Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Internal capital markets and competitive threats
(USC Thesis Other)
Internal capital markets and competitive threats
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INTERNAL CAPITAL MARKETS AND COMPETITIVE THREATS
by
Garrett Swanburg
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BUSINESS ADMINISTRATION)
May 2015
Copyright 2015 Garrett Swanburg
Internal Capital Markets and Competitive Threats
1
Acknowledgements
I would like to acknowledge the following people for their constructive feedback and support:
John Matsusaka, Oguzhan Ozbas, Yongxiang Wang, Maria Ogneva, Gordon Phillips, Kenneth
Ahern, Sakya Sarkar, and Zhishan Guo. I would also like to thank the seminar participants at The
Federal Reserve Bank of Chicago, San Diego State University, and The University of San
Francisco.
Internal Capital Markets and Competitive Threats
2
Table of Contents
Acknowledgements
1
List of Figures
3
List of Tables
4
Abstract
5
Introduction
6
Chapter 1: Theory on Financial Strength and Competition
A. Predation
B. Aggressive Pricing
C. Entry Deterrence
D. Investment
E. Internal Capital Markets
13
13
14
15
15
16
Chapter 2: Data and Methods
A. Segment Data
B. Import Tariff Data
C. Covariate Balance
19
19
20
21
Chapter 3: Investment Decisions
A. Investment Behavior
B. Resource Allocation
23
23
28
Chapter 4: Pricing and Performance
A. Price-Cost Margins
B. Market Share
C. Stock Market Performance
32
32
33
35
Conclusion
38
References 56
Internal Capital Markets and Competitive Threats
3
List of Figures
Figure 1: Number of Tariff Cuts over Time
40
Figure 2: Conglomerate Segment Propensity Score Distribution
41
Figure 3: Investment Growth Around Tariff Cut Events
42
Figure 4: Investment Growth and Net Debt
43
Figure 5: Diversified Firm Cash Holdings Around Tariff Cut Events
44
Figure 6: Stock Returns to Diversified and Synthetic Diversified Firms 45
Internal Capital Markets and Competitive Threats
4
List of Tables
Table 1: Summary Statistics
46
Table 2: Investment Growth with Competition Shocks
47
Table 3: Asset Growth with Competition Shocks
48
Table 4: Industry Exit Following Tariff Cuts
49
Table 5: Investment in Unrelated Segments
50
Table 6: Investment Growth and Alternative Investment Opportunities
51
Table 7: Price-Cost Margin
52
Table 8: Market Share Growth
53
Appendix A: Two Year Window for Treatment Variable
54
Appendix B: Variable Definitions 55
Internal Capital Markets and Competitive Threats
5
Abstract
Theory suggests that a diversified firm’s internal capital markets can help or hinder its ability to
respond to competitive threats: internal capital markets can help by allowing the firm to move
resources quickly to threatened markets (resource flexibility), but can hurt by making it easier for
the firm to exit an industry (providing weaker entry deterrence). This paper examines how
diversified firms respond to heightened competition following a tariff reduction in order to assess
the competing theories. The main finding is diversified firms reduce their investment in
threatened industries relative to non-diversified firms, suggesting that internal capital markets
weaken the ability to compete. The evidence also shows that diversified firms increase the share
of total investment funds to non-threatened industries following a competitive threat, indicating
that they are transferring resources out of the competitive industry.
Internal Capital Markets and Competitive Threats
6
Introduction
This paper examines the effect of diversification on investment and pricing strategies when firms
face competitive threats in product markets. Existing theory leads to predictions of either more or
less aggressive behavior by diversified firms with respect to these investment and pricing
decisions. Conglomerates may respond more aggressively by using their internal resources and
access to external capital to maintain market share when faced with new competitors (Faure-
Grimaud and Inderst (2005), Telser (1966)). Alternatively, they may respond less aggressively
because they have an easier way to redeploy resources if they abandon threatened segments
(Matsusaka and Nanda (2002), Cestone and Fumagalli (2005)). This paper uses instances of
unanticipated import tariff reductions as a quasi-natural experiment to test which of these effects
is more impactful when competitive threats arise. The main finding is that internal capital
markets are more likely to be used to pull resources from threatened segments than to strengthen
them. This is demonstrated by lower investment, lower sales growth, and less aggressive pricing
when compared to single-segment rivals.
There is a fairly long history of papers that document the determinants of divisional investment
in conglomerate firms. The early studies (e.g., Shin and Stulz (1998), Rajan et al. (2000)) test
theories of efficient capital allocation based on resources available to the firm and the
opportunities that exist in the segments. However, the question of how diversification impacts
strategic decisions in the context of competitive product markets is not substantially addressed
by prior work. Additionally, I add to a more recent strand of the literature that studies decisions
made by diversified firms within a natural experiment setting (Natividad (2013), Seru (2014)).
Internal Capital Markets and Competitive Threats
7
This approach allows for an identification of the effects of diversification, rather than cross-
sectional correlations of divisional characteristics.
The purpose of this paper is to offer the first broad-based evidence on how diversified firms
respond to competitive threats. While there are ample theories suggesting that diversification can
strengthen or weaken the ability of firms to respond to an increase in competition, little empirical
evidence is available to disentangle the different views. The most relevant prior study was by
Khanna and Tice (2001), who tracked the actions taken by diversified and focused discount
retailers following entry by Wal-Mart into the local market. They found that diversified firms are
quicker to exit, but those that stay invest more aggressively. They also note that the firms act
efficiently by demonstrating a sensitivity of investment to ROA. My findings in manufacturing
industries are consistent with those of early exit, and I also observe internal capital markets
shifting away from the industries that have become less profitable. Boutin et al. (2013) measured
the rates of entry into industries with incumbents affiliated with business groups. They found that
high levels of liquidity among group partners are associated with lower levels of entry in a
particular firm’s home industry. In this paper, I also find that liquidity plays a role in the
competitive response, as the business segments with more cash at the firm-level are able to
maintain higher investment growth when faced with the competition shocks. However, I observe
that the overall effect of diversification is a reduced rate of investment.
Other studies have shown an effect of diversification in responding to changes in investment
opportunities. Guedj and Scharfstein (2004) observe differences in the setting of FDA drug trials
where some firms are diversified in the sense that they have multiple projects. They found that
biopharmaceutical firms with multiple potential drugs are more likely to abandon one of them if
difficulties arise in the clinical trial process than firms with just one drug. The results that follow
Internal Capital Markets and Competitive Threats
8
in this paper add to this prior evidence of winner-picking and refocusing, but in a multi-industry
setting. Another related paper by Maksimovic and Phillips (2008) divided industries by their
rates of growth and examined differences by organizational form. Their setting of industry
growth essentially represents the flip-side of the heightened competition I investigate. They
found that in growth industries, conglomerates are more active in acquisitions than single-
segment firms and seem to be less likely to exhibit signs of financial dependence.
In this paper, I estimate how investment and pricing policies differ for diversified firms
compared to specialized firms when faced with an increase in product market competition. My
identification strategy is to examine decisions by firms following an exogenous increase to
competition within their industries. Following Fresard (2010), I make use of instances of large
import tariff reductions to form a quasi-natural experiment. By using a panel of business segment
data, I am able to perform difference-in-difference tests that specifically examine outcomes in
the affected industries of diversified firms.
As with any study investigating effects of diversification, the issue of endogeneity should be
considered. The assignment of conglomerate status is not random and could therefore be related
to certain characteristics of the firm. In this paper, the concern would be that there are
fundamental differences between stand-alone segments and segments of diversified firms. For
that reason, I use propensity score analysis to ensure that there is sufficient overlap in covariate
values between the two groups in the sample: diversified and focused firms’ segments.
Additionally, I test the effects of segment characteristics alongside the effect of diversification
during the shocks and find that only diversification has explanatory power.
Internal Capital Markets and Competitive Threats
9
I begin by examining investment behavior of firms threatened by new market entrants. For the
main hypothesis – how the investment response is affected by diversification – a difference-in-
difference estimate from a panel regression shows a more than 13% shortfall in segment-level
investment attributable to the firm’s diversified organizational form following a tariff cut. No
significant difference can be seen in placebo tests prior to the events, suggesting that these
shocks are in fact the cause of the observed effect. Diversified firms also fall behind in asset
growth by about 6% in the affected industries. This evidence indicates that diversified firms cut
back on investment in threatened industries relative to their focused rivals. I next examine
whether these diversified firms are shifting the unused resources to other segments or reducing
financing. I find that unrelated segments to those hit by the shock have little change in
investment growth, but receive a nearly 1.5% increased share of their firms’ total investment
funds. This result supports the hypothesis that diversified firms will tend to shift resources away
from a segment that is threatened rather than solidifying it by drawing on resources from the
unrelated segments. Additionally, cash reserves of affected diversified firms increase by 4%,
demonstrating that some of the cutbacks are saved for future opportunities.
Next, I examine how pricing behavior changes in this setting. Conglomerates may use their
access to external financing and internal capital markets to undercut prices in the presence of
competitive threats (Faure-Grimaud and Inderst (2005)). Lowering pricing margins can drive out
competitors and increase market share. While the results do show diversified firms cut their
margins slightly, focused firms cut theirs 1.5% more. This suggests that diversified firms are not
using their internal capital markets to subsidize aggressive pricing strategies. Leaving margins
high accommodates increased market entry by foreign competitors. The less aggressive pricing
and relatively low investment by diversified firms appears to have an impact on performance in
Internal Capital Markets and Competitive Threats
10
the product market. I find evidence of lower market share gains by segments of diversified firms
when tariffs are lowered. They exhibit sales growth around 4% lower than the growth of focused
firms when adjusted for the sales growth of their respective markets.
