Cross-National and Temporal Differences in Business Group Prevalence:
Do They Matter1?
Michael Carney*
John Molson School of Business
Concordia University
Montréal Québec
Michael.Carney@Concordia.ca
Marc Van Essen
Darla Moore School of Business
University of South Carolina
Saul Estrin
London School of Economics
Daniel Shapiro
Beedie School of Business
Simon Fraser University
Date: 16 March 2015
*corresponding author
1
The authors gratefully acknowledge discussions with Tomek Mickiewicz and Sumon Bhaumik, as well as
excellent empirical assistance from Adeline Pelletier. Any remaining errors are their own.
1
Cross-National and Temporal Differences in Business Group Prevalence:
Do They Matter?
Abstract:
The business groups (BGs) literature focuses on the financial performance of its affiliates,
relative to non-affiliated firms in the same country. In this study we take a different approach, we
focus on country-level performance outcomes and BG prevalence, both across countries and over
time. We base our analysis on contrasting views of BGs provided by the institutional voids and
entrenchment perspectives. We build a unique database by extracting estimates of BG prevalence
for multiple countries between 1978 and 2012. We find that BGs tend to persist, consistent with
the entrenchment view, but find no evidence that they limit country performance.
Keywords: business group prevalence, foreign direct investment,
2
Introduction
Research on business groups (BGs) has accumulated over the past two decades but the vast
majority of empirical work is focused on the question of whether group affiliation has a negative
or positive impact on the financial performance of member firms (Carney et al., 2011). Moreover
both theory and evidence regarding the effects of BGs and group affiliation on national economic
performance has not yet reached a consensus. On one end of the spectrum is a stream of literature
associated with institutional voids theory (Khanna & Rivkin, 2001; Khanna & Yafeh, 2007), which
takes a second best view in arguing that group membership has a net positive effect on various
aspects of an affiliated firm's performance, with the implication that the filling of institutional
voids is socially valuable. However, that literature also suggests that BGs will become less
prevalent as development proceeds and institutional voids are reduced. At the opposite end of the
spectrum are entrenchment and tunneling theories which depict the pyramidal structures and
opaque governance of BGs as purpose built and designed for self-dealing and minority investors
expropriation, thus emphasizing the negative net effects of affiliation (Bertrand, Mehta, &
Mullainathan, 2002; Morck, Wolfenzon & Yeung, 2005; Young et al., 2008). In this view, BGs
not only persist, but they limit a country’s development potential.
The motivation for the study arises from the fact that BG research has not yet responded to
the calls for greater scrutiny of group impact on their host economies. Granovetter (2005:445) calls
for researchers “to look closely at how business groups have responded to changes in the
economies they inhabit and how we understand their capacities and the way they change over
time”. Similarly, Khanna & Yafeh (2007: 364) call for research on “the validity of cross-sectional
explanations for the ubiquity and performance of business groups with time-series-based
3
perspectives” in particular they pose the question “do groups evolve in a fashion that is consistent
with missing institutions?”
Despite these gaps in the literature, there are few studies that provide explicit cross-national
comparisons of BG prevalence and its impact on country level socio-economic outcomes over time
(see Belenzon, Bekovitz, & Rios, 2013; and Masulis, Pham & Zein, 2011) along with studies that
look at change over time in a single country (e.g. Khanna & Palepu, 2000; Lee, Peng, & Lee, 2008,
Zattoni, Pedersen & Kumar, 2009). Although such studies could in principle provide a broader
perspective on BGs, they are rare, perhaps due to the difficulties of assembling a comprehensive
set of data. In this paper we address this problem in a novel way by considering difference in BG
prevalence across countries and over time, in addition to the effects of the prevalence of BG
affiliation upon a number of country level performance measures, including inward and outward
FDI flows, innovativeness, and financial market development. We tackle the data problem using
a fresh approach that combines the results from multiple single-country studies into a single multicountry study by extracting data from all published articles on BGs in particular countries. We use
this data set to provide an exploratory perspective on BG prevalence across countries and over
time, as well as the impact of BG prevalence on a number of country level outcomes.
Central to our study is the notion of BG prevalence, which varies across countries.
Prevalence measures the number of firms affiliated with a BG as a proportion of publicly listed
firms in an economy. Prevalence varies widely across emerging markets; for example 65% of
Indonesia’s publicly listed firms are affiliated with a BG but the affiliation rate is only 23% in
Israel and 25% in the Philippines (Khanna & Yafeh, 2007). Scholars initially believed BG
affiliation to be a prominent feature of emerging markets where the group structure addresses
institutional weaknesses and serves as a substitute for poorly developed market supporting
4
structures such as weak property rights or the absence of external finance (Ghemawatt & Khanna,
1998). However, recent research shows that group affiliation is also common in some highly
developed markets with strong market supporting institutions such as Belgium, Germany, and
Sweden (Belenzon, et al, 2013). Consistent with the theme of this special issue, we focus on the
understudied relationship between group prevalence and a country’s participation in the global
economy. Specifically we consider whether high levels of BG affiliation is associated with
increased or reduced levels of inward and outward FDI, and with factors that might contribute to
these flows, such as innovation capacity and access to financial resources.
To assemble our data, we surveyed the empirical BG literature to identify studies that
compared attributes of group affiliates and unaffiliated firms. Each study typically uses a dummy
variable to distinguish between the two, and for the total sample the dummy is used to compute
the percentage of publicly listed firms that are affiliated with a group (Claessens, Fan, & Lang,
2006). In this way, we are able to identify BG prevalence in the population of firms in a country
at a specific point in time. In some countries, BGs have been studied intensively so there are several
estimates of the prevalence of group affiliation taken at different points in time, and this allows us
to identify trends in the levels of group affiliation. We then explore the relationship between these
estimates and country level performance indicators.
The data we have assembled cover a large number of countries, as well as within-country
data over time, and thus afford an opportunity to provide a ‘big picture’ on the world's diverse
population of BGs as well as an overview of the impact of group prevalence on country-level
performance measures. We find considerable evidence that BG prevalence not only varies across
countries, but also that BG persistence varies across countries. Thus we find no consistent evidence
that BG prevalence will diminish over time as institutional voids are reduced. Moreover, we do
5
not find support for the view that BG prevalence is associated with “negative” country-level
performance, though we do identify some evidence that it is associated with “positive”
performance. In particular, we find consistent evidence suggesting that FDI outflows and financial
market development are positively associated with BG prevalence, Thus, we provide a more
nuanced view of view of BGs from that provided by much of the extant literature that focuses on
firm-level single-country analysis. Our results also suggest that there is considerable heterogeneity
in BG persistence and effects on outcomes, in turn suggesting that BGs can evolve in a variety of
ways. However, as a note of caution, our BG data has limitations, and our work is therefore
necessarily exploratory at this stage.
We begin by describing our sample and then we consider the overall trends in BG
prevalence in several economies. Secondly, we explore the relationship between prevalence and
outward and inward FDI. Third, due to their effect on national competitiveness we analyze the
effects of group prevalence on levels of national innovation and stock market development. We
conclude by considering the impact of the findings for future research on BGs and in particular the
internationalization potential of the BG structure.
