Ch. 8: Conglomerate Firms and Internal Capital Markets 439
Thus, Stein suggests that diversification is value increasing when valuation errors
are small and when the returns of projects within an industry are highly correlated,
Diversification is value reducing when valuation errors are likely to be large and when
the payoffs of projects within industries are likely to have a low correlation.^18
3.2. Conglomerates and organizational competencies
Matsusaka (2001)develops a matching model to explain why conglomerate firms exist.
In his model firms have different organizational competencies. The organizational com-
petencies are somewhat transferable across industries. When sales decline in an industry
it is not optimal for firm to go out of business. Instead it should diversify into new lines
of business in order to find a good match between their organizational competence and
the line of business. If they find a good match they may transit into the new industry and
exit their original industry.
Matsusaka’s (2001)elegant framework generates several predictions. Diversifying
firms trade at a discount because on average the match between their organizational
competence and their existing main divisions is bad. Because the match in the new
industry may also turn out to be bad, many diversification attempts are in fact re-
versed. However, the announcement of a diversification is a signal that the firm is worth
maintaining, resulting in a positive announcement effect. The theory also predicts that
successful diversifiers quit their original industry. Thus the theory is quite consistent
with the early evidence on the diversification discount (e.g.,Lang and Stulz, 1994, and
Berger and Ofek, 1995), as diversification results from a poor match between industries
and firm’s organizational competence, and on announcement returns (e.g.,Schipper and
Thompson, 1983; Hubbard and Palia, 1998) which document positive or non-negative
returns to changes in the level of diversification.
3.3. Diversification and the failure of corporate governance
Given the message from the early literature that diversification destroys value, the obvi-
ous question is why we observe so many diversified firms. One plausible answer is that
while diversification destroys investor value it benefits the managers of corporations.
Thus diversification might arise as a result of a failure of corporate governance which
should be penalizing managers who diversify inappropriately.
Jensen (1986)andStulz (1990)argue that managers may obtain increased status and
perquisites when they diversify their firms. Diversification allows managers to act on
a broader stage, and in particular may allow them to participate in “hot” and exciting
industries. It may also be easier to skim from a diversified firm (Bertrand and Mul-
lainathan, 2001).
(^18) The model does not analyze the possibility that a focused firm may rank projects correctly but over- or
under-invest in the aggregate because the valuation errors it makes are correlated across projects.