A theory of economic policy and the role of institutions

Nicola Acocella published a paper in the Journal of Economic Surveys (a free, preliminary version is available here) on the development of the theory of economic policy. Acocella is clearly fully aware of the differences between classical political economics and marginalism (neoclassical economics).* And he dismisses the pre-margnialist views on economic policy as being unsystematic and devoid of general principles.
In his words:

Most classical writers and the marginalists had suggested cases where public intervention was in order. This had been so for Smith (1776), Ricardo (1817), Mill (1848), Marshall (1890), Walras (1874-1877, 1898). But these cases were mainly what Walras called ‘examples of empirical policy’ rather than consistent policy. They were certainly dictated on the basis of an analytical evaluation of the circumstances suggesting them, but were not part of a systematic and consistent assessment of the foundations and the articulation of public policy.

In his view, economic policy as an autonomous discipline, meaning one based on theoretical principles, only started with Sidgwick,  Marshall and Pigou, that is, with Welfare Economics or Social Choice Theory, which apparently now is referred to as Implementation Theory (I didn't know that, I might confess). So it is in the debate of the 1930s, with Robbins and his followers and critics, and more fundamentally with Arrow's Impossibility Theorem that he sees the development of the main themes of a theory of economic policy.

He also suggests that before the development of a the theory of economic policy proper "the ‘night-watchman’ position became an exception as most classical and marginalist economists tended to state a number of specific or general cases where government intervention was in order." I'm not fully convinced about that. Sure there were some views about government intervention, but for the most part the Victorian consensus, among economists (not necessarily in practice, meaning actual economic policy) was for free trade, adherence to the rules of the Gold Standard, and sound finance. The idea of the minimal state was probably dominant, and the exceptions, particularly among marginalists that dominated in the UK at least, were associated to market failures.

The two major obstacles to this dominant view that government action was possible, in Acoccela's view were due to Arrow and Lucas. In his words, these 'vital failures' were: "the impossibility of taking people’s preferences as a reference for public action, underlined by Arrow (1951) and ‘radical’ objections to effectiveness of public action of the kind raised by Lucas (1976)." I should note that it seems that, in his view, the major critiques have been successfully dealt with, by Sen and his followers in one field and New Keynesians in the other, and the current consensus would be that government action is possible and desirable.

In this view, then, the history of the development of of the field is one in which the profession goes from being in favor of government action up to the development of the theory (essentially the from the beginning of marginalism to the 40s), broadly speaking, to being critical from the 50s to the 80s, with a slow return to a more interventionist consensus ever since. And maybe this describes well the field of Implementation Theory. But it strikes me as being at odds with conventional views of what the profession thinks about the role of government over time.

I mean, normally you would think of the profession as being essentially for laissez-faire, the reason why sometimes there is a conflation of classical and marginalist authors, and then after the Great Depression and the Keynesian Revolution (even if not complete, and dominated by an imperfectionist view) there would be an interventionist turn, which starts to be contested in the 1970s, with the Conservative Revolution and the Great Inflation. The apotheosis of the return of the market would come with the fall of the Soviet Union, shock therapy and the so-called Washington Consensus. The question would be to what extent this neoliberal dominance has been undermined by the last Global Financial Crisis (I would say less than what people think).

I also think that the dismissal of the classical political economy approach to economic policy, which was based on historical understanding of the specific conditions of a particular situation, is a mistake, and that there is a lot be learned from that tradition. Note that for classical political economy authors history and institutions, and, hence, policy analysis, were at a different level of abstraction, and had a different relation with the core of the theory (value and distribution). At any rate, this is an interesting paper worth reading.

* For example, he is careful to note that the meaning of the invisible hand in Adam Smith is often out of context. However, he argues that: "we use the term ‘invisible hand’ as a metaphor of the Smithian position as well as of later theories, in particular neo-classical thinking, which has then prevailed, even if the latter are deprived of some social aspects of the working of the market that certainly were in Adam Smith." This reading of Smith leaves room for intervention in the case of imperfections, and seems to follow the ideas of Acoccela's mentor, the famous Italian economist Federico Caffè.