Thayer Watkins

Public Finance

The Size of Government

The United States' economic system is a variety of corporatism. Every major economy is now of this form. The world "corporatism" has nothing to do with "corporations," instead it comes from the Latin word corpus, meaning body. A corporatist economy is one in which the major share of economic production is handled through the private sector but the government has the power to regulate and direct the private economy. In corporatist economies there is usually a significant segment of economic production directly controlled by the government. In the U.S. this segment has included the postal service, public schools, dams and waterways, highways, trash collection, research, police and fire departments, public housing and medical care. In other corporatist economies the public sector has been much more extensive, including telephone service, television and radio, airlines and so forth. In contrast, the socialist economies of the Stalin and Mao attempted to have virtually all production carried out in the public sector and under the direct control of the goverment.

Often the corporate economic system is called a "mixed economy," but that term disguises the fact that there is logic and ideology of corporatism which is distinct and separate from socialism and the pure market economy.

The U.S. is unique in that it has three distinct levels of government. There is the Federal Government on the national level, each state is an independent unit of goverment and there are about ninety thousand local governments. This includes about three thousand county goverments, untold city governments and tens of thousands of special districts such as school districts, water districts and fire districts. These are not subject to the control of the governments of the cities and counties within which they are located but are independent units. In contrast, the local city government in the United Kingdom are merely branches of the national government. In the U.S. there are also some quasi-public corporation such as the Federal Reserve Banks and The Tennessee Valley Authority which are, in fact, government units with an important degree of autonomy.

In 1993 the government employed 16 per cent of the nonfarm, civilian labor force and about 20 per cent of the total labor force. This gives a rough idea of the direct involvement of government in the economy. But the government's role goes far beyond the employment of about 23 million workers in 1993. In 1996 the total spending by government in the U.S. amounted to about 32 per cent of the GDP. This was an enormous amount, about equal to the entire GDP of Germany.

Although the share of U.S. production controlled by government spending is large, about one-third, it is relative small compared with other major industrial countries. In France and Italy in 1994 government spending was about 55 per cent of GDP. The government share in Japan was relatively low, about 35 per cent, compared with the seven nation average of 46 per cent.

Government spending as a proportion of GDP increased significantly during the 20th century, rising from a level of about 7 percent in 1902 to a high of 43 per cent in 1945 during World War II. This share dropped back under 25 per cent during the period from 1955 to 1965. It then rose to the one third level and has remained at that level ever since. The growth in the spending of the govenment in the twentieth century was largely the growth of the Federal Government spending. State government spending also increased but its absolute size is a small fraction of federal spending. Spending by local governments, as a share of GDP, stayed roughly constant over the century.

The receipts and expenditures of the three levels of government can be summarized roughly as follows:

Level of GovernmentSource of RevenueMajor Areas of
FederalPersonal Income Tax
Payroll Tax for Social Security
Corporate Profit Tax
National Defense
Health Care
State GovernmentsPersonal Income Tax
Sales Tax
Health Care
Local GovernmentsProperty Tax
Sales Tax
Public Safety

The Changing Structure of the Federal Tax Structure

There has been a political emphasis placed upon the personal income tax. Cuts in the income tax rates or at least not raising them have been a key item in evaluating the fiscal policy of presidential candidates. Because of this emphasis the share of Federal revenue coming from personal income taxes has not stabilized at about 45 per cent. In constrast the share of funds coming from the payroll tax has rose from about 10 per cent in 1955 to 35 per cent in 1995. In 1993 the Federal got 44.7% of its revenue from the personal income tax and 37.5% from the payroll tax. In contrast the share of Federal revenue coming from the Corporate profit tax declined from 26.5% in 1955 to 10.3% in 1996.

Other countries rely heavily upon a Value-Added Tax but this has never been part of the American tax structure. Other countries often have taxes on consumption to raise revenue and encourage saving by the tax payers. This has not been part of the tax system in the U.S.

Accounting in the Public Sector

The U.S. Constitution provides for the publication of a "regular Statement and Account of the Receipts and Expenditures." This prescription is carried out by the annual publication of the Budget of the United States Government by the Office of Management and Budget. This document, which is roughly the size of the telephone directory of a major city, does not tell the whole story. Although this budget is called the unified budget it There are certain receipts and expenditures of the Federal government which are not included in the budget totals. For example, most of the transaction connected with Social Security are not included in the total and are said to be "off budget" items. In fiscal 1995 there were $335 billion in receipts and $290 billion in expenditures that were designated as off budget items. Another example is the billions of dollars the Federal Government paid to bailout the savings and loan industry. A major part of these expenditures were designated as off budget item. There the official deficit or surplus of the Federal Government is not necessarily an accurate or meaningful figure.

