BLAIR’S BRITAIN, 1997–2007
Page 79
usually means ‘whatever the governing party’s whips can persuade their
19 See McLean and McMillan, State of the Union, tables 8.3 and 8.11; John Curtice, ‘Restoring
Confidence and Legitimacy? Devolution and Public Opinion’, in Trench, Has Devolution
Made a Difference? , figure 9.8.
10 George Tsebelis, Veto Players: How Political Institutions Work (Princeton, NJ: Princeton
University Press, 2002), pp. 2, 8–9.
2.00
1.50
Labour
Lib Dem
1.00
SNP
Ratio
Conservative
0.50
Other
0
HC1997
SP1999
HC2001
SP2003
HC2005
SP2007
Year
Figure 22.1. Seat/vote ratios in Scotland
2.00
1.80
1.60
1.40
1.20
Labour
Ratio
Lib Dem
1.00
Plaid Cymru
0.80
Conservative
Other
0.60
0.40
0.20
0.00
HC1997
NAW1999
HC2001
NAW2003
HC2005
NAW2007
Year
Figure 22.2. Seat/vote ratios in Wales
followers to vote for’. Therefore the win set, in a Westminster system, is
large. This means that the governing party can make radical policy
changes in any direction it likes, so long as it controls its own MPs and the
House of Lords does not veto the change. Coalition government contracts the win set, because a majority requires support from more than
one party. As a consequence, policy is more stable. Stability is neither
good nor bad in itself, but merely a characteristic of the parliamentary
setting. In their cross-national survey of the impact of constitutions on
policy, Persson and Tabellini calculate that proportional regimes spend
more on welfare policy and have higher budget deficits than majoritarian
regimes such as the UK parliament. This, they argue, is because under PR
there are more veto players who could veto any reduction of welfare benefits for their client groups.11
However, the situation in Scotland and Wales is more complicated.
Both countries had new powers in 1999 to do things they could not do
before. Therefore there could have been majorities that were suppressed
before 1997, which could change policy radically even under a coalition government. In Scotland, there were. The Scottish Parliament has
power to alter Scots law on domestic matters, and in some notable
areas it has done so. It has abolished feudalism and ‘poinding and
warrant sales’ (a form of recovery of assets from debtors). And it
has established the right to roam in the countryside.12 On the other
hand, in the face of homophobic hostility, it ducked the opportunity to
legislate for civil partnerships in Scotland, passing the parcel hastily to
Westminster, which enacted a Scottish section in the Civil Partnerships
Act 2004.
Policy initiatives in Wales have been more limited – by the National
Assembly’s powers among other things. The hurried design of the new
institutions was radically faulty. The primary legislation / secondary legis-
lation distinction completely fails to map on to the reserved powers /
devolved powers distinction. Therefore, the Assembly lacks the legal
power to do a lot of the things its majority might like to. The Assembly’s
Richard Commission recommended granting more powers to the
National Assembly. The Government of Wales Act 2006 (2006 c.32),13 a
notable constitutional statute hardly noticed outside Wales, does some of
this with effect from the National Assembly taking office in 2007. It abolishes the ‘county council’ model for Welsh government in the 1998 Act,
and substitutes a ‘government and opposition’ model. It empowers the
National Assembly to make quasi-Acts (‘Assembly Measures’) in the areas
for which it is responsible. Devolution, as Ron Davies was fond of saying,
is a process not an event, and the process will continue beyond the 2007
election in unpredictable ways.
11 Torsten Persson and Guido Tabellini, The Economic Effects of Constitutions (Cambridge,
MA: MIT Press, 2005), tables 6.4 and 6.7.
12 Abolition of Feudal Tenure etc. (Scotland) Act 2000 asp 5; Abolition of Poindings and
Warrant Sales Act 2001 asp 1; Land Reform (Scotland) Act 2003 asp 2.
13 See the Assembly government’s website explaining the Act at http://new.wales.gov.uk/
gowasub/gowa/?langϭen. For the Richard Commission on the Powers and Electoral
Arrangements of the National Assembly for Wales (2004) and the former ‘county
council’ model, see http://image.guardian.co.uk/sys-files/Politics/documents/2004/03/
31/richard_commission.pdf and the Constitution Unit’s Wales monitoring reports for
2004 and 2005.
However, over most of the domestic agenda, the changes made are
more incremental than the abolition of feudalism and of poindings. And
some important developments have been non-changes, where policy has
changed in England.