Operationalizing BG Prevalence
Economic theories of organization (Arrow, 1974; Williamson, 1991), which underpin the
institutional voids approach, suggest that the prevalence of an organizational form is a direct
function of its comparative efficiency relative to alternative organizational forms performing
similar transactions. Thus, high levels of BG prevalence will at best be neutral with respect to
national economic performance and, to the extent that BGs have strong performance capabilities
over time, there could be a positive association (Guillén, 2000). On the other hand, the
6
entrenchment argument which stresses rent seeking behavior and the expropriation of surpluses
into the hands of self-interested elites, argues that high levels of BG prevalence have seriously
negative consequences, including restricting international capital flows (Morck et al., 2005),
inhibiting innovation (Mahmood & Mitchell, 2006) and retarding the emergence of external capital
markets (Almeida & Wolfenzon, 2006). While hypotheses about BGs are frequently constructed
in abstract and generic terms, BGs around the world are actually very heterogeneous with respect
to ownership, corporate structure, and vertical and horizontal scope (Yiu, Lu, Bruton, &
Hoskisson, 2007). Such heterogeneity is reflected in the widely accepted definitions that BGs are
“a collection of firms bound together in some formal and informal ways” (Granovetter, 1994:454)
and that “a business group is a set of firms which, though legally independent, are bound together
by a constellation of formal and informal ties and are accustomed to taking coordinated action”
(Khanna & Rivkin, 2001:47). Scholars agree that BGs are complex social economic phenomena
serving several purposes and the relationship of a firm to the group may be highly variable and
multiplex (Khanna & Rivkin, 2006). It is therefore not surprising that there is considerable
variation across the BG literature in the exact definition of a BG, as well as differences in sample
size resulting in considerable variation across studies in the estimates of prevalence.
Empirical studies typically operationalize BG affiliation as a dummy variable dividing the
national population of listed firms into two distinct categories. Data on affiliation are frequently
provided by country specific institutions or stock market guides such as the classification devised
by the Center for Monitoring the India Economy; Dodwell Marketing Consultants classification
of Industrial Groupings in Japan; Taiwan’s biennial directory Business Groups in Taiwan (BGT);
compiled by the China Credit Information Service; Thailand's Brooker Group publication: Thai
Business Groups. In Korea, the population of BGs is usually defined by Korean Foreign Trade
7
commission’s annual determination of firms affiliated with the top 30 Chaebols. While these are
likely to reflect country specific idiosyncrasies, such definitions typically produce reliable yearto-year estimates of affiliation in a specific jurisdiction.
Alternatively, other studies define affiliation by ownership structure, such as a specific type
of ownership relationship or a threshold level of ownership cash flow or control rights (Claessens,
Fan, & Lang, 2006). Depending upon ownership structure and threshold selected, the estimated
prevalence of group affiliation can be quite variable. For example, in a study of European corporate
groups Belenzon & Berkovitz (2009) apply a holding company ownership definition with a 20%
voting rights threshold and find very high levels of group affiliation in countries not typically
associated with high prevalence of BGs, including Finland (69%), the Netherlands (67%), and UK
(68%). In contrast, Masulis et al (2011), in a multi-country study which partially overlaps with
Belenzon & Berkovitz (2009), examined the prevalence of family controlled BGs using a pyramid
definition of ownership structure and found significantly lower levels of BG affiliation across all
countries. Thus, while Belenzon & Berkovitz (2009) found group affiliation to be 28%, 65% and
32% for Belgium, Denmark and Germany respectively, Masulis et al (2011) found affiliation for
the same three countries to be 4%, 21%, and 9%. Hence, assessing the prevalence of group
affiliation in such studies and across countries and over time requires judgment and caution.
To compile our sample, we used five complementary search strategies to identify the
population of studies that provide statistics of prevalence of group affiliates and nonaffiliated
firms in particular country (Heugens, Van Essen, & Van Oosterhout, 2009). First, we consulted
several review and meta-analytic articles (Carney et al., 2011; Khanna & Yafeh, 2007). Second,
we searched five major electronic databases (ABI/INFORM Global, EconLit, Google Scholar,
JSTOR, and SSRN) by using the following search terms: using the following search terms:
8
“business group,” “business houses,” “chaebol,” “grupos economicos,” “guanxiqiye,” “hongs,”
“keiretsu,” “oligarchs,” “pyramids,” “qiye jituan”, and “zaibatsu”. Third, we conducted a manual
search of journals in the disciplines of economics, finance, and management that periodically
publish articles related to BGs. Fourth, after collecting an initial set of studies, we used a
“snowballing” technique (Von Hippel, Franke, & Prügl, 2009) that involved backward-tracking
all the references reported in the articles and tracing forward all articles that cite the original articles
by using Google Scholar. Fifth, we directly contacted authors of one or several papers relevant to
this topic who did not report information on prevalence. After conducting these five steps and
removing any manuscripts that used data identical to those of other studies, we arrived at a final
sample of 150 BG studies that have compared group affiliates and nonaffiliated firms between
1978 and 2012. Each study provides one or multiple country-year estimates of prevalence of BG
affiliation in the study. Collectively, these studies yield 351 prevalence/year estimates for 65
countries. For 26 countries, such as Algeria, Lebanon, Poland, Sri Lanka, and Ukraine we have
only a single estimate of prevalence. For 39 others, we have multiple estimates, and many Asian
and Latin American countries are intensively studied. For instance we find (64) estimates for
prevalence in South Korea; Japan (36), India (28) and Chile (26) and seven each for Brazil and
Colombia.
We expected the full set of prevalence estimates to be statistically ‘noisy’ given the likely
group heterogeneity and variation in the definition and operationalization of group affiliation in
the population of studies. Therefore to explore the relationship between prevalence and country
performance we constructed two distinct samples. The first sample is the full set of all observations
including all of the estimates of prevalence in all studies. We created a second sample, which we
9
describe as ‘all good observations’2 producing a sample of 266 prevalence year estimates. Here
we excluded country estimates that provided ambiguous affiliation definitions or too little
information on either definition or calculation method. We also excluded prevalence estimates
using a strict pyramid definition of group affiliation. Khanna & Yefeh (2007) suggest pyramidal
equity ties are atypical of group ties in many countries and are therefore likely to underestimate
the prevalence of groups; for instance, they found only 50% of Chilean BGs were characterized
by pyramidal ownership ties. In the next section, we briefly consider two contending accounts of
expected trends in BG prevalence and we then depict cross-country and within-country specific
trends based upon the data reflected in our sample.
Country-Specific Trends in BG Prevalence
The institutional voids and entrenchment hypotheses represent two prominent theories of
the life cycle of BGs but they offer competing views about expected group longevity and impact.
The institutional voids theory has two components: the first suggests that BGs materialize to fill
voids in market supporting infrastructure (Khanna & Yafeh, 2007). The group structure enables
affiliates to acquire resources that would be otherwise unavailable via arms-length contracting.
The second suggests that with the development of market supporting infrastructure arms-length
contracting will become feasible and the value of affiliation will erode (Khanna & Yafeh, 2007).
As their functional advantages erode, BGs are expected to focus, restructure and loosen their ties.
Ultimately groups should then unravel and disappear as it becomes more profitable for firms to be
unaffiliated than affiliated because markets are able to provide the resources and capabilities more
2
For the purposes of reliability one author collected and coded all the prevalence data. Two other authors checked
the data and independently coded whether the observation should be included in second sample ‘all good
observation. We obtained a high inter-rater agreement of .98 (Cohen's kappa coefficient; Cohen, 1960).
10
efficiently and cheaply than from within the BG. Under this scenario, group affiliation should
become progressively less prevalent following institutional and economic development
(Hoskisson, et al, 2005; Lee, et al, 2008). Based upon their confidence in the institutional voids
hypothesis, Khanna and Palepu (1999) build their policy advice on the expectation that BGs will
fade away when they advocate “governments in developing countries must focus on building up
those market institutions in the long term. The dismantling of business groups will, we believe,
follow naturally once these institutions are in place” (1999: 126). Thus, as a functional economic
response to underdevelopment, institutional voids theory predicts that business groups are a
transitional phenomenon.