These findings are distinct from those of papers studying efficiency of internal capital markets,
such as Maksimovic and Phillips (2002). They show both theoretically and empirically that
conglomerate firms move investment funds to divisions that have the most profitable
opportunities. In other words, conglomerates demonstrate optimal profit-maximizing behavior in
their resource allocation decisions. While their findings improve our understanding of
diversification, the question of how diversified firms respond to competitive threats should be
viewed differently than research on responses to demand shocks. Competition brings about
strategic considerations which can possibly raise the marginal benefit of investment whereas
demand reductions represent a purely diminished return on investment. For example, a model by
Grenadier (2002) and an empirical study by Akdogu and MacKay (2008) show that competition
can cause firms to invest more quickly and at higher levels. In sum, the implications of
competitive shocks introduce more possibilities for different responses by diversified and single-
segment firms.
The findings in this paper can provide some insights into the magnitude of previously explored
issues within a diversified firm. The fact that diversified firms cut investment in the threatened
industries lends support to the “winner-picking” effect (Weston (1970), Williamson (1975), and
Stein (1997)). In these models, the firm’s headquarters uses its information on the ROA of
investment in its business segments to allocate investment funds. The observation of firms
allocating a larger portion of firm investment resources to unaffected industries seems to
demonstrate this effect. The findings do not support models of corporate socialism (e.g.,
Internal Capital Markets and Competitive Threats
11
Scharfstein and Stein (2000) and Rajan et al. (2000)). There does not seem to be a subsidization
of threatened units, which is what would be expected from a firm exhibiting socialism where
investment allocation is insensitive to ROA.
Overall, it is difficult to conclusively state whether the observed actions by diversified firms of
lower investment growth and higher pricing are value-creating or value-destroying. If
conglomerates are more vulnerable to predation due to the ability to redeploy resources, then
their firm value will be at risk when threats arise. If, on the other hand, they benefit from an
ability to easily exit a threatened industry, then they will be able to maintain their value through
these shocks. To shed light on this issue, I perform tests of stock returns using the causal
inference method called synthetic controls. This test shows that returns of diversified firms after
the shocks are nearly exactly the same as returns of the single-segment firms that form the
synthetic controls. This suggests that the redeployment of resources is not value-destroying since
their returns keep pace with their single-segment rivals across all industries.
In terms of identification strategy, this paper is similar to Fresard and Valta (2012). As with my
study, they use exogenous tariff cuts to identify shocks to competition. They study firm-level
changes in financing policies brought about by the competition and find that firms affected by
increased foreign competition become more conservative as seen by lower capital expenditures
to assets and higher cash holdings. Additionally, treated firms prefer financing through equity
rather than debt, thereby experiencing a reduction in leverage. I also document reductions in
investment in this paper; however, my study focuses on the role of organizational structure on
competitive responses. To that end, I use a panel of segment-level data to compare differences
between single-segment firms and divisions of diversified firms solely in the industry that is
Internal Capital Markets and Competitive Threats
12
affected. Additionally, I perform tests that consider actions and characteristics of segments
operating in industries not affected by the shock within a firm that is.
A concurrent working paper by Bai (2014) also uses tariff rate shocks to assess firms’ responses
to competition. The paper’s focus is on plant-level investment decisions of manufacturing firms,
including the likelihood of plants to be shut down or sold. It finds that conglomerate firms have
an ability to offset financial constraints to maintain spending in their individual factories. In
contrast, I investigate decisions made at the business segment level to understand how diversified
firms decide to allocate their funds across their industries of operation.
Internal Capital Markets and Competitive Threats
13
Chapter 1: Theory on Financial Strength and Competition
There are several existing models which give a framework for the relationship between
competition and cash. This line of theory is highly relevant in analyzing the effects of
diversification as these multi-segment firms can pool their cash together in dealing with
competitors in one particular segment. Although competition and cash may not be the primary
focus of the models presented, interpretation of their main results or varying parameters within
can shed light on the topic. Here, I outline how five inter-related aspects of competition can are
directly influenced by the strength of firms’ finances.
A. Predation
The literature on predation is one of the most important foundations in studying the relationship
between product market competition and finances. The seminal paper by Bolton and Scharfstein
(1990) begins by deriving the optimal terms of a contract for a firm to raise external funds in a
two-period setting with two firms. However, the resulting contract shuts down the firm (no re-
financing is given) when it encounters a state of low revenues in the first period. This solution is
no longer optimal when a competing firm can pay a fixed cost to increase the probability of the
firm having low revenue, then leaving the competitor with monopoly profits in the second
period. If firm A has a refinancing contract and firm B has an opportunity to prey, the condition
for the predation to occur is:
(β
2
– β
1
)(μ – θ)(π
m
– π
d
) > c,
where (β
2
– β
1
) is the increase in probability of firm A receiving refinancing given high
repayment, (μ – θ) is the reduction in probability of high repayment of firm A to its investors
Internal Capital Markets and Competitive Threats
14
given predation, (π
m
– π
d
) is the increased profitability to firm B if A is not refinanced, and c is
the cost of predation paid by firm B. This modeling is relevant to considerations of the number of
competitors faced. While their model is designed to describe two firms, it can be thought of as
the two most similar firms within a larger product market. Thus, the more competitors that
operate in the market, the more closely related each firm is likely to be to its nearest rival. When
the product offerings of firms are overlapping, firms can more easily influence their rivals’ sales
through strategies like pricing, imitation, and advertisement. That implies a lower value of c and
a higher value of μ – θ since it is less costly and easier to have influence. Based on the condition
given in the Bolton-Scharfstein model, this increases the likelihood of a firm facing predation.
Having cash reserves can counteract this by allowing for continuation of the firm even if
revenues are low by either repaying investors from cash or financing investments with it. Thus,
having deep pockets is advantageous in this setting when the barriers to entry are lowered in an
industry.
A predation model which incorporates the value of cash reserves was developed by Chi and Su
(2013). They begin with costless external financing and no competitive threats where the
marginal value of cash is one; meaning paying out cash has no impact on value to investors.
Adding financial constraints raises the value of cash and subsequently allowing predation at a
convex cost to the rival further increases the value of cash since strong financials reduce the
optimal choice of the rival’s predation.
B. Aggressive Pricing
The case for cash reserves being necessary to increase market share by aggressive pricing
strategies was formulated by Telser (1966). By maintaining a “long purse” a firm can threaten to
Internal Capital Markets and Competitive Threats
15
undercut its competitors. This results in either driving the rival out of the market or forcing it to
open itself to acquisition. The aggressive practices will not be taken in equilibrium since the cash
reserves themselves are enough to give credibility to the threat. Additionally, the potentially
targeted firms have an incentive to hold cash to gain bargaining power in takeover negotiations.
Their ability to last longer in competition should negotiations fail allows them to demand a
higher price in the deal. Thus, in highly competitive markets, all firms have an incentive to
maintain cash reserves.
C. Entry Deterrence
Liquidity can be important for incumbent firms to prevent potential entrants from joining their
market space. The model of Benoit (1984) outlines the progression of competition with an
entrant who, with positive probability, will fight as long as possible to survive should the
incumbent choose not to accommodate the entry. Liquidity is an important factor to both the
existing monopolist and the entrant because those determine the willingness of the former to
fight a rival and the “staying power” of the latter (i.e., the amount of time it can survive given
efforts by the incumbent). In their model, the decision rule of the potential entrant is r(0) > 1.
The function r(0) is strictly increasing in (L – N) where L is the number of periods the incumbent
is willing and able to fight the entrant and N is the number of periods the entrant can survive the
fighting. Thus, each firm has an incentive to achieve a high level of liquidity.
D. Investment
Firms that face heavy competition cannot approach investment decisions the same way
monopolists do. The traditional view of investments is that opportunities contain an option to
wait. That allows the firm to resolve uncertainty about the cash flows of the project and raise
Internal Capital Markets and Competitive Threats
16
necessary funding. This also leads to an execution threshold where the minimum NPV required
to execute a project is significantly greater than zero. However, the presence of competitors with
access to the same opportunities removes this inherent option. The continuous-time Cournot
model of Grenadier (2002) shows that a large number of competitors makes the firm take action
quickly to avoid being left behind and also pushes the execution threshold closer to zero. Thus,
as the number of competing firms increases, each firm is required to make investments with very
short notice, the payoffs of which are less certain. Lyandres and Palazzo (2012) construct a
model that ties the cash reserves of a firm to its R&D and production decisions. The optimal
level of cash in the beginning stage is chosen to commit to producing output later on since
internal funds are less costly than external ones. Since output by the firm lowers the profitability
of its competitors, the firm can deter competition in R&D preemptively by raising cash reserves
initially. These models are relevant to the study of cash holdings because of the liquidity
requirement of firms that take on many projects with little lead time. If firms must invest no later
than their rivals do, there may not be sufficient time to raise funds externally. Large cash
reserves can give the firm the means as well as the credibility to invest aggressively.
E. Internal Capital Markets
There is a healthy theoretical literature on the connection between diversification and
competition. Perhaps the most popular view is that conglomerate firms have a competitive
advantage due to their financial flexibility.
1
This advantage stems from their access to external
financing and their ability to pool resources across divisions. Coinsurance models show that the
combination of multiple business units into one firm can reduce borrowing costs (Lewellen
1
Anti-trust authorities often consider the resources of a conglomerate to be a source of market power, as was
cited in the European Commission’s denial of the merger of GE and Honeywell (Case no. COMP/M 2220, July 2001)
Internal Capital Markets and Competitive Threats
17
(1971), Hann et al. (2013)). This access to financial resources can improve product market
performance. Regarding resource flexibility, Faure-Grimaud and Inderst (2005) directly
identified the potential strength of conglomerates using the predation framework of Bolton and
Scharfstein (1990). By incorporating a pooling of resources, they demonstrated that a
conglomerate with homogeneous divisions seeking external financing has a higher probability of
obtaining refinancing than if its segments were individual firms. They showed that segments of a
conglomerate are less likely to face predation threats by rivals than if they were stand-alone
firms and are more likely to be predatory themselves.
