Written evidence submitted by Professor
Michael Keating, University of Aberdeen
SUMMARY
The taxation proposals Scotland Bill are flawed and,
if enacted, will need to be amended before long. There is a case
for further fiscal devolution. If English and Scottish preferences
on the size of the state diverge, larger powers will be necessary.
There are always problems with fiscal devolution and compromises
must be made. It is difficult to address the overall effect of
the proposals in the absence of proposals for the reform of the
Barnett Formula.
1. The present position, whereby the Scottish
Parliament has a large degree of policy autonomy but almost no
fiscal autonomy, is anomalous but the Bill will contribute only
partially to the aims of autonomy, policy innovation and accountability.
The way in which income tax is to be devolved differs little in
principle from the present 3p power, although the ceiling is raised.
There is rather little scope to use taxes as an instrument of
policy innovation. I have the impression that the motivation for
the new 10p tax power is more the political one of forcing the
Scottish Parliament to make an explicit tax decision, than to
give it more autonomy.
2. A radical measure of fiscal autonomy, with
the Scottish Parliament responsible for the bulk of taxes but
paying a share to Westminster for UK services would provide greater
policy autonomy and flexibility for Scotland and spread the cyclical
risk of tax fluctuations, at least to some degree. This would
resemble the Basque system.
3. On the other hand, I am inclined to say that
VAT should remain a UK tax, since: it is important that Westminster
too have a direct link to the taxpayer/ voter; it cannot be varied
under EU rules; there are problems in assigning it, since in principle
it is a tax on each stage of the production process, collected
for convenience at the point of sale. For this reason, it is normally
a federal tax in federal systems (as in Germany, Canada and Australia).
VAT is remitted on goods leaving the common tax area; this also
applies to the EU.
4. There is a case for further devolution in
social security, although in the Basque Country this is not devolved.
This, however, plays into the current debate about reform of welfare
and simplification of benefits, taking us well beyond the scope
of this bill.
5. It has been argued that that various taxes
should not be devolved given there are economic constraints on
variation. I am not convinced by these arguments. On the contrary,
it would be salutary Scottish Parliament were to face these constraints
directly. If taxes are raised too much, the country becomes uncompetitive;
if they are lowered too much, it loses revenue. This is a calculation
for Scottish policy-makers to undertake.
6. Fiscal devolution would allow Scotland to
raise or lower the overall burden of taxation, engaging in fiscal
competition with the rest of the UK. There is a strand of opinion
in Scotland that wants to engage is a strategy of radical tax
reduction in order to attract business investment and mobile skilled
labour. Suggestions that such tax-cutting will pay for itself
are wishful thinking; cutting taxes leads to reduced revenues.
Some sectors of political opinion would welcome the consequent
shrinking of the state. Others fear that such tax competition
will represent a "race to the bottom" with reduced
social services and note that much public spending supports economic
growth. Tax competition may lead to all governments underspending
on social policies. International evidence for this is rather
mixed. Transition or developing economies, who do not have to
support mature welfare states have tried this. There is tax competition
in Switzerland and in the United States, especially among municipal
governments. Elsewhere, governments realise that public spending
supports economic growth and that there are trade-offs involved.
In mature welfare states there is little scope for aggressive
tax-cutting without radical reduction in the scope of welfare.
These judgements about the balance of taxes and services, however,
is something that the Scottish Parliament should be make for itself.
7. Governments may not be able to make proper
decisions, however, if tax competition is so strong that they
are constrained to cut taxes to the level of their neighbours.
While it is unlikely that tax competition will destroy the welfare
state, it could have damaging effects. These can be limited by
a concordat on business taxation analogous to that on industrial
assistance; there are agreements in the Basque Country and something
similar has been proposed at European level although strongly
opposed by successive UK governments.
8. Devolved excise taxes would give Scotland
room for policy autonomy. If the UK government should commit itself
to a radical shrinking of the public sector and the Scottish Parliament
choose to maintain the existing welfare state, then more radical
devolution of taxation will be necessary. This could make tax
harmonization and concordats more difficult. The dilemma is inescapable
but such divergence will severely test the devolution settlement.
9. There might be some cross-border trade distortions
arising from differences in excise taxes but the geography of
the United Kingdom is such that small but significant variations
could be possible without provoking massive cross-border shopping.
In my experience of Canada, a minority of consumers behave irrationally,
spending a whole day driving hundreds of miles to save a few dollars,
but most consumers do their calculations and value their own time.