The entrenchment hypothesis is based on the assumption that concentrated economic
(monopoly) power in the hands of rent seeking entrepreneurs will have negative consequences for
a country as a whole (Morck et al., 2005). This has also two components. Morck et al (2005)
propose that BGs are often initially formed by states to orchestrate a’ big push’ toward industrial
modernization. However, with few exceptions, dominant BGs use their monopoly power to
entrench themselves by capturing the commanding heights of the economy. Subsequently, these
incumbents seek to defend their sources of rents by retaining the conditions that favor their
economic and social prominence. Given their important role in the economy and the political
powers that it entails, BG entrepreneurs and elite politicians can attain a stable equilibrium
(Schneider, 2009). As entrenched actors, economic and political, they are able successfully to resist
entry from outsiders and to extract surplus from minority shareholders and other potential
stakeholders. In this view, BGs persist and remain prevalent among the economy's leading firms.
Which of these contending hypotheses is best supported by our prevalence data? Based on
our ‘all good observations’ sample, we depict overall trends in prevalence in each of 18 countries
11
for which we had multiple ‘good’ prevalence observations by calculating linear least squares
estimates of prevalence over time, reproduced in Figure 1. Overall the sample shows slight upward
trends in prevalence in Belgium, Brazil, China, India, Indonesia, Israel, Japan, Malaysia, Peru,
Singapore, Taiwan, and Thailand and downward trends in Chile, Hong Kong, Philippines, Russia,
and Turkey. To gain a finer grained picture we separately plotted the least-squares line for several
intensively studied countries. Figure 2 depicts individual trends in 7 countries. Each point in Figure
2 represents a study’s estimate of prevalence for a particular year. The line is the best fitting
regression for those estimates. Between 1998 and 2010 BG affiliation in China appears to have
grown from 20% to 50%. In India since the early 1990s BG affiliation appears to be of slight
upward trend in a band between 40 and 55%. Israel and Japan show similar slight upward trends.
Affiliation appears to be relatively constant in Taiwan. Chile shows a slight long-term decline.
Only in South Korea does there appear to be a steep decline in affiliation.
Because economic development is highly correlated with many indicators of institutional
development, we plot prevalence against GDP per capita in Figure 3. We also provide the best fit
regression line, which is curvilinear. Most of the observations are in the less than $10,000 GDP
per capita band since many of the observations are taken from countries at the early stages of
economic development. Figure 3 suggests that BGs remain prevalent at relatively high levels of
income, and show no clear downward trend over time, as predicted by institutional voids theory.
Indeed, the overall picture suggests that BGs tend to persist on a global scale, albeit not in every
country. Thus we see some support for the entrenchment perspective and Granovetter’s conclusion
that “BGs have typically defied predictions of their imminent demise surviving the conscious
attempts by politicians to break them up and the impact of financial crises” (2005: 445). Moreover,
group affiliation remains prevalent at high levels of GDP per capita (a variable closely associated
12
with institutional development), implying that factors other than the filling of institutional voids
explains the persistent prevalence of group affiliation.
Figure 1: BG prevalence trends in 18 countries
Figure 2: BG prevalence trends in 7 countries
Figure 3: BG prevalence and gross domestic product
Prevalence and Country Level Economic Performance While financial effects of BG affiliation has attracted the lion’s share of the empirical
attention, the impact of the prevalence on national economic performance remains hotly debated,
even though evidence on the question is sparse. Advocates of a positive view imply there are at
least two ways that the prevalence of group affiliation can benefit country level economic
performance. The first is through simple aggregation logic: affiliation has positive effects so the
greater the number of firms associated within BGs, the greater these positive effects. Specifically,
in the context of institutional voids, BG's member firms enjoy superior allocative efficiency
compared with unaffiliated firms, since affiliation gives the firm access to group mediated
resources, such as investment capital and managerial know-how3. Second, there will be spillover
effects, in the form of trained managers, fostering the creation of competent stock analysts,
building industrial capacity, communications and transportation infrastructure, which lead BGs to
3
But this scenario provides only a partial equilibrium (Kali, 1999). Full equilibrium requires every firm in the
economy to join a group. However, there are increasing coordination costs with group scale and scope (Hoskisson et
al., 2005) where the addition of new firms outweigh the available benefits therefore some proportion of firms fail to
join groups producing a partial equilibrium outcome at some level of prevalence.
13
accelerate institutional development (Fisman &Khanna, 2004) and to lower the costs of doing
business for all firms including those unaffiliated with a group.
In contrast, the entrenchment and varieties of capitalism (Hall and Soskice, 2001)
perspectives suggest two ways in which BG prevalence can harm country level economic welfare.
Both perspectives concede that BGs might provide positive welfare effects in the early stages of
economic development. Both agree that, at later stages of economic development, BGs
subsequently actively suppress or passively inhibit sustained growth and competitiveness. The
entrenchment perspective provides an elite capture theory that considers BGs prevalence as
facilitating the concentration and coordination of economic power in the hands of a few families
and oligarchs. In this view, institutional development becomes endogenous (Morck et al., 2005);
concentrated economic power is wielded to entrench the status and positions of elite actors and to
reduce positive spillovers by actively inhibiting the entry of business rivals and preventing the
emergence of institutions favoring market competition.
Secondly, the varieties of capitalism perspective views BG prevalence as an outcome of
complementarities between labor, the state and corporate elites. These groups are constructed at
the early stages of development to address national economic coordination problems, such as wage
setting, but subsequently these relationships become locked in and so foreclose the emergence of
new complementarities such as, arms-length contracting and flexible labor markets, as the
economy becomes more developed (Schneider, 2009). BG prevalence will therefore be associated
with negative performance effects through a ‘crowding out’ mechanism in which the dominance
of a single organizational form limits the evolution of diversity and economic specialization in the
economy. In this view, the mere existence of BGs may passively hamper the emergence of
alternative corporate forms by depriving them of niches in which to specialize and these negative
14
effects will be amplified when BGs are highly prevalent. Thus the BG competitive advantages
comprise generic capabilities such as project execution and mass production (Amsden & Hikino,
1998) and comparatively efficient resource allocation (Khanna & Palepu, 1999). However, BG
corporate structures are incompatible with the kinds of firm-specific organizational and
technological capabilities supportive of advanced product and process innovations (Kock &
Guillen, 2001). Accordingly, if too many firms are group affiliated, competitive selection
processes will be muted, BGs will crowd out other more specialized forms of organization and so
dampen the incentives and capacities associated with the development of the firm-specific
capabilities required for international competitiveness. One implication of the idea that BGs
passively retard the emergence of more advanced institutions and specialized organizations is that
economies with a high prevalence of BGs can become trapped in a middle-income equilibrium
(Schneider, 2010).
