Comparing Public Bureaucracies: Problems of Theory and Method

5. The Pursuit of Power

The Behavior of Public Officials institution in budgetary politics is its ability to expand in numbers and more particularly its ability to add personnel in upperechelon positions. As I noted above, some scholars have argued that the budget-maximizing bureau chief should prefer adding labor rather than capital. Furthermore, additional personal staff is one of the few real benefits that a bureau chief may be able to appropriate, given rather uniform pay and perquisites in the civil service. Table 19. Federal Civilian Employment, 1950-1984 Year 1950 1955 1960 1965 1970 1975 1980 1981 1982 1983 1984 Number 1,934,040 2,371,462 2,370,826 2,496,064 2,884,307 2,848,014 3,065,672 2,891,844 2,860,754 2,863,071 2,901,137 Percentage of Labor Force 3.1 3.6 3.4 3.4 3.5 3.1 2.9 2.7 2.6 2.6 2.6 Source: u.s. Bureau of the Census (annual-b). As table 19 shows, the total federal civil service has not been expanding rapidly in absolute numbers in the past decade, after significant increases in the 1960s and 1970s. It has certainly not been expanding as a proportion of the total labor force in the United States (see also Peters, 1985b) and has been declining steadily for the past several years-even before the Reagan administration. Again, this trend does not suggest an institution capable of protecting and expanding its own interests. In the face of what is apparently substantial political opposition, it has been losing its position in the economy, both in terms of the salaries being offered and the numbers of people being employed . 137 The Behavior of Public Officials Sheer numbers of civil servants, however, may not be a sufficient measure of the behavior of the budget-maximizing bureau chiefs in enhancing their own positions in public sector employment. One possible measure raised by Hood, Huby, and Dunsire (1984) is the ratio of chiefs to Indians, or the ratio of the number of senior positions relative to lower-level positions in civil service. As table 20 shows, however, the percentages of employees in different grades within the career civil service has remained relatively stable over the fourteen years in question . In particular, the proportion of very top civil servants (G.S. 16 and above) has not increased. If we add executive-level appointments (not in the General Schedule), the proportion of "chiefs" increases slightly, but only slightly, over this period. This relative stability has occurred during a period in which information processing and other technological advances might have been expected to reduce the number of openings for lower-level employees and in which the increasing professional and technical thrust of the federal government might have been expected to require more highly skilled, and highly ranked, personnel. Elsewhere in their discussion of chiefs and Indians, Hood, Huby, and Dunsire stated that the hiring of additional personal secretaries (a very important perquisite in the British civil service ) to assist the "chiefs" indicates bureaucratic power. It is difficult to obtain comparable information for secretarial and clerical positions in the United States, but one surrogate meaTable 20. Federal Civil Service by Grade, 1970 and 1984 (percent) Grade 1970 G.S. 1-6 41.3 G.S.7-10 24.0 G.S. 11-12 21.0 G.S. 13-15 13.0 G.S. 16+ 0.7 1984 36.3 24.1 24.3 14.7 0.6 Source: Office of Personnel Management (biennial). 138 The Behavior of Public Officials sure could be the grading of the total secretarial staff in General Schedule employment. We will assume that secretaries who were given posts as executive or personal secretaries would receive higher rankings than would the remainder of those employed in the same personnel category of "secretary.11 When we examine the civil service gradings across time (Office of Personnel Management, biennial), however, we find that over this decade and a half the average gradings of secretaries have declined slightly, and the number of very highly graded secretaries (at G.S. 12 and above) has been reduced to virtually zero. If this surrogate measure has any validity at all, then there is additional evidence that either bureaucratic entrepreneurs do not want to maximize the size of their budgets and personal staffs or, if they want to do so, they are not terribly successful. Another...