The innovative performance of Group-affiliated firms (GAFs):
Evidence from Bangladesh, India and Pakistan
University of Groningen Faculty of Economics and Business
Bachelor Thesis International Economics and Business
Abstract
This paper provides a comparison on the innovative differences of group-affiliated firms (GAFs) and stand-alone firms (SAFs) using the entry barrier thesis formulated by Castellacci (2015). Its focus is on comparing the differences
in innovation between GAFs and SAFs and the impacts of the financial, labor market and legal institutions. The data used for empirical analysis is from the World Bank Enterprise Survey from the period 2013-2014 and it focuses on a sample of manufacturing firms across Bangladesh, India and Pakistan. The results are insignificant indicating that further empirical research needs to be done to link the role of institutions with regards to the innovativeness of GAFs
vis-avis SAFs.
Name Student: Elger Lodewijk Student ID number: S1983202
Student email: Elger.lodewijk@gmail.com Date Thesis: 02/06/2015
Introduction
Diversified businesses groups usually consist of legally independent firms based in several separate markets and are bound together by both formal and informal ties (Khanna and Rivkin 2001). The term group is defined as the area over which a decision making authority has influence (Hazari 1966). Business groups exist both in emerging economies and in developed economies. The structure of the groups vary considerably, in some cases they are bound by equity ties, in other cases they are vertically controlled groups. Apart from these formal ties there are also informal ties within business groups in the form of social and familial ties or a common sense of identity (Khanna and Yafeh 2005). These informal ties are usually pre-existing and formulate the norms and morals within the group which help the coordination of activities from within (Granovetter 2005). Especially in emerging economies business groups often account for a large share of the
nation’s employment and of the value added (Castellacci 2015). Because business groups are vital in developing economies, they are subject to a growing body of research interested in topics such as the ownership structure, how the groups formed, the economic performance of the groups and how they are integrated (Khanna and Yafeh 2010). The emphasis of the research is often placed on the corporate governance of the business group (La Porta et al., 1998). The majority of the focus has been on their performance economically and financially, but not necessarily on their organizational strategy. As of yet it is unclear how the innovative activities of firms interact with the institutions in which they do business.
Far less research has been done with regard to the organizational strategy of innovation, and even less research has been done studying innovation in
of the country. However, they can also help close the inequality gap by being more innovative and thus help make the domestic economy more innovative. Recent studies have found empirically that GAFs are more innovative than SAFs. The cause for this is that GAFs have more resources to their disposal and have the advantage of spillovers from foreign parts as well as within the group (Mahmood and Mitchell, 2004; Mahmood and Lee, 2004; Mahmood et al., 2011). If this is the case then it is in the interests of nations to help facilitate group affiliated firms to increase the innovativeness of the domestic economy.
The majority of Asian nations are in a developmental transition, they are moving from being classified as developing economies to being classified as developed economies, this process is called market development (Cuervo-Cazzura and Dau 2009). The interest of this paper lies in how this market development and thus the improved institutions affect the innovativeness of GAFs vis-à-vis SAFs. Are GAFs comparatively more innovative as the institutions become stronger or are GAFs less innovative vis-à-vis SAFs when the institutional quality increases? This paper will investigate this question from the entry barrier thesis (Castellacci 2015). The entry barrier thesis postulates that GAFs are more innovative in economies where institutions are better developed, efficient and stronger (Chari and David 2012). The entry barrier thesis is a new theoretical point of view contrasting the Institutional Barrier Thesis and it offers a contrasting opinion of the role GAFs play on the innovativeness of an economy.
This paper will use the most recent data of the World Bank enterprise Surveys (WBES) dataset using the data provided for the years 2013-2014. This data set was chosen as it offers rich and extensive survey datasets of thousands of businesses in developing countries including information on their strategies, economic performance and characteristics as well as their view on the
ways to regulate large corporations, offer new policies governments can enact to quicken the economic transition and help their domestic economy become more innovative.
This paper will be divided into five different sections. The first section titled theory explains the entry barrier thesis and the relevant literature relating to business groups in Asia and innovation. The second section is titled hypotheses and will elaborate on the hypotheses this paper seeks to investigate and the reasoning behind these hypotheses. The third section is titled Data and Variables, which states the dataset used and which variables are used from that data set. The fourth section is titled Econometric model and results, which will discuss the model used to analyze the data and the results from the data. The final and fifth section is titled Conclusion and discussion which will evaluate the results and discusses possible future research. Each section will also have subsections to help distinguish points.
Theory
Due to their influence business groups are of great interests to scholars and politicians. There is no consensus on the role of business groups; the contention towards business groups oscillates between a social-welfare reducing view and a social-welfare enhancing response to imperfect markets (Khanna and Palepu, 2000).
The two contrasting ideas are formulated in two competing theories on group affiliation and their effect on innovation, namely the established Institutional Barrier Thesis and a newly formed Entry Barrier Thesis. By studying innovation, insight can be gathered on the influences that business groups have. The two theories differ in how they regard the role of business groups in emerging economies, in the theoretical framework underpinning the theories, in the role changes of the groups over time and in their cross-country pattern. This paper will focus on the Entry Barrier Thesis postulated by Castellacci (2013).
In the Institutional Void Thesis GAFs are viewed as paragons, playing a vital role by filling in institutional voids lacking in developing nations. In this view
business groups originate and prosper where institutions are weak and thus group performance makes up for the lack of institutions. Chang et al. (2006), find that the positive effects of group affiliation on innovation is more pronounced in less developed economies supporting the Institutional Void Thesis.
