Phillips Andrews' Contribution and the Publication of Manufacturing Business

Manufacturing Business was published in 1949 in a very specific context. It constituted, on the one hand, a reaction to the well-known Cambridge Cost Controversies of the 1920s and 1930s and was, on the other hand, to a large extent a continuation of the famous Hall and Hitch empirical investigation which appeared in 1939.

In addition to this theoretical background, it is relevant to recall that Manufacturing Business emerged from the initial Courtauld Inquiry and was also an attempt to provide some “practical” tools and empirical evidence for the few existing theories on the internal organisation of the firm. Andrews made clear his theoretical inspiration:

This mention of a wider experience gives me an opportunity to pay a tribute to a major element in my education as an industrial economist—my association with the pre-war Oxford Economists’ Research Group... It was the work of this Research Group that developed so strongly the conviction that the behaviour of business men was consistent, and that, accordingly, even though, on many points, it might not seem directly explicable by generally accepted economic theory, there was hope that one would arrive at a consistent theory by studying individual businesses (Andrews 1949: xv).

In the volume, Andrews used an approach based on observed industrial realities at the expense of elementary mathematical formalism. Andrews’ first objective was to illustrate the combination of both deductive and inductive approaches, and to emphasise their complementarities.

At the same time, however, Andrews tried to develop analytical founda­tions to go with Hall and Hitch’s empirical results. In fact, he was strongly in favour of an integrative approach, combining the full-cost principle (reshaped as “normal cost”) with a revival of the Marshallian framework. Andrews’ main idea was that in his analysis of the short run, Marshall could clearly be inter­preted with the help of marginal tools. His analysis of the long period was, however, considered to be incompatible with these tools and their individual­istic foundations.^[17] Andrews’ interpretation of Marshall’s theory in particular stressed the existence of long-run supply curves, including economies of scale. The expansion of a firm’s operations over the long run could not be supported by a marginal approach, which only admitted increasing average costs across such a time period.

Andrews’ expression of normal costs in the long run was, to a large extent, influenced by Marshall’s long-period theoretical framework, and especially by his concept of the representative firm. By contrast with the marginal interpretation of the representative firm, which considered this concept as an equilibrium firm, Andrews considered it as a firm which represented the real­ity of industry. He made it clear in the following note:

This [concept of the representative firm] was his [Marshall’s] new semi-historical concept which he brought into his analysis. In Book IV, Ch. XIII, p. 317 when he refers to the long period, he talks about normal expenses of production and says that for these we must refer to the representative firm, not to any particular competitive firm (Andrews’ Lecture Notes IV, 2 December 1968, Andrews and Brunner Archive, LSE: 3; underlining in original).

Hence, Andrews refused to see Marshall’s contribution to economics as a “static marginalist equilibrium theory”^[18] extended to the long run. Marshall’s representative firm was rather an industrial concept and ‘in effect he [Marshall] is saying that we must refer the industrial supply curve to industrial conditions and not disaggregate it to purport to get long run marginal cost curves for individual businesses’ (ibid.: 4). Thus, the content and methodology of Manufacturing Business was a direct attack on the mar- ginalist theory of the firm.

From Manufacturing Business emerged a series of further work on industries which led, in turn, to the establishment of industrial economics as an aca­demic discipline at Oxford.

4