Ch. 8: Conglomerate Firms and Internal Capital Markets 463
waste resources by diversifying into industries in which they do not have a comparative
advantage.
To distinguish between these interpretations,Maksimovic and Phillips (2002)ex-
amine how firms respond to industry demand shocks. According to their model, firms
should grow the segments in which they are particularly productive and that have re-
ceived a positive demand shock. They should reduce the growth of segments in which
they are not efficient and which have received a negative demand shock.
To see this, recall from Section3.5that the output of a conglomerate firmiin in-
dustry 1 depends onv 1 i=d 1 ip 1 −r 1. Suppose that there is positive demand shock in
industry 1. First, output prices increase,p 1 >0. Second, the price of capacity in the
industry also increases,r 1 >0. Productive firms in the industry increase output in
industry 1. More formally, they experience an increase in theirv 1 ibecause the marginal
positive effect of a price rise,d 1 i×p 1 , outweighs the effect of an increase in the cost
of capacityr 1. The higher the abilityd 1 i, the more capacity a firm adds in response to
a positive price shock.
The effect of a price shock in industry 1 on the marginal producers not is more com-
plex. If the effect of the expansion by the productive firms onr 1 is minor, then the
marginal firms may also expand, although at a slower rate than the more productive
producers. However, if the price of capacity is bid up sufficiently high so that for some
firms with smalld 1 i,d 1 i×p 1 −r 1 <0, then these marginal producers will sell some
capacity to more productive producers and focus instead in industry 2. These firms’ op-
erations in industry 1 decline not only relative to those of more productive firms, but in
absolute size as well.^36
The MP model also generates another testable prediction regarding cross-segment
effects. When managerial capacity is a fixed factor of production, investment decisions
by conglomerate firms in one segment create opportunity costs for investments in other
industries in which they operate. Thus, segments’ investment decisions will depend on
the relative demand growth across all the industries in which the conglomerate operates.
Specifically, suppose that there is a large positive demand shock in one of the industries
in which a conglomerate operates. If the conglomerate’s segment in that industry is
highly productive, it will grow relatively fast. This growth increases the conglomer-
ate’s costs in other segments, thus decreasing the other segments’ optimal size. Thus,
a conglomerate which has a productive segment in an industry that has received a pos-
itive demand shock will grow more slowly than it otherwise would have in itsother
segments.
Suppose, instead, that the conglomerate had an unproductive segment in a fast grow-
ing industry. Then the conglomerate may find it optimal to divest or reduce operations in
the high growth industry for two reasons. The positive demand shock in the industry will
have increased the value of its capacity, increasing a low productivity conglomerate’s
(^36) Here we assume that the effect of a demand increase in industry 1 affectsv 1 only and does not affectv 2.
Appendix Adiscusses both effects.