|Cato Policy Analysis No. 250||February 28, 1996|
by Richard K. Vedder
Richard K. Vedder is a professor of economics at Ohio University and John M. Olin Visiting Professor of Labor Economics and Public Policy at the Center for the Study of American Business at Washington University in St. Louis.
As ancient Rome lived in splendor off the tribute raised in the provinces, so in modern America the political capitals are prospering economically at the expense of the rest of the nation. The productive, private citizens in outlying regions of our nation and states are financially burdened to pay for a parasite public economy of lawmakers, lobbyists, contractors, and bureaucrats in the political centers. Several statistics support those claims:
Average annual pay of workers in the District of Columbia exceeds the national average by 48 percent. Nation-wide, income per person in counties with state capitals tends to be nearly 10 percent higher than in other regions.
The income differential between Washington and the rest of the nation rose from 25.9 percent in 1980 to 32.1 percent in 1990. The poorest states have capitals with per capita income levels about 17 percent above state averages; the richest states show virtually no income differential, which suggests that government income redistribution may contribute to poverty rather than enhance wealth.
Although unemployment in the Washington, D.C., metropolitan area has been increasing, it remains almost 30 percent below the national average. Unemployment rates in counties containing state capitals average about 20 percent lower than in other counties.
That evidence is consistent with the hypothesis that those who make up the "parasite economy" have been successful at improving their economic well-being at the expense of those working in the productive private economy.
At the time of the birth of Christ, Rome was a prosper- ous, thriving place, with a standard of living that was probably significantly higher than that of any other city in the western world, if not on the entire planet. Ancient Rome did not become relatively rich through capital formtion and technological advance, the usual causes of economic progress and prosperity. It thrived as a government center living largely off the tribute (taxes) of the provinces and conquered lands. An ever-growing share of the income that was earned through the work of people living in the provinces was transferred each year to those living in the capital. Rome did not produce wealth; its prosperity came at the expense of individuals living in Spain, France, northern Africa, and elsewhere. The parasitic nature of Rome's opulent existence contributed to its eventual downfall.
Is a similar redistribution of income and wealth taking place in contemporary America? This study looks at data on incomes, job opportunities, and population movements for recent years and concludes that, by and large, American political centers do benefit at the expense of surrounding areas. Resources are being extracted from the general taxpaying public to feed the bureaucrats and government officials who tend, other things being equal, to concentrate themselves in state or national capitals. Moreover, the tendency to enrich the political center at the expense of the rest of the polity has, on balance, grown over time. America's government centers have become parasite economies.
The Growth of the Parasite Economy
It is useful to describe the United States as having a dual economy. On the one hand, there is the private-sector economy, which produces goods and services in response to the wants of consumers and businesses. On the other hand, there is the government sector, which largely redistributes income. Typically, government is said to redistribute income from rich to poor or from young workers to older retirees, but increasingly there is evidence that individuals use government to redistribute income from the general taxpaying public to themselves. Author Jonathan Rauch has termed Washington, D.C., a "parasite economy." Of course, not all governmental activity is of a parasitic nature. The protection of rights--through police, courts, and national defense--is an essential function of government. Many traditional functions of government such
as providing roads and schools do yield benefits to taxpayers, (though some of those services might be produced more efficiently in the private sector). However, as government as grown larger, the parasite economy--that is, the people who derive their livelihood from government spending and taxes--has flourished. The amount of government spending defines the potential "pie" from which the parasites ("rent seekers" to economists) try to siphon off income for themselves.
Regulations that transfer wealth are also part of parasite economies. Table 1 shows the maximum possible size of the parasite economy over the past 40 years. Total government expenditures (excluding intergovernmental transfers) have risen from less than one-fourth of total output to two-fifths. The growth in spending has been relent- less, with government's share of total output occasionally flattening but never significantly declining. Increases in governments' (and parasites') share of total output are observed during both Republican and Democratic administrations in Washington. Clearly, the potential income available for siphoning into the pockets of the parasites has grown over time. In 1994 the figure for total government spending--federal, state, and local--was roughly $2.7 trillion.
The Relative Economic Status of Washington, D.C.
How much of that spending was of a parasitic nature-- how much of those funds enriched people involved in govern- ment, rather than citizens who are supposed to be helped? One way to answer that question is to examine how rapidly government centers are growing relative to nongovernment centers.
