No "One Best Way" to Manage Change: Developing and Describing Distinct Administrative Reform Dimensions across the Fifty American States

INTRODUCTION Administrative Reform (AR) is in some respects like the 57 ingredients in Heinz ketchup or the 31 flavors of Baskin Robbin's ice cream. It may be imported, marketed, sold, and consumed by a diverse array of jurisdictional and/or organizational customers. Thus the character and content of the commodity as well as the complexion of the consumers produces a complex collage on the intellectual and practical landscapes of public management. Kettl (2005) sketched the global market for reform and offered two dominant patterns, the Westminster (found in the United Kingdom, Australia, Canada, and most comprehensively in Zealand's approach) and American (reinvention in three phases during the 1990's). The scope and significance of AR is demonstrated by his claim (2005, 60) that, Perhaps never before have so many governments tried to change so much so fast in such similar ways. The continued currency of AR is exemplified by Light's updating of his tides of reform (Light 1997, 2006). Moynihan (2006) recently assessed the use of Managing for Results (MFR) among the American states in the late 1990's as a component of the New Public (NPM). He found significant discrepancies in the states between an emphasis on performance (results) and enhanced managerial authority, especially less of the latter. Previous work across state agencies explored contrasting top-down and bottom-up approaches to administrative reform (Burke and Wright 2002), as well as the measurement and explanation of reform implementation (Brudney, Hebert, and Wright 1999). In this article, a different line of inquiry is pursued, that compares and contrasts performance/results-oriented reform techniques with structural/process reforms. Using a distinctive statistical approach--Confirmatory Factor Analysis (CFA) --these two contrasting dimensions of reform are first hypothesized and then emerge from the analysis. We define, describe, and measure these dimensions for each of the fifty states. They are approximate parallels to Moynihan's MFR on the hand and increased management discretion and systems analysis on the other. This approach yields new insights into the depth and character of Administrative Reform across the American states in the 1990s. In general, we find that many states adopted both performance/results-oriented reforms and structure/process reforms simultaneously, leading to triumphs of reform effort. In a few states the two dimensions represented alternatives or tradeoffs, with some opting for over the other. Across all states the analysis reveals that performance/results reforms outpaced structural/process strategies with respect to adoption, a finding consistent with Moynihan's analysis. Regardless of the tradeoffs or triumphs, a major finding is the absence of a single or one best way to reform state administration in the 1990's. The dual dimensional character of our results runs counter to Peters' (2005, 356) incompatibility hypothesis where picking reform may preclude implementing others effectively. The article concludes with observations about the impact and implications of the separate reform dimensions. REFORM, REINVENTION, AND THE AMERICAN STATES The public management field faces a quandary in grappling with the cluster of issues associated with contemporary AR. The issues involve, among others, the identification, classification, measurement, explanation, and prescription (or advocacy) of various and specific administrative reforms. Since the 1980's and 1990's, many national and subnational governments now function with updated techniques, approaches, and cultural underpinnings. Difficulties remain, however, in estimating the character, content, extent, and results of the so-called reform revolution (Kettl 2005). The assessment and categorization of Public Management (NPM) reforms is fraught with methodological challenges (Poister and Streib 1999; Calista 2002; Thompson 2002). …

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.