Review of Industrial Organization (2006) 29:171–191 © Springer 2006
On the Merits of Vertical Divestiture
DAVID E. M. SAPPINGTON
Department of Economics, University of Florida, P.O. Box 117140, Gainesville, FL
32611-7140, USA. E-mail: sapping@uﬂ.edu
Abstract. This paper demonstrates that vertical divestiture may increase consumer welfare
even when the divestiture eliminates substantial scope economies and precludes only lim-
ited sabotage. More generally, the merits of vertical divestiture are shown to vary with:
(1) the type and the intensity of competition in the retail market; (2) the locus of scope
economies under vertical integration; and (3) the relative social values of consumers’ sur-
plus and proﬁt.
Key words: regulation, sabotage, vertical divestiture.
In settings where a vertically integrated provider (VIP) supplies vital inputs
to its retail competitors, the VIP may have both the incentive and the abil-
ity to disadvantage (or “sabotage”) rivals and thereby hinder industry per-
formance. For example, in today’s telecommunications industry, VIPs often
supply local loops and/or switching services to competing suppliers of retail
Similarly, owners of railroad track, electricity
distribution systems, and natural gas pipelines often provide retail services
in direct competition with service providers that employ the owners’ infra-
structure. In such settings, a VIP may be able to exploit its position as the
supplier of bottleneck wholesale services to disadvantage its retail compet-
itors. By reducing the quality of the inputs it supplies to retail competitors
or otherwise increasing their operating costs via some form of sabotage, the
VIP may be able to secure a competitive advantage in the retail market,
and thereby increase its proﬁt.
The local loop is the line that runs from the telecommunications supplier’s central
ofﬁce to the customer’s premise.
Economides (1998) provides a seminal analysis of sabotage. Sibley and Weisman
(1998), Mandy (2000), and Beard et al. (2001), among others, provide additional
conclusions regarding the likely incidence and magnitude of sabotage in vertically related
industries. Reiffen et al. (2000) and Zimmerman (2003) provide empirical evidence of