There are also theoretical arguments why internal capital markets may make conglomerates
weaker competitors. Matsusaka and Nanda (2002) developed a model of internal capital markets
and costly external financing in which financial flexibility of internal capital allocation can be a
competitive disadvantage because it prevents the firms from committing resources to be invested
in a particular segment. In contrast, a focused firm’s resources are “locked in” to be spent in that
market since it is costly for them to be redeployed. This inability of diversified firms to commit
to an industry can lead to predation by rivals. Similarly, when Faure-Grimaud and Inderst (2005)
considered firms with more profitable alternative investment opportunities, they found that the
business segments become vulnerable to predation since the firm finds it more profitable to
engage in winner-picking. A similar effect is demonstrated in a model of business group internal
capital markets by Cestone and Fumagalli (2005). They showed that the conventional wisdom –
that subsidiaries making monopolistic profits subsidize others in the business group facing new
competition – is not necessarily the optimal choice when resources are limited. Instead, the group
exercises winner-picking by only funding businesses in less competitive industries since it is too
costly to subsidize the threatened business.
Internal Capital Markets and Competitive Threats
18
Thus, there are two separate lines of theory regarding the impact of internal capital markets in
product market competition. The former implies that conglomerate firms would respond to
competition more aggressively, pulling on their firm-level cash holdings, free cash flow, and
external financing to fight off potential entry. This would be evident by higher levels of
investment and more aggressive pricing in the affected industry. The latter would predict a
winding down or exiting from the affected industry, exhibiting low investment and less
aggressive pricing. In the following chapters, I investigate which effect is most dominant
empirically.
Internal Capital Markets and Competitive Threats
19
Chapter 2: Data and Methods
A. Segment Data
I use the Compustat Segments data file for the classification of conglomerate firms. I began by
obtaining all business segments having non-missing primary SIC codes and segment annual sales
of at least $1 million. I then collapse each firm’s segments having common primary 4-digit SIC
industry codes and aggregate their financials. To adequately cover the history of tariff cut events,
I collect data from 1976-2005.
2
A firm is then considered “diversified” if it has two or more
business segments with different industry classifications at the 4-digit level and “single-segment”
if it has one. The results I present are robust to alternative definitions of diversification.
3
This
results in 28% of manufacturing firms being considered diversified and 49.6% of segments
labeled as belonging to a diversified firm. Business segments with undefined industry codes,
ending in “0” or “9”, are included in classification of organizational form, but not used in
regressions.
I am also able to obtain sales, assets, capital expenditures, and operating profit data from the
segments file. These are used in multivariate models and are defined in Appendix B. Summary
statistics are presented in Table 1. Following Dimitrov and Tice (2006), I calculate a measure of
the segments’ price-cost margins by taking the ratio of segment operating profit to segment net
sales. While this gives the average – rather than marginal – profit per unit, it can still give
insights into changes that occur over time. All variables are winsorized at 1% and 99%.
Additionally, observations having an investment growth rate of more than 300% are not included
2
FASB Statement of Financial Standards No. 131 expanded the reporting of business segments beginning in 1998.
Restricting the sample to years prior to 1998 does not qualitatively change the results.
3
The main results are robust to alternative definitions of diversified, including use of the 2- or 3-digit SIC code
rather than 4-digit.
Internal Capital Markets and Competitive Threats
20
in models with investment growth as the dependent variable.
4
For regression models, I consider
only segments that have at least four years of data and that operate in industries with at least two
peers.
B. Import Tariff Data
To identify industry years with substantial exogenous shocks to competition, I follow the method
used by Fresard (2010) in utilizing a series of large, industry-level import tariff reductions. I
begin by collecting import, export, and tariff rate data. Feenstra (1996) provides the time series
of data with concordances from the 10-digit Harmonized System to 4-digit SIC industry
classification for the period 1974-1992. Additionally, an update of the data provided by Schott
(2010), using concordances from Pierce and Schott (2009), covers the period 1989-2005.
5
These
datasets span 449 manufacturing industries, 134 of which appear in Compustat.
As described by Feenstra et al. (2002), the tariff rate can be calculated as the total duties
collected divided by the total customs for each 4-digit SIC industry in every year of the
combined period, 1974-2005, unless directly given. Similar to Fresard and Valta (2012), I
classify a large tariff cut as a reduction exceeding three times the average absolute change
experienced by the industry in the time series of data.
6
To ensure that the observed cut is not a
transitory anomaly, the cut will not be counted if it is offset with an equally large raise in rates
over the following three year period. Since the tariff data covers manufacturing industries, all
tests I perform are restricted to SIC codes beginning with “2” or “3”. SIC codes ending in “0”
and “9” are also excluded, following Clarke (1989) and Fresard (2010).
4
Keeping these observations does not qualitatively change the results. However, point estimates become much
higher in magnitude when they are included.
5
I thank The Center for International Data and Peter Schott’s International Economics Resource Page for providing
access to these datasets.
6
Results are robust to alternative definitions, including 2.5 and 3.5 times the mean.
Internal Capital Markets and Competitive Threats
21
After merging with the business segment file, a total of 79 industry-years in 67 unique industries
are identified as undergoing a tariff reduction. Figure 1 displays the count of segments affected
by tariff cuts over the sample period. Fortunately, the cuts are not isolated to one period of time.
This allows for an analysis using a large set of panel data spanning three decades. Additionally,
the large number of affected business segments provides the power necessary to disentangle
differences that take place between types of organizational form. There are 18 years in the
sample that include at least one tariff cut. However, some instances of trade liberalization are
evident, where over 15 industries undergo a cut within a two-year period. Fresard (2010) shows
that the tariff reductions correspond to large jumps in imports in the affected industries, which
justifies their use as a valid instrument for competition. Additionally, there is little evidence that
the cuts are chosen in a manner correlated with firm characteristics and they do not appear to be
anticipated. In regression results that follow, I perform placebo tests two years prior to the cuts in
order to confirm this characteristic.
C. Covariate Balance
The fact that diversification is an endogenous decision by the firm has been well documented by
prior research (e.g., Campa and Kedia (2002), Hyland and Diltz (2002)). It is, however,
important to note that financial data in this paper is reported at the segment level. This may
alleviate concerns about fundamental differences that exist with firm-level data. However, the
classification of belonging to a diversified firm or being a stand-alone firm does invite questions
about endogeneity, primarily that there could be large differences in the distribution of covariates
between groups. To address this issue, I generate a propensity score for each segment-year to
estimate its probability of being part of a multi-division firm as opposed to being stand-alone. As
covariates in the prediction, I use segment ROA, size, and age. Using a trimming method similar
Internal Capital Markets and Competitive Threats
22
to that proposed by Crump, Hotz, Imbens, and Mitnik (2009), I drop all observations which have
a propensity score that deviates from the mean of 0.496 by more than 0.2. Ideally, the final
distribution of propensity scores should have a large mass at the mean. Figure 2 gives a
histogram of the original estimation followed by a second histogram after trimming has taken
place. This process ensures that there is substantial overlap in the values of covariates between
segments of diversified and focused firms and gives more confidence to the results of the
regression models that follow.
7
7
Without eliminating extreme propensity scores, most tests have similar results, often with more significance
since they utilize more observations.
Internal Capital Markets and Competitive Threats
23
Chapter 3: Investment Decisions
A. Investment Behavior
In this section, I examine the differences in investment behavior between diversified and single-
segment firms. Investment is a channel through which managerial decisions can affect
performance in the current period as well as in the future by innovating, increasing capacity, and
lowering future production costs. If diversified firms wish to commit themselves to their
industries of operation, they will maintain investment levels at or above those of single-segment
rivals following shocks to competition. If they instead see it best to shift resources to other
segments, their investment levels will fall short of their rivals and their competitiveness in the
affected industries will be lessened.
I begin by viewing the univariate differences in investment between diversified and stand-alone
segments surrounding competition shocks. In Figure 3, the average investment growth rates for
segments affected by tariff reductions are presented for a three year period centered at the year of
the cut. A large discrepancy is apparent at time 0 where single-segment firms heavily outspend
their diversified rivals who actually cut investment growth from the previous year where the two
were roughly equal. The relationship persists for the following year as well, although there is
substantial convergence.
The results demonstrated graphically seem to give support for the theory that diversified firms
shift resources away from threatened segments. However, to make causal statements, it is
important to conduct a research design that accounts for segment characteristics and includes
actions of firms not affected by the competition. In Table 3, I perform difference-in-difference
Internal Capital Markets and Competitive Threats
24
panel regressions to estimate the influence of organizational form on the response to competition.
The regression models follow the form:
Investment Growth
t+1
= α
i
+ η
t
+ δ
1
×Diversified
i,t
+ δ
2
×Diversified
i,t
×Tariff Cut
i,t
+ δ
3
×Tariff Cut
i,t
+ β’X
i
+ ε
i,t
where each observation is a business segment, i, at time t. The dependent variable is the
percentage increase in capital expenditures from year t to t+1. The dummy variables
“Diversified” and “Tariff Cut” identify whether the segment belongs to a diversified firm and
whether the segment’s industry experienced a tariff cut at year t, respectively. In Appendix A, I
extend the window to two years (indicating a cut at year t or t-1) and find similar results to those
with a one-year window. The coefficient of interest in this model is δ
2
, the estimate of the effect
of being diversified during a period of increased competition. Each regression includes segment
and year fixed effects. Segment fixed effects de-mean each segment’s time series of data,
removing any characteristics that do not vary over time. This does, however, dictate the
interpretation of the diversified firm dummy variable. With fixed effects, it now represents the
time series – rather than cross-sectional – difference in investment associated with organizational
form. In other words, it is only relevant for segments that begin as a single-segment firm and
later become diversified or vice versa. The interaction between diversification and the tariff cut
indicator is not similarly affected by fixed effects since no segment experiences a cut every year
of its existence in the dataset.
