The Linkage Between Implementation Processes and Policy Outcomes: an Analysis of Hud's Administrative Agency Experiment.

Abstract

Program implementation is a research area that has gained a great deal of attention over the last two decades from political scientists and policy-makers. It has been recognized that policies are not self-executing and that program implementation can play an influential role in shaping public policy. Nevertheless, while the literature on program implementation has given attention to describing how programs are set in motion, the crucial issue of what difference does implementation make in terms of program impacts is often overlooked or reserved for program evaluators. The purpose of this dissertation is to explore this important issue. The leading question to be addressed is what difference do variations in implementation procedures have on program impacts. Implementation is viewed here as a necessary but not sufficient condition for program success. The study is based on data from the Administrative Agency Experiment of the Experimental Housing Allowance Program (EHAP) performed by the Department of Housing and Urban Development during the early 1970's. EHAP was designed to test the effects of housing vouchers on solving the housing problems of low income families. The data base allowed for both an evaluation of program impacts and an assessment of the implementation procedures used by the eight agencies running the program. The study shows the limitations of program implementation in fostering desired program outcomes. For example, it is shown that improvements in housing quality was more a function of the housing market that participants resided in than the stringency of inspection st and ards employed by agency personnel. Similarly, the percent of enrollees becoming recipients at each site was more a function of the availability of st and ard housing in the area than the supportive services offered by the agencies. In sum one finds that implementation analysis when combined with evaluation research can give a more complete picture of why programs fail. Here the findings suggest that environmental context played an important role in determining program outcomes. In addition, the study shows that varying degrees of implementation performance would not have a significant impact on program outcomes.