the interests
of clarity, values for the years when official estimates were
not produced (those years not ending in either a 1 or a 6) have
been interpolated. Back
46
See para 43-48. Back
| 