I now present results of panel regressions with models following the design outlined above. All
regressions include segment and year fixed effects with standard errors that are robust to
heteroskedasticity and clustered by firm. In Table 2, column 1 presents the most basic test by
Internal Capital Markets and Competitive Threats
25
only using segment and year fixed effects and estimating the difference-in-difference parameter.
With no control variables, we observe a highly significant -13.46% effect of diversification
during competition shocks. This is consistent with the univariate representation of investment
and, again, supports the theory that diversified firms shift away from industries threatened by
entrants. It is also a sizeable magnitude since it is near the sample mean capital expenditure
growth rate of 18.16% across all manufacturing segments. In column 2, I include segment
covariates, each of which is highly significant in explaining investment growth. I find a very
similar coefficient on the interaction term to what was found in the previous test, at -13.69%.
The model in column 3 controls for the potential time-varying effects of segment characteristics
by including interaction terms between segment covariates and the tariff cut indicator. This setup
is used to rule out the possibility that the underlying driver of the effect is a characteristic of the
segment that is correlated with diversification and also allows for a comparison of the magnitude
of the effect of diversification to other segment characteristics. Again, the model estimates a
nearly 14% reduction in investment growth compared to single-segment firms also experiencing
a tariff cut. It is important to see the diversification interaction remain highly statistically
significant whereas none of the other three interaction terms seem to have explanatory power. By
allowing the effect of ROA, size, and age to vary during the competitive shock, it is clear that it
is actually diversification and not a correlated segment characteristic driving the result.
There are two important assumptions that must be made to have confidence in the validity of the
coefficients in models 1-3. First, it must be the case that single-segment firms and segments of
diversified firms are comparable. To address this, I estimated the propensity of being diversified
based on size, ROA, and age, as was outlined in the previous section. This ensures sufficient
overlap of covariates between the two groups. Further, I include regression models that interact
Internal Capital Markets and Competitive Threats
26
the tariff cut indicator with each of the segment covariates to ensure that none of them is driving
the estimate of the effect of diversification during the cut.
Second, the industries affected by tariff cuts must not be fundamentally different from others in
a way that would bias the results. Fresard (2010) shows that industry cash, leverage, and
performance are not correlated with selection for tariff cuts. He also gives evidence that the
shocks are unanticipated and lead to significant increases in imports. I also address this issue by
performing placebo tests where the variable Placebo Cut
t
identifies segments that will be affected
by a cut in two years. If the results were driven by industry characteristics, the coefficient on the
placebo interaction term would be similar to the one in the actual tests. Conversely, low
significance on the placebo estimate supports the validity of the tariff cut as an instrument.
Columns 4 and 5 test for differences in investment using a placebo date for the shock, moving it
two years prior to the actual date. Column 4 gives a small and insignificant difference-in-
difference estimate of 1.27%. Column 5 adds segment covariates and gives a similar estimate of
1.75%. Thus, the same segments exhibiting a large differential in investment following the tariff
cut show no such effect two years prior. This is evidence that industry selection is not driving the
significant estimates in the first three tests.
The theoretical models that predict reduced investment spending on threatened business
segments rely on the assumption that there is a constraint to firm liquidity and raising outside
funds is costly (Matsusaka and Nanda (2002), Cestone and Fumagalli (2005)). In order to
confirm that liquidity is playing a role in the cutbacks observed by diversified firms, I calculate
the firm-level net debt (total debt minus cash holdings, divided by asset) of all diversified
business segments in the year of a tariff reduction. Firms that are close to their debt capacity and
Internal Capital Markets and Competitive Threats
27
have low levels of cash on hand are the ones predicted to be forced to cut back on investment
spending. Figure 4 graphs the average investment growth across terciles of industry-adjusted net
debt, where tercile 3 can be described as the sub-sample of liquidity constrained firms. Indeed,
this group experiences the lowest rates of investment growth, 16.8% lower than the mean of
tercile 1. The difference has a t-statistic of 2.48. Clearly, a large portion of the reduced spending
by diversified firms can be attributed to those with liquidity constraints.
The tests in Table 2 related to capital expenditure outlays. Theory would suggest that firms
committed to the industry would make investments that can cut costs of production. In order to
understand how the spending is allocated, I next consider changes in fixed assets. To measure
this, I calculate the annual growth rate of identifiable assets at the segment level. This represents
an increase in infrastructure that could lead to more efficient production, thereby cutting costs
and increasing capacity in the future. After changing the dependent variable to segment asset
growth, estimates of the differences between conglomerate segments and stand-alone segments
are presented in Table 3.
The estimates are qualitatively similar to those with capital expenditures. The model in column 1
shows that multi-segment firms are not keeping pace with asset growth of single-segment rivals,
as they experience growth 5.9% lower than their focused rivals. After adding segment covariates
in column 2, the difference-in-difference estimate is nearly unchanged at -6.07%, despite the fact
that the covariates themselves are highly relevant. In column 3, the effect of segment size during
the cut seems to be significant in explaining asset growth. Again, the influence of diversification
persists with an estimated value of -5.21%. The placebo estimate in column 4 is much smaller in
magnitude and not close to statistical significance. Adding covariates in column 5 further reduces
the magnitude of the estimate.
Internal Capital Markets and Competitive Threats
28
These results on asset growth provide a robustness interpretation to the tests of capital
expenditures. Essentially, examining changes in investment spending alongside growth in fixed
assets provides two measures of the firm’s commitment to the industry. The change in assets in
the segments of diversified firms represents both new cash outlays as well as the shifting of
assets between business segments.
Since the previous tests only include firms that remained in the industry following the shocks,
survivorship is a potential concern in interpreting the estimates. In order to examine the exit of
segments in industries experiencing large tariff cuts, I track the firms that were assigned a value
of one with the Tariff Cut variable for two years after the cut. Here I do not consider firms
entering the industries in order to follow the original firms only.
Table 4 shows that there is a sizeable amount of exit that takes place in the two years following
the tariff cuts, with more than 20% of segments no longer listed in the industries. However, the
rates of exit between diversified and focused firms do not greatly differ. This does not help to
answer which theory of conglomerate response dominates, but it does help to interpret the
regression results. It suggests that analysis of investment decisions is not biased by survivorship
issues since each group exits at a similar rate. If anything, the slight decrease in the proportion of
segments belonging to diversified firms could imply the ones remaining are those more
committed to performing strongly in the industry. Thus, results showing lower investment by
conglomerates should be viewed as conservative estimates in this regard.
B. Resource Allocation
Theory suggests that if diversified firms wind down operations in threatened industries, they will
focus on others in which they operate. This implies a reduction in the affected industry, but little
Internal Capital Markets and Competitive Threats
29
change in investment by the firm as a whole. Since it appears that multi-segment firms do cut
investment to segments impacted by competition, I next approach the question of whether all
segments are given reduced investment or if the managers shift firm-level resources to
unaffected business segments. A previous study by Lamont (1997) found that all segments of a
conglomerate show less investment due to cash flow shortfalls when a negative demand shock
impacts its primary industry. With this competing cash flow effect in mind, I investigate the level
and relative amount of investment allocation between segments of a diversified firm to determine
whether all are proportionally cut or if resources are shifted to alternative segments.
I restrict the sample to multi-segment firms that experience a tariff cut in at least one of their
industries at any point in time. With this panel of diversified firm data, I perform regressions
predicting both segment-level investment growth and the segment’s allocation of total firm
investment. The treatment variable Unaffected indicates that the segment is in a firm
experiencing a tariff cut in a different segment in that year. Therefore, its coefficient represents
the indirect effect of competition to the rest of the firm.
Column 1 of Table 5 shows that growth in capital expenditures by these unaffected segments is
slightly negative, but not statistically significant. The multivariate model in column 2 again
shows a slightly negative but insignificant coefficient on the treatment variable. Industry Q, the
average Tobin’s Q of single-segment firms in that segment’s industry, is a strong positive
determinant of future investment with an estimated 10% increase in investment for each
additional unit increase of Q. The interaction term between Unaffected and Industry Q shows
that segment-level investment is more sensitive to investment opportunities in the years when the
firm is shifting resources away from a threatened segment, although the t statistic is just above 1.
Note that Industry Q has been demeaned to maintain interpretability of the treatment variable.
Internal Capital Markets and Competitive Threats
30
For columns 3 and 4, I calculate each segment’s share of the firm’s investment in each year and
use the change from time t to t+1 as the dependent variable. Column 3 shows that unaffected
segments gain 1.37% of total firm-level investment funds in the year following a competition
shock. After adding covariates in column 4, the estimate moves to 1.6% and is still highly
statistically significant. Although neither has a significant t statistic, the effect of the segment’s Q
is positive and its magnitude is over 50% higher in the periods following a cut, as seen in the
interaction term. These results imply that diversified firms do not cut all investments equally
when they experience foreign import competition. Rather, they allocate firm investment
resources to other segments. This effect is observed in spite of the previously noted cash flow
effect documented by Lamont (1997). The results support the implications of the model
presented by Matsusaka and Nanda (2002) where a diversified firm optimally invests in other
segments when one encounters a negative shock.