The biggest incremental changes in Scotland have been two expensive
spending commitments: the non-adoption of ‘top-up’ fees for university
students domiciled in Scotland and the provision of free social care for
the elderly. Both of these followed independent reports and were supported by coalition majorities.14 They confirm Persson and Tabellini’s
claim that coalition government is associated with higher welfare spending. The Scots did both of these because they could: their formula
funding, examined below, allowed up to 20% per head more spending on
domestic services than in England.
Otherwise, policy change has arisen because the Scots and Welsh have
not made changes that have been made in England. Neither Scotland nor
Wales has league tables of school or hospital results, nor of local authority performance. Neither has introduced the quasi-market reforms of
the NHS introduced in England during the second Blair administration
(see chapter 18). On the whole, the results of policy divergence must
be highly satisfying to Tony Blair. In England there has been much pain
and many complaints from the providers of health and education. But,
if the measurements can be trusted, standards have risen. Waiting
times have dropped sharply, and school results have improved. In education, it is difficult to make comparisons, just because the Scots and
Welsh have refused to publish league tables. But in health, they have
stood still or gone backwards, even though more is spent per head on
health in Wales and Scotland than in England as a whole, and (in
Scotland) more than in any region of England except London. Victims of
these disparities include National Assembly Health Minister Jane Hutt,
although her sacking in 2005 was not directly linked to the relative failure<
br />
of the NHS in Wales,15 and three successive chief executives of the Welsh
14 The (Scottish) Cubie Report on student finance, and the (UK) report of the Sutherland
Royal Commission on social care, both published in 1999. The UK government rejected
Sutherland’s recommendation to make social care free for those whose medical condition
required it, but the Scottish Executive accepted it. See Rachel Simeon, ‘Free Personal Care:
Policy Divergence and Social Citizenship’, in Robert Hazell (ed.), The State of the Nations
2003: The Third Year of Devolution in the United Kingdom (Exeter: Imprint Academic,
2003), pp. 215–35. For full details of both policies, see the Scottish monitoring reports of
the Constitution Unit.
15 John Osmond (ed.), Labour’s Majority in Doubt: Monitoring the National Assembly
December 2004 to April 2005 (Cardiff: IWA, and London: Constitution Unit, 2005), p. 10.
Ambulance Service, who resigned or were dismissed in quick succession
in 2006.16
The two ‘wicked issues’ of devolution are finance and representation
in the House of Commons. Before devolution, Scotland, Wales and
Northern Ireland were each funded by a block grant from the UK
Treasury. Since the early 1970s, this has been by way of the now-notorious ‘Barnett formula’. It takes its name from Joel (Lord) Barnett, who was
Chief Secretary to the Treasury from 1974 to 1979, although the Treasury
was already using it under the previous Conservative government.
Barnett served two purposes. It substituted a block grant for annual bargaining over each and every service, which the Treasury suspected the
‘Celts’ of using to force public spending up to an unacceptable level. In
this they were backed by their territorial Secretaries of State as part of
killing Home Rule by kindness. And it was designed to bring very gradual
convergence towards equal public spending per head for each territory of
the UK. In 1976 the Treasury also conducted a ‘needs assessment’ with the
grudging agreement of the territorial departments. This showed that
Scotland and Northern Ireland, but not Wales, were receiving public
spending allocations for the services that would have been devolved
under the (abortive) Scotland and Wales Bill(s) ahead of their ‘needs’.
In the short run Barnett protects this relative overspending (although
if the Treasury’s numbers were reliable it should never have been applied
to Wales). Therefore it made sense for the Constitutional Convention to
say that it should continue. Its continuation was promised in the White
Papers preceding the 1998 Scotland and Wales Acts, but is not in the Acts
themselves, so it could be altered without legislation. In the long run, it
would cause spending in Scotland and Wales to crash down below whatever their relative needs now are to a level equal per head to spending in
England. This would not be fair, and would not be in the interests of
either Wales or Scotland.17
In the long run, as Keynes said, we are all dead. However, the long run
has not yet arrived. Barnett has been running for thirty years. But the
16 ‘Ambulance Reform “to Cost £140m”’, BBC Wales, 26 September 2006, at
http://news.bbc.co.uk/1/hi/wales/5378974.stm; Wales Audit Office, ‘Ambulance Services
in Wales’, December 2006, www.wao.gov.uk/assets/englishdocuments/Ambulance_
Inquiry.pdf. For a book-length discussion of relative performance to 2003, see Scott Greer,
Territorial Politics and Health Policy: UK Health Policy in Comparative Perspective
(Manchester: Manchester University Press, 2004).
