With such an important finding, there are of course many potential explanations as to why the discount exists. A far from complete list includes:
- Poor managerial incentives
- Poor Monitoring and a lack of transparency (hard to tell who is doing what, so why not shirk)
- Inefficient internal capital markets (so money is wasted through misallocation)
- A lack of loyalty on mangers' and employees' behalves---this would lead to reduced performance and higher expenses.
There really is little doubt that all of these play a role in the discount. Moreover, we should still consider the possibility that there may not be a real discount since the firms do freely chose to become conglomerates. This endogeneity may be the result of a discount that would have been larger had the firms not become conglomerates. (A view which I doubt, but do consider worthy of attention).
The authors report what every upper and middle level manager in the world already knows: that politics matter.
More specifically McNeil and Smythe write that lobbying by divisional managers plays a role in the allocation of capital. This is a problem because if the internal market were perfect, the allocation decision would be based soley on the merits (the risk and returns) of each project.
In their words:
"To test the Lobbying Power Hypothesis, we examine the sensitivity of business
segment capital expenditures to segment manager characteristics expected to
contribute to a manager's lobbying power for a sample of firms that have
identifiable division/segment managers....There are several division manager
characteristics, such as tenure as suggested by Wulf (2002b), that could be
connected to lobbying power. We collect information on division manager tenure
with the firm, time in position, salary level relative to the CEO, membership on
the board of directors, age, and whether the manager is one of the firms top
five executives. Each characteristic could indicate and/or impact the degree of
a manager's lobbying power. In the analysis, we examine the association between
division capital expenditures and each of the aforementioned division manager
and the findings?
"We find evidence that segment level capital expenditures are associated with
division manager characteristics which, we argue, reflect division manager
lobbying power. For example, the results indicate that relatively high q
segments receive lower capital expend itures when headed by a manager with low
tenure or by a manager competing with multiple top executive/segment