Taxing Questions


How to Limit Government Spending, by Aaron Wildavsky, Berkeley: University of California Press, 1980, 197 pp., $8.95.

The Power to Tax, by Geoffrey Brennan and James M. Buchanan, Cambridge: Cambridge University Press, 1980, 231 pp., $22.50.

Both of these books are about constitutional limitations on government, and both are in favor of the idea. That's all they have in common, however.

Wildavsky is, to judge by his book, passionately devoted to the middle of the road; his book is, perhaps as a result, consistently superficial. One example is his explanation of the failure of the ambitious social legislation of the '60s. The problems government undertook to solve were simply too difficult. "No matter how much money is spent, reading, health, and recidivism rates do not improve, because there is no known way of doing these things." It does not occur to him to ask whether government, being unable to teach, cure, or reform, ought perhaps to get out of the schooling, health, and rehabilitation businesses. That might force him to question the belief implicit throughout the book that the optimal level of government spending must be within 10 percent of the present level of government spending.

Another example is his attempt to explain the growth of government by political developments of the early '60s—primarily the abandonment of the goal of a balanced budget. He somehow neglects to mention that by 1960 the federal government was already spending 18½ percent of GNP, up from 2½ percent in 1890; it is currently spending about 22½ percent. Causes usually precede their effects.

How to Limit Government Spending is irritatingly superficial, rarely pursuing any argument beyond a few sentences, but it is by no means useless. Its discussion of a wide range of arguments for and against current proposals to limit government expenditure will be useful to anyone involved in that controversy; if it had treated them in much greater depth, it would have been a much longer book.

While Buchanan and Brennan make a few bows to the "tax revolt" in their preface, The Power to Tax has little direct connection with current political controversy; it is a product of an approach to public finance that has occupied one of the authors since long before Proposition 13 was thought of. To understand what that approach is and why it is both radical and important, one must first understand the orthodoxy it opposes.

The traditional literature in public finance seeks to determine how government can extort a given quantity of resources at the least cost to the society. Stated in this way, it may seem a harmless and even useful exercise. But buried in that approach is the assumption that the government is a rational and benevolent agency, which, having decided how much it ought to spend, requires scholarly advice on how to do as little damage as possible in the process. Government is treated as a benevolent despot—an agency that merely needs to be told how best to do good. This approach to the economics of government is unfortunately not unique to public finance.

To Buchanan and Brennan, on the other hand, government is an agency with its own objectives, and generalized benevolence is not likely to be one of the more important ones. The scientific question of interest is not "What can government, given the power, do that is good?" but rather, "Given certain powers, how will government choose to use them and what will be the consequences?" In the case of public finance this becomes the question, "How will the range of taxes the government can impose affect both the amount the government chooses to collect and the cost to the rest of us inflicted by the process of collection?"

In answering that question, Buchanan and Brennan adopt the working hypothesis that, while government may be limited by constitutional restrictions (such as limitations on what taxes it can collect), it is not otherwise limited by the political process. They further assume that government's own objective is to maximize its revenue. They then proceed to analyze the behavior of a government that, given the ability to impose certain sorts of taxes, will always choose those tax rates that yield the maximum revenue. The result is to reverse many of the traditional policy recommendations.

Orthodox public finance favors taxes that are simple to collect and have low "excess burden"—taxes that do as little damage as possible beyond the inevitable damage of making someone poorer by the amount of the tax. Excess burden is associated with the ability of the taxpayer to avoid the tax (at some cost); an income tax imposes excess burden because it induces taxpayers to shift time from working for income to leisure or home production—even when working for income is more productive than the alternatives. But to Buchanan and Brennan the ability of taxpayers to avoid a tax is often desirable even if the avoidance is costly: the more readily taxpayers can avoid the tax, the lower the level of tax at which the government discovers that increased rates lead to decreased revenue. Since a revenue-maximizing government will generally want to collect and spend more than the populace wants it to collect and spend (Buchanan and Brennan assume that at least some of the money goes to provide services the taxpayers want—otherwise the ideal constitution would permit no taxes at all), this is an advantage. Hence Buchanan and Brennan prefer, as a rule, just those sorts of taxes the traditional literature opposes.

Orthodox public-finance economists are, in effect, philosophers advising a prince. While they might suggest that he limit his expenditure, they would never advise him to choose taxes designed to limit the revenue he can collect: if he were willing to limit his expenditure he would not need to be forced; if not, he would not take the advice. Buchanan and Brennan are advising potential victims—citizens setting up a government, which, once established, will be out of their hands.

The Power to Tax is a good book for anyone who wishes to understand what government is and how we may best live with it. It has, however, several serious weaknesses. One is a tendency to overuse the technical tools of economics (both authors are economists) in contexts where they add little to the argument. The book is filled with diagrams of the sort used to ornament microeconomics texts; in most cases they tell us only what the verbal argument has already demonstrated—that a particular result (the superiority of one tax to another) is likely to occur but might not. One occasionally has the impression that the authors, having produced a long article's worth of interesting ideas, had to find something to fill up the rest of the book.

The book's most serious weakness, however, is not what is in it but what is not. If at the constitutional level we can impose binding restraints on government, why should we restrict tax collections in a clumsy and expensive way by restricting the government to certain limited taxes instead of restricting taxes and expenditure directly? The authors raise this question but do not resolve it. Until they do—until they offer some serious explanation of what sorts of restrictions can or cannot be enforced upon government—they are open to the charge of having done an ingenious job of explaining the wrong way to do the right thing.

David Friedman is a visiting professor of economics at the University of California, Los Angeles.