Although the federal government has facilities throughout the nation and beyond, the largest single concentration of its resources is in and around the nation's capital. With the growth of government, many of the most important federal agencies have moved from the District of Columbia to nearby Virginia or Maryland, and a larger number of federal workers live in those suburban areas than in the capital itself. Accordingly, a good case can be made that the appropriate definition of the capital is the Washington metropolitan area, including the Virginia and Maryland suburbs. At the same time, however, the District of Columbia itself is the most purely governmental part of the metropolitan area, so the most relevant comparison is the district.
As Figure 1 indicates, income per capita in the District of Columbia itself in 1993 was $29,836, some 43.6 percent higher than the average for the rest of the United States, $20,761. Despite its lack of any significant manufacturing or services companies, per capita income in the District of Columbia exceeded that of every single state in the Union. Income per capita in the capital was more than 26 percent higher than in the Baltimore metropolitan area, which is near Washington. That is despite the fact that Washington contains deep pockets of poverty among those not associated with government.
Moreover, Washington's income disparity, relative to the rest of the nation, has risen dramatically in recent years as the parasitic forces have extended their grasp over the nation's income, as Figure 2 indicates. After rising steadily in the 1980s (from 25.9 to 32.1 percent) the differential between the District of Columbia and the rest of the nation has soared (from 32.1 to 43.6 percent) in the Bush-Clinton era. That reflects the fact that in absolute dollars income per capita has risen more than twice as much since 1990 in the District of Columbia as in the rest of the nation.
Data for the Washington metropolitan statistical area (including parts of Maryland, Virginia, and even a corner of West Virginia) show that per capita income in 1992 was $26,817, 26 percent above the U.S. metropolitan area average of $21,247, and that Washington ranked 11th of 310 metropolitan areas, topped only by parts of the New York City metropolitan area, some wealthy Florida retirement communities (for example, Palm Beach), and San Francisco and Silicon Valley.
Clearly, many people in Washington, D.C., are prospering as a result of government. The statistical evidence supports the view that a geographic redistribution of income occurred between the rest of America and the nation's capital, since the capital is largely sustained economically through governmental expenditures in large part financed by taxes levied on individuals living in the outside areas. Relatively few people in the Washington area directly produce goods and services--many are lawyers, legislators, lobbyists, and bureaucrats who live off the tax dollars that flow to Washington.
There are other clear signs that the parasite economy in Washington is growing. The number of lobbyists has grown far out of proportion to the growth in the American labor force, as Figure 3 shows. Between 1976 and the early 1990s, the number of American workers serving as registered U.S. Senate lobbyists more than doubled. The "income creators" have grown numerically less rapidly than the "income redistributors" (parasites), thereby draining some vitality and productivity from the national economy.
In theory, the observed income differential for Washington might be explained by high levels of property income among Washingtonians or by greater levels of labor force participation. Data on average annual pay of workers, however, suggest that such is not the case. Comparing data for the District of Columbia, the metropolitan Washington area, and the United States for 1993 shows that the nation's capital has relatively high pay levels for its workers (Figure 4). The District of Columbia had higher average annual pay in 1993 than any one of the 50 states, more than $6,000 above highest ranked, Connecticut. Pay in the district exceeded the national average by more than $12,800, or more than 48 percent. Among the 310 metropolitan areas, the Washington area ranked 10th, exceeding the average of metropolitan areas by 20 percent.
Moreover, the disparity between Washington and the nation has actually been growing in recent years, when economic theory suggests that the equalizing qualities of the market should have actually narrowed the wage differential. As Figure 5 shows, real annual pay growth from 1984 to 1993 in the District of Columbia was 12.2 percent, nearly four times the growth (3.3 percent) for the nation as a whole. The capital-area workforce has prospered relative to ordinary working people living outside Washington.
The relatively high pay level in the nation's capital is not surprising considering the recent findings of Wendell Cox and Samuel Brunelli that suggest that civilian employees of the federal government earn 26 percent more than their private-sector counterparts. Taking account of greater fringe benefits, the private/federal differential is even greater (over 45 percent.) Moreover, the private/public differential grew in the 1980s, as pay for federal employees, especially those in the military, rose several times faster than pay in the private sector.