One shortcoming of the analysis of unaffected segments’ spending is that the theory does not
provide predictions regarding the timing of resource shifting. Diversified firms may alter their
financial activities in the interim while planning for future investment projects. For that reason, I
track the cash holdings of diversified firms to better understand where the unspent funds are
going. Since the absolute changes in investment by other segments do not seem to offset the
large reduction experienced by the affected segment, the firm’s cash reserves can be
hypothesized to increase. In Figure 5, mean cash holdings percentages are presented for
diversified firms that undergo a tariff cut to a segment accounting for at least 20% of firm sales.
Clearly, cash holdings rise after the event year, moving from 9.4 to 13.4 percent of assets.
Another approach that tests the theory of shifting to more promising industries is to estimate the
effect of alternative investment opportunities rather than diversification itself. This also provides
Internal Capital Markets and Competitive Threats
31
robustness to the results that rely specifically on the diversification dummy indicator. I first
calculate the average value of Tobin’s Q of single-segment firms in each industry-year as a
measure of a segment’s investment opportunities. Then, the variable Alternative Q
t
is set equal to
the average Q of other industries in which the diversified firm operates. For single-segment
firms, it is set to one since their immediate alternative is to hold cash. Thus, rather than testing
the effect of diversification, the following models in Table 6 test the effect of having promising
alternative investment opportunities. The use of the Q measure as opposed to the diversification
dummy also gives more information about the time-series variation within diversified firms.
While fixed effects limit the interpretation of the dummy variable, they simply remove the mean
from the alternative Q variable, allowing it to explain investment decisions over time.
Column 1 gives a highly significant estimate of the effect of alternatives during a shock to
competition. For every one unit increase in the Q of alternatives above the baseline of one,
investment in the segment is cut 13.63%. After controlling for covariates in column 2, the
estimate of the effect of alternative investment opportunities is significantly negative in the full
sample (outside of tariff cut shocks) at -3.53%. This means that having outside opportunities
draws away resources in general, but the effect is boosted by 13.11% in years when competition
increases, for a combined effect of nearly 17%. The result persists with the inclusion of more
interactions in column 3 as none of the interaction terms is statistically significant. There is also
no significance for placebo years prior to the tariff cuts, with an estimated effect of -2.04 in
column 4 and just -0.96 with segment covariates in column 5. These results give support to the
theory of shifting resources to more promising opportunities when a business segment is
threatened with competitors.
Internal Capital Markets and Competitive Threats
32
Chapter 4: Pricing and Performance
In this section, I examine the pricing, market share, and valuation changes by organizational
structure. The results from the previous section indicated that diversified firms invest less than
focused firms in industries affected by exogenous increases to competition. I now examine
pricing decisions and trace the impact of investment and pricing policies on product market and
stock market performance.
A. Price-Cost Margins
There are theoretical foundations to predict that either conglomerate segments or stand-alone
segments would price more aggressively in periods of heightened competition. As described by
Faure-Grimaud and Inderst (2005), conglomerates can be committed to their industries and use
their resources and access to external capital to undercut competitors. On the other hand,
Chevalier and Scharfstein (1996) model how financially constrained firms have an incentive to
lower prices and increase short-term revenue during downturns. Single-segment firms can be
thought of as being more constrained since they lack internal capital markets and their borrowing
costs may be higher.
I conduct panel regressions with the pricing margin as the dependent variable in Table 7. As a
measure of pricing, I divide the segment’s operating profit by its net sales for the year. This is
meant to capture the average price-cost margin for the segment’s goods sold. Note that ROA is
not used as a control variable since operating profit is already on the left-hand side of the
equation. There are two important results in column 1. First, the differential between diversified
firm segments’ margins and those of focused firms widens by 1.5% following shocks to
competition. Second, it is estimated that single-segment firms reduce margins by nearly 2% in
Internal Capital Markets and Competitive Threats
33
this period, as seen in the coefficient on Tariff Cut. That means both groups are cutting margins,
but focused firms are cutting theirs significantly more. In column 2, the difference-in-difference
estimate is similar at 1.54% and both groups again are seen lowering margins. In column 3, the
interaction of covariates with the tariff cut indicator adds little to the model and the estimated
effect of diversification remains significant at 1.62%. The placebo test in column 4 shows no
significant differential between groups, nor in the overall effect of the tariff cuts, suggesting the
pricing change is caused by import competition. Adding size and age covariates to the placebo
test gives similar results in column 5.
These results contradict theories of conglomerate firms competing on pricing with lowered
margins and support the idea that single-segment firms are more aggressive in their pricing when
threatened. The evidence does not support the idea that conglomerates are using funds from other
business operations to subsidize performance in the threatened industry. One clear limitation of
the analysis is that it is only possible to estimate the average – rather than marginal – pricing
strategy of the firms.
B. Market Share
The previous results have shown that diversified firms are likely to maintain higher margins than
their focused rivals when facing increased competition. Assuming the demand for their products
is fairly elastic, this pricing should result in significant changes in market share. Additionally, the
lower levels of capital expenditures and asset growth likely result in less production capacity
Since higher pricing and lower investment imply a relatively lower rate of sales growth, I now
turn my attention to changes in market share. Observing differences in market share would
Internal Capital Markets and Competitive Threats
34
provide more evidence that the previous findings are real and have an impact in the product
markets.
To determine the relative changes in ales performance, I run tests with the same structure as was
performed with investment growth.
8
The dependent variable is a measure of the change in
market share as defined by Fresard (2010) in his study of the impact of cash holdings on
performance. It is calculated as the segment’s percentage growth in sales minus the growth of
sales in the industry. Note that these figures estimate the change in the domestic market share
since data on foreign firms is not included.
In Table 8, the tests show that conglomerate segments tend to lose market share following tariff
cuts. In column 1, the estimated effect of being diversified following the shock is a 4.84%
shortfall in sales relative to the industry. The effect remains significant at -4.25% when
covariates are included in column 2. While inclusion of the interaction terms in column 3 reduces
the estimate of the effect of diversification to -3.3%, it remains statistically significant at the 10%
level. The placebo tests in columns 4 and 5 show that there was no significant difference prior to
the cuts. These results are consistent with the theory of conglomerates shifting focus out of
threatened industries and link the observed shortfalls in investment spending and higher pricing
margins to market share losses.
One source of ambiguity in interpreting results based on sales growth is that sales are determined
by the product of price and quantity. Since there are estimates showing that diversified firms
maintain higher prices, the fact that they lag in sales growth is most likely caused by a
significantly lower rate of growth in actual quantity produced and sold. In fact, with the observed
8
Results are economically and statistically more significant when considering changes up to two years after the
shock rather than one. See Appendix A for the two-year window.
Internal Capital Markets and Competitive Threats
35
differential in prices, these estimates of changes in sales growth should provide a lower bound
for the difference in quantity.
C. Stock Market Performance
Ultimately, it is important to understand whether the observed differences in investment and
pricing between diversified and focused firms have an impact on their market value. Identifying
the effect on valuation allows for interpretation of whether the actions taken are value-creating or
value-destroying. The theories cited regarding behavior by conglomerates do not give direct
predictions of differences in stock returns. While diversified firms seem to be more willing to
pull out of industries facing increased imports, they can maintain their value by concentrating on
other industries. Conversely, focused firms seem to be more committed, but they must withstand
lower pricing margins and put forth capital outlays to grow their domestic market share. In this
section, I formally test the impact of the decisions made by diversified firms to address these
open questions.
In this setting, comparisons of stock returns are difficult since only one segment of a diversified
firm is exposed to the industry-level shock, but valuation data is only available at the firm-level.
In order to analyze the change in value of a conglomerate firm, an estimate of the counterfactual
aggregation of its segments’ values as stand-alone firms is needed. To address this issue, I apply
the synthetic controls method, originally introduced by Abadie et al. (2010). It constructs a
“synthetic” version of each treated observation using a weighted average of selected control
observations. The weights for the controls making up the synthetic are calculated by minimizing
the difference between treated and synthetic values in the pre-treatment period. The treatment
Internal Capital Markets and Competitive Threats
36
effect can then be estimated by comparing differences between the actual and synthetic values
following implementation of the treatment.
I consider each diversified firm impacted by a tariff cut to be treated beginning in January of the
year of the tariff reduction. For each diversified firm, I collect all single-segment firms that
operate in the industries it spans and identify these as the pool of potential control firms. A set of
weights is then calculated for the control firms in order to minimize the difference in monthly
stock returns between the diversified firm and its synthetic diversified firm prior to the year of
the event. This gives a synthetic version of each conglomerate firm comprised of single-segment
firms, allowing for an estimate of the necessary counterfactual value. After the weights have
been determined, I sum the weights of control firms in the affected industry in order to measure
the diversified firm’s exposure to the shock. Combining weights with stock returns in the event
year gives both an estimate of the unobservable return to unaffected segments of the
conglomerate and the return to single-segment firms in the affected industry. Thus, the
difference between treated firms and synthetic controls should reveal the difference in valuation
due to diversification following the competition shock.
I conduct the above process for 872 diversified firms; however, I only consider cases where
control firms in the affected industry form at least 20% of the synthetic firm. Figure 6 graphs the
average stock return (above the market) for 512 diversified firms as well as their corresponding
synthetic controls. The vertical line indicates where the matching process ends and the treatment
year begins.
9
Note that the figure combines firms from various industries and years, so the actual
points when the tariff reductions are enacted will be spread through the year. Additionally, only a
portion of the treated and control firms is actually exposed to the shocks. Therefore, the main
9
Similar results are found when the end of the matching period is moved 6 and 12 months prior to the event year.
Internal Capital Markets and Competitive Threats
37
aspect of interest is the difference between the treated and synthetic values rather than their
individual movements. In the end, there appears to be very little difference in returns between the
treated and control firms in the year of the shocks. In fact, the difference in cumulative returns
over the year is just 0.2% with a t statistic of 0.07.