BG Prevalence and Outward FDI
These conflicting views about prevalence and economic performance are echoed in the
literature concerning inward and outward FDI. We first consider arguments regarding the effects
of group affiliation on outward FDI. The argument for positive effects reflects the institutional
voids approach together with elements of RBV theory. It is well-established in the institutional
voids perspective that, in the absence of market supporting institutions, affiliation with a BG can
provide superior access to a range of resources including capital, skilled human resources and
technological and organizational know-how that are unavailable to unaffiliated firms. Other things
equal, this perspective implies that affiliates have competitive advantages relative to unaffiliated
firms, which is consistent with the RBV perspective on firm competitiveness. Thus, group
affiliation is credited with boosting national industrial capacity by organizing export-oriented firms
15
and helping firms acquire advanced technology (Guillen, 2000) and providing project management
skills to help firms enter new markets (Amsden & Hikino, 1994). Over the past decade scholars
have documented the pioneering effects of group affiliates in engaging in FDI, becoming
emerging-market multinationals (Matthews, 2006). Since, many emerging market multinational
enterprises invest overseas to acquire strategic assets rather than to exploit firm specific
capabilities, BG affiliation may provide a springboard to foreign markets (Luo & Tung, 2007: Yiu,
et al 2005; Ramamurti & Singh, 2009).
More recently, research has emphasized BG affiliation as a mechanism for competing in
globally deregulated industries by developing advanced capabilities consistent with international
standards of competitiveness. In this view, multiple ties among affiliates including buyer supplier
equity and director ties facilitate individual and complementary combination effects with respect
to the development of the R&D capabilities required to be internationally competitive (Mahmood,
Zhu, & Zajac, 2011). Relatedly group affiliation allows firms to access and leverage the market
knowledge and international connections in their sister affiliates (Lamin, 2013). In a longitudinal
study of Indian firms, Chari (2013) found that group affiliates were more likely to engage in FDI
than were unaffiliated firms. The consensus arising from this stream of research is that BG
affiliates are in the vanguard of economic development and represent the country's most
competitive firms.
These perspectives and findings suggest unaffiliated firms will be less able to access the
resources needed to internationalize their activities in globally competitive industries. Collectively
this literature suggests that there is a greater likelihood that firms affiliated with groups will have
access to the resources and capabilities needed to support outward FDI. Therefore, by the logic of
16
aggregation discussed above this reasoning suggests that the greater the degree of BG affiliation
in an economy, the greater the level of outward FDI.
However, there is no unanimity on the view that affiliation will enhance the international
competitive capabilities of group members in a way that would encourage outward FDI. We have
noted that a substantial body of literature highlights potentially negative attributes of group
affiliation, and how these might suppress incentives and capacities to internationalize activity.
First, controlling shareholders may undermine the competitive potential of their affiliates by
extracting or tunneling resources from them (Bertrand, Mehta, & Mullainathan, 2002). Others
emphasize the BG potential to exploit linkages with political elites for personal gain where a
relatively small number of business families control significant shares of the economy (Claessens,
Djankov, & Lang, 2000). Oligarchic family BGs can use their economic power to extract resources
from government (Faccio, 2006) through corporate bailouts (Faccio, Masulis, & McConnell,
2006), rent seeking (Fisman, 2001) and by inhibiting competition from domestic and international
rivals (Fogel, 2006) humor. Receiving an assured stream of rents, these business owners have
limited incentive to seek investment opportunities beyond their borders where the possibilities for
rent extraction are less auspicious and the likely returns much lower. Indeed, successful rent
seeking strategies may displace emphasis on developing alternative sources of competitive
advantage.
More recently studies have identified several factors that inhibit BG affiliates’ outward
FDI. Specifically, due to the dense multiplex linkages with other businesses in their domestic
environments, group affiliates can become overly embedded in a set of country specific institutions
(Pederson & Stucchi, 2015). A related argument suggests family BGs engage in domestic product
diversification on the basis of the reputation of a family/reputable entrepreneur. However,
17
reputations are highly localized phenomena and not easily transferable across borders. Indeed,
product diversification and international expansion may present firms with an inherent trade-off
(Kumar, Guar, &, Pattnaik, 2012) and locally diversified firms may be required to refocus their
business portfolio before they engage in international expansion (Meyer, 2006). Tan & Meyer
(2010) find that extensive political and business ties with local actors drive domestic growth and
inhibit international expansion. In particular they find that government ownership in any group
discourages internationalization of group affiliated firms. Similarly, Chari (2013) in a sample of
Indian firms finds that the affiliates of highly diversified BGs have lower levels of FDI compared
with more focused BGs. Implicit in these perspectives is the idea that unaffiliated firms are not
exposed to the negative impact of group membership on internationalization. Hence, the
prevalence of BG affiliated firms is expected to be associated with lower levels of outward FDI.
We undertook a regression analysis to provide greater insight into which of these two
perspectives better describes the impact of BGs on outward FDI. The results are reported in Table
1. We regress each country-level economic performance measure, in this case outward FDI, on a
measure of BG prevalence, and a number of control variables. For these, we include GDP per
capita (lagged one period) to control for a variety of country effects including macro-economic
policy and other institutional variables. We also included specific measures of institutional
strength, and here report results including a variable for the “Rule of Law” (Globerman & Shapiro,
2003). Finally, we attempt to control for the fact that the prevalence data may be measured
differently across studies by including a variable indicating the paper from which the data was
taken (noting that several papers provided measures for more than one country or multiple year
observations). In addition, we note the size of the firm population over which prevalence is
measured. We do so because it is generally the case that firms affiliated with BGs are on average
18
larger than the average for the population of unaffiliated firms. Therefore, when the prevalence of
BGs is measured on a smaller population (usually the largest firms), as opposed to a larger
population (all firms), the percentage of BGs will be higher in the former case.
Table 1: BG prevalence and performance outcomes
The results for outward FDI (OFDI), measured in logarithms because of the non-normality of the
distribution,, are reported in columns (1) and (2), for the full and “good” samples respectively. We
note that the full sample size is reduced from that reported above because of missing values.
Regardless of sample, the coefficient on the BG prevalence term is positive and statistically
significant. .Indeed in the full sample we find a strongly positive relationship between prevalence
and outward FDI with the implied possibility that BGs are capable of generating internationally
competitive capabilities. We find no evidence supporting the view that higher levels of BG
presence are associated with reduced OFDI flows from a country.
It is important to note that these positive results are obtained while controlling for GDP per
capita, which has a consistently positive and statistically significant effect on OFDI flows, as well
as to controls for institutional quality In this sense our results are robust to heterogeneity across
countries. However, we do note that the positive BG effect does not stand up in the presence of
country fixed effects; this is probably because the BG variable is fairly constant across time, and
therefore is closely related to the country specific effect. However, we reiterate that we cannot
claim causality for these results, which do not derive from a formal model linking development
process, institutional development and BG evolution. Even so, our results suggest that at the very
19
least BG prevalence on average has a positive association with a country’s OFDI position at any
point in time. Our evidence does not therefore favor the entrenchment view.
BG Prevalence and Inward FDI Discourse on the relationship between BG prevalence and inward FDI is equally divided.
Perhaps the most salient argument is elite capture, an argument that emphasizes the suppressing
effects of oligarchic BGs dominating the economy. The argument is a direct corollary of the
entrenchment hypothesis (Morck et al., 2005). In this view politically connected entrepreneurs
gain access to national monopolies through quid-pro-quo agreements with political actors. The
dependence on domestic markets is exacerbated and provides BGs with a strong incentive to resist
encroachment on their domestic terrain. BGs are keen to minimize competition and they will be
particularly concerned with erecting barriers to the entry of foreign rivals. BGs will oppose
institutional and regulatory developments that threaten their dominant positions (Rajan &
Zingales, 2003). Hence, higher levels of BG prevalence should be associated with lower levels of
inward FDI and a reduced presence of multinational enterprise.
Nevertheless, many emerging market countries seek to accelerate their economic
development by encouraging inward FDI to stimulate the transfer of technology and managerial
expertise. The international economics literature finds that such investment can contribute
relatively more to growth than domestic investment (Sabirianova, Svejnar, & Terrell, 2005).
