Нужен перевод explaining the scope of activities pursued by the modern business enterprise is clearly central to our understanding of the organization of industry. yet, as ronald coase points out, the received theory of industrial organization is unable to explain why general motors is not a dominant factor in the coal business, or why a & p does not manufacture airplanes [coase (1972, p. 67)]. nor does the received theory explain why aircraft manufacturers are now producing missiles and space vehicles, why union oil is producing energy from geothermal sources, or why exxon is looking for uranium. one reason for this neglect is suggested by nelson's observation that microeconomic analysis views the enterprise as little more than a black box, and the distribution of economic activity between markets and firms is taken as datum [nelson (1972, p. 37)]. while sometimes it suffices to take institutions as pre-existing entities and model economic phenomena in familiar demand and cost curve terms, there are other circumstances where it is instructive to begin with more elemental units of analysis. firms, after all, do not come in predetermined shapes, and neither do markets. rather, 'both evolve in active juxtaposition with one another,the object being to reach a. complementary configl,lration that economizes on (production and) transactions costs' [williamson (t978)]. the purpose of this paper is to explore some comparative institutional considerations which surround the scope of the business activities engaged in by the modern business enterprise. specifically, the' paper explores an efficiency rationale of corporate diversification. i it turns out that the theoretical framework developed by williamson to explain vertical integration [williamson (1975, ch. 5)] can be extended readily to explore multiproduct diversification. this is because the principal differences between vertical integration and diversification relate simply to the types of transactions being internalized. whereas vertical integration involves internalizing the supply of tangible inputs (such as components and raw materials) to a single production process, the integration of interest here involves internalization of the supply of knowhow and other inputs common to two or more production processes. it turns out that diversification can represent a mechanism for capturing integration economies associated with the simultaneous supply of inputs common to a number of production processes geared to distinct final product markets.