This test does not indicate that conglomerate firms lose value compared to their single-segment
rivals in industries experiencing new competition. By shifting resources away from the impacted
segments, they seem to earn returns roughly equal to those of focused firms that stay and invest.
While their individual decisions differ, they may very well be operating efficiently given their
respective investment opportunities and constraints.
Internal Capital Markets and Competitive Threats
38
Conclusion
Most of the research regarding diversification focuses on the financial and operational effects on
the firm itself without considering the effects on strategic behavior in competing with rival firms.
Theories supporting both the efficiencies and inefficiencies of conglomerates are often extended
to comment on diversified firms’ competitiveness in product markets. However, the overall
effect of internal capital markets and multi-divisional management structures has not undergone
thorough empirical testing in the setting of product market interactions. In this paper, I directly
test the differences in responses to competition shocks between segments of diversified firms and
single-segment firms. The evidence from testing implies that diversified firms are less committed
to their performance in threatened product markets as they exhibit low investment relative to
focused rivals when competitors are given access to the industry.
The idea that diversified segments are weaker competitors when dealing with competition shocks
does not imply that internal capital markets are inefficient. In fact, it can be optimal for a
diversified firm to channel resources into more promising industries than one experiencing a
flood of international competition. The main insight revealed is that internal capital markets are
more likely to be used to channel resources away from a threatened segment than to bolster it.
This implies that conglomerates have an inherent vulnerability to predation, given their ability to
redeploy resources to other industries.
The findings in this paper are meant to build upon prior research on the resource allocation of
conglomerate firms. The results support a form of winner-picking where a conglomerate that
faces heightened competition in one industry shifts funds to its other segments and does not
subsidize the threatened segment. As demonstrated by tests of stock market returns, this effect is
Internal Capital Markets and Competitive Threats
39
not associated with reduced firm value compared to focused rivals. Conglomerates appear to
preserve value by maintaining focus on their more promising opportunities.
Internal Capital Markets and Competitive Threats
40
Figure 1: Number of Tariff Cuts over Time
This graph presents the number of business segments affected by significantly large tariff cuts over the period 1976-
2005. The condition for classification as a large cut is a rate reduction that is greater than three times the mean of the
absolute value of all rate changes in that industry. Industries are defined at the 4-digit SIC level.
Internal Capital Markets and Competitive Threats
41
Figure 2: Conglomerate Segment Propensity Score Distribution
These histograms give the distribution of propensity scores in the dataset of business segments. The propensities are
based on a probit regression where the dependent is 1 if the segment is part of a multi-segment firm and 0 if it is a
stand-alone segment. Segment ROA, size, and age are used as independent variables.
Internal Capital Markets and Competitive Threats
42
Figure 3: Investment Growth around Tariff Cut Events
The graph gives the average percentage growth in segment capital expenditures for segments in industries affected
by tariff cuts. Year 0 corresponds to the year of the cut.
Internal Capital Markets and Competitive Threats
43
Figure 4: Investment Growth and Net Debt
The chart shows growth of capital expenditures for segments of diversified firms affected by a tariff cut. The
segments are divided by tercile of industry-adjusted net debt (total debt minus cash holdings divided by assets) at the
firm level.
Internal Capital Markets and Competitive Threats
44
Figure 5: Diversified Firm Cash Holdings Around Tariff Cut Events
Cash holdings as a percentage of assets are presented for diversified firms with at least 20% of sales in an industry
affected by a tariff cut. Year 0 represents the year the cut took place.
Internal Capital Markets and Competitive Threats
45
Figure 6: Stock Returns to Diversified and Synthetic Diversified Firms
The synthetic control method constructs a “synthetic” version of a treated observation using a weighted combination
of observations in the control group. In this figure, diversified firms affected by a tariff reduction are matched to
single-segment firms operating in its industries based on monthly stock returns in the year preceding the shock
(month -11 to month 0). The vertical line is placed at the end of this year. The lines plot above-market average stock
returns for diversified firms and their synthetic controls.
Internal Capital Markets and Competitive Threats
46
Table 1: Summary Statistics
This table list the mean, standard deviation, and distribution percentiles for segment variables used in the study.
Percentiles 1 and 99 are presented since winsorizing is performed at the 1% level.
Variable Mean Std. Dev. 1% 25% Median 75% 99%
ROA (%) 10.50 13.33 -24.42 2.93 10.40 18.06 45.55
Size (log) 4.38 2.07 0.13 2.86 4.29 5.80 9.34
Age (log) 1.68 0.84 0.00 1.10 1.79 2.30 3.22
Investment Growth (%) 18.16 73.24 -91.24 -32.14 3.45 49.55 262.96
Asset Growth (%) 14.03 41.31 -57.91 -4.31 5.95 19.78 245.80
Price-Cost Margin (%) 5.82 15.30 -92.89 2.17 7.52 12.65 33.53
ΔMarket Share (%) 3.79 31.88 -66.44 -10.61 -0.02 11.84 164.45
Industry Q 1.44 0.48 0.82 1.10 1.32 1.66 3.03
Alternative Q 1.25 0.41 0.80 1.00 1.00 1.42 2.78
Internal Capital Markets and Competitive Threats
47
Table 2: Investment Growth with Competition Shocks
The models cover a panel of manufacturing segment data from 1976-2005 where the dependent variable is the one-
year percentage growth of segment capital expenditures. Tariff Cut
t
is a dummy variable equal to one if a tariff cut
took place in year t in the segment’s industry. Placebo Cut
t+2
is equal to one if that segment will experience a tariff
cut two years later. All variables are defined in the appendix. Year and segment fixed effects are included in all
models. Standard errors are robust and clustered by firm.
Dependent: Investment Growth
t+1
(%)
Variable (1) (2) (3) (4) (5)
Diversified Firm
t
1.42 -6.54*** -6.52*** 1.17 -6.80***
(1.54) (1.53) (1.53) (1.55) (1.53)
Diversified
t
×Tariff Cut
t
-13.46*** -13.69*** -14.34***
(5.14) (5.05) (5.15)
Tariff Cut
t
3.20 4.90 1.16
(3.96) (3.87) (7.77)
Segment ROA
t
0.97*** 0.97***
0.97***
(0.03) (0.03)
(0.03)
Segment Size
t
-18.60*** -18.62***
-18.61***
(0.74) (0.74)
(0.74)
Segment Age
t
-7.39*** -7.41***
-7.42***
(1.07) (1.07)
(1.07)
ROA
t
×Tariff Cut
t
0.04
(0.20)
Size
t
×Tariff Cut
t
0.65
(1.21)
Age
t
×Tariff Cut
t
0.59
(3.44)
Diversified
t
×Placebo Cut
t
1.27 1.75
(5.43) (5.36)
Placebo Cut
t
-2.32 -3.61
(4.03) (3.97)
Year Fixed Effects Yes Yes Yes Yes Yes
Segment Fixed Effects Yes Yes Yes Yes Yes
Observations 64,047 64,047 64,047 64,047 64,047
R-squared 0.14 0.17 0.17 0.14 0.17
Internal Capital Markets and Competitive Threats
48
Table 3: Asset Growth with Competition Shocks
The models cover a panel of manufacturing segment data from 1976-2005 where the dependent variable is the one-
year percentage growth of segment total identifiable assets. Tariff Cut
t
is a dummy variable equal to one if a tariff
cut took place in year t in the segment’s industry. Placebo Cut
t+2
is equal to one if that segment will experience a
tariff cut two years later. All variables are defined in the appendix. Year and segment fixed effects are included in all
models. Standard errors are robust and clustered by firm.
Dependent: Asset Growth
t+1
(%)
Variable (1) (2) (3) (4) (5)
Diversified Firm
t
6.74*** -2.13** -2.15** 6.66*** -2.22**
(0.97) (1.05) (1.05) (0.96) (1.05)
Diversified
t
×Tariff Cut
t
-5.93** -6.07** -5.21**
(2.90) (2.57) (2.56)
Tariff Cut
t
4.09* 5.10** 12.71***
(2.36) (2.04) (4.59)
Segment ROA
t
0.64*** 0.64*** 0.64***
(0.02) (0.02) (0.02)
Segment Size
t
-26.93*** -26.89*** -26.93***
(0.64) (0.64) (0.64)
Segment Age
t
-11.95*** -11.91*** -11.96***
(0.71) (0.71) (0.71)
ROA
t
×Tariff Cut
t
-0.01
(0.11)
Size
t
×Tariff Cut
t
-1.51**
(0.63)
Age
t
×Tariff Cut
t
-1.22
(1.81)
Diversified
t
×Placebo Cut
t
-1.66 -0.59
(2.97) (2.68)
Placebo Cut
t
2.97 1.85
(2.25) (1.99)
Year Fixed Effects Yes Yes Yes Yes Yes
Segment Fixed Effects Yes Yes Yes Yes Yes
Observations 70,664 70,664 70,664 70,664 70,664
R-squared 0.22 0.35 0.35 0.22 0.35
Internal Capital Markets and Competitive Threats
49
Table 4: Industry Exit Following Tariff Cuts
This table presents the number of business segments operating in industries affected by tariff cuts, followed by the
number of those firms that remained operating in those industries one and two years after the cut took place. New
entrants are not included. The number of diversified firm segments gives the number of segments that are associated
with a diversified firm as of the year of the cut.
Year
Total
Segments
Diversified
Firm
Segments
Focused
Firm
Segments %Diversified %Focused
Tariff Cut (0) 2288 1053 1235 46.02% 53.98%
1 2019 922 1097 45.67% 54.33%
2 1781 798 983 44.81% 55.19%
Internal Capital Markets and Competitive Threats
50
Table 5: Investment in Unaffected Segments
The models cover a panel of segment data from 1976-2005. The dependent variable in columns 1 and 2 is the
segment investment growth rate (%). In columns 3 and 4, it is the change in the percent of firm capital expenditures
spent in the segment, expressed as a percentage. Industry Q is demeaned for interpretability purposes. Only
diversified firms that experience a tariff cut at some point are included. The models segments in include
manufacturing industries (first SIC digit of “2” or “3”). Year and segment fixed effects are included in all models.
