Targeting Benefits for Electoral Gain: Constituency Marginality and the Distribution of Grants to English Local Authorities
| Author | Hugh Ward,Peter John |
| Published date | 01 March 1999 |
| Date | 01 March 1999 |
| DOI | http://doi.org/10.1111/1467-9248.00186 |
| Subject Matter | Article |
Targeting Bene®ts for Electoral Gain:
Constituency Marginality and the
Distribution of Grants to English
Local Authorities
HUGH WARD
University of Essex
AND PETER JOHN*
University of Southampton
We model the spatial allocation of resources over constituencies as an optimization
problem in which governing parties face uncertainty about voter preferences, but seek
to increase their chances of getting re-elected. We show that a rational government
should allocate extra resources to marginal constituencies and especially favour
opposition-held marginals. We test this hypothesis on data about central government
grants to larger English local authorities. We consider whether Conservative con-
trolled and `¯agship' local authorities also bene®t. Our empirical results suggest that
the government allocated around £500 million more to local authorities containing
marginal constituencies and around £155 million more to `¯agship' local authorities
than they could have been expected to get on the criteria of social need and
population.
The idea that politicians design policies to win votes is central to contemporary
political science. The consequence of this simple idea is the profusion of theor-
etical and cross national empirical work on spatial models of party competition
and political business-cycles. It is also possible, however, that politicians
compete for votes in ways which are not so readily observable, but are no less
important. When a party can gain power by winning a small number of
marginal seats and where a small swing of the vote can change the outcome,
political leaders may try to allocate bene®ts towards electorally vulnerable or
winnable areas. Indeed, the use of distributive politics for political ends may be
more advantageous than manipulating the economy. Governments risk their
* We thank Rita Hale of Rita Hale Associates, London and the Association for Metropolitan
Authorities (now Association for Local Government) for supplying the data. We are grateful to
George Boyne, John Gibson, David Halpern and David Sanders for written comments. We thank
John Preater for ideas on the modelling and Sue High for advice on the statistical tests. We also
thank participants both of the Department of Political Science and International Studies seminar,
Birmingham University, who heard the paper on 16 November 1995, and of the Political Studies
Association Rational Choice Group meeting, 23 January 1996 for their reactions. We are also
grateful to Carol Ward for help in the preparation of the data.
# Political Studies Association 1999. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main
Street, Malden, MA 02148, USA.
HUGH WARD AND PETER JOHN
33
reputation for good economic management when they in¯uence the macro-
economic aggregates to maximize short-run electoral gain. The executive's
future freedom of decision in economic policy may be curtailed if a re¯ationary
policy triggers a price-wage cycle which has to be damped down in later years.1
While room for manoeuvre on the macro economy has always been limited by
international economic competition,2 the globalization of ®nancial markets
during the 1980s further limits the possibilities for creating a pre-election boom
by a tax cut and the reduction of interest rates. Instead, governments may prefer
policy instruments that do not disturb macro-economic aggregates but win
elections by redistributing bene®ts.3 The tactic may also avoid the need to adopt
policies which are widely supported among the electorate as a whole but are
inconsistent with the policy preferences of the party elite.
Empirical studies of the political determinants of grant distribution generally
assume rather than demonstrate that it is rational to direct resources towards
marginal constituencies.4 Johnston, for example, argues that the incentive for
competitive parties to please the median voter is stronger in marginal con-
stituencies,5 but he fails to model the constraints on the total resources that can
be distributed.6 There is no study which shows that distributing resources to
marginals is a solution to a well-de®ned problem of constrained vote maxim-
ization.7 Yet we demonstrate below that such a distribution is rational within a
stylized model of Westminster-style political systems. In contrast to the assump-
tion in the empirical literature that the relationships between marginality and
bene®ts are continuous and linear, we predict a discontinuous relationship
between seat marginality and bene®ts. Counter-intuitively, we show that the
government favours marginals held by the opposition parties.