Federal loan and loan guarantee programs were for a long time not reported as a component of the budget. For example, in 1979 the Chrysler Corporation was in financial difficulty and could not borrow except at high interest rates because the lenders feared bankruptcy for the firm. Chrysler sought and received a Federal loan gurantee which meant that if Chrysler did not repay the loans the Federal Government would repay them. This enabled Chrysler to borrow at much lower interest rates. As it happened Chrysler did repay the $1.5 billion in loans and the Federal Government did spend anything. But if Chrysler had had to get a private organization to accept the risk of nonpayment it would have had to pay many millions of dollars for that insurance. Thus the Federal Government gave something to Chrysler, a loan guarantee, which was worth many millions to Chrysler. The 1992 Federal Credit and Reform Act now requires the cost of Federal credit programs to be estimated and budgeted. In 1994 the costs of such direct loan programs at subsidized interest and loan guarantee programs was estimated to be between $123 billion and $223 billion.

Another type of cost of Federal Government activity that does not show up in the Budget is that of regulation. The annual cost of such regulation in 1994 was estimated to be $392 billion, but only the operating costs of $12 billion for the agencies with their 113,000 employees was shown on the Budget. Major sources of these regulatory costs are OSHA (Occupational Safety and Health Administration) and the ADA (Americans with Disabilities Act).

The Role of Government

Economic theory has been developed over the past 225 years for analyzing questions of public policy. This economic theory supports the following fundamental principle:

The operation of a competitive market yields an efficient allocation of available economic resources. Government intervention in such a market process does not produce, on balance, an improvement in the economy but instead on balance makes the economy worse off.

There are key words in the statement of this principle. Competitive usually means that there are many independent buyers and sellers, so many that no one of them can influence the market price. Efficient has the meaning given to it by the Italian economist-sociologist Wilfredo Pareto (1848-1923) who said that a situation is efficient (this is also known as Pareto-optimal) if there is no way to improve the well-being of some people without making others worse off.

The usual statement of the Principle assumes that there are no externalities. An externality is something like air or water pollution. There is result known as the Coase Theorem which says that even where externalities are involved there may be no need for governmental intervention. The Coase Theorem will be treated later.

Welfare Analysis

Economists have developed various techniques of analysis to examine questions of public policy. Some are quite complex so we will start with the simplest possible version. The first is the notion of the marginal benefit of another unit of consumption of a good. This is also called the marginal willingness to pay MWTP. Suppose a consumer is consuming 52 quarts of milk per year. Consider now a situation in which that consumer is offered one more quart of milk at various costs. If the additional benefit the consumer would get from that quart of milk is greater than the additional benefit he or she would get from the other goods and services that could be purchased with that amount of money then the consumer will buy the quart because this will make the consumer better off. On the other hand, if the benefit of the additional quart is less that the benefit that could be derived from spending the money elsewhere then the quart will not be purchased because doing so would make the consumer worse off. Between the prices at which the consumer wants to buy another quart of milk and the prices at which the consumer does not want to buy one is a price at which the consumer is indifferent about whether he or she buys the additional quart or not. He or she is indifferent because the additional benefit of the additional quart of milk is exactly balanced by the benefit from other things the money would buy that the milk costs. This is the definition of the marginal benefit of that additional quart of milk. The other name for this concept is the marginal willingness to pay.

The marginal benefit from an additional unit of a good would depend upon how much is already being consumed. In principle, we can construct for any individual a graph of the marginal benefit or MWTP as a function of the amount consumed, such as shown below.

Each person's marginal benefit curve could be different.

An individual's demand curve is the quantity which would be consumed for various price level for the good, assuming all other influences are held constant. For any price the consumer increases his or her consumption up to the point where marginal benefit equals the price, as shown in the following graph.

If one were to plot the quantity of milk consumed by an individual versus the price; i.e., the red dots in the above graph, the result would be the reconstruction of the marginal benefit curve for that individual. In other words, the demand curve and the marginal benefit curve (or marginal willingness to pay curve) are exactly the same thing for each individual. The market demand curve for milk is obtained by adding together the demand curves for milk of all of the individuals in the market. Thus the market demand curve is the marginal benefit curve for all consumers in the market. This is an important result. It means that the concept of marginal benefit which is an individualistic construct can be identified with the market demand curve which is something which, in principle, can be measured objectively or determined statistically from the behavior of consumers in a market.

The increase in benefit that a consumer experiences when consumption increases from Q1 to Q2 is the area under the marginal benefit curve from Q1 to Q2, as shown in the following diagram shows.