Standard errors are robust and clustered by firm.
CAPX Growth %Firm CAPX
Variable (1) (2) (3) (4)
Unaffected Segment
t
-2.70 -1.35 1.37*** 1.60***
(2.67) (2.75) (0.50) (0.58)
Industry Q
t
10.03***
0.81
(3.10)
(0.62)
Unaffected
t
×Q
t
6.69
1.33
(6.45)
(1.57)
Segment ROA
t
0.55***
0.07***
(0.05)
(0.01)
Segment Size
t
-20.09***
-3.49***
(1.61)
(0.30)
Segment Age
t
-6.92***
-0.15
(2.09)
(0.35)
Year Fixed Effects Yes Yes Yes Yes
Segment Fixed Effects Yes Yes Yes Yes
Observations 19,081 19,081 20,120 20,120
R-squared 0.12 0.15 0.06 0.08
Internal Capital Markets and Competitive Threats
51
Table 6: Investment Growth and Alternative Investment Opportunities
The models cover a panel of manufacturing segment data from 1976-2005 where the dependent variable is the one-
year percentage growth of segment capital expenditures. Tariff Cut
t
is a dummy variable equal to one if a tariff cut
took place in year t in the segment’s industry. Alternative Q is the average Tobin’s Q of other industries in which
diversified firm operates and is equal to one for single-segment firms. Placebo Cut
t+2
is equal to one if that segment
will experience a tariff cut two years later. All variables are defined in the appendix. Year and segment fixed effects
are included in all models. Standard errors are robust and clustered by firm.
Dependent: Investment Growth
t+1
(%)
Variable (1) (2) (3) (4) (5)
Alternative Q
t
0.11 -3.53*** -3.52*** -0.16 -3.81***
(1.24) (1.23) (1.23) (1.24) (1.23)
Alternative Q
t
×Tariff Cut
t
-13.63** -13.11** -13.56**
(5.86) (5.89) (6.21)
Tariff Cut
t
13.89* 14.87* 11.81
(8.14) (8.10) (9.51)
Segment ROA
t
0.98*** 0.98*** 0.98***
(0.04) (0.04) (0.04)
Segment Size
t
-18.54*** -18.55*** -18.55***
(0.76) (0.76) (0.76)
Segment Age
t
-7.85*** -7.87*** -7.86***
(1.10) (1.10) (1.10)
ROA
t
×Tariff Cut
t
-0.04
(0.20)
Size
t
×Tariff Cut
t
0.79
(1.28)
Age
t
×Tariff Cut
t
0.48
(3.59)
Alternative Q
t
×Placebo Cut
t
-2.04 -0.96
(6.72) (6.61)
Placebo Cut
t
0.51 -1.79
(8.79) (8.66)
Year Fixed Effects Yes Yes Yes Yes Yes
Segment Fixed Effects Yes Yes Yes Yes Yes
Observations 61,815 61,815 61,815 61,815 61,815
R-squared 0.14 0.17 0.17 0.14 0.17
Internal Capital Markets and Competitive Threats
52
Table 7: Price-Cost Margin
The models cover a panel of manufacturing segment data from 1976-2005 where the dependent variable is the price-
cost margin of the business segment, estimated as operating profit divided by sales. Tariff Cut
t
is a dummy variable
equal to one if a tariff cut took place in year t in the segment’s industry. Placebo Cut
t+2
is equal to one if that
segment will experience a tariff cut two years later. All variables are defined in the appendix. Year and segment
fixed effects are included in all models. Standard errors are robust and clustered by firm.
Dependent: Price-Cost Margin
t
(%)
Variable (1) (2) (3) (4) (5)
Diversified Firm
t
1.07*** 1.21*** 1.20*** 1.10*** 1.24***
(0.27) (0.27) (0.27) (0.27) (0.27)
Diversified
t
×Tariff Cut
t
1.50** 1.54** 1.62**
(0.72) (0.72) (0.73)
Tariff Cut
t
-1.99*** -1.99*** -2.50**
(0.65) (0.65) (1.15)
Segment Size
t
0.35** 0.36** 0.36**
(0.18) (0.18) (0.18)
Segment Age
t
-1.15*** -1.17*** -1.15***
(0.20) (0.20) (0.20)
Size
t
×Tariff Cut
t
-0.19
(0.15)
Age
t
×Tariff Cut
t
0.74
(0.50)
Diversified
t
×Placebo Cut
t
-0.37 -0.29
(0.76) (0.76)
Placebo Cut
t
0.69 0.64
(0.66) (0.66)
Year Fixed Effects Yes Yes Yes Yes Yes
Segment Fixed Effects Yes Yes Yes Yes Yes
Observations
80,047 80,047 80,047 80,047 80,047
R-squared
0.65 0.65 0.65 0.65 0.65
Internal Capital Markets and Competitive Threats
53
Table 8: Market Share Growth
The models cover a panel of manufacturing segment data from 1976-2005 where the dependent variable is the one-
year percentage growth of segment sales minus industry sales growth. Tariff Cut
t
is a dummy variable equal to one
if a tariff cut took place in year t in the segment’s industry. Placebo Cut
t+2
is equal to one if that segment will
experience a tariff cut two years later. All variables are defined in the appendix. Year and segment fixed effects are
included in all models. Standard errors are robust and clustered by industry.
Dependent: ΔMarket Share
t+1
(%)
Variable (1) (2) (3) (4) (5)
Diversified Firm
t
2.98*** 1.61** 1.58** 2.90*** 1.53**
(0.68) (0.69) (0.68) (0.67) (0.68)
Diversified
t
×Tariff Cut
t
-4.84** -4.25** -3.30*
(2.06) (2.02) (1.90)
Tariff Cut
t,t-1
3.24* 2.87 6.55*
(1.78) (1.83) (3.92)
Segment ROA
t
-0.22*** -0.22*** -0.22***
(0.03) (0.03) (0.03)
Segment Size
t
-9.56*** -9.55*** -9.56***
(0.60) (0.61) (0.60)
Segment Age
t
-13.23*** -13.21*** -13.24***
(0.98) (0.98) (0.98)
ROA
t
×Tariff Cut
t
-0.09
(0.09)
Size
t
×Tariff Cut
tt-1
-0.49
(0.50)
Age
t
×Tariff Cut
t,t-1
-0.79
(1.35)
Diversified
t
×Placebo Cut
t
-0.12 0.72
(1.89) (1.84)
Placebo Cut
t
2.85 2.65
(1.89) (1.94)
Year Fixed Effects Yes Yes Yes Yes Yes
Segment Fixed Effects Yes Yes Yes Yes Yes
Observations 71,268 71,268 71,268 71,268 71,268
R-squared 0.23 0.27 0.27 0.23 0.27
Internal Capital Markets and Competitive Threats
54
Appendix A: Two Year Window for Treatment Variable
This table alters regression models previously presented by using a two year window for the Tariff Cut treatment
variable. It now takes the value of 1 for the year of the tariff cut and the following year. The models cover a panel of
manufacturing segment data from 1976-2005 where the dependent variable is the one-year percentage growth of
segment capital expenditures (1), segment assets (2), and segment industry-adjusted sales growth (3). All variables
are defined in the appendix. Year and segment fixed effects are included in all models. Standard errors are robust
and clustered by firm.
Dependent: Investment Growth (%) Asset Growth (%) ΔMarket Share (%)
Variable (1) (2) (3)
Diversified Firm
t
-6.52*** -0.97 1.10*
(1.53) (0.89) (0.66)
Diversified
t
×Tariff Cut
t
-9.39*** -3.03* -4.71***
(3.58) (1.64) (1.47)
Tariff Cut
t,t-1
2.70 1.41 3.41***
(2.74) (1.29) (1.12)
Segment ROA
t
0.97*** 0.59*** -0.16***
(0.04) (0.02) (0.02)
Segment Size
t
-18.60*** -20.01*** -8.29***
(0.75) (0.50) (0.34)
Segment Age
t
-7.23*** -10.17*** -10.77***
(1.06) (0.59) (0.49)
Year Fixed Effects Yes Yes Yes
Segment Fixed Effects Yes Yes Yes
Observations 64,034 70,156 70,975
R-squared 0.17 0.33 0.26
Internal Capital Markets and Competitive Threats
55
Appendix B: Variable Definitions
Asset Growth
t
= percentage change in segment identifiable assets, 100% * (ias
t+1
– ias
t
)/ias
t
Alternative Q
t
= mean of other segments’ Industry Q values if segment is part of a multi-segment
firm and 1 if it is a single-segment firm.
Change in Share of Firm Investment
t+1
= (segment capital expenditures
t+1
(capxs
t+1
) / total firm
capital expenditures
t+1
) - (segment capital expenditures
t
(capxs
t
) / total firm capital
expenditures
t
)
Diversified
t
= 1 if the segment belongs to a firm having more than one business segment at time
t, 0 otherwise
Industry Q
t
= mean of single-segment firm Tobin’s Q in the industry-year. Q is defined as: (total
assets (at) + market value of common equity (prcc_f * csho) – book value of common
equity (ceq) – deferred taxes (txdb)) / (.9*total assets + .1*market value of assets)
Investment Growth
t+1
= percentage change in segment capital expenditures, 100% * (capxs
t+1
–
capxs
t
)/capxs
t
ΔMarket Share
t+1
= percentage increase in segments net sales minus industry increase in sales,
100% * (sales
t+1
– sales
t
)/sales
t
– industry sales growth%
Price-Cost Margin
t
= segment operating profit (ops
t
) / segment net sales (sales
t
)
Segment Age
t
= natural log of the number of years the business segment has appeared in the
Compustat segments data up to time t
Segment ROA
t
= segment operating profit (ops
t
) / segment total identifiable assets (ias
t
)
Segment Size
t
= natural log of segment total identifiable assets (ias
t
)
Tariff Cut
t
= 1 if the segment operates in an industry experiencing a tariff cut at time t, 0
otherwise
Internal Capital Markets and Competitive Threats
56
References
Abadie, Alberto, Alexis Diamond, and Jens Hainmueller, 2010, Synthetic control methods for
comparative case studies: estimating the effect of California’s tobacco control program, Journal
of the American Statistical Association 105, 493-505.
