Public Finance — Local Public Finance and Fiscal Federalism

Public Finance — Who Collects and Who Spends Changes the Outcome

The public goods analysis of Chapter 1 treated government as a single entity. In reality, central government and several tiers of local government share taxing and spending. This chapter looks at the principles of that division, and at how the way central government passes money to local governments changes local choices.

1. Decentralization respects differences in preferences

Oates’s decentralization theorem: if the benefits of a public good stay within one region, preferences differ across regions, and central government would have to supply the same level everywhere, welfare is higher when each region sets its own level.

Suppose residents of two regions, A and B, want park budgets of 2 billion and 4 billion won respectively. If central government gives each 3 billion won, A is oversupplied by 1 billion won and B undersupplied by 1 billion won. Letting each locality decide eliminates this loss.

Conversely, functions whose benefits cross regional borders (regional transport, infectious disease control), that have large economies of scale (defence), or whose effects leak as people move, such as redistribution, are better handled centrally.

2. Voting with one’s feet: the Tiebout hypothesis

Tiebout argued that if people move to jurisdictions offering the combination of taxes and public services that suits their preferences, local public goods can be supplied as efficiently as in a market. People reveal their preferences by moving rather than by voting.

This conclusion rests on strong assumptions: no moving costs, job locations that do not constrain where people live, enough jurisdictions, and no spillovers. In reality, traces of the Tiebout mechanism appear as capitalization. House prices in areas with good schools reflect the value of those services, and house prices in areas with high property taxes reflect the present value of those taxes.

Tiebout sorting also produces segregation. When residents with similar incomes cluster, rich areas get good services at low tax rates while poor areas get poor services even at high rates. This is why fiscal equalization across regions is needed.

3. The form of grants changes local choices

Central government passes money to local governments in two main ways.

• Unconditional (general) grants: given with no restriction on use. They have only an income effect, shifting the locality’s budget line outward in parallel.
• Matching grants: for every won a locality spends on a particular programme, central government adds a set proportion. This adds a price effect that lowers the relative price of that programme.

With a 1:1 matching grant, in which central government adds 1 won for every won the locality spends on a public service, the price of a won’s worth of service to the locality is half a won. Spending on that service rises by more than with the same amount given unconditionally. In terms of the externality logic of Chapter 1, giving a matching grant equal to the share of benefits that spill over to other regions is a Pigouvian subsidy.

4. The flypaper effect: money sticks where it lands

In theory, 1 won of unconditional grant is the same as 1 won of residents’ income. If residents spend 10% of their income on local public services, an unconditional grant of 10 billion won should raise spending by about 1 billion won, with the remaining 9 billion won returned to residents through tax cuts.

Most empirical studies find that 1 won of grant raises local spending by much more than 1 won of residents’ income. Because the money sticks to the government budget where it first lands, this is called the flypaper effect. Proposed explanations include fiscal illusion (residents are not well aware of the grant), bureaucrats’ incentive to expand budgets (Chapter 3), and expectations that the grant will be temporary.

5. Fiscal imbalances and soft budget constraints

Vertical imbalance is a structure in which tax bases are concentrated at the centre while spending responsibilities lie with local governments. Horizontal imbalance is the difference in tax bases across regions. Grants fill both gaps, but when localities spend money they did not raise themselves, the accountability that links taxes and spending weakens.

The bigger risk is the soft budget constraint. If local governments expect central government to bail them out in a fiscal crisis, they overspend and overborrow in advance. The more credible the central government’s promise of rescue, the bigger the problem. That is why prior discipline, such as approval of local bond issues and designating fiscally distressed authorities, is needed.

Check your understanding

Central government adds 3 won for every won a locality spends on libraries. What price does the locality face for library services? $1/(1+3)=0.25$: it buys a won’s worth of services for a quarter of a won. If 20% of library benefits go to residents of other regions, the price that corrects the spillover is 0.8 (a matching rate of $m=0.25$). A price of 0.25 is far below that, so the grant goes beyond the corrective level and induces oversupply.

References

• Wallace Oates, Fiscal Federalism (1972)
• Charles Tiebout, “A Pure Theory of Local Expenditures,” Journal of Political Economy (1956)
• James Hines and Richard Thaler, “The Flypaper Effect,” Journal of Economic Perspectives (1995)
• Harvey Rosen and Ted Gayer, Public Finance, ch. 22