In Defense of Bureaucracy

Abstract

Managerial capacity, meant as available potential for managerial resources to be deployed when needed, can be considered ‘slack’ in a public organization during normal times, but recent developments in the research literature of public administration suggest that such capacity can sometimes contribute to public program performance. Does managerial capacity help to dampen or eliminate the effects of sizeable and negative budget shocks on the outcomes of public organizations? This question is investigated in a set of 1,000 organizations over an eight-year period. For the most part, and largely due to managerial adjustments, budgetary shocks of 10 percent or more have only limited or no negative impacts on performance in the short term. They do, however, cause a drop in performance for certain outcome measures, both immediately and in the following year. Sufficient managerial capacity, however, mitigates these negative performance effects. The findings point toward a key question with which public managers must wrestle: how to balance the costs of slack against the benefits that capacity-as-slack can generate when environmental shocks threaten to disrupt the operation of public programs.

Key words:

• Budget cuts
• education
• environmental shocks
• management capacity
• performance
• slack

Acknowledgement

The article is part of an ongoing research agenda on the role of public management in complex policy settings. That agenda has benefitted from the helpful comments of George Boyne, Stuart Bretschneider, Gene Brewer, Amy Kneedler Donahue, Sergio Fernández, H. George Frederickson, Carolyn Heinrich, Patricia Ingraham, J. Edward Kellough, Laurence E. Lynn, Jr, H. Brinton Milward, Sean Nicholson-Crotty, David Peterson, Hal G. Rainey and Bob Stein on various aspects of this research program. Sergio Fernández and Hal Rainey also offered useful comments on an earlier draft of this article. Needless to say, this article is the responsibility of the authors only.

Notes

1 ‘“The dog did nothing in the night-time.”“That was the curious incident,” remarked Sherlock Holmes’ (Conan Doyle, 1894: 50).

2 By an environmental shock we mean any sort of disruption emanating from outside the administrative system and buffeting the core organization.

3 Capacity, broadly speaking, can be considered to consist of two dimensions: quantity of available managerial resources and also managerial quality. We focus on the former here. In earlier work, Meier and O'Toole (2002) explored the relationship between managerial quality and public organizational performance and found that the two are indeed related: the higher the quality, the stronger the performance. The concept of managerial capacity, essentially a structure variable in organization theory, is explored more thoroughly later in this article.

4 Huge and continuing shocks, whether budgetary or otherwise, would undoubtedly take their toll.

5 Other studies show that capacity interpreted in this fashion has been shown to suppress performance losses from other managerial action (Hicklin et al. 2008) and assist in overcoming non-budgetary shocks to administrative systems (Meier et al. 2010).

6 Doig and Hargrove (1990: 3), quoting March (1981: 28–9), consider the notion that ‘[a]dministrators are vital as a class but not as individuals … What makes an organization function well is the density of administrative competence.’ Doig and Hargrove's own position is partially different from this one, but one of their dimensions of leadership behavior has relevance for the kind of function we are examining: ‘systematically scan[ning] organizational routines, and points of internal and external pressure, in order to identify areas of vulnerability… followed by remedial action’ (emphasis in original).

7 Other measures of managerial capacity might consider education levels, skills sets of managers or even years of experience.

8 We attempted to create a measure that somehow incorporated the skills or quality of the central office staff by adjusting this percentage based on how well paid the individuals were. This ‘quality’ measure had no impact in any of the models presented here.

9 There were no statistically significant losses in any of the ten performance measures in the third or fourth year.

10 The TAAS was replaced by the Texas Assessment of Knowledge and Skills (TAKS) in 2003.

11 The fiscal measures are in thousands to make the regression coefficients relatively comparable. An increase of one dollar in the average teacher salary is a trivial amount; an increase in $1,000 in the average teacher salary is a noticeable amount.

12 The set of control variables used here is similar to those in most education production functions and cover the key areas of resources and constraints. Some production functions control for enrollment or growth in enrollment. When we add logged enrollment to the TAAS equation, it increased the R² by only .0002. It increased the R² for high end performance by .0276 but the relationship is positive rather than negative as the literature predicts and is driven primarily by the poor performance of very small rural school districts. The inclusion of logged enrollment in this equation, however, did not affect the findings presented in Table 4. Using percent change in enrollment in both equations had no impact, increasing explained variation by only .0001 in each case.

13 The findings are also robust to the inclusion of a set of management variables, including managerial quality, managerial networking and managing upward (results not shown).

14 Salaries paid to regular education teachers are lower than salaries paid for specialists in bilingual education, compensatory education and special education. By replacing specialists who leave the district with generalists, the district can still cover classes in the face of budget reductions. Doing this successfully, however, means that management needs to figure out how to use its remaining specialists more effectively.

15 In some preliminary analysis, we find that increased staffing of the central office leads to some tradeoffs likely to benefit more disadvantaged students at the expense of students with greater advantages. The relative size of the central office staff is positively associated with increased student attendance and negatively associated with dropout rates, and both of these are desired outcomes. There is also a negative relationship between central office staff and the percent of teachers the district employs in gifted programs. In essence there appears to be a clear tradeoff: build managerial capacity if the goal is to help the disadvantaged students or build specialized teaching capacity if the goal is to improve the performance of high achievers.