This Could Be the Start of Something Big: Linking Early Managerial Choices with Subsequent Organizational Performance
研究英格兰和威尔士的犯罪与失序减少伙伴关系，发现早期领导者的合作优先级选择会影响近十年后的犯罪率，优先促进合作行动比改善态度更有效。
The influence of early events in the history of a country, a social phenomenon, or an organization on later developments has received significant attention in many social science disciplines. Often dubbed “path dependence,” this influence occurs when early events influence later outcomes even when the original events do not reoccur. “Path dependence,” however, has received little theoretical or empirical attention in public administration. This article discusses how early events in an organization’s history can come to influence later outcomes. The article then empirically tests for the presence of path dependence using data from Crime and Disorder Reduction Partnerships in England and Wales, a cross-organizational collaboration inside local government. We find that early choices by the leader of the collaboration about which activities to prioritize to create collaboration set in motion a path creating collaborations that were more successful and less successful, producing differences in crime results almost a decade later. The most successful early priorities involved getting partner organizations to act in collaborative ways, rather than working to improve the attitudes of these organizations toward collaboration. We argue that path dependence should be examined in public administration research from a, prospective, prescriptive perspective, to learn more about what early managerial actions can produce better later results. It is commonplace in many social sciences to argue that events or choices early in the history of a nation, a social phenomenon, or an organization can influence what happens decades or even centuries later. The European countries that are predominantly Protestant today are those whose kings became Protestant during the Reformation. Differences in the prevalence of choral societies in Italian regions in the second half of the 19th century are associated with the extent to which Italian regional governments answer citizen information letters in the second half of the 20th century. A powerful health insurance industry created by a World War II tax law change designed to help hold back nominal wage increases during an inflationary period made it politically more difficult to create government-run health insurance decades later. The authors are grateful for research support provided by The Roy and Lila Ash Center for Democratic Governance and Innovation at Harvard University’s John F. Kennedy School of Government. Address correspondence to the author at steve_kelman@harvard.edu. JPART 25:135–164 at Y osei U niersity on Feruary 1, 2015 http://jpaordjournals.org/ D ow nladed from Journal of Public Administration research and Theory 136 Such observations appear in a number of social science classics. Marx (1951, 224, originally published 1869) begins The Eighteenth Brumaire of Louis Napoleon: Men make their own history, but they do not make it just as they please; they do not make it under circumstances chosen by themselves, but under circumstances directly encountered, given and transmitted from the past. The tradition of all the dead generations weighs like a nightmare on the brain of the living. Weber (1949, 171, 183), in Methodology of Social Sciences notes that a different outcome of the Battle of Marathon between Athens and Persia would have increased the odds of an eventual development of a “theocratic-religious” culture in the West by creating “‘loaded’ dice” favoring one kind of civilization over another. More recently, arguments about the “shadow of the past” on the present have been made in so many fields—political science, organizational sociology, economics, even business history— that David (2001, 15) suggests they represent a quest for a “historical social science.”1 The basic mechanism posited in this work is the same: “[S]ome initial event or process generates a particular outcome, which is then reproduced through time even though the original generating event or process does not recur” (Pierson 2004, 45, emphasis in original). Persistence of the original behavior occurs without recurrence of the original cause, continuing unless something new disrupts its continuation (Jepperson 1991; Stinchcombe 1968). The term used to describe the shadow of the past has varied in different disciplines; political scientists have referred to “critical junctures” (Lipset and Stein 1967), organizational sociologists (Marquis and Tilcsik 2013; Stinchcombe 1965) to “organizational imprinting.” However, those using this reasoning have increasingly come to use the phrase “path dependence,” coined in a famous article by the economic historian Paul David (1985) describing—of all things—persistence in use of the QWERTY keyboard even after the original reason for it had disappeared, and its continued use was dysfunctional. In this article, we examine the influence of early history on later outcomes in the context of a cross-organizational collaboration in government called Crime and Disorder Reduction Partnerships (CDRP’s), established in all local authorities in England and Wales in 1999 with the aim of reducing crime.2 We test for the influence of managerial choices the collaborations’ first leaders made on the collaboration’s crime performance a decade later. Our central result is that early choices about how to promote collaboration among the different member organizations establish better, or worse, patterns and habits of collaboration that in turn produce a noticeable impact on later crime rates. Our interest relates to the central concern in public administration 1 In a short paper directed to organizational scholars, Lawrence (1984) draws a distinction between “historical research” and “historical perspective.” “Using written documents and artifacts to study attitudes during the Depression” exemplifies the former, whereas “using historical information about the Depression to explain differences in attitudes today” exemplifies the latter. 2 CDRP’s consist of the police (autonomous from local government), Probation Service (a central government agency working with released prisoners), Youth Offending Service (a central agency dealing with young people at risk of crime), Fire Service (an autonomous local agency), and local government service units (e.g., street lighting, parks, and inspection services). CDRP’s frequently organize initiatives the CDRP runs itself; for a further description, see Kelman, Hong, and Turbitt (2013). at Y osei U niersity on Feruary 1, 2015 http://jpaordjournals.org/ D ow nladed from Kelman and Hong This Could Be the Start of Something Big 137 research and practice with organizational performance. Thus, the question that concerns us is to what extent the early choices that organizational managers make influence how successful the organization eventually will be in achieving its purposes. The more later outcomes are path dependent, the more early choices matter. There has been relatively less attention in public administration than in other social science traditions to path dependence, and the modest attention the topic has received has almost all involved the least-novel version of path dependence theory. This is an example of problems the field faces of insufficient connection to social science research (Kelman 2007). We thus seek to direct attention of public administration scholars to this research and also suggest (and test in one context) a theory about how path dependence ideas can usefully be used in public administration scholarship.