Ch. 8: Conglomerate Firms and Internal Capital Markets 471
were misallocating resources prior to the spinoff. Using three-different approaches to
control for endogeneity they show that refocusing decisions does not necessarily cause
improvements in efficiency. In particular, firms that choose to spin-off and divest divi-
sions are larger, more diversified, and subject to more serious problems of asymmetric
information. Further, the spun-off segments tend to be in fast growing industries with a
great deal of IPO and corporate control activity. Finally, they appear to have experienced
recent unanticipated shocks to profit. They find that although spin-offs and divestitures
may be associated with improvements in investment efficiency, they do not cause these
improvements. When they control for measurement error, they also show that the sensi-
tivity of investment to both industry Tobin’sqdoes not significantly change following
the refocusing decision.
- Conclusions: What have we learned?
There have been a substantial number of careful empirical papers on internal financial
markets in the last few years. Any summary of what has been learned is bound to be
subjective and reflect the interests of the authors. With that caveat in mind, we can
summarize the existing evidence about internal capital markets.
- The early work established clearly that, using single-segment firms as benchmarks,
there exists a conglomerate discount. - Initial attempts to explanation the discount focused on agency conflicts and conflicts
among divisions that led to overinvestment in divisions with poor prospects and un-
derinvestment in divisions with highqs. - Conclusions drawn from econometric studies of segment capital expenditures, which
use the Tobin’sqs of single-segment firms to proxy for segment investment opportu-
nities, are subject to measurement error and may not be valid. - Diversified firms rely more on acquisitions than single-segment firms. Thus, studies
that focus on capital expenditures may miss important components of investment by
diversified firms. - A conglomerate discount is not, by itself, evidence of agency or inefficiency—it may
be due to the fact that single segment and diversified firms operate on different regions
of the production function. - A simple neoclassical model that recognizes that the decision to diversify is en-
dogenous and that firms grow fastest in industries where they have a comparative
advantage in response to positive demand shocks in those industries is consistent
with the growth patterns of diversified firms. - The sales of plants by firms are also consistent with a simple neoclassical profit-
maximizing model. - Much of conglomerate discount can be explained by sample selection. Firms that
choose to diversify, or to stay diversified or to be acquired by diversifiers inherently
differ from single-segment firms.