Inward, FDI may be particularly effective in encouraging industrial restructuring (Dunning &
Narula, 2003), which is an important antecedent of outward FDI by BGs (Meyer, 2006). However,
several East Asian economies such as Japan and Korea actively discouraged inward FDI in the
early stages of development as a means of developing domestic capabilities (Amsden, 1989;
20
Gerlach, 1992). Therefore, the relationship between levels of BG prevalence and inward FDI may
well depend on the ability of BGs to influence government inward FDI policy.
Recent research in the varieties of capitalism perspective framework suggests BG
prevalence and high levels of inward FDI produces a self-reinforcing dynamic resulting in strong
complementarity between MNEs and BGs, who together form a dominant stable coalition in many
emerging markets (Schneider, 2009; Nolte & Vliegenthart, 2009). Schneider (2009) finds that
MNEs are especially visible amongst Latin America's largest firms in the concentrated
manufacturing and capital intensive sectors where they transfer foreign manufacturing
technologies into a host market. To gain access to an economy MNEs will seek to link up with
prominent politically connected entrepreneurs. Alternatively, government may require MNEs to
enter joint ventures with domestic partners as a condition of entry. In other cases, MNEs enter the
country by acquiring, and subsequently delisting, a local firm. Either way, Schneider (2009)
predicts that linkages and inward investments by MNEs push BGs into new sectors and expand
the scope of their diversification. Eventually, these economies achieve a stable equilibrium where
MNEs specialize in medium manufacturing industries while BGs focus on local agricultural and
services, such as finance and telecommunications. Nolte & Vliegenthart (2009) depict similar
dynamics in the central Europe's transitional economies which they describe as ‘dependent market
capitalism’ where local diversified BGs coalesced with MNEs to create global assembly platforms
for medium-range consumer durables such as auto parts and electronics. In both cases BGs
affiliated firms are highly prevalent alongside substantial inward FDI.
Thus, the entrenchment and varieties of capitalism arguments make contrary predictions
about the relationship between BG prevalence and inward FDI. Our evidence on this matter is
21
found in Table 1, columns (3) and (4) where IFDI has replaced OFDI as the dependent variable,
with the same independent variables as were described above.
Although the results are not as strong as those for OFDI, we are again unable to find any
evidence that GB prevalence is in any way associated with lower inward FDI flows. In the larger
sample the BG coefficient is positive and statistically significant, while in the smaller sample it is
negative, but not statistically significant. Once again, the evidence is not consistent with the
entrenchment view.
BG Group Prevalence and Technological Innovation
Related to the issue of inward and outward FDI is BG prevalence with respect to an
economy’s innovation activity. Several studies in the institutional voids tradition propose that BGs
protect member property rights and so facilitate market-oriented innovation among affiliates
(Chang & Hong, 2000; White, Hoskisson, Yiu & Bruton 2008). These perspectives expect that
high-performing BGs will support innovative activities that strengthen their competitiveness. One
reason for this is that profitable BG's may substitute for other capital market weaknesses, for
example by functioning as venture capitalists using their free cash flow to fund new innovative
startups (Masulis et al., 2011). Such arguments are consistent with the view that BGs develop
proprietary capabilities that would support sustainable outward FDI.
Conversely, BGs can be ‘two faced’ with regard to innovation (Mahmood & Mitchell,
2004). In particular, when BGs account for only a small share of the economy they are more likely
to facilitate innovation by filling institutional voids as suggested above. However, when BG's
account for a large share of the economy's activities, they can hinder innovation by erecting
barriers to entry for new firms who experiment with new business models (Mahmood & Mitchell,
2004). Facing less competition from unaffiliated firms also allows BG to become entrenched.
22
Under these circumstances BG's may seek to protect existing sources of rent which will inhibit
their incentives for innovation (Chang, Chung, & Mahmood, 2006). BGs will have little incentive
to fund R&D expenditures and other activities such as patenting new technologies because, at high
levels of prevalence, there is an increased probability that one affiliate’s innovation will
cannibalize another affiliate’s products (Mahmood & Mitchell, 2004). Similarly, at low levels of
BG prevalence, affiliates will face stronger competition from unaffiliated firms and have a greater
incentive to invest in innovation (Mahmood & Mitchell, 2004). This argument suggests that high
levels of prevalence will reinforce the selection environment for BGs generic capabilities and
weaken selection for firm specific advanced product and process innovations (Kock & Guillen,
2001).
This view is strengthened when there is strong path dependency. To the extent that BGs
are successful in imitating and diffusing technologies among their membership that originate
beyond their borders, they may create and perpetuate external dependencies on foreign technology
(Nolte & Vliegenthart, 2009). If BGs meet their technology needs through joint ventures with
foreign companies, they will have limited interest in supporting the development of intellectual
property rights that incentivize local scientific endeavor (Schneider, 2009). Hence, the prevalence
of BG affiliation is likely to hamper national innovation efforts.
Our data does not provide much support for either point of view. In Table 1, Columns (5)
and (6) we explore the impact of BG prevalence on the logarithm of patents granted (at the country
level) – a commonly used indicator of national innovation efforts. The results with respect to BGs
23
show no statistically significant relationship. Thus we find no evidence that BG prevalence is
related to innovative activity one way or the other4.
BG Prevalence and the Development of Stock Markets
Finally we consider the effect of BG prevalence on the development of external capital
markets. A significant strand of the literature points to potential negative economic consequences
associated with extensive group affiliation and their associated internal capital markets. Early
research associated BG internal capital markets with inefficient investment outcomes as groups
were argued to support weak underperforming firms at the expense of better external opportunities
(Scharfstein & Stein, 2000). Almeida & Wolfenzon (2006) develop a theoretical model suggesting
that substantial group affiliation in an economy hampers the allocative efficiency of external
capital markets thereby contributing to their under-development. In this view high levels of
affiliation produce strong feedback levels in the rest of the economy because when new firms
expect others to join BGs, they are more likely to do so as well since there are few alternative
sources of capital to fund new ventures. This further limits the emergence of efficient external
capital markets.
However, other research affirms the beneficial effects of BGs in the context of weak capital
markets. Under these circumstances BGs offer their affiliates preferential access to group mediated
resources that are unavailable through arms-length transactions in external capital markets. Thus,
where capital markets are illiquid and underdeveloped, BGs’ internal capital markets provide
venture capital to startups (Masulis et al., 2011) and extend credit to distressed firms (Jia, Shi, &
4
We note that in this set of equations, the Rule of Law variable is unexpectedly negative and statistically significant.
To the extent that this variable is associated with the protection of property rights, this result is hard to explain and
suggests some misspecification with respect to institutions and innovation
24
Wang, 2013). BGs internal markets are found to be especially valuable for capital intensive firms,
with a high need for external finance, in countries where equity markets are poorly developed
(Belenzon, et al, 2013). These arguments support the view that BGs may be associated with less
developed capital markets, not because they are inefficient, but because they are efficient, both in
funding startups and in extending credit to high risk firms.
Even as other market supporting institutions develop with the progress of economic
growth, path dependence is believed to have an especially prominent effect with respect to the
development of liquid capital markets. Several scholars document that, despite acceleration in
economic growth, external capital markets failed to develop because economies dominated by BGs
develop reliance upon internal capital markets to facilitate economic growth (Schneider, 2009).