When an area is perceived as relatively economically desirable, people tend to move to it, increasing population. Similarly, when an area is perceived as having a relative lack of economic opportunities, people tend to move out. The growing economic disadvantage of communism forced the East Germans in 1961 to build the Berlin Wall to prevent massive emigration. The growing economic advantage of life in America led this nation in the 1920s to restrict immigration.
Unemployment and Working Conditions
Economists have developed the concept of compensating differentials, which suggests that sometimes wage or income differentials are explainable in terms of undesirable aspects of the employment experience. Factory workers typically are paid higher wages than are workers in service industries, in part because of somewhat greater risk of accidents, and in part because there is a greater probability of layoff. In other words, factory workers trade off some job security for additional income.
There is no evidence that higher pay in Washington is a result of jobs' being otherwise undesirable. The Cox and Brunelli study on employee compensation suggests that in many respects federal employees have greater nonwage perks, including better fringe benefits and fewer work hours. Area-specific data on workplace safety are not readily available, but some national data suggest that it is unlikely that employees in the Washington area face greater risks. Data on employment accidents suggest that disabling accidents are not unusually high among government workers, despite the current slang term "going postal" used to refer to a colleague's becoming hysterical or violent. The accident rate is even lower in nongovernmental service occupations associated with the parasite economy, such as lobbying. Pay is not high in the Washington area because the work is unusually dangerous.
The evidence is clear that employment security is much greater in Washington than in the average American locale. Table 2 lists the Washington, D.C., metropolitan area unemployment rate for six dates, along with that for the nation as a whole. In every year the unemployment rate in Washington was significantly below that for the entire nation. Averaging the six years, the Washington unemployment rate of 3.83 percent was fully 40percent less than the national average of 6.42 percent.
Moreover, the geographic area surrounding Washington is not inherently a low-unemployment area. Table 2 includes the statistics for the adjacent Baltimore area. In every single year, unemployment in Washington was materially lower than in Baltimore.
Not only is unemployment in general much lower in the Washington metropolitan area; the variations in unemployment tend to be less as well. Thus residents of the nation's capital and its suburbs do not face the worries associated with economic fluctuations that concern the bulk of the American population. There appears to be much truth to the old quip that Washington is recession proof.
The overall desirability of federal employment is revealed in statistics on "quit rates," the number of emloyees who quit their jobs in a given month per 100 employ-ees. Dissatisfied, underpaid workers tend to quit work in relatively high numbers, while satisfied, relatively high paid workers tend to quit less often. The quit rates for federal employees are dramatically lower than for private-sector workers. Indeed, the recent quit rates in federal employment tend to be more than 20 percent lower than were quit rates in private-sector manufacturing during the midst of the Great Depression, when workers were desperately seeking any kind of job available.
In short, in our nation's capital income and working conditions are markedly higher and better than the national average. Washington today is one of the wealthiest areas in the country; it has attracted inflows of population to a much greater extent than is typical of other areas, and it offers its residents a very comfortable life style with high job security and low unemployment. Like ancient Rome, our nation's capital has prospered relative to the provinces surrounding it. The nearly 99 percent of the American people living outside Washington have paid increasingly high taxes (or interest rates or prices, where nontax financing has been used) to pay for the affluence of a privileged class in the capital.
The State Capitals
Washington, D.C., is not America's only parasite economy. Increasingly, state capitals have been siphoning off wealth from the residents within their states' borders.
The Growth of State Governments
Whereas Table 1 suggested that state and local governments grew rapidly over the past 40 years, the relevant consideration in terms of the prosperity of state capitals is spending by state governments alone. In particular, the key statistic is state direct general spending other than grants and transfers to local governments, since those intergovernmental transfers presumably benefit outlying regions of each state. Table 3 shows the growth in that restricted category of state spending, also excluding spending for such business operations as liquor stores, utilities, and insurance funds.
In real terms (using the gross domestic product price deflator to capture changing prices), state direct general expenditures (other than intergovernmental transfers) quintupled from 1950 to 1990. As a proportion of total GDP, they rose from less than 3.8 percent to over 6 percent. States siphoned an increasing proportion of the incomes of their citizenry to finance expenditures that potentially enriched the capital area relative to the rest of the state.
