Privatization and public employment: an essay on the current status and the stakes.

• Author: Donahue, John D.

Is conventional public service being swept away by a tidal wave of outsourcing? If replacement by private providers is eroding public employment, how (if at all) should the interests of government workers, and government employment as part of America's social landscape, be considered as factors in the debate about privatization? This essay engages these two questions--the first with some specificity, the second far more tentatively.

IS PRIVATIZATION SHRINKING THE PUBLIC WORKFORCE?

The rhetoric of privatization opponents and enthusiasts alike suggests a major siege on public employment as an institution. Gerald W. McEntee and William Lucy of the American Federation of State, County, and Municipal Employees ("AFSCME") warn of a "coordinated campaign to privatize government at every level [that] far exceeds anything we've seen in the past." (1) At the other end of the ideological spectrum, the Reason Public Policy Institute declared triumphantly that "privatization moves ahead in breadth and depth" and is "thriving in the United States." (2) Such organizations issue regular publications packed with examples (which are, respectively, denounced or celebrated) of services formerly provided by public workers being shifted to private delivery.

One might infer, from the volume and tenor of talk about privatization, that American public employment (at least as we know it) is on the verge of withering away. As a preliminary to any inquiry into the effect of outsourcing on public employment, it is useful to get a sense of the current status of the government workforce. The question of whether public jobs are many or few turns out to be far more complex than is commonly supposed. The following section calibrates the scale of public employment from several perspectives, including relative to the past, scaled to the size of the population and workforce, and relative to public spending.

As of 1999--the most recent year for which complete Census Bureau data are available--there were roughly 20 million government workers. (3) Around 2.8 million of these were civilians working for the federal government, of which the largest group (876 thousand) consisted of postal workers, and the second-largest (713 thousand) were civilians involved in national defense and international affairs. (4) The federal government also employed 1.4 million uniformed military personnel. (5) The states, in the aggregate, employed somewhat more people than the federal government: 4.8 million overall, of which 2 million were involved in higher education. (6) The local government workforce exceeded federal and state combined, at 10.7 million. (7) More than half of these (5.7 million) worked in primary and secondary education. (8) The remaining 5 million local government workers were scattered across a score of categories, with no single category except police services claiming more than half a million. (9)

Is 20 million public workers a big or small number? An obvious point of reference is the total count of American workers, which was approximately 135 million at the end of 1999, (10) making public employment 15 percent of the total. But this is only useful for assessing whether privatization is contributing to a downward spiral of government employment by reference to the past. Figure 1 traces the number of government workers (on a full-time equivalent basis) from 1948 to 1998. Public employment roughly tripled over that half-century, but by no means smoothly. There were sharp surges in net hiring (in the early 1950s, late 1960s, and late 1980s) alternated with plateaus of little growth, and the public payroll has been close to steady since around 1990. This simple headcount of government workers, though, tells us little on its own.

[FIGURE 1 OMITTED]

What about the scale of the public workforce relative to America's population? Total government employment climbed from about 6.5 percent of the population in the early 1960s to around 8 percent by the late 1960s, and has stayed fairly close to that level. From 1975 through 1999, public workers never accounted for less than 7.7 percent of the population, and never accounted for more than 8.2 percent, with no strong trend discernible in the quarter century's data. (12)

Total population, however, may not be a particularly relevant denominator. The population figures include many groups (such as young children or undocumented aliens) the members of which are not very likely to be employed by the government or a government contractor. A more germane point of reference is total employment. Figure 2 traces government employment relative to total employment, economy-wide, with both categories measured on a full-time equivalent basis, with part-time workers counted in proportion to their work hours. Two half-time teachers' aides, for example, would count as one education worker on a full-time equivalent basis. As noted earlier, the current share is around 15 percent, and Figure 2 could be interpreted as tracing significant erosion of the governmental workforce. Public employment's share of the total was down by roughly a quarter from the peak of more than 20 percent it reached in the years just before and just after 1970. This is consistent with downward pressure on the public workforce caused by outsourcing and other forms of privatization. But a closer look suggests a more complicated story. The public-sector share of employment is lower than it was in the 1970s, but about the same as it had been in the middle of the last century, suggesting that the early 1970s (not the 1990s) may have been the aberration. More tellingly, perhaps, is that more than half of the decline from the peak occurred prior to the mid-1980s, when the privatization debate first gained much prominence. Unless privatization was a major force before it drew much attention (and far less influential as the debate grew noisier) it seems unlikely that outsourcing explains much of the ebb in the public share of the workforce.

[FIGURE 2 OMITTED]

Separating the governmental workforce into its major components provides some additional perspective, at the expense of what turns out to be inevitable complexity in the story of American public employment. Public workers' total share of employment was about the same at century's end as it had been fifty years earlier. But its composition has changed radically. Figure 3 divides the public workforce into four largely-exhaustive groups: the armed services, other federal workers (including postal workers), state and local workers involved in education, and all other state and local workers. In 1948, these four groups were close to the same size, each claiming between 3 and 4 percent of American employment. Over the next half-century, military employment dwindled (for well-known and perfectly logical reasons). Other federal employment shrank as well, while state and local employment, both education-related and not, surged as a share of the workforce. There is a story to be told here, (14) but it is not--at least not in any obvious way--one of rampant outsourcing. Figure 3 underscores a significant point: privatization decisions at the state and local level will have a far larger impact on the big picture of American public employment than will decisions at the federal level, where the stakes are simply smaller.

[FIGURE 3 OMITTED]

Perhaps the more revealing comparison is not to population, or total employment, but to the scale of the government's mission. The public workforce can be called large or small, growing or shrinking, only by reference to the amount of work the government has to get done. If the headcount shrinks while the mission swells, it is more plausible that the anecdotes of outsourcing do indeed sum to a meaningful privatization trend.

Although government's share of America's gross domestic product ("GDP") has varied rather less than one might infer from strident arguments about the size of government, it has risen and fallen with changes in economic climate and political fashion. The budgets of all levels of government combined climbed from roughly one-quarter of GDP in the early 1960s to something over 30 percent in the mid-1970s, staying at or slightly above 30 percent from 1980 through 1995, then declining somewhat to around 28 percent for the last half of the 1990s. (17) One intuitively appealing measure of the public workforce is the number of government...