Masulis and his colleagues state that (2011: 3558): “Across various lines of analysis, our evidence
consistently indicates that the continuing prevalence of family groups and their choice of
organizational structures reflect not only control motivations, but also their ability to support
high-risk, capital-intensive firms that could otherwise find it difficult to attract external funding,
especially in weak capital markets”. Hence, whether one accepts a positive or negative
interpretation of BG functioning there seems to be significant consensus on both ends of the
spectrum that the prevalence of BGs will be associated with less developed external equity
markets.
Our data suggest otherwise. In Table 1, columns (7) and (8) we explore the relationship
between stock market capitalization and our list of independent variables. Holding GDP per capita
constant, we find a positive and statistically significant association between stock market
capitalization and BG prevalence. Thus, we find no evidence to support the idea that the internal
25
capital markets typically associated with BGs are also associated with under-capitalized external
financial markets.
Discussion: Reflections on the BG Prevalence Debate
It is now apparent that BGs are a prominent feature of the business landscape in both
developing countries, where they have been accorded more attention, but also in more mature
economies. Nevertheless, both the degree of their persistence and their impact upon country
economic performance remain unsettled areas of research. As such, there is no consistent
evidence to support either side of the competing narratives that we have described in this paper.
In particular there is very little comprehensive evidence regarding the impact of BGs on
economic performance at the country level.
In this paper, we take a different approach to thinking about this problem. By culling
evidence from 150 existing studies about the level and persistence of BG prevalence across
countries over the last 30 years and exploring their associations with country level indicators of
economic performance, we offer an original contribution to the debate. We find that, while levels
of prevalence do indeed vary significantly around the world, for the most part these persist over
long periods of time. We find no evidence that BGs disappear and only in few cases does group
affiliation appear to decrease in importance for the national economy, even though income levels
have increased considerably over the period and institutions have in many cases also
strengthened. At the same time we find no evidence that high levels of prevalence have harmful
effects on the country level indicators we examine. Indeed, there is more evidence that
prevalence has a positive effect on inward and outward FDI as well as the development of capital
markets.
26
Seemingly our data provide support for both the positive and negative narratives
surrounding group prevalence. The positive institutional voids narrative holds that BGs
materialize to fill institutional voids with consequent positive effects on economic development
but, subsequently, groups should unbundle and disappear once market supporting institutions
emerge to support the effective functioning of freestanding firms. In so far as our data show
nonnegative or positive associations between group prevalence and capital flows as well as
capital market development, then they are consistent with the positive institutional voids
narrative. However, our finding that BGs do not unbundle and disappear but persist over time is
inconsistent with the same narrative. Indeed, evidence of persistence would seem to support the
negative narrative articulated by entrenchment and varieties of capitalism perspectives. However,
the fact that high levels of group prevalence are also found in mature economies contradicts the
negative narrative that groups are associated with ensnaring the economy into a middle-income
trap. Similarly, the positive association of BG prevalence with various country performance
indicators also speaks against the negative narrative.
As a point of departure for resolving this apparent contradiction, we would suggest that
both the positive institutional voids and negative entrenchment narratives share a common
consensus that the ideal ‘endpoint’ of economic and institutional development will approach a
‘standard model’ (Hansmann, & Kraakman, 2004) comprising a population of focused,
freestanding firms embedded in an architecture of market supporting institutions of the type
found in mature liberal market economies. In these perspectives, departures from the standard
model such as BGs represent a suboptimal outcome consisting of second-best institutions
(Rodrick, 2008).
Our reflection on the prevalence debate suggests the need for an alternative approach,
27
based on the proposition that BGs are a flexible and adaptable form of organization, likely to
take on diverse ownership and structural characteristics in different national settings. In addition,
this approach should recognize that BGs are likely embedded in local business ecosystems that
support a variety of co-existing organizational forms.
There is already evidence that when BGs emerge they are likely to vary across
economies with respect to ownership, organizational structure, and horizontal scope (Yiu, Lu,
Bruton, & Hoskisson, 2007). We suspect there where groups emerge and attain significant scale
they will display adaptive capacities that help their affiliates adjust faster to global opportunities
than do local institutions. The observed outcome of group salience is the emergence of
nonstandard (compared to the ideal standard model) complementarities between the population
of firms and its evolving institutional context. This complementarity appears to facilitate both
inward and outward FDI, such that BGs may not block inward FDI but they do assemble
resources to facilitate outward FDI. Indeed, the prevalence of the group form may provide
spillovers to freestanding firms to facilitate a comparable outward investment
opportunities. Thus, in terms of our ‘big picture’ overview, we conclude that the prevalence of
BGs seem on balance to facilitate their economies participation in the global economy rather
than locking them into autarkic isolation, and this points to the need for a theoretical framework
that recognizes this (potential) adaptability.
In this regard, because BG prevalence is different across countries and time, and because
BG prevalence at any point in time does not on average appear to impair country economic
performance, we entertain the possibility that there are a variety of equilibrium outcomes in
terms of institutional and organizational configuration. The idea of multiple possible optima is
consistent with the idea of several broadly efficient varieties of capitalism (Hall and Soskice,
28
2001), suggesting that differences in the economic performance of institutional modes may be
quite small. The notion that efficiency differences between alternative institutional modes are
small has important implications for theories advocating the virtues of the standard model. The
creation and diffusion of new institutions in the economy is costly and if their benefits are small
and difficult to quantify policy makers and executives may be reluctant to incur the costs. Thus,
the path dependent development of groups and institutions emphasized in the varieties of
capitalism perspective may result in the establishment of different but complementary bundles of
organizations and institutions that yield equivalent long-term economic performance outcomes.
Although we reject the idea that business group prevalence declines automatically as
institutional voids decline and market supporting institutions improve, the pattern of group
persistence we document in this paper does not appear to be caused by phenomena at the heart of
entrenchment theory, namely elite capture whereby the owners and political allies of dominant
business groups control the economy for their own benefit but to the detriment of broader
economic welfare (Morck et al 2005). The idea of multiple optima suggests a context of path
dependent development such that business groups can become embedded in an economic system
in a way that allows BGs to facilitate, or least not inhibit, internationalization and capital market
development importance. The polarized paragon/parasite (Khanna & Yafeh, 2007) depiction in
prominent BG theories therefore misses an important element of what BG's actually do in the
developmental process.
Our results also reinforce Granovetter’s (2005) call for researchers to examine the way
BGs change over time. Our finding that BGs prevalence not only persists on average, but does so
in conjunction with economic and institutional development suggests adaptability within the
parameters of the group structure. This suggests that groups do not evolve in a fashion consistent
29
with missing institutions in the manner predicted by institutional voids theory. Rather, we
speculate that BGs persist in layers of co-existing corporate systems (Meuer, Rupietta & BackesGellner, 2015). Thus rather than ‘crowding out’ alternative configurations of institution and
corporate structures, BGs simply occupy a nodal point, albeit at times a rather prominent one, in
a shifting and developing corporate landscape. This could occur because BGs develop along their
own path dependent trajectory continuing to perform valuable functions in certain sectors of the
economy. Simultaneously, continuing economic and institutional development generates a
dualistic, or heterogenous, industrial and corporate structure. For instance, in the manner
suggested by Belenzon and her colleagues (2013), German corporate groups continue to
coordinate industrial activities in technology and capital-intensive sectors of the economy
alongside a large and diverse population of autonomous midsized Mittlelstand firms (Herrigel,
2000). A similar parallel is evident in the US economy where a population of publicly listed
managerially controlled firms, organized along the lines of the standard model, intermingle with
a significant body of family controlled firms with each achieving comparable levels of financial
performance (Anderson & Reeb; 2003). Thus, while we do not speak directly to Granovetter’s
question concerning how groups adapt over time our findings that many do persist over time in
both emerging and mature economies points to a need for research about the apparent BG
adaptive capability to changing circumstances.
