« Icahn's Eleven (Er, Ten) | Main | An Open Letter to the Yale Law Journal: What is Your Real Policy? »

Thursday, May 15, 2008

How does one measure "governmental centralization"?

At the risk of wearing out readers with my shilling for my ALEA panelists' articles, I mention Ezra Friedman's first-rate empirical piece investigating the relationship between governmental size and willingness to redistribute wealth. Friedman finds that counties with more decentralized expenditures are less likely to redistribute wealth. That is, county Leviathans are more friendly to the poor.

One might, however, quarrel with Friedman's measure of decentralization. He looks to the fraction of county expenditures that are spent by subcounty governments -- towns, villages, cities, etc. He acknowledges that the measure is imperfect: The towns might get all of "their" revenue from the feds, the state, or the county, and those higher-levels might dictate the spending decisions by "puppet" local governments.

But the problem is even greater than he acknowledges. Many local governments are created for the sole purpose of keeping taxation low. Such municipalities are often formed defensively to avoid annexation by a neighboring municipality. (The practice was comprehensively documented by Gary J. Miller, Cities by Contract: The Politics of Municipal Incorporation (Harvard Press 1981) and Nancy Burns, The Formation of American Local Governments: Private Values in Public Institutions (Harvard Press 1994)). Such municipalities are often little more than five unpaid trustees and an answering machine: They purchase minimal services from the county, bargaining for low fees and low service levels, because their constituents dislike taxes more than they like services. The decision about revenue in such a county is decentralized in the sense that tiny towns are dictating low taxes and low expenditures. But such a county will come across as highly centralized in Friedman's measure, because the county budget might dwarf those of the anti-tax towns.

This comment is not meant as a criticism of Friedman (whose collection and analysis of data is truly impressive) but rather as a lead-in to a more general question: How does one measure decentralization quantitatively? Some scholars have simply despaired of finding any reliable quantitative measure (see, e.g., Edward C. Page, Localism and Centralism in Europe: The Political and Legal Bases of Local Self-Government (1991)).

Such a measure would be enormously useful in analyzing how various legal arrangements affect centralization levels. (Think of the bromides one reads in law reviews about federalism's being protected through the national political process: A good quantitative measure of true federalism -- that is, state control of money or power -- would help confirm or disconfirm such statements).

Alas, I have never encountered a reliable measure. Has anyone else had better luck?

Posted by Rick Hills on May 15, 2008 at 01:23 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference How does one measure "governmental centralization"?:


Post a comment