Interorganizational Coordination, Administrative Consolidation, and Policy Performance

Abstract

Although administrative and policy reformers often urge coordination as the key to more effective delivery of public services, there is little beyond anecdotal evidence to suggest that most approaches to coordination can actually improve public services in multiorganizational delivery systems. Indeed, critical accounts of bureaucratic behavior suggest that it is quite difficult to get organizations to cooperate with each other, much less to do so effectively. Extended scholarly analyses have pointed out the great difficulty of obtaining coordination in the absence of the consolidated administration of programs. For this reason, reformers would often prefer to see programmatic and administrative consolidation, hoping that the enhanced bureaucratic control they anticipate will improve program outcomes. This paper examines the effect of coordination patterns and administrative arrangements on the accomplishment of policy goals in the delivery of employment and training services under the Job Training Partnership Act (JTPA). JTPA is carried out in an intergovernmental, multiorganizational setting in which state and local JTPA administrative entities must obtain the support of a variety of other administrative agencies if JTPA goals are to be achieved. The act requires that employment and training coordination plans be written at both the state and local levels to link the activities of these disparate organizations. The organizations themselves have primary missions that overlap, but are not identical. Requirements and written plans are no guarantee that coordination will occur; even if it does, there is no guarantee that it will improve performance. Jennings (1994) found that increasing coordination has positive effects on program outcomes, but his analysis relied strictly on a perceptual measure of performance as reported by state agency administrators. The current study combines Jennings's data with objective measures of program performance, such as placement rates and wages. This is a critically important step since subjective and objective measures of performance can be at variance with each other (Brudney and England, 1982; Brown and Coulter, 1983; Parks, 1984). This article seeks answers to several questions. First, what is the relationship between subjective and objective measures of JTPA performance? Second, how are the objective measures of performance routinely reported to the U.S. Department of Labor related to each other? Third, does the analysis of the effects of coordination and administrative consolidation on performance depend on whether objective or subjective measures are used? Fourth, do coordination and consolidation affect different dimensions of performance differently? As the article pursues answers to these important questions, it tests two central hypotheses: [H.sub.1] Increased levels of coordination lead to higher levels of performance. [H.sub.2] Administrative consolidation of the JTPA administrative agency and the state employment service increases JTPA performance outcomes. The article proceeds as follows. First, we discuss the theory of coordination as reflected in the literature of public administration. We discuss coordination as it relates to interprogram and interorganizational coordination. We then take note of the legal requirements and formal provisions for coordination under JTPA. A discussion of administrative organization follows. We then present a multivariate model of employment and training program outcomes and discuss the measures that are employed in the analysis. Sections analyzing the data follow the discussion of coordination. We examine correlations among the variables, looking in particular at the relationship between objective and subjective measures of performance. We examine the relationships among the different objective measures of performance. We then estimate multiple models of JTPA performance, using the objective indicators of performance as the dependent variables. …