A Short Essay on Local Public Goods
David M. Nowlan
The term "local public
goods" was introduced into the economics literature by Charles Tiebout in
a still frequently cited 1956 paper.
Until this paper changed the focus, public or social goods were
conceived of simply as goods that, if available to one person, were available
to all. National defense was a commonly cited example. Richard Musgrave's somewhat awkward 1969
characterization of such goods as exhibiting "non-rivalness" in consumption
and "non-excludability" from consumption has remained in the
literature.
The non-excludability of public
goods has an important consequence, namely that a decentralized mechanism to
achieve their optimal provision cannot be found; that is, it is not generally
possible to find a way to get individuals to reveal their true valuation for
public goods. This is the public goods
problem, and it's one upon which Paul Samuelson laid considerable stress in a
well known 1954 paper on public goods.
The public goods problem
highlighted by Samuelson led to Tiebout's 1956 response. Tiebout argued that
there was a class of public good, the local public good, for which a
decentralized mechanism for achieving optimal allocations did indeed exist. His paper, along with others published in
reaction to Samuelson's article, focussed on the fact that many public goods
are subject to congestion. This is
especially true, it was argued, of public goods provided by local governments. A city park, a stretch of roadway, a fire
department, a school; these are available to everyone in the community, but for
any given level of infrastructure the more people who use the facility the more
crowded it becomes and the less it is available or useful to others. Using
Musgrave's terminology, local public goods exhibit non-excludability but not
non-rivalness; they are partially rival (or partially non-rival).
Tiebout's model is one in which
each local community or jurisdiction provides a mix of public goods. Those who live in the jurisdiction receive
the benefits of these goods and pay for them through a tax levied equally on
each taxpayer. There are no interactions between jurisdictions.
The key to Tiebout's mechanism lies
in the assumed mobility of people. If
people can costlessly move from one jurisdiction to another, they will move to
the jurisdiction in which the mixture of services and tax level provides them
with the greatest net benefit. Other
locational attributes, such as the availability of jobs, are ignored.
Those who like better schools and
are prepared to pay more for them will be in one community; those who would
prefer to pay less and make do with poorer schools will be in another
community. Taxes are paid according to
benefits received, a situation that has come to be called the
"benefits" view of local taxation.
With enough variety among the jurisdictional offerings, each community
will end up with people having identical preferences.
Because of congestion, there will
be some optimum or least-cost size for each community. This size will occur where the benefit of
sharing the infrastructure costs with another taxpayer will be just equal to
the crowding cost imposed by the new person.
If there are more people with a given preference pattern than can fit within
an optimal community, then their needs will be met in another identical
jurisdiction.
Within each jurisdiction, residents
will pay the same tax, consume the same amount of public goods, and agree on
the desirability of expanding or contracting any given service. The problem of providing a mechanism to
achieve the efficient provision of public goods would appear to have been
solved, at least for local public goods.
In Tiebout's words, "Spatial mobility provides the local
public-goods counterpart to the private market's shopping trip."
Tiebout's theory has a close
counterpart in the theory of clubs, which was introduced in a 1965 paper by
James Buchanan, and it is through the development of club theory that a number
of aspects of local public goods have become better understood. Like Tiebout's communities that provide local
public goods only to tax-paying members of the jurisdiction, clubs are able to
restrict the provision of club services to those who are members. Club goods are thus excludable, like private
goods, but like local public goods they are only partially rivalrous.
One of the best known club
theoretic results is that a group of individuals with heterogeneous preferences
would be better off forming a separate club for each homogeneous preference
group rather than forming one club for the larger group. In terms of the local
public-goods model, this implies that if there were two types of people, the
rich with one set of preferences and the poor with another set, both types
would be better off in their own separate Tiebout jurisdiction than they would
be together in a mixed jurisdiction.
This conclusion rests on a number
of strong assumptions. The first is that
local taxes must be the same for everyone.
Using the example of the above paragraph, if poor
people in a mixed jurisdiction pay lower local taxes than rich people while
having the same access to local public goods, then poor people will not
necessarily be better off in their own jurisdiction.
The second assumption is that there
must be no beneficial interaction among the different members of a
heterogeneous group: there must be no benefit in having people of different
skills or cultures together in one community.
The third assumption is that an
integer number of optimally sized jurisdictions exists
for each homogeneous group. If the best
sized community for some set of local public goods is 50,000 and there are
75,000 people sharing a preference for this mix of public goods, then the
Tiebout solution cannot exist.
This so-called "integer"
problem may be less important than it at first appears. An optimally sized jurisdiction occurs at the
bottom point on a U-shaped cost curve (where costs include congestion as well
as infrastructure costs) where returns to scale are constant. Constant returns to scale in this case means,
for example, that equal proportionate changes in infrastructure and in facility
use would not change the degree of congestion.