To examine the degree to which capitals have exploited surrounding areas, income and population data for the 1980s were gathered for each county containing a state capital, as well as each state as a whole. Subtracting the capital county statistics from the state aggregates yields information on the state outside the capital county. The county was used as the frame of geographic reference for several reasons. The use of municipal boundaries seemed too restrictive, inasmuch as a large proportion of people employed in many capitals lives in surrounding suburbs. On the other hand, there were problems with a metropolitan area designation: some state capitals are not in metropolitan areas; some metropolitan areas extend into other states.
Figure 7 shows that the unweighted mean per capita personal income in 1989 for the counties with state capitals was $17,866, $1,468 higher than the mean per capita income of $16,398 in noncapital counties. While that is not as striking as the national data, typically the capital county had about 10 percent more income than the surrounding area. In 35 of the 50 states, income was higher in the capital county.
To be sure, as Table 4 shows, the situation in individual states varied widely. If we look for wide income differentials, we find 15 states where the capital county had income levels at least 20 percent higher than the other counties: Alaska, Arizona, Arkansas, Georgia, Hawaii, Idaho, Iowa, Louisiana, Mississippi, New Mexico, North Carolina, Oklahoma, Tennessee, Virginia, and West Virginia. By contrast, only one state, Delaware, had per capita income in the capital county more than 20 percent below the average of the other counties.
|Per Capita Personal Income for Capital County and rest of State, 1989|
|State||Capital County||Rest of State||Capital County as Percentage of Rest of State|
Source: Author's calculations based on data from U.S. Department of Commerce, Bureau of Economic Analysis.
One characteristic of states where the capital counties had much higher incomes than the outlying counties is that they tended to be relatively poor. Twelve of the 15 states mentioned above had per capita incomes below the national average. That suggests that states where income was redistributed the most to the capital area were states that were comparatively poor.
Accordingly, the states were divided into two categories, "rich" and "poor." The rich states were defined as those with the 25 highest per capita incomes, and the poor states were defined as those with the 25 lowest per capita incomes. The capital/noncapital income differentials of the two groups were then compared.
The results are startling. The typical (median) rich state had a capital county with an income level that differed hardly at all (0.4 percent higher) from that of the other counties. By contrast, in the poor states, the capital counties typically (at the median) had markedly higher (more than 17 percent, or $2,000 per person) income levels than did the surrounding areas. Indeed, in almost half of the cases, the capital/other area income differential exceeded 20 percent, or about $3,000 per person. As a generalization, then, it seems that where capital cities are relatively opulent compared with nearby areas, those surrounding areas typically are poor relative to the rest of the nation.
That suggests that the very act of redistributing income from the general public to public employees living in capital counties may have an impoverishing impact on the population of a state. That is consistent with a view that taxing relatively highly productive individuals in the private sector in order to subsidize less productive public-sector workers is debilitating to economic development.
An important question is, are the poor states (with relatively rich capitals) also states with substantial governmental tax burdens? The answer is yes. In the poor states, the median level of taxes per $1,000 was $206.93, more than 11 percent higher than the median for rich states ($185.51). Poor states tended to burden their citizens more with taxes (relative to the ability to pay) and tended to have big income disparities between the capital and the surrounding areas. Rich states, by contrast, tended to have relatively low tax burdens and little economic difference between the capital and surrounding areas.
That brings us back to ancient Rome. It appears that in relatively poorer, less developed states, income does flow from the "provinces" to the capital, and that redistri- bution of income is viewed as a path to wealth. What we have in those states is a kind of Robin Hood-in-reverse syn- drome. State governments redistribute income from the less affluent areas outside government centers to the more affluent state capitals.
There are two ways to obtain wealth or income: create it or take it (legally or illegally) from someone else. States that have succeeded the most economically (as measured by per capita income) seem to have less outward evidence of income being taken, at least with respect to the measure used here. Those states that have succeeded the least, however, seem to tax their populations more and have a lot of income redistribution activity. It is no wonder that in Mississippi, the nation's poorest state, the capital county had an income level more than one-third higher than that of the remainder of the state.
By contrast, in those states where the spirit of enterprise seems to be flourishing to a relatively high degree, the state capital is not a place of unusual opulence. Indeed, all four examples of state capital counties with relatively low incomes (more than 10 percent below the noncapital average) were found in relatively high-income states.
