Viewed historically, there is no end to the things – for good and ill – that states have undertaken and succeeded in doing. States can obviously achieve (under favorable circumstances, at least) goals of the most varied kind. As mentioned earlier, implementation researchers have mainly produced long lists of factors associated with the successful implementation of public programs. In order to bring some clarity into this discussion, I shall make use of a simple analytical outline. The idea is to sort out the various factors on a general level. One reason for this approach is to facilitate an analysis of these issues in terms of principle (in a higher degree, at any rate, than has marked the production of those unsorted check lists). Another purpose of the outline is to assist in the formulation of a more general theory of policy implementation – of what, in other words, the state can and cannot do.

To begin with, we may distinguish among factors associated with implementation failure in the following manner: do the causes of failure lie (a) in the design of the program as such, or (b) in the stage of execution, that is, in the organization of the implementation process? The basic idea is simple enough – that any program, however cleverly designed it may be, will fail if its implementation is entrusted to an organization unsuited to the purpose.