If the bottom part of the U-shaped cost curve were relatively flat,
implying that constant returns to scale prevailed approximately for some
distance to either side of the optimum, then a range of community sizes would
be approximately optimal and so it would be easier to provide an integer number
of communities for a group of like-minded people of any size. If constant returns to scale existed
everywhere, then the integer problem would completely disappear: any person
with a unique preference profile could have their own optimally sized
jurisdiction of one.
The existence of constant returns
to scale has another interesting implication, one first explored within the
context of optimal road pricing. If
exclusion is possible with respect to a local public good, so that a user fee
may be charged, then a fee set optimally to equal the external congestion cost
imposed on other users will generate exactly the amount of resources needed to
cover the infrastructure cost, if the infrastructure is of optimum size.
Tiebout's article has generated an
enormous and continuing literature, much of it dealing with two principal
normative implications of his model: that smaller homogeneous local
jurisdictions are better than larger integrated municipalities,
and that politics is unnecessary at the local level.
Some see the absence of politics as
a strength of the mechanism, while others are critical
of the model precisely because it does not deal with the politics of municipal
finance. The absence of politics leaves
the Tiebout model with a gap on the supply side of local public goods. How do the optimal jurisdictions arise? The issue of supplying optimal clubs has been
the subject of much club-theoretic literature, where reasonable dynamics have
proved difficult to establish. This
difficulty carries over to the local public-goods model of Tiebout, although
some argue that the maximization of land value in a local jurisdiction provides
the proper signal for a developer of an urban community to achieve the Tiebout
result.
The issue of smaller homogeneous
jurisdictions versus larger integrated ones is currently in high relief, both
in the theoretical literature and in practice.
Those who favour smaller jurisdictions rely almost entirely on the
arguments set forth by Tiebout, combined with the standard view of fiscal federalism
that local governments should not engage in redistribution but should focus on
supplying local public goods optimally.
Those who argue differently do so for a variety of reasons. One is that local governments do and should
have a redistributive role to play.
Larger, heterogeneous jurisdictions are better positioned to play that
role than smaller homogeneous ones.
A second reason for favouring
larger jurisdictions is based on more theoretical considerations. Tiebout did not take into account the effect
on land prices as more people try to access a local public good available only
at a given geographic location. Recent
research suggests that when land markets are taken into account, the
Tiebout-type decentralization of local public good provision requires larger
metropolitan governments supplying a range of goods and services.
Local Public Goods – Bibliography
Buchanan, J.M. "An Economic Theory of
Clubs." Economica. 32 (1965), 1-14. The original exposition of
club theory and quite accessible to non-technical readers.
Cornes, Richard and Todd Sandler.
The Theory of Externalities, Public Goods, and Club
Goods.
Henderson, J. Vernon. "The Tiebout
Model: Bring Back the Entrepreneurs."
Journal of Political Economy. 93.2 (1985), 248-264. Shows how the Tiebout
outcome could occur with land-value-maximizing community developers and no
politics.
Hochman, Oded, David Pines and
Jacques-François Thisse. "On the Optimal
Structure of Local Governments."
The American Economic Review. 85.5 (1995), 1224-1240. Reviews some of the literature incorporating
land markets into the Tiebout model and argues that
efficient local governments should be metropolitan in scope, embracing a range
of overlapping locally provided public goods.
Mieszkowski, Peter and George R. Zodrow.
"Taxation and the Tiebout Model: The
Differential Effects of Head Taxes, Taxes on Land Rents, and Property
Taxes." Journal
of Economic Literature. 27
(September 1989), 1098-1146. This survey
article examines the conditions under which a local property tax would be a Tiebout-type benefits tax, and it discusses the
difficulties of testing empirically whether or not the local property tax is in
reality a benefits tax. It provides a
nice linkage between Tiebout's contribution and more
recent theories of local public finance.
Musgrave, R.A. "Provision for Social
Goods." In
Public Economics, edited by J. Margolis and H. Guitton,
pp. 124-144. International
Economics Association,
Rose-Ackerman, Susan. "Beyond Tiebout:
Modelling the Political Economy of Local
Government." In Local Provision
of Public Services: The Tiebout
Model After Twenty-five Years, edited by George R. Zodrow, pp. 55-83.
Samuelson, Paul A. "The Pure Theory of
Public Expenditure." The Review of Economics and Statistics. 36.4 (1954),
387-389. This short paper provoked a lot
of reaction, including Tiebout's 1956 paper.
Tiebout, Charles
M. "A Pure Theory
of Local Expenditures." Journal of Political Economy. 64 (October 1956), 416-424. Still frequently referred
to, especially in the context of the debate over the best size for municipal
government.