We test our model with data on central government grants to UK local
authorities. The key feature of the system is the centralization of power in
Whitehall in the hands of government ministers, such as the Secretary of State
for the Environment and the Ministers of State for Wales, Scotland and
1 K. Shultz, `The politics of the political business cycle', British Journal of Political Science, 25
(1995), 85±7.
2 J. Alt, `Political parties, world demand and unemployment: domestic and international sources
of economic activity', American Political Science Review, 79 (1985), 1016±40; R. C. Bryant, Money
and Monetary Policy in Interdependent Nations (Washington DC, Brookings, 1980).
3 See the critique of business cycle models by M. S. Weatherford, `An Agenda Paper. Political
business cycles and the process of economic policymaking', American Politics Quarterly, 16 (1988),
99±136 and the review A. Alesina, D. Cohen and N. Roubini, Macroeconomic Policy and Elections
in OECD Countries, NBER Working Paper no. 3830 (Cambridge MA, National Bureau of
Economic Research, 1991). The more precise modeling of the business cycle of Schulz is consistent
with the line of argument presented here.
4 R. J. Johnston, `The aggregation of federal money in the United States: aggregate analysis by
correlation', Policy and Politics, 6 (1978), 279±97; J. N. Reid, `Politics, Program Administration,
and the Distribution of Grants-in-aid: a Theory and a Test', in B. S. Runquist (ed.), Political
Bene®ts (Lexington MA, Lexington, 1980); H. Raimond, `The political economy of state inter-
governmental grants', Growth and Change, 14 (1983), 17±23.
5 R. J. Johnston, Political, Electoral and Spatial Systems: an Essay in Political Geography
(Oxford, Oxford University Press, 1979), pp. 86±7.
6 A. Downs, An Economic Theory of Democracy (New York, Harper-Row, 1957), p. 52.
7 The closest is G. Wright, `The political economy of New Deal spending: an econometric
analysis', Review of Economics and Statistics, 56 (1974), 30±8 who models the distribution of
resources to US states. He uses, though does not solve, a constrained maximation model to create
an index of the vote `productivity' of a given amount of public spending in a state. We solve the
constrained maximization problem Wright considers prohibitive.
# Political Studies Association, 1999
34
Targeting Bene®ts for Electoral Gain
Northern Ireland. Local authorities do raise some revenue through local taxa-
tion. However, about 80 per cent of local authority expenditure is now funded
from national government, most of which is determined by a centrally decided
formula which distributes the revenue support grant. Although there is still
some local ®nancial discretion, central government has the power to limit the
total amount local authorities spend and, to a large extent, to determine how
the money is spent. Grants to local authorities are about a quarter of total
public expenditure, so they are big enough to ensure that small adjustments to
their distribution create large dierences in the spending opportunities of
local authorities and thus to the services delivered to citizens or to the level of
local taxation. Other forms of government expenditure, for example transfer
payments made under national welfare programmes, are governed by uniform
eligibility criteria which are dicult to adjust to favour marginal constituencies.
It is thus plausible to suggest that the UK national government adjusts its
®nancial allocations to local authorities for electoral advantage.8
There already exists some support for the hypothesis. Despite the convention
that the determination of central government decisions follows publicly agreed
de®nitions of the need for local services generated through a process of debate
between central and local government and interest group consultation,9 there is
some limited evidence that political variables enter into central grant alloca-
tions.10 Moving beyond central allocation of grants to local authorities, govern-
ments target economic resources and avoid policies which have adverse
economic ± particularly employment ± impacts in electorally vulnerable
seats.11 Other political factors may also enter the geographical distribution of
resources. `Flagships' ± local authorities which are exemplars of the Tory
philosophy of eciency and the application of market forces at local level ± are
important for the Conservatives to win. They also in¯uence national elections as
voters in constituencies with ¯agship local authorities have electorally rewarded
the government.12 The ¯agship London municipality Wandsworth received a
favourable grant allocation so it could deliver a low local tax in 1990 and in
subsequent years.13 There is also a folk wisdom in local government that central
government favours councils of its own...
To continue reading
Request your trial