It is possible that the observed phenomenon of capital county prosperity developed decades ago, and that in recent years the income disparity has actually been disappearing. If that is so, the argument that bureaucratic parasites are redistributing income loses some of its persuasiveness: the ability to redistribute income would be eroding over time.
One way to test that hypothesis is to compare the growth rate of capital counties with the growth rate of noncapital counties. The evidence shows that income was rising faster in the capital counties in the 1980s, and that the income differentials were widening. The mean growth of real income per capita in capital counties was 28.2 percent, well above the 23.4 percent figure for noncapital counties. In 28 of the 50 states (including 11 of the 13 largest states) income grew faster in the capital counties.
Economic theory would predict that market forces would narrow income differentials over time, and indeed a signifiant narrowing of differentials has occurred nationally over the generations as labor has moved to high-income areas and capital has moved to low-income areas. The tendency for capital county income per capita to rise more than noncapital county income reinforces the view that nonmarket factors (successful redistribution of income by the political process to the public sector) have played not only a historical but a continuing role.
The general tendency for capital cities to prosper relative to outlying areas (especially in poorer states) is confirmed by other evidence. For example, on average, population growth was about 20 percent greater in counties with state capitals than in other counties (Figure 8). After accounting for the natural growth of population because of birth rates exceeding death rates, it is likely that the migration of population from other places in the United States and from overseas to the capital cities was almost twice as great as it was to other places in the United States. People voted with their feet and moved to the relatively high-income areas surrounding the centers of state political power.
Unemployment rates tended to be persistently lower in the capital counties than elsewhere. Data on unemployment were taken from the 1970 and 1980 censuses, as well as Bureau of Labor Statistics annual estimates for three additional years: 1982, 1986, and 1991. In each of those five years, unemployment rates were significantly lower in the capital counties (5.6 percent) than in the rest of the nation (7.1 percent)--a 26 percent difference. Not only do incomes tend to be higher in capital counties, but the probability of income disruption from unemployment also appears to be significantly lower.
Detailed analysis of the data by state for each date reveals the persistence of lower unemployment in capital counties. The unemployment rate was lower in capital counties in at least 40 states (80 percent of the cases) for every date examined. Indeed, for 70 percent of the states, the unemployment rate was lower in the capital county in every single year. There was no case, by contrast, in which the capital county unemployment rate was above the average unemployment rate for the rest of the state in every single year. As has Washington, D.C., state capitals seem to have become insulated from economic fluctuations; they, too, are now largely recession proof.
Why Are the Capitals Richer?
Government expenditures are often justified on the basis of helping less fortunate citizens--the poor, the disabled, children, the uninsured. But this study shows that an increasingly large component of government expenditures is captured by the parasitic economy of capital cities. The winners tend to be lawyers, lobbyists, legislators, government contractors, bureaucrats, and others who are more affluent than the average taxpayer who bears the financial burden of sustaining the parasite economy in Washington and the 50 state capitals. Public-choice economics argues that individuals attempting to increase their satisfaction in life will try to extract income from government. They are rent seekers who attempt to get a payment without providing any service in return. Special-interest groups do that by obtaining subsidy payments of various kinds, having taxes or burdens imposed on competitors (such as tariffs on international competitors), and the like.
One formidable special-interest group is the unionized governmental workforce. Government workers attempt to use the political process to increase their compensation levels above what purely market considerations would dictate. Numerous studies suggest that public employees have been, on the whole, rather successful. Continuing studies by Douglas Adie showed that postal workers were paid considerably more than comparable private-sector workers, and that low turnover rates among postal employees demonstrated that compensation levels were too high for those governmental employees.
The pay of public school teachers is dramatically higher than the average compensation of private school teachers. In the 1990-91 school year, the average earned income of elementary public school teachers was $31,868, more than 67 percent above that of their private school counterparts ($19,050). It seems extremely unlikely that the pay differential is justified by considerations of productivity.
The most comprehensive study of private/public pay differentials is that by Cox and Brunelli for the American Legislative Exchange Council. As Figure 9 shows, inflation-adjusted compensation of government employees--federal, state, and local--grew substantially faster than that of private-sector employees from 1980 to 1991. In every state in 1991, Cox and Brunelli estimate that average hourly compensation was higher for state employees than for private-sector employees.
