Administrative Autonomy Among American State Agencies: An Empirical Analysis of Fragmentation and Functionalism

Abstract

Administrative (or bureaucratic) autonomy has been an issue discussed in many different political and policymaking settings. Since administrative agencies are actively involved in the formation and implementation of public policy the issue of administrative autonomy is an important contemporary issue. We measure the concept of administrative autonomy empirically and systematically among state administrative agencies along two general features or components: (1) fragmentation and (2) functionalism. Each of two features of autonomy is subdivided into institutional and perceptual dimensions. The former is based on organizational and position characteristics. The latter derives from attitudes of individual administrators. Data are derived from responses by over 1000 state agency heads from the 1998 American State Administrators Project (ASAP). Among the prominent findings are: (a) there are positive relationships between the institutional and perceptual dimensions for both fragmentation and functionalism, and (b) variations in institutional and perceptual dimension of administrative autonomy among different types of agencies and different selection methods of the agency heads.

Acknowledgments

We wish to acknowledge support from the Earhart Foundation of Ann Arbor, Michigan and the Odum Institute for Research in Social Science of the University of North Carolina at Chapel Hill.

Notes

^aThis proportion has been fairly constant around 70–75% since 1978. The rapid extension of state agency involvement with federal aid transpired in the 1960s and 1970s. The percentages rose from 34% in 1964 to 54% in 1968, and continued upward to 63% in 1974 and 76% in 1978.

^bThis proportion has also remained fairly constant through the 1980s and 1990s, but it too rose sharply in the 1960s and 1970s.

^cIn comparison to earlier decades, the percentages indicating no skewing effects in 1994 and 1998 (36% and 48%) are well below the proportions in earlier decades. In 1948 the proportion responding that there were no distortion effects from federal aid was 70% (Council of State Governments, 1949). In 1988 the percentage was 57%.

^dThis question was first asked in the 1964 ASAP survey. These survey percentages are as follows: 1964, 49%; 1968, 44%; 1974, 47%; 1978, 48%; 1984, 53%; 1988, 58%; 1994, 54%; 1998, 48%). Perhaps the most noteworthy finding from the above array is the relative stability over four decades. All proportions are slightly above or below 50%.

^eThese questions were first asked in 1948 and have been included in ASAP surveys since 1974. As in the case of attitudes on several other federal aid items, there has been considerable consistency in the responses. The percentages are fairly stable in the sense that they generally cluster in the 60–70% range.

^fWe can use data from prior ASAP surveys starting in the 1970s for an analysis similar to the one reported here.