10. It has been argued that fiscal autonomy itself
produces economic growth. The evidence does not convince. There
are far too many other variables involved and the comparative
literature arrives at no reliable conclusion. Sometimes it is
claimed that growth ensues because autonomous governments sustain
low taxes but there is no systematic evidence that low-tax jurisdictions
grow faster than higher-taxed ones. There is, indeed, a high-cost
path to growth based on good infrastructure, education and training,
health, and reducing the social burden of deprivation. Others
argue that fiscal autonomy allows governments to make better spending
priorities, but this is also true under a system of block grants
as is presently the case in Scotland. A stronger argument is that
governments dependent on their own tax bases will prioritize policies
that maximize tax take and promote employment but I do not know
of systematic evidence to support this.
11. The 10p variation gives little scope for
variation in practice and it is likely that Scottish governments
will default to the UK rate. As no UK government has raised income
tax since the 1970s, it has become politically a very difficult
instrument to use. I would favour devolution of the bands, rates
and allowances, in order to provide room for pursuing social and
economic policy objectives. Under the present proposals, where
the UK Government alters the bands or allowances, the Scottish
take would be affected automatically, possibly to its detriment.
If this is to be compensated by ad hoc adjustments to the block
transfer, this undermines the idea of Scotland being responsible
for its own finances. It also opens up the issue of fiscal equalization,
which the present proposals do not mention (see below). Short
of a radical devolution of taxes, there is at least a case for
assigning a share of VAT and corporation tax on a population
basis to ensure the Scotland does not lose out from a shift in
taxes. This is common in other devolved systems.
12. The framework would not have performed well
in the recent recession since the Scottish Parliament does not
have borrowing powers. The loss of revenue would have had to be
met directly by spending cuts, as happens in US states with balanced
budget contributions, such as California where there are huge
budget and spending fluctuations. In Spain, autonomous communities
have suffered revenue shortfalls in previous recessions and central
government has helped them out by altering the transfer formula,
a benefit they kept into better times, creating an imbalance.
In the current recession, the central government is taking a stronger
line. Italian governments have also, contrary to the spirit of
decentralization laws, bailed out local and regional governments.
13. Allowing the Scottish Parliament to levy
other taxes will broaden its scope. Other countries allow sub-state
governments to levy any new tax that does not duplicate an existing
tax, which makes sense.
14. Borrowing powers are needed for three purposes:
to deal with short-term shortfalls, which the present proposals
provide for; to deal with cyclical downturns; and for capital
investment. There should be more scope for borrowing to deal with
cyclical downturns. If petroleum revenues were to be devolved,
there should also be a fund to accumulate surpluses in years of
high prices. The Scottish Government should also be allowed to
borrow for capital investment. International experience gives
many ways to do this, including borrowing on the market, government
taxpayer-backed bonds and revenue-bonds, which could be used for
things like bridges or rail improvements. There could be a limit
of total government debt such as exists (albeit rather theoretically
so far) in the EU.
15. International experience shows that it is
very difficult to predict the outcome of fiscal decentralization.
The present proposals contain problems that have been highlighted
by critics and which could prove destabilizing. If they are implemented,
we will return to the issue before long.
16. In many federal and devolved systems, government
has increased fiscal decentralization; an exception is Germany
where such proposals have not prospered. It usually happens that
one fiscal reform is followed by another, as the system develops.
Other countries have not relied so much on devolution of a single
tax. Rather they have transferred, by devolution or assignment,
a share of income, sales and business taxes. Tax variations may,
for economic reasons, be rather marginal but, spread across a
range of taxes, represent a significant degree of autonomy. There
is no perfect system of fiscal devolution and every system contains
anomalies. The objectives of avoiding market distortions, fiscal
responsibility and equity are by no means always consistent and
the outcome is always a political compromise.
17. One issue that is not addressed at all in
the proposals is that of fiscal equalization, with the Barnett
formula continuing for the non-devolved part of the grant. We
know that Barnett has been subject to constant refinements and
there have been so many instances of by-pass that the convergence
in spending levels implicit in it has repeatedly been postponed.
The inevitable anomalies in the proposed system will doubtless
lead to further tweaking of the Barnett allocations. There will
need to be a transparent mechanism to ensure that any restructuring
of UK taxation will not have a detrimental effect on Scotland.
18. Inter-territorial distribution of taxes and
expenditure has become a major issue in all devolved and federal
countries and has not been resolved anywhere. Other countries
have experimented with systems of revenue-sharing based on needs
and resources. This is always contentious and the outcome is a
compromise between objective calculation and political management.
Governments have often dealt with issues sequentially or postponed
difficult decisions but in times of economic crisis, as at the
present, have been forced to make harder decisions.
25 January 2011
| 