Defenders of the differentials might claim that publicsector employees are more highly educated and trained, on average, and accordingly should be paid more. School teachers make up a large proportion of government employees, for example. However, a large hourly compensation differential holds even when instructional employees are excluded. Moreover, the dramatic widening of the public/private pay differential during the 1980s is difficult to explain. Such a widening conceivably could be justified by rising productivity of government workers relative to private-sector workers; evidence that the pay increase was "catch-up" pay and that in 1980 government workers earned below-market rates of compensation; a marked change in government jobs--toward higher paid, professional positions.
In reality, however, there is no evidence that any of those things happened. Data on governmental employee productivity are very limited, in part because public "output" is not easily measured since services are not sold in competitive markets. At least two indicators, however, suggest that public-sector productivity has not risen relative to that of the private sector. Indeed, the reverse may be the case.
First, the federal government attempts to measure productivity for some classes of federal employees. Productivity growth for all measured federal employees from 1970 to 1990 is not appreciably different from the estimated productivity growth for the nonfarm business sector. However, those data are confined to limited aspects of federal government employment.
A second hint that governmental productivity over the long run may be rising less than that of the private sector is provided by the government purchases component of the GDP price deflator. Between 1960 and 1993, the government purchases price index rose 5.86 times, compared with 4.78 times for the total GDP price deflator. The cost of government services was rising relative to prices generally, which reflects higher relative compensation increases in the public sector, slower productivity growth in that sector, or a combination of those two factors.
A look at the literature does not suggest that there was a serious labor shortage in the public sector around 1980 that would justify rapid increases in public-sector pay in the ensuing decade. Similarly, there is no evidence that structural shifts in the public-sector labor force were more profound than in the private sector.
We are left, then, with the public-choice explanation that rent-seeking public employees were extremely successful in using the political process to improve their relative economic status in the 1980s. During that decade nominal public expenditures more than doubled; they rose substantially in real terms as well. Total federal, state, and local expenditures for purchases of goods and services, for example, went from $507.1 billion in 1980 to $1,042.9 billion in 1990, an increase of over 30 percent in real terms. Public employees were apparently quite successful in capturing a significant proportion of the increase in real expenditures for themselves in the form of a higher standard of living.
The comparative affluence of residents of capital cities takes the form of higher incomes, faster income growth, and more job security. Part of that affluence results from successful bureaucratic efforts to redistribute income from the general taxpaying public, largely living in outlying areas, to the political center. Those efforts have apparently been increasingly successful, since the differential between capital cities and the outlying regions has been growing.
Redistributionist activities exact a high price on the economies in which they occur, however. It is not a coincidence that a majority of the state capitals that have gained the most in relative affluence are in the poorest states. The taxes that finance affluent capitals crowd out privatesector activity that is relatively more efficient and productive than governmental activity. The affluence of government centers is making the rest of America poorer. "Capital crimes" are occurring with ever greater frequency and are gradually sapping the nation of needed economic vitality.
 The definitive account of ancient Rome is Edward Gibbon, The Decline and Fall of the Roman Empire (New York: Modern Library, 1932). On the Roman Republic, see Michael I. Rostovtzeff, Rome, trans. J. D. Duff (New York: Oxford University Press, 1960). For a good sense of the role that tribute played in Roman life, read the historical novels of Colleen McCullough, for example, First Man in Rome (New York: Morrow, 1990), or The Grass Crown (New York: Morrow, 1992).
 Jonathan Rauch, "The Parasite Economy," National Jour- nal, April 25, 1992, p. 980. For a revealing discussion of the parasite economy, see Jonathan Rauch, Demosclerosis: The Silent Killer of American Government (New York: Time Books, 1994).
 The size of government is not an all-encompassing measure of the size of the parasite economy. Spending regulations that transfer wealth are also part of it.
 A detailed summary of the growth of government in the United States in the 20th century is provided in Stephen Moore, "Government: America's Number 1 Growth Industry," Institute for Policy Innovation, Lewisville, Texas, 1992.
 A series of supposedly temporary "crises" may have been responsible for sudden surges in government spending in various periods. See Robert Higgs, Crisis and Leviathan (New York: Oxford University Press, 1987).
 Sen. Robert Byrd, as chairman of the Appropriations Committee, moved many government offices and facilities to his home state of West Virginia.
