Socializing Capital: The Rise of the Large Industrial Corporation in America (review)

A generation ago, Paul Feyerabend delivered a provocative, indeed ferocious, attack on the scientific/objective bases of knowledge ( Against Method [London: New Left Books, 1975]). In that demolishing spirit, William Roy’s contribution to the ongoing debate about the origins and significance of the large corporation in American society might well have been dubbed Against Efficiency, or more narrowly, Against Chandler. In this thickly written monograph, Roy aims to shatter the reigning “efficiency” interpretation of big business’s development from 1830 to 1920, targeting Alfred Chandler’s “functionalist” reading of intersecting technological change, market opportunity, and organizational form as insufficient, if not positively misleading. Whereas Chandler relied upon economic and technological referents, Roy argues, often effectively, for the greater salience to outcomes of power, property relations, and institutional histories, each located in an environment of ideological influence, contested state activities, and mimetic rather than efficiency-based duplication of corporate structures. Roy positions his work as “historical sociology,” and, given the book’s distinctive organization and approach, I would urge T&C readers to seek out reviews in journals of sociology, economics, and business history as complements or correctives to these reactions. Socializing Capital may prove to be a “big book” in some quarters and a messy intervention in others. I lean a bit toward the latter interpretation, but with some qualms.

Though Chandler’s account of large corporations’ advance to leadership of the American economy is the nominal target, Roy equally challenges the explanatory utility of the new institutionalism in economics (e.g., Oliver Williamson, Markets and Hierarchies [New York: Free Press, 1975]). He considers the latter a relatively desperate and insufficient effort to rescue rational efficiency motifs by tacking them onto (and inside of) organizational processes. From a perspective based broadly in political sociology, Roy would have us pay far more attention to power and property relations, the articulation of legal contexts, and the conjunctural synergies of capital market formation, personal connections among business leaders, and strategies for developing product markets, firmly sidelining the imperatives of technology and bureaucracy-building so central to Chandler’s framework. This opening out of the interpretive range is wholly commendable, yet Roy’s alternative account remains problematic, not so much in its conclusions as in its delivery system.

The heart of the critical dissection rests in chapter 2, wherein Roy undertakes to operationalize, in quantitative social science terms, an assessment of core claims that underpinned Chandler’s analysis. Roy sensibly notes that his antagonist’s “greatest methodological weakness” lay in “reasoning backward from examining the common characteristics of what he consider[ed] successful companies rather than comparing successful companies or industries with a control group” (p. 30). Using manufacturing census data to show statistically that “growth, productivity, and change in productivity—variables widely cited as major factors stimulating the rise of corporations—explained virtually no [sectoral] variation” (p. 28), Roy argues that only initial size (circa 1880) and capital intensity correlated with the construction of dominant enterprises. The first three variables, representing efficiency, washed out; the latter two, which relate to power and access to capital (feeding and fueling technical advances), ruled outcomes. This is heavy artillery brought to bear on a major interpretation: efficiency seems to have been a peripheral factor in the rise of major industrial firms. As Roy later summarizes: “Large corporations were found in industries with large firms and high capital intensity, but were no more likely to be in highly productive industries than in low productive industries and no more likely in quickly growing industries than in stagnant ones” (p. 221).

The problem here is that Roy makes little effort to explain to a narrative-based community (Chandler’s choir and most historians of technology) how these quantitative analyses can claim decisive weight and why the data on which they are based should be accepted as reliable. This is a matter of evidence and epistemology. While I...