Organizational Structure and the Diversification Discount

11 August 2008 at 2:04 am Peter G. Klein Leave a comment

| Peter Klein |

Do diversified conglomerates trade at a discount relative to more specialized firms? A huge literature in strategy and corporate finance emerged over the last couple of decades devoted to this question. Early studies claimed to find a substantial diversification discount, though more recent papers claim that the observed discount is due to measurement error, self-selection, and other characteristics, not a harmful effect of diversification per se. (For a good overview of this literature, now slightly dated, see this roundtable report edited by Belén Villalonga. Some of my own contributions are here and here.)

Seemingly lost in the search for a diversification discount, however, is a related question: What is being discounted? Potential benefits of diversification, according to the literature, include access to internal capital markets and more efficient redeployment of distressed assets; potential costs include inefficient rent-seeking, bargaining problems, and bureaucratic rigidity. But these benefits and costs have little to do with industry or geographic diversification per se — they apply to the management of any multi-unit organization, even if its activities do not span different industries or regions.

In a new paper, “Organizational Structure and the Diversification Discount: Evidence from Commercial Banking,” Marc Saidenberg and I try to distinguish the effects of diversification and organiztaional complexity by studying multi-unit firms within a single industry, commercial banking.

We look at 1990-94 period, immediately before the implementation of the Riegle–Neal Act; during this period, banks were forbidden from branching across state lines, so a bank wishing to operate in multiple states was required to establish a bank holding company with independently chartered subsidiaries in each state. This resulted in organizations that produce a single “product” (financial services) yet have complex, multidivisional structures.

We find evidence for an “organiztional complexity discount,” controlling for firm-specific unobservables and using instruments derived from differences in state-level banking regulation to mitigate endogeneity. Our results suggest that at least part of the discount reported in the diversification literature reflects organizational complexity, not industry diversification.

The paper is forthcoming in the Journal of Industrial Economics. You can get the working-paper version at SSRN. Here’s the abstract:

This paper provides evidence on organizational structure, geographic diversification, and performance at bank holding companies (BHCs). First, we show that a BHC’s member banks benefit from access to the parent organization’s internal capital market. Second, we ask if the benefits of internal capital markets are best realized within loosely structured, decentralized organizations or more consolidated, centralized firms. We find that BHCs with many subsidiaries are less profitable and have lower q ratios than similar BHCs with fewer subsidiaries. However, because we study multi-unit firms in a single industry, our results suggest that the valuation discount reported in the diversification literature in empirical corporate finance reflects not only industry diversification, but also organizational structure.