My recent contributions to public choice.

AuthorNiskanen, William
PositionAssociation Lecture
  1. Introduction

    Some years ago, Henry Kissinger asked Zhou Enlai, "What have been the long-term effects of the French Revolution?" After a moment's reflection, Zhou responded, "It is too early to tell." Some months ago, Jim Gwartney asked me to present a paper on the long-term effects of public choice. After reflecting about this issue for too long, I responded similarly that it is too early to tell, even though the first major contributions to public choice shortly predate the French Revolution. For that reason, my paper summarizes my own primary recent contributions to public choice. You are invited to make your own judgment about the status of public choice based on the substance of these contributions.

  2. Autocratic, Democratic, and Optimal Government

    My book Autocratic, Democratic, and Optimal Government was published in 2004 and was developed from and was an update of my article with the same title published in 1997. The book (and article) develops models of the fiscal behavior of these three major types of government, incorporates estimates of the parameters of these models based on U.S. data, and then estimates the fiscal and economic outcomes of each type of government. Autocratic government is assumed to choose a combination of domestic government spending and a tax rate that maximizes the total return to the autocracy. Democratic government is assumed to choose a fiscal policy that maximizes the income of the median voter after taxes and transfers. Optimal government is assumed to choose a fiscal policy that maximizes the average income after taxes. (Per capita spending for defense and interest is assumed to be exogenous and the same for each type of government.)

    The next step is to estimate the parameters of these three models. Each model requires estimates of the elasticities of output per potential worker with respect to the level of non-defense government consumption spending per potential worker (G) and to the after-tax rate (I-R). The democratic model, in addition, requires estimates of the distributions of income, taxes, and transfer payments. For this purpose, my estimates of the two elasticities are those that were implicit in the actual levels of G and R in the United States in 1996, given the model of democratic government and the estimates of the several distributional parameters from census data.

    Autocratic governments are motivated to provide some services to the general population but only to increase the net amount that they can extract from the economy--in this case only about 58% of the services per potential worker provided by the United States government in 1996. The tax rate is the revenue-maximizing tax rate, in this case about 23 percentage points higher than in the United States. The economic outcomes are characteristically grim for the general population; in this case, the average net income is about 36% of that in the United States.

    An optimum government, by contrast, provides about 7% more government services per potential worker than the United States. The tax rate is about 15 percentage points lower than in the United States, even though it faces the same per capita expenditures for defense and interest payments, primarily because a risk-neutral population will not authorize any transfer payments behind the veil of ignorance; the average net income is about 22% higher than in the United States.

    Variations of the parameters of the democratic model make it possible to estimate the effects of changing the voting rule, the form of taxes, and the fiscal horizon with some surprising results.

    A political measure that reduces the voting franchise, such that the pre-tax income of the median voter is closer to the average pre-tax income, interestingly, would have broadly beneficial effects. This measure would increase expenditures for general government services, reduce the average tax rate, and increase both average and median net income.

    Conditions that increase the marginal effect of the after-tax rate on total output would have a mixed effect. Such conditions include an increased mobility of labor and capital among jurisdictions or an increase in the progressivity of tax rates. An increase in this parameter would reduce the relative size of government by reducing both expenditures for general government services and the average tax rate with only a small net effect on total output. Although an increase in this parameter would lead to a small increase in average net income, it would also lead to a reduction in transfer payments and a substantial reduction in median net income. A broad-based tax reform, in contrast, would have the opposite effects, increasing the relative size of government, unless complemented by a constitutional amendment that would require a super-majority rule on subsequent...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT