Hierarchies and the Organization of Specialization

This paper studies both theoretically and empirically hierarchies’ role in organizing the division of labor. We develop an equilibrium model of the depth and breadth of knowledge individuals acquire and of the organizational implications of these choices. We show that as the size of the economy increases, the value of the arbitrage between fields performed by generalists declines, and this leads more agents to field-specialize. We then show that organizational changes should accompany increases in field-specialization. As individuals’ knowledge becomes narrower but deeper, organizing in hierarchies becomes more valuable because hierarchies raise the utilization rate of costly knowledge. It follows that both the share of individuals who work in hierarchies and “spans of control” within hierarchies should be greater in larger economies. We then analyze the hierarchical organization of law offices using confidential data from the 1992 Census of Services. Our main findings are that: (1) the share of lawyers working in hierarchies increases with local market size; (2) this change takes the shape of a move from low leverage/unleveraged generalists to leveraged specialists; (3) all of the increase in this share is accounted for by an increase in the share of lawyers who are associates; and (4) changes in hierarchical structure are stronger at the partner/associate margin than at the lawyer/non-lawyer margin, a margin where there are legal constraints on the division of labor. These patterns are consistent with the view that hierarchies play an important role in organizing specialized knowledge within law offices. *University of Chicago Graduate School of Business and CEPR **University of Chicago Graduate School of Business and NBER We thank Pedro Vicente for useful comments. The research in the paper was conducted while the authors were Census Bureau research associates at the Chicago Research Data Center. Research results and conclusions are those of the authors and do not necessarily indicate concurrence by the Bureau of the Census. This paper has been screened to insure that no confidential data are revealed.