17 This is a brutally concise summary. A whole book can be, and has been, written about the
Barnett formula: Iain McLean, The Fiscal Crisis of the United Kingdom (Basingstoke:
Palgrave, 2005).
Table 22.4. UK identifiable expenditure and relative GVA by country
and region, per head, 2005–6, excluding social protection
Region
£ per head
Index (UKϭ)
Index GVA
Scotland
5,093
119.9
96
Wales
4,648
109.4
78
Northern Ireland
5,457
128.4
80
England:
4,099
96.5
102
North-east
4,530
106.6
79
North-west
4,415
103.9
88
Yorks and
4,109
96.7
87
Humberside
E. Midlands
3,648
85.9
93
W. Midlands
3,947
92.9
89
E. England
3,475
81.8
107
London
5,288
124.5
136
South-east
3,578
84.2
115
South-west
3,747
88.2
94
UK
4,249
100.0
100
Source:
Cols. 1 and 2: Public Expenditure Statistical Analysis (London: HM Treasury
2007), calculated from table 9.11.
Col. 3: Office of National Statistics, Headline Gross Value Added (GVA) at
current prices by region.
Correlation between col. 2 and col. 3 Ϫ0.05.
latest public spending relativities, published by the Treasury in March
2007, show that Scotland still has public expenditure per head on
devolved services18 almost 20% ahead of the UK average (table 22.4).
Wales has spending about 9% ahead of the UK average. But Wales is a relatively poor region and Scotland has roughly average income per head. If
spending were designed to counter poverty, one would expect the correlation between public spending per head (table 22.4, col. 2) and gross value
added (GVA) per head (col. 3) to be strongly negative – approaching Ϫ1.
18 Table 22.4 excludes ‘social protection’, most of which comprises pensions and social security benefits. These are not devolved and are payable at uniform rates throughout the UK.
Some other identifiable expenditure, notably part of that on agriculture and fisheries, is
not controlled by the devolved administrations, but exclusion of that does not materially
affect table 22.4.
In fact it is Ϫ0.05. There is almost no correlation between the wealth of a
UK region and the public spending it gets. Scotland had ample cushion,
and Wales had a little, to increase welfare spending compared to England,
as Persson and Tabellini predict.
The problem of representation has been acidly called ‘what those with
short memories call the West Lothian Question’.19 Gladstone, who had a
long memory, wrestled with it for seven years and failed to find an answer
in either of his Government of Ireland Bills (1886 and 1893). The
problem is: how (if at all) should a te
rritory which has a devolved government be represented in the House of Commons? If every territory had
devolution, the problem would be simple. But England does not, and the
attempt to offer an elected assembly in the north-east was turned down
overwhelmingly in a referendum in 2004. The House of Commons therefore doubles as the elected part of the government of the UK and of the
government of England. How many MPs from Scotland and Wales
should sit there, and what powers should they have?
One possibility would be to exclude them. But that would not be fair.
Scotland and Wales are not independent countries. Taxation, social security, foreign affairs and defence are not devolved. For the people of
Scotland and Wales to be excluded would be taxation without representation – the slogan of the rebellious American colonists in 1776.
Another possibility is what Gladstone called the ‘in and out solution’ –
that Scottish and Welsh MPs could vote on non-devolved matters such as
defence and social security, but not on devolved matters such as health
and education. This is current Conservative policy. But it is unworkable.
Whenever the UK party majority in the House of Commons differed
from the English party majority, the government of the day would be
unable to carry half its legislation. Either it would be formed by the party
with a majority of seats in England, which could not tax or run foreign
affairs; or it would be formed by the party with a UK majority, which
could not carry its English health or education measures.
A possible solution to the West Lothian Question is to reduce the
numbers but not the powers of MPs from Scotland and Wales. This has
been done in a minor way for Scotland but not for Wales. Scotland has
come down from seventy-two MPs to fifty-nine, which is only slightly
above its population share. Wales has forty, which is far above its population share. Northern Ireland was reduced to about two-thirds of its population share of MPs between 1920 and 1979 because it had a devolved
19 By the constitutional lawyer and Northern Ireland specialist Brigid Hadfield.
assembly. But when devolution there ended, its under-representation
ended too. On none of the occasions since then when devolution was
restored there, including the latest time in 2007, has anybody dared
mention the idea of cutting its Westminster representation back again.
Were this Northern Ireland solution to be again adopted, its representation in the Commons would be cut to about twelve, that of Scotland to
about forty, and of Wales to about twenty-four.