For small increases this area is just the increase in consumption times the market price as shown below:

The area of the pink rectangle behind the red trapezoid has just about the same area as the red trapezoid. The area of the pink rectangle is just the market price P1 times the increase in consumption.

The previous result, that the increase in benefits is the area under the demand curve, may be used to determine the total benefit of consumption of an amount of milk of Q, as shown below. The area shown in red is the benefit of increasing the consumption of milk from 0 to Q.

The above benefit does not take into account the money paid for the Q units of milk. The difference betweeen the total benefit gained from consumption of Q and the money expended for it is called Consumers' Surplus, as is shown in the diagram below. Suppose the quantity of milk consumed is in response to a market price of milk of P. The money expended is equal to the price times the quantity which is the area shown in light blue.

If a project or policy results in an increase in consumption from Q1 to Q2 and fall in the price from P1 to P2 then there is an increase in consumers' surplus which shows the net benefit of the project or policy that brought about the change, as shown below.

The Impact of Taxes on Price

Private Goods and Public Goods

The crucial question is what goods and services should the government provide and what should it not attempt to provide. Goods and services can be categorized as Private Goods and Public Goods. Most goods are private goods, meaning that for a given level of production of that good the more one person consumes the less is available for the others to consume. Thus each individual can choose their level of consumption of that good. Public goods, on the other hand, are one such that one person's consumption does not diminish any other person's consumption. A public good is generally something such that if anyone get it everyone gets it. The usual example given of a public good is national defense against enemy invasion. If one person is protected in a community against outside attack then everyone in that community is protected. Knowledge is also sort of a public good in that if someone has a bit of knowledge that does not diminish the availability of that knowledge to others.

Within the category of public goods some distinctions can be made which make the concept more precise. A good is a rival good if one person's consumption affects the availability of the good for consumption. A nonrival good is one in which costs of providing the good to others once some have it is zero. Another distinction is whether a good is excludable or nonexcludable. For a nonexcludable good if one person gets it everyone gets it usually in equal measure. An excludable good is one in which it could be provided for some but others can be excluded. Thus a pure public good would be one that is both nonrival and nonexcludable. In the previous paragraph national defense against enemy invasion was cited as the common example of a public good. But a little thought reveals that while national defense is nonrival it is not completely nonexcludable. Obviously people living outside of the country, citizens as well as noncitzens, do not enjoy the benefits of the national defense of a country. Even within a country it may be possible to provide national defense for some regions without providing it for all regions. For example, it might be possible to defend an enclave in the Rockies and leave the coastal regions and the midwest defenseless. A better example of a pure public good would be a program to detect asteroids that are on a collision path with Earth and provide for their deflection. Since the collision of an asteroid would produce life-destroying temperatures around the world no one could get this protection from asteroids without everyone getting it. And one person's protection would not diminish everyone else's protection.

The refinement of the concept of public good to include nonexcludability as well as nonrivalness is important when considering such things as knowledge or enjoyment of works of artisitic creation. Such goods may be nonrival but they are not nonexcludable. It may be necessary to have the decision about whether or not to produce a nonrival, nonexcludable good (a pure public good) to be made in the public sector but clearly decisions about excludable goods, whether they are rival or nonrival, can be made in the private sector. In the case of excludable goods the question is whether it is more efficient to have the decision made in the private or public sector.

The decision about how much of any good should be produced is a matter of finding what level of production results in the maximum net social benefit, net social benefit being the difference between social benefit and social cost. The optimal level of output is the one at which the marginal social benefit of another unit of output is equal to the marginal social cost of another unit of output. The matter of external benefits or costs will be left until later.

In the absence of external benefits or costs from the consumption of the good the marginal benefit curve is the market demand curve. The market demand curve is obtained by summing up the individual demands. In terms of a graph this is the horizontal sum. Likewise, in the absence of externalities in the production of the good, the marginal cost curve for the product is the market supply curve, which again is the horizontal sum of the individual producers' supply curve. The optimal output in this case is just the output that would be the market output in a competitive market.

The situation for the pure public good is quite different. The marginal social benefit curve can also be called the marginal-willingness-to-pay curve. For the pure public good each consumer is getting the same amount the marginal benefit of another unit of the public good and therefore the marginal social benefit is the sum of the marginal benefits of all the consumers. Thus in this case the marginal social benefit curve is the vertical sum of the marginal benefit curves of all the consumers. It is this curve which needs to be compared with the marginal social cost curve to determine the optimal output of the public good.

(To be continued.)

HOME PAGE OF Thayer Watkins