The feasibility of social policies

The notion that, given sufficient ingenuity and resources, any social problem can be solved, in a manner consistent with the announced policy objectives and with the accepted institutional framework, has been a major source of failures and frustrations. This paper suggests that policy analysis should concentrate on the investigation of the conditions of feasibility of public programs. Feasibility is defined in terms of all the relevant constraints: social, political, administrative, and institutional, as well as technical and economic. The emphasis on the pure logic of choice and optimal decision rules, so characteristic of normative theorizing in decision theory, management science, and welfare economics, has tended to obscure the fact that in the field of social policies, the range of feasible choices is severely limited by a variety of constraints. Most bad decisions are not just suboptimal; on closer examination, it usually turns out that they were not even feasible at the time they were made. Only by understanding the reasons why, under certain circumstances, a given goal cannot be achieved, can we hope to gain better knowledge about the working of social institutions. Just as the essence of the scientific method is the critical analysis and refutation of proposed theories, rather than the accumulation of evidence in favor of pet solutions, so the most important task of policy analysis consists in submitting plans and objectives to the most stringent tests of feasibility. Also in analytic case studies, the search for constraints that were ignored by the decisionmakers has considerable heuristic value, as I try to demonstrate with a critical discussion of some aspects of the British National Health Service.Paper presented at the Workshop on Models and Cases in Administration Decision Making, Joint Sessions of Workshops of the European Consortium for Political Research, Strasbourg, France, March 28–April 2, 1974.