Akdogu, Evrim and Peter MacKay, Investment and competition, Journal of Financial and Quantitative
Analysis 43, 299-330.
Bai, John, 2014, Organizational form and firms real actions: Plant-level evidence from import tariff
shocks, working paper.
Benoit, Jean-Pierre, 1984, Financially constrained entry in a game with incomplete information, RAND
Journal of Economics 15, 490-499.
Bolton, Patrick and David Scharfstein, 1990, A theory of predation based on agency problems in
financial contracting, American Economic Review 80, 93-106.
Boutin, Xavier, Giacinta Cestone, Chiara Fumagalli, Giovanni Pica, and Nicolas Serrano-Velarde, 2013,
The deep-pocket effect of internal capital markets, Journal of Financial Economics 109, 122-
145.
Campa, Jose Manuel, and Simi Kedia, 2002, Explaining the diversification discount, Journal of Finance
57, 1731–1762.
Cestone, Giacinta and Chiara Fumagalli, 2005, The strategic impact of resource flexibility in business
groups, RAND Journal of Economics 36, 193-214.
Chevalier, Judith and David Scharfstein, 1996, Capital-market imperfections and countercyclical
markups: theory and evidence, American Economic Review 86, 703-725.
Chi, Jianxin and Xunhua Su, 2013, Product market predatory threats and the value of cash holdings,
working paper.
Clarke, Richard, 1989, SICS as delineators of economic markets, Journal of Business 62, 17-31.
Crump, Richard, V. Joseph Hotz, Guido Imbens, and Oscar Mitnik, 2009, Dealing with limited overlap
in estimation of average treatment effects, Biometrika 96, 187-199.
Dimitrov, Valentin and Sheri Tice, 2006, Corporate diversification and credit constraints: real effects
across the business cycle, Review of Financial Studies 19, 1465-1498.
Duchin, Ran, 2010, Cash holdings and corporate diversification, Journal of Finance 65, 955-992.
Faure-Grimaud, Antoine and Roman Inderst, 2005, Conglomerate entrenchment under optimal financial
contracting, American Economic Review, 95, 850-861.
Internal Capital Markets and Competitive Threats
57
Feenstra, Robert, 1996, U.S. imports, 1972-1994: Data and concordances, NBER Working Paper 5515.
Feenstra, Robert, John Romalis, and Peter Schott, 2002, U.S. imports, exports, and tariff data, NBER
Working Paper 9387.
Fresard, Laurent, 2010, Financial strength and product market behavior the real effects of corporate cash
holdings, Journal of Finance 65, 1097-1122.
Fresard, Laurent and Philip Valta, 2012, Competitive pressure and corporate policies, working paper.
Grenadier Steven, 2002, Option exercise games: an application to the equilibrium investment strategies
of firms, Review of Financial Studies 15, 691-721.
Guedj, Ilan and David Scharfstein, 2004, Organizational scope and investment: Evidence from the drug
development strategies and performance of biopharmaceutical firms, NBER Working Paper
10933.
Hann, Rebecca, Maria Ogneva, and Oguzhan Ozbas, 2013, Corporate diversification and the cost of
capital, Journal of Finance 68, 1961-1999.
Hyland, David, and David Diltz, 2002, Why firms diversify: An empirical examination, Financial
Management 31, 51–81.
Khanna, Naveen and Sheri Tice, 2001, The bright side of internal capital markets, Journal of Finance
56, 1489-1528.
Lamont, Owen, 1997, Cash flow and investment: Evidence from internal capital markets, Journal of
Finance 52, 83-109.
Lewellen, Wilbur, 1971, A pure financial rationale for the conglomerate merger, Journal of Finance 26,
527-537.
Lyandres, Evgeny and Dino Palazzo, 2012, Strategic cash holdings and R&D competition: theory and
evidence, working paper.
Maksimovic, Vojislav and Gordon Phillips, 2002, Do conglomerate firms allocate resources inefficiently
across industries? Theory and evidence, Journal of Finance 57, 721-768.
Maksimovic, Vojislav and Gordon Phillips, 2008, The industry life cycle, acquisitions and investment:
Does firm organization matter?, Journal of Finance 63, 673-708.
Matsusaka, John and Vikram Nanda, 2002, Internal capital markets and corporate refocusing, Journal of
Financial Intermediation 11, 176-211.
Natividad, Gabriel, 2014, Financial capacity and discontinuous investment: Evidence from emerging
market multibusiness firms, Review of Financial Studies 26, 2317-2410.
Internal Capital Markets and Competitive Threats
58
Opler, Tim, Lee Pinkowitz, Rene Stulz, and Rohan Williamson, 1999, The determinants and
implications of corporate cash holdings, Journal of Financial Economics 52, 3–46.
Ozbas, Oguzhan, 2005, Integration, organizational processes, and allocation of resources, Journal of
Financial Economics 75, 201-242.
Pierce, Justin and Peter Schott, 2009, A concordance between 10-digit U.S. harmonized system codes
and SIC/NAICS product classes and industries.
Rajan, Raghuram, Henri Servaes, and Luigi Zingales, 2000, The cost of diversity: The diversification
discount and inefficient investment, Journal of Finance 55, 35-80.
Scharfstein, David and Jeremy Stein, 2000, The dark side of internal capital markets: divisional rent-
seeking and inefficient investment, Journal of Finance 55, 2537-2564.
Schott, Peter, 2010, U.S. manufacturing exports and imports by SIC or NAICS category and partner
country, 1972-2005.
Seru, Amit, 2014, Firm boundaries matter: Evidence from conglomerates and R&D activity, Journal of
Financial Economics 111, 381-405.
Shin, Hyun-Han and Rene Stulz, 1998, Are internal capital markets efficient?, Quarterly Journal of
Economics 113, 531-553.
Telser, L.G., 1966, Cutthroat competition and the long purse, Journal of Law and Economics 9 259-277
Weston, J. Fred, 1970, Diversification and merger trends, Business Economics 5, 50-57.
Williamson, Oliver, 1975, Markets and hierarchies, Analysis and antitrust implications, New York, NY:
The Free Press.
Abstract (if available)
Abstract
Theory suggests that a diversified firm’s internal capital markets can help or hinder its ability to respond to competitive threats: internal capital markets can help by allowing the firm to move resources quickly to threatened markets (resource flexibility), but can hurt by making it easier for the firm to exit an industry (providing weaker entry deterrence). This paper examines how diversified firms respond to heightened competition following a tariff reduction in order to assess the competing theories. The main finding is diversified firms reduce their investment in threatened industries relative to non-diversified firms, suggesting that internal capital markets weaken the ability to compete. The evidence also shows that diversified firms increase the share of total investment funds to non-threatened industries following a competitive threat, indicating that they are transferring resources out of the competitive industry.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
How large are the coinsurance benefits of mergers?
PDF
Essays in corporate finance
PDF
Cash holdings and corporate diversification
PDF
Essays on information and financial economics
PDF
Share repurchases: how important is market timing?
PDF
Three essays on board structure
PDF
Slashing liquidity through asset purchases: evidence from collective bargaining
PDF
Managing product variety in a competitive environment
PDF
Essays in financial intermediation
PDF
Competing across and within platforms: antecedents and consequences of market entries by mobile app developers
PDF
Essays in empirical asset pricing
PDF
Voluntary disclosure responses to mandated disclosure: evidence from Australian corporate tax transparency
PDF
Accrual quality and expected returns: the importance of controlling for cash flow shocks
PDF
Workplace organization and asset pricing
PDF
Does mandatory adoption of International Accounting Standards reduce the cost of equity capital?
PDF
Comparative study on Asian financial capitals' competitiveness: focused on strengths and weaknesss of the city of Seoul
PDF
Three essays in derivatives, trading and liquidity
PDF
How social and human capital create financial capital in crowdfunding projects
PDF
Three essays on young entrepreneurial firms
PDF
Green strategies: producers' competition and cooperation in sustainability
Asset Metadata
Creator
Swanburg, Garrett D.
(author)
Core Title
Internal capital markets and competitive threats
School
Marshall School of Business
Degree
Doctor of Philosophy
Degree Program
Business Administration
Publication Date
04/14/2015
Defense Date
03/17/2015
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
conglomerate,diversification,internal capital markets,OAI-PMH Harvest,product market competition
Format
application/pdf
(imt)
Language
English
Contributor
Electronically uploaded by the author
(provenance)
Advisor
Matsusaka, John G. (
committee chair
), Ogneva, Maria (
committee member
), Ozbas, Oguzhan (
committee member
), Wang, Yongxiang (
committee member
)
Creator Email
swanburg@gmail.com,swanburg@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c3-546589
Unique identifier
UC11298459
Identifier
etd-SwanburgGa-3292.pdf (filename),usctheses-c3-546589 (legacy record id)
Legacy Identifier
etd-SwanburgGa-3292.pdf
Dmrecord
546589
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Swanburg, Garrett D.
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA
Tags
conglomerate
diversification
internal capital markets
product market competition