Pressman and Wildavsky's Implementation occupies center stage in the developing literature about policy implementation, in part because of the analogy they drew between implementation processes and the multiplicative model from probability theory. This paper takes the relevance of probability theory further and considers the additive model from probability theory and conditional probabilities as well as the multiplicative model. This expanded coverage of probability theorems (1) leads to markedly increased optimism about the likelihood of successful implementation, (2) encompasses empirically reasonable tactics such as persistence, packaging of clearances, engineering bandwagons and policy reduction, and (3) generates advice to hopeful implementers – some of it non-obvious.

Inadequate problem formulation leads to inadequate, alas persistent, solutions (programs). Program evaluation is commonly held as a means to put programs and policies to the test. The best of controlled experimentation, however, cannot serve as the ‘potential falsifier’ of problem formulation, only of the adequacy of means to solve a pre-conceptualized problem. An analysis of disadvantaged youth policy in Israel is provided to illustrate and support these contentions.

An important issue in the discussion of fiscal policy is the contention that the public sector could expand only at the expense of the private sector, which must contract to provide the necessary room. This paper is concerned with ‘financial’ crowding out, which relates to the financing of public expenditure, rather than resource crowding out, which relates to the size of public expenditure. The paper attempts to determine empirically, using the National Institute of Economic and Social Research (NIESR) macroeconomic model of the UK economy, whether fiscal actions under different modes of finance affect some strategic economic variables. The paper utilises techniques of optimal control, which are considered superior to simulation. The main conclusion of the paper is that there is no significant crowding out in the NIESR model; it is, nevertheless, important to distinguish between money-financed and bond-financed increases in government expenditure.

Policy analysts typically presume that future payoffs should be discounted relative to present ones, and that this discounting should proceed at the same rate for all goods and all periods. Closer inspection of four arguments for discounting, however, shows these practices to be problematic. Only two of those arguments provide plausible justifications for time-discounting at all; and neither of those justify a constant rate throughout all periods or for all goods. Indeed, there is an important class of ‘nontradable’ goods which can be discounted only in their own terms. The overall conclusion is that policymakers ought, logically as well as morally, to weight the interests of the future far more heavily than in ordinary discounting procedures.