Definitions and other foundational issues are summarized elsewhere.
Kenyon reviews six theories that "give a flavor of the range" of existing theories, starting with
Tiebout (1956)
The classic voting-with-your-feet model.
The key actors in Tiebout’s (1956) model are individuals (consumer-voters) who decide which of many local governments to locate in, based on their demands for government services and the public service/tax packages offered by the various governments. Tiebout assumes that individuals have full knowledge of the various government revenue and expenditure packages; that individuals may choose among a large number of communities; and that individuals are fully mobile (for example, that jobs place no locational constraints on them). Furthermore, he assumes that no intercommunity pillover effects are present and that each community is able to attain its optimal size, at which the average cost of producing its particular package of public services is minimized.
She says this best describes IJC for households in a large metro area with many suburban governments. Tiebout's conditions are restrictive (complete information, full mobility, no intercommunity spillovers, optimal sizing), but to the extent they are met, local govts will exhibit allocative (produce the right services in the right amounts) and productive efficiency (produce them at least cost). In this model, local taxes are benefit taxes rather than ability-to-pay taxes.
Besides its restrictive assumptions, Tiebout doesn't account for decisions by business firms.
Oates-Schwab (1991)
This is a model of competition for mobile capital in the Tiebout tradition.
In this model, jurisdictions compete for a mobile capital stock by lowering taxes and providing public inputs to firms, such as roads and police and fire protection. In return for a larger capital stock, residents receive higher wages. A community must, however, weigh the benefits of higher wages against foregone tax revenues and the cost of public services (1991, p. 129).
Oates and Schwab assume that the local government’s objective is to maximize the welfare of its constituents, subject to the applicable budget constraints. They also assume that no beneficial or negative spillovers occur and that a sufficient number of local governments exist to approximate a competitive market. Furthermore, they assume that communities have full information about the wage benefits provided by the location of business firms in their communities, and that firms can correctly evaluate the tax and expenditure packages offered by the various communities. An assumption implicit in their model is that economic development efforts by local governments (that is, the attempts to determine optimal tax–public service packages for business firms, and bargaining with business firms) are costless. (p. 20)
In the Oates-Schwab model taxes on both households and businesses become benefit taxes. Firms are neither subsidized nor exploited, they pay the cost of the services they use. As in the Tiebout model, there is no redistribution. Service provision is allocatively and productively efficient.
Key criticisms of Oates-Schwab:
Full knowledge -- how can officials evaluate the benefits of a new firm to the community? Or the cost of services (or their benefits) to the firm?
Bargaining power and political considerations -- it doesn't account for the bargaining power of a large, high-profile firm. Officials may find the community cost of paying too much less than the personal political cost of letting a big one get away.
McGuire (1991)
McGuire's informal model suggests that states engage in "destructive competition."
She assumes that individuals “have preferences for redistribution and thus choose revenue systems that rely on ability-to-pay taxes” (p. 154). McGuire further assumes that the nation’s population is heterogeneous in terms of income and mobility. (p. 21)
The first assumption seems to rest mainly on McGuire's presumed preference for redistribution and ability-to-pay taxes. The second just means that peoples' incomes and mobility vary.
According to McGuire's model an optimal level of public services could be computed and could be obtained if there was no mobility of people or firms. But the optimal can't b obtained because any single jursidiction has an incentive to cut taxes for wealthy and mobile individuals and businesses to lure them to relocate. The tax-cutting govt will hope to use revenue gained from the incoming individual or firm to cut taxes or increase services for existing residents. However, all localities have the same incentive to cut. Eventually all the tax cuts cancel out, there is little net relocation, and communities are left with lower tax revenue or else have to raise revenue on immobile residents and businesses. According to McGuire allocative efficiency can't be achieved. [This conclusion would seem to rest on her assumption of a preference for redistribution.] But productive efficiency holds; jurisdictions seek to minimize tax burden for a given level of public services.
McGuire notes a similarity between her theory and the prisoners dilemma. Collectively states would be better off if none of them offered incentives. But individually they each have an incentive to offer them.
If your competitor offers no incentive, you can gain the upper hand by offering one. If he does offer, then you need to also in order to compete. Thus individually rational decisions produce a collectively irrational result.
In the case of states, since they all have equal autonomy, any sort of noncompetitive agreement would require outside (i.e. federal) enforcement.
Wolkoff (1992)
Wolkoff uses game theory in an attempt to explain seemingly irrational public policies.
In his model, governments use subsidies to induce potentially mobile firms to not move. Not all firms are potentially mobile and governments do not have perfect info about which are mobile and which are not. Both the firms and the governments engage in strategic behavior. (That is they calculate the response of the other and take that into account in choosing its own action.) The community decides whether and what size subsidy to offer. The firm decides the size of subsidy to request.
Wolkoff sets up a decision tree with all possible outcome states and their probabilities. Then he calculates the expected value of each course of action. A community is assumed to choose the coures that produces the largest expected value.