None of this is meant to suggest that the entrenchment and institutional voids approaches
completely lack validity. Indeed, one could reasonably interpret the Russian experience as being
consistent with the former and the South Korean with the latter. Our results tend to suggest,
however, that while these are possible outcomes, they are not necessarily general. Thus, we do
not reject either of these approaches; rather we argue that they are special cases. The challenge is
30
therefore to untangle the conditions of path dependence that lead to heterogeneous outcomes,
including the special cases.
Conclusion
So are BGs paragons or parasites with respect to their host countries economic development and
integration with the global economy? We suggest that this is the wrong question. Indeed, our main
conclusion is that the two apparently competing narratives that lead to this question should rather
be seen as special cases of a broader theory that is as yet elusive. Although the evidence presented
in this paper suggests that on balance BGs are development friendly, we recognize that there is
considerable cross-country variation in both BG organization and effects on country performance.
Thus we suggest a theoretical approach to BGs that focuses on the evolution of institutional and
organizational ecosystems within countries, with no convergence to a standard global model. As a
prevalent economic and social institution BGs are likely bear a complex and multifaceted
relationship with their host economies and we suspect the identification of their merits and failings
will occupy corporate governance and international business scholars for some time to come.
31
Bibliography
Almeida, H., & Wolfenzon, D. (2006). Should business groups be dismantled? The equilibrium
costs of efficient internal capital markets. Journal of Financial Economics, 79, 99-144.
Amsden, A. (1989). Asia's next giant: South Korea and late industrialization. New York: Oxford
University Press.
Amsden, A. H., & Hikino, T. (1994). Project execution capability, organizational know-how and
conglomerate corporate growth in late industrialization. Industrial and Corporate
Change, 3, 111-147.
Anderson, R. C., & Reeb, D. M. (2003). Founding family ownership and firm performance:
Evidence from the S&P 500. Journal of Finance, 58(3), 1301-1328
Arrow, K. J. (1974). The limits of organization. New York: John Brockman Associates.
Belenzon, S., & Patacconi, A. (2009). When is a Nexus of Contracts More Firm-Like? Theory
and Evidence from Business Groups. http://ssrn.com/abstract=1476546
Belenzon, S., Berkovitz, T., & Rios, L. A. (2013). Capital markets and firm organization: How
financial development shapes European corporate groups. Management Science, 59,
1326-1343.
Bertrand, M., Mehta, P., & Mullainathan, S. (2002). Ferreting out tunneling: An application to
Indian business groups. Quarterly Journal of Economics, 117, 121-148.
Carney, M., Gedajlovic, E. R., Heugens, P. P., Van Essen, M., & Van Oosterhout, J. H. (2011).
Business group affiliation, performance, context, and strategy: A meta-analysis. Academy
of Management Journal, 54, 437-460.
Chang, S.-J., Chung, C.-N., & Mahmood, I. P. (2006). When and how does business group
affiliation promote firm innovation? A tale of two emerging economies. Organization
Science, 17, 637-656.
Chang, S. J., & Hong, J. (2000). Economic performance of group-affiliated companies in Korea:
Intragroup resource sharing and internal business transactions. Academy of Management
Journal, 43, 429-448.
Chari, M. D. R. (2013). Business groups and foreign direct investments by developing country
firms: An empirical test in India. Journal of World Business, 48, 349-359
Claessens, S., Djankov, S., & Lang, L. H. P. (2000). The separation of ownership and control in
East Asian corporations. Journal of Financial Economics, 58, 81-112.
Claessens, S., Fan, J. P. H., & Lang, L. H. P. (2006). The benefits and costs of group affiliation:
Evidence from East Asia. Emerging Markets Review, 7, 1-26.
Dunning, J., & Narula, R. (2003). Foreign direct investment and governments: Catalysts for
economic restructuring: Routledge.
Faccio, M., Masulis, R. W., & McConnell, J. (2006). Political connections and corporate
bailouts. Journal of Finance, 61, 2597-2635.
Faccio, M. (2006). Politically connected firms. American Economic Review, 96, 369 -386.
Fisman, R. (2001). Estimating the value of political connections. American Economic Review,
91, 1095-1102.
Fisman, R., & Khanna, T. (2004). Facilitating development: The role of business groups World
Development, 32(4), 609-628.
Fogel, K. (2006). Oligarchic family control, social economic outcomes, and the quality of
government. Journal of International Business Studies, 37, 603-622.
Gerlach, M. L. (1992). Alliance capitalism: The social organization of Japanese business.
Berkely, CA: University of California press.
32
Ghemawat, P., & Khanna, T. (1998). The nature of diversified business groups; A research
design and two case studies. Journal of Industrial Economics, 46, 35-61.
Globerman, S & Shapiro, D.M. “Governance Infrastructure and U.S. Foreign Direct Investment”,
Journal of International Business Studies, 34, 2003, pp. 19-39
Granovetter, M. (2005). Business groups and social organization. In N. J. Smelser & R.
Swedburg (Eds.), The handbook of economic sociology (2nd ed., pp. 429-450).
Princeton: Princeton University Press
Guillen, M. F. (2000). Business groups in emerging economies: A resource based view. Academy
of Management Journal, 43, 362-380.
Hall, P., & Soskice, D. (2001). Varieties of Capitalism: the institutional foundations of
comparative advantage. Oxford: Oxford University Press.
Hansmann, H., & Kraakman, R. (2004). The end of history for corporate law. In J. N. Gordon &
M. J. Roe (Eds.), Convergence and persistence in corporate governance (pp. 33-68).
Cambridge: Cambridge University Press.
Herrigel, G. (2000). Industrial constructions: The sources of German industrial power (Vol. 9):
Cambridge University Press
Heugens, P. P., Van Essen, M., & Van Oosterhout, J. H. (2009). Meta-analyzing ownership
concentration and firm performance in Asia: Towards a more fine-grained understanding.
Asia-Pacific Journal of Management, 26, 481–512.
Hoskisson, R. E., Johnson, R. A., Tihanyi, L., & White, R. (2005). Diversified business groups
and corporate refocusing in emerging economies. Journal of Management, 31, 941-965.
Jia, N., Shi, J., & Wang, Y. (2013). Coinsurance within business groups: Evidence from related
party transactions in an emerging market. Management Science, 59, 2295-2313.
Kali, R. (1999). Endogenous business networks. Journal of Law Economics and Organization,
15, 615- 636.
Khanna, T. (2000). Business groups and social welfare in emerging markets: Existing evidence
and unanswered questions. European Economic Review, 44, 748-761.
Khanna, T., & Palepu, K. (2000). The future of business groups in emerging markets: Long run
evidence from Chile. Academy of Management Journal, 43(3), 268-285
Khanna, T., & Palepu, K. (1999). The right way to restructure conglomerates in emerging
markets. Harvard Business Review, 77, 125-134.
Khanna, T., & Rivkin, J. W. (2006). Interorganizational ties and business group boundaries:
Evidence from an emerging economy. Organization Science, 17, 333-352.
Khanna, T., & Rivkin, J. (2001). Estimating the performance effects of business groups in
emerging markets. Strategic Management Journal, 22, 45-74.
Khanna, T., & Yafeh, Y. (2007). Business groups in emerging markets: paragons or parasites?
Journal of Economic Literature, 45, 331-372.
Kock, C., & Guillen, M. (2001). Strategy and structure in developing countries: business groups
as an evolutionary response to opportunities for unrelated diversification Industrial and
Corporate Change, 10, 77–113.
