The Political Economy of Federal Government Growth, by James T. Bennett and Manuel H. Johnson, College Station: Texas A&M University Press, 1980, 151 pp., $12.95/$4.95.
Is the federal government growing in size? The common perception is that it is, by leaps and bounds. Four cabinet-level departments have been created since 1959. New agencies and commissions have been cobbled together to regulate everything from health and safety to environmental quality to wages and prices. Food stamps, student loans, Medicare and Medicaid, and dozens of other social programs have proliferated like rabbits in the last 20 years.
Yet the growth of government turns out to be hard to document. In fact, published statistics suggest that if the federal government is growing at all it's doing so at a snail's pace. The total number of people employed by the government was higher in 1966 than it was in 1978. Government spending as a fraction of our gross national product has shown only a modest increase: from 18.7 percent in 1959 to 21.9 percent in 1978. As far back as 1952, government spending was 20.6 percent of GNP.
The official statistics, then, are in direct opposition to the public perception. Can this paradox be resolved? That is the objective of a new book by Profs. James Bennett and Manuel Johnson, two economists at George Mason University.
The book represents a highly ingenious bit of detective work. Bennett and Johnson have looked beyond the government's official report of its size and found that the publicized statistics bear little relation to reality. Their work represents a landmark effort to gauge the true role of government in our society.
The authors begin by showing that federal employment statistics reflect a great deal of duplicity. For one thing, the official numbers only count full-time employees, excluding anyone who is only working part-time. So each March 31, when the agencies report their employment statistics, they switch thousands of individuals from full-time to part-time status. So pervasive is this practice that these full-time/part-time bureaucrats are known throughout the federal government as "25-and-ones": for 25 of the 26 federal pay periods each year, these workers are classified as full-time, but in the one pay period in which the head count is taken, they become part-time.
The official statistics also count only those who work directly for the government, excluding all who work for outside firms and agencies, only to have some or even all of their salaries paid by the government. In 1979 the Department of Health, Education and Welfare "employed" 144,256 workers. Yet it paid the salaries of almost eight times that many people. The "nonemployees" were people working under government contracts in think tanks, universities, and state and local government.
If the official employment statistics were corrected to eliminate such distortions, they would be much more consistent with the perception people have of the government's actual size. Indirect and part-time employment may bring the number of nonmilitary federal employees up to 10 percent of the civilian labor force, rather than the 2.4 percent that is officially reported.
Government spending statistics also suffer from distortion: they fail to take into account financial commitments whose effects are delayed in time. From 1967 to 1977, the federal government's liabilities (public debt plus interest, checks outstanding, and accounts payable) rose by over 7 percent per year. Obligations such as funds not yet paid and goods not yet delivered grew by 14.5 percent per year. Long-term contracts for goods and services rose 12 percent per year, and contingencies such as insurance commitments, Social Security payments, and other annuity programs climbed 30 percent per year.
Bennett and Johnson show that when these financial commitments are considered, the public's perception of the government is again a lot closer to reality than are the official reports. In 1977, the total of liabilities not included in federal expenditure figures was $7.38 trillion—that's 7.38 million million dollars!
Official employment and spending statistics are misleading for another reason too: they fail to reflect the increasing tendency of the government to force the private sector to perform the government's job. Take the withholding tax, for example. By law, employers are forced to become tax collectors for the government. Yet the cost to employers of collecting withholding taxes is not counted as part of government spending.
Over the last 20 years this may be the most important way in which government has experienced hidden growth. Instead of spending dollars, the government simply requires private individuals to spend them. Instead of employing people to do a job, the government requires businesses to employ them.
The total size of this type of government activity is staggering. The authors estimated that the total cost of government regulations was $66 billion in 1976, including $25 million in paperwork costs alone. Individual businesses pay 95 percent of these regulation costs.
Having documented rapid growth as a very real fact, Bennett and Johnson explain the phenomenon: the self-interest of the bureaucrat, self-interest of the politician, and self-interest of the voter are played off against one another in a complex web of give and take. First, the bureaucrat can only rise in power, prestige, and salary if he supervises more employees. The number of employees he supervises will only increase if he is able to convince legislators to appropriate more money for his department. The legislators will only do that if they think such action will be embraced by their own constituents. Therefore, the bureaucrat has an incentive to apportion benefits to specific groups, preferably those which have strong lobbyists and make up a sizable voting block.
The bureaucrat also has an incentive to promote and prolong a crisis atmosphere (the energy crisis, the health care crisis, the environmental crisis) to ensure that the public is made aware of the agency's importance. He must gain and keep public support for a government program to "do something" immediately. The media often cooperate willingly, because "crises" sell their product, also.
There seems to be a new mood in Washington, this time to "do something" to curtail the growth of the federal government. Yet to carry this mood beyond one budget-cutting session, we must truly understand the nature of this growth. Bennett and Johnson have made an extremely valuable contribution to this enterprise.
John Goodman teaches economics at the University of Dallas and is the author of several books on health policy.
This article originally appeared in print under the headline "How Big Is Big Government?".