Wolkoff ’s model explains two types of seeming irrationalities in existing economic development programs. Suppose that all firms request the same subsidy, whether they are potentially mobile or not. The community then has no way of distinguishing between the two types of firms. It turns out that the most advantageous strategy for the community will be for it to offer modest subsidies to all firms. The inevitable result is that some firms with no potential for relocation will receive a subsidy. What seems like a waste of funds from the community’s perspective is rational maximizing behavior.
An alternative scenario outlined by Wolkoff is based on a community’s effort to separate potentially mobile from immobile firms. To do this, the community makes subsidy awards uncertain. Immobile firms then reduce the size of their subsidy requests. The community ends up avoiding providing large subsidies to firms that have no possibility of relocating. However, at the same time, the community rejects the requests of, and thereby loses, some mobile firms. When looked at in isolation, the fact of providing insufficient economic development subsidies to certain mobile firms appears irrational. Wolkoff ’s point is that we cannot look at such phenomena in isolation. According to Wolkoff, “the apparent irrationality at the micro level is resolved when one understands these decisions as being part of a more general subsidy strategy for all firms” (1992, p. 352). (p. 23)
There are also political and organizational explanations for seemingly irrational policies.
Wolkoff argues that the political credit to be gained from an economic development program that is only partially effective may be great. As Wolkoff states, “the symbolic benefits of being able to point to the operation of a policy actually may be more valuable to political officials than the fiscal benefits tied to the investment itself ” (1992, p. 343). A second explanation is organizational. Wolkoff points out that economic development officials are generally charged with fostering economic growth, but typically they are not responsible for the consequent effects on a jurisdiction’s budgets. Particularly when some program costs are difficult to identify and measure, one ends up with behavior that is rational from the perspective of the economic development official but wasteful from the perspective of the community at large. (p. 23)
These points are similar to some raised by Rubin and others.
Wolkoff doesn't paint a pretty picture of economic development policy:
One finds in Wolkoff ’s paper that communities
sometimes award subsidies to firms that will
be unaffected by such subsidies, that communities
sometimes neglect to offer subsidies to certain firms
sufficient to entice them to stay, that political appearances
may be more important to a public official than
economic growth generated, and that the structure of
economic development programs within governments
may create a systematic ignorance of the costs of
economic development initiatives. (p. 23)
Well, no one said politics was pretty!
Kenyon notes as shortcomings that Wolkoff doesn't consider the rationality of e.d. programs from the wider perspective of the country nor does he consider the separate roles of individuals and businesses.
Besley and Case (1995)
In the Besley-Case model voice, rather than exit, is the key to accountability of officials; in other words, yardstick competition. They also account for imperfect information.
Politicians know more about cost of service than their constituents; voters use info about tax changes in other jurisdictions to evaluate their own incumbents.
Some (bad) politicians engage in rent-seeking by raising taxes above the cost of service; other (good) politicians do not.
Politicians use strategic behavior in their tax-setting in order to influence voters’ beliefs regarding whether they are good or bad politicians. Voters fail to reelect incumbents whom they judge by their tax changes, relative to the tax changes of neighboring jurisdictions, to be bad politicians. (p. 24)
Kenyon says that the Besley-Case model is most applicable to state-level competition or perhaps in metro areas with a relatively small number of suburbs.
It's not clear to me what this has to do with competition for firms. Kenyon says it is difficult or summarize the effects of IJC in this model. It seems to be more of a model of how politicians change taxes and what consequences they suffer.
Besley and Case find their empirical evidence to be consistent with their theory. They find that “own tax changes increase the probability of incumbent defeat, and neighbors’ tax changes reduce the probability.” Further, they find that “when a neighboring state increases/decreases taxes by one dollar, the home state will increase/decrease taxes by roughly 20 cents” (p. 36). (p. 24-25)
Breton (1996)
Breton offers a general theory of competitive government. From Kenyon's description it appears to encompass much more than just competition for firms. He assumes that individuals seek to maximize utility and governments seek to maximize consent.
He includes both implicit and yardstick competition. He notes that in a perfect Tiebout world their will be no yardstick competition; everyone would be sorted into their own perfectly-matched community. Other governments wouldn't provide a yardstick for comparison.
Breton has a benign view of competition. He sees it as a way to reveal the demand for government supplied goods and services.
He does recognize the possibility of instability. For example, state efforts to attract business through lenient corporate charter laws might produce a race to the bottom as each state tries to match the most lenient state. Another example is urban crisis, in which high inner city tax rates drive wealthy residents to the burbs, leading to more inner city tax hikes, etc.
Breton sees a role for the national government in monitoring state and local competition.
Those are the six theories Kenyon reviews.
Kenyon, D.A. (1997). Theories of interjurisdictional competition. New England Economic Review, March/April, 13-35.
Posted by Chip on June 14, 2004 at 08:55 PM | TrackBack