Kumar, V., Gaur, A. S., & Pattnaik, C. (2012). Product diversification and international
expansion of business groups. Management International Review, 52, 175-192.
Lamin, A. (2013). The Business Group as an Information Resource: An investigation of business
group affiliation in the Indian software services industry. Academy of Management
Journal, 56, 1487-1509.
33
Lee, K., Peng, M. W., & Lee, K. (2008). From diversification premium to diversification
discount during institutional transitions. Journal of World Business, 43(1), 47-65
Luo, Y., & Tung, R. L. (2007). International expansion of emerging market enterprises: A
springboard perspective. Journal of International Business Studies, 38, 481-498.
Mahmood, I., & Mitchell, W. (2004). Two faces: Effects of business groups on innovation in
emerging economies. Management Science, 50, 1348-1365.
Mahmood, I. P., Zhu, H., & Zajac, E. J. (2011). Where can capabilities come from? Network ties
and capability acquisition in business groups. Strategic Management Journal, 32, 820848.
Masulis, R. W., Pham, P. K., & Zein, J. (2011). Family business groups around the world:
Financing advantages, control motivations, and organizational choices. Review of
Financial Studies, 24, 3556-3600.
Matthews, J. A. (2006). Dragon multinationals: New players in 21st century globalization AsiaPacific Journal of management, 23(1), 5-27.
Meuer, J., Rupietta, C., & Backes-Gellner, U. (2015). Layers of co-existing innovation systems.
Research Policy, 44(4), 888-910
Meyer, K. E. (2006). Global focusing: From domestic conglomerates to global specialists.
Journal of Management Studies, 43(5), 1109-1144.
Morck, R., Wolfenzon, D., & Yeung, B. (2005). Corporate governance, economic entrenchment,
and growth Journal of Economic Literature, 43, 655-720.
Morck, R., & Yeung, B. (2004). Family control and the rent seeking society. Entrepreneurship
Theory and Practice, 391-409.
Nolte, A., & Vliegenthart, A. (2009). Enlarging the varieties of capitalism: The emergence of
dependent market economies in East Central Europe. World Politics, 61, 670-702.
Pedersen, T., & Stucchi, T. (2014). Business groups, institutional transition, and the
internationalization of firms from emerging economies. In R. Ramamurti & A. CuervoGazurra (Eds.), Understanding multinationals from emerging markets (pp. 224-240).
Cambridge: Cambridge University Press.
Rajan, R. G., & Zingales, L. (2003). The great reversals: The politics of financial development in
the 20th century. Journal of Financial Economics, 69, 5-50.
Ramamurti R. & J. V. Singh (Eds.), Emerging multinationals in emerging markets (pp. 64-77).
Cambridge: Cambridge University press
Rodrik, D. (2008). Second-best institutions: National Bureau of Economic
Sabirianova, K., Svejnar, J., & Terrell, K. (2005). Distance to the efficiency frontier and foreign
direct investment spillovers Journal of the European Economic Association, 3(2-3), 576-586
Scharfstein, D. S., & Stein, J. C. (2000). The dark side of internal capital markets: Divisional
rent-seeking and inefficient investment. Journal of Finance, 55(6), 2537-2564.
Schneider, B. R. (2009). Hierarchical market economies and varieties of capitalism in Latin
America. Journal of Latin American Studies, 41, 553-575.
Schneider, B. R. (2010). Business groups and the state: the politics of expansion, restructuring,
and collapse. In A. M. Colpan, T. Hikinio & J. R. Lincoln (Eds.), The Oxford Handbook
of Business Groups (pp. 650-669). Oxford: Oxford University Press.
Tan, D., & Meyer, K. E. (2010). Business groups' outward FDI: A managerial resources
perspective. Journal of International Management, 16, 154-164.
Von Hippel, E., Franke, N., & Prügl, R. (2009). Pyramiding: Efficient search for rare subjects.
Research Policy, 38(9): 1397–1406.
34
White, R. E., Hoskisson, R. E., Yiu, D., & Bruton, G. D. (2008). Employment and market
innovation in Chinese business group affiliated firms: The role of group control systems.
Management and Organization Review, 4(2), 225-256
Williamson, O. E. (1991). Comparative economic organization: The analysis of discrete
structural alternatives. Administrative Science Quarterly, 36, 269-296.
Yiu, D. W. (2011). Multinational advantages of Chinese business groups: A theoretical
exploration. Management and Organization Review, 7, 249-277.
Yiu, D., Bruton, G. D., & Lu, Y. (2005). Understanding business group performance in an
emerging economy: Acquiring resources and capabilities in order to prosper. Journal of
Management Studies, 42, 183-206.
Yiu, D. W., Lu, Y., Bruton, G. D., & Hoskisson, R. E. (2007). Business groups: An integrated
model to focus future research. Journal of Management Studies, 44, 1551-1579.
Young, M. N., Peng, M. W., Ahlstrom, D., Bruton, G. D., & Jiang, Y. (2008). Corporate
governance in emerging economies: A review of the principal–principal perspective.
Journal of Management Studies, 45, 196-220.
Zattoni, A., Pedersen, T., & Kumar, V. (2009). The Performance of Group-affiliated Firms
during Institutional Transition: A Longitudinal Study of Indian Firms. Corporate
Governance: An International Review, 17(4), 510-523
35
Figure 1: BG: prevalence trends in 18 countries
36
37
38
Figure 2: Prevalence trends in seven countries
Percentage BG firms
0.9
0.8
0.7
R² = 0.0999
0.6
0.5
0.4
0.3
0.2
0.1
0
0
5000
10000
15000
20000
25000
Figure 3: business group prevalence and per capita GDP
39
30000
35000
40000
45000
Table 1: group prevalence and economic outcomes
VARIABLES
Business group
prevalence
Paper number
Firm (year)
observations
gdpcaplag
Rule of law_t
Observations
R-squared
(1)
(2)
FDI outward mns US$ (log)
Large sample
Good obs
sample
(3)
(4)
FDI inward mns US$ (log)
Large
Good obs
sample
sample
(5)
(6)
Patent activity grants (log)
Large
Good obs
sample
sample
(7)
(8)
Stock market cap (log)
Large
Good obs
sample
sample
2.345***
1.222*
1.329**
-0.058
0.152
0.149
2.262***
2.312***
(0.681)
0.009***
(0.002)
0.000
(0.683)
0.008***
(0.002)
-0.000
(0.568)
0.007***
(0.002)
0.000
(0.594)
0.006***
(0.002)
-0.000
(0.705)
0.006***
(0.002)
0.000**
(0.785)
0.002
(0.002)
0.000
(0.595)
0.007***
(0.002)
0.000**
(0.573)
0.005***
(0.002)
0.000
(0.000)
0.000***
(0.000)
-0.027
(0.252)
(0.000)
0.000***
(0.000)
-0.013
(0.245)
(0.000)
0.000**
(0.000)
0.028
(0.206)
(0.000)
-0.000
(0.000)
0.085
(0.213)
(0.000)
0.000***
(0.000)
-0.908***
(0.266)
(0.000)
0.000***
(0.000)
-1.284***
(0.292)
(0.000)
0.000***
(0.000)
0.046
(0.215)
(0.000)
0.000***
(0.000)
-0.139
(0.206)
162
0.438
153
0.396
150
0.260
166
0.243
153
0.254
166
157
158
0.166
0.098
0.139
Standard errors in parentheses
*** p<0.01, ** p<0.05, * p<0.1