 More than 35 percent of the gross product of the Dis- trict of Columbia is generated from government, whereas in no state in the Union is the proportion as much as 25 percent. See Survey of Current Business, December 1993, for more details.
 Data on personal income by location were obtained from the U.S. Department of Commerce, Bureau of Economic Analysis. In the calculations that follow, the income and population totals for Washington, D.C. (either the district or the broader metropolitan area), were subtracted from the national totals. That permitted the calculation of personal income per capita for the Washington, D.C., area and for the rest of the United States. The data used are for 1992 and 1993.
 The employment/population ratio in the District of Columbia in 1993 was below the national average; more than two-thirds of the states had higher labor force involvement. See U.S. Department of Commerce, Bureau of the Census, Statistical Abstract of the United States, 1994 (Washington: Government Printing Office, 1994), p. 399.
 U.S. Department of Labor, Press Release 94-516, October 20, 1994.
 In the private sector, capital resources will move to areas where wages are low, while human resources (workers) will move to areas where wages are high. That will tend to equalize the amount of capital available per worker, reducing productivity and wage differentials.
 Wendell Cox and Samuel A. Brunelli, America's Protected Class III (Washington: American Legislative Exchange Council, April 1994), p. 1.
 Ibid., p. 3.
 Ibid. The authors conclude the same thing with respect to state and local government employment. U.S. Department of Commerce, Bureau of the Census, p. 436.
 For example, in September 1990 the quit rate among federal employees was 0.84, compared with 1.10 for manufacturing workers in 1933, the year with the highest unemployment in American history. See Office of Personnel Manage- ment, Employment and Trends as of September 1990, Report, Fiscal Year 1990 (Washington: Government Printing Office, 1992), p. 65; and U.S. Department of Labor, Bureau of Labor Statistics, Employment and Earnings Statistics for the United States, 1909-68 (Washington: Government Printing Office, 1968), p. 53.
 The data used in this section were obtained from U.S. Department of Commerce, Bureau of Economic Analysis, Local Area Personal Income, 1984-89 (Washington: Government Print ing Office, 1991). Earlier editions of the same publication were also used in this study.
 Calculated by the author from data in ibid.
 That is consistent with the findings in Richard K. Vedder, "Economic Impact of Government Spending: A 50-State Analysis," National Center for Policy Analysis, Dallas, Texas, April 1993. Using econometric techniques, I estimated that excessive increases in state and local government employee compensation in the 1980s lowered personal income by over $200 billion by 1990.
 The U.S. population grew roughly 11 percent in the 1980s, with 8 percent attributable to natural population increase and 3 percent to immigration. Assuming that natural population growth (births exceeding deaths) was 8 percent in both capital and noncapital counties, population grew 4.86 percent in capital counties through net migration, compared with only 2.67 percent in noncapital counties.
 The original work using the public-choice approach described here is James M. Buchanan and Gordon Tullock, The Calculus of Consent: Logical Foundations of Constitutional Democracy (Ann Arbor: University of Michigan Press, 1962). For more recent work, see Robert E. McCormick and Robert D. Tollison, eds., Politicians, Legislation and the Economy: An Inquiry into the Interest-Group Theory of Government (Boston: M. Nijhoff, 1981); or James P. Gwartney and Richard E. Wagner, eds., Public Choice and Constitutional Economics (Greenwich, Conn.: Jai Press, 1988).
 Douglas K. Adie, An Evaluation of Postal Service Wage Rates (Washington: American Enterprise Institute, 1977); and Douglas K. Adie, Monopoly Mail: Privatizing the United States Postal Service (New Brunswick, N.J.: Transaction Publishers, 1989).
 National Center for Education Statistics, Digest of Education Statistics: 1993, p. 82.
 Cox and Brunelli, p. 11.
 Data for federal productivity were obtained from the LABSTAT database of the U.S. Department of Labor, Bureau of Labor Statistics, and data on business-sector productivity come from the 1994 Economic Report of the President (Washington: Government Printing Office, 1994), p. 322.
 Ibid., pp. 272-73.
 1992 Economic Report of the President (Washington: Government Printing Office, 1992), pp. 390, 364, 303.
© 1996 The Cato Institute
Please send comments to webmaster