Entry Barrier Thesis
This theory offers a parasitic view on the role of business groups for developing nations. This view propagates the idea that business groups are rent seeking, inefficient, are the cause of agency problems, suffer from capital miss-allocations, are oligopolistic producers and erect entry barriers in their domestic markets. There are two types of theoretical foundations on which the entry barrier thesis is founded. The theoretical framework is based on the evolutionary economics and on agency and corporate governance.
transition through market liberalization and GAFs have more resources and better capabilities to take advantage of the opportunities and. They also benefit more from having close ties to the government giving them privileged
information on new policies. This gives them an advantage to abuse their market position and adapt quicker than competitors.
The first pillar is the evolutionary economics perspective (Nelson and Winter, 1982; Nelson, 1995), which postulates that according to evolutionary logic, groups act in self-interests by raising entry barriers to ensure domestic market dominance and by doing so reduce competition, cause heterogeneity and block the entry of new firms and their new possible innovative capabilities (Mahmood and Mitchell, 2004). According to this view the development of technology is expensive, it is also difficult to imitate competitors and therefore the market is characterized by cumulativeness and persistence. During economic transition this could lead to a further strengthening of the larger and more dominant GAFs to the detriment of smaller, less competitive stand-alone firms. There are four corroborating arguments to support the view that better institutions will reduce the inefficiencies of GAFs.
The first argument is according to Fortune and Mitchell (2012); as the domestic market opens up and becomes more competitive, through evolutionary selection the less efficient and less competitive companies are driven out. This process is called dissolution. The more efficient companies are purchased and acquired by larger companies or by groups. This process is called acquisition. The larger firms and groups will grow through the acquisition of domestic companies, thus strengthen their capabilities and their capacity to innovate when contrasted to the smaller companies. GAFs will undertake adaptive measures to the change whereas SAFs will more likely experience dissolution.
Secondly, GAFs will be better equipped to deal with the knowledge spillovers from MNE's which enter the market during market liberalization (Schneider, 2008). They are better equipped to use the knowledge spillover than SAFs because they have better imitative and absorptive capacities.
competition and innovation could be positive, especially so when there is fierce competition between firms. Given that GAFs are more innovative than SAFs the higher degree of competition will strengthen their innovative lead. This is especially the case in developing economies where a lack of institutions is favorable for group affiliated-firms so they will get a head start.
The fourth corroborating arguments is that GAFs are rent-seeking with close ties to the government (La porta, Lopez-de-Silanes, and Schleifer. 1999; Rajan & Zingales 2003). Groups are formed with support from the government and can better take advantage of these connections through privileged information. During institutional change the relationship between the governments changes from formal ties to informal ties. However, Khanna and Yafeh(2005), and Chung Mahmood and Mitchell (2004) find that despite the change in ties, this is still an important way in which GAFs can get privileged information, lobby their interest and continue their rent-seeking activities to entrench their power.
The final corroborating argument is the mechanism-illustrated best by Chakrabarti et al (2011); Asset reconfiguration is important for resource development, two outcomes of asset reconfiguration are retrenchment and growth. Retrenchment strategies reduce the size and diversity of operations to cut expenses to become financially more stable and growth strategies aims for increasing the firm’s market share. During market reform the stronger firms will focus on growth whereas lesser firms will focus on retrenchment. When the institutions are stronger and more efficient the growth strategy is more effective than the retrenchment strategy. This means that more firms will focus on growth leading to more innovation. The stronger firms will focus on growth; they will be better candidates to join GAFs which leads to an increased innovative difference when compared to lesser firms who use the retrenchment strategy.
interests between the minority shareholders and the majority shareholders. Another agency issue is called tunneling, where value can be moved from one part of the pyramid to another to serve the interests of majority shareholders. These issues cause capital miss-allocations and depress the investments in innovation.
GAFs also face the self-destruction effect (Morck et al., 2005). This postulates that when innovation is created by one firm in the group, it could be competing with innovation from another firm from within the group. The innovation will
negatively affect each other wasting resources and reducing the incentives to invest.
Stronger institutions would lessen the agency problem of pyramids and could make groups more efficient and innovative. According to Fogel (2006), familial control of groups and the rent seeking actives of groups are more common in nations with less strong institutions. The implication is that as institutions improve, groups have to adjust their structure making them more innovative. Studies from Korea find that groups have a stronger incentive to increase efficiency than stand-alone firms (Kim et al. 2010)
Business groups in Asia
The topic of business groups and their innovative abilities with regard to their institutions is of great interest for emerging economies. Many Asian economies are in the process of economic transition and as predicted by the Institutional Void thesis there is a prevalence of business groups in these nations (Ghemawat and Khanna 1998). Many Asian economies have undertaken significant changes to their institutions and have undergone economic reform through privatization or trade liberalization to open up their markets and make them more efficient. It is important to study how business groups and stand-alone firms differ in
strategies in response to the changing markets.
firms. The families own the majority of the value listed of the groups and are often in high ranking positions in the firms. (Claessens et al., 2002).
The industrial sectors in emerging Asian economies have a high level of group affiliation. Claessens et al (2002) also finds that the dominance of business groups in Asia lies in the privileges they receive from the government. These privileges could include exclusive rights for importing or exporting, the
protection foreign competition which grants them effective monopoly power in the local market.
India in particular is interesting as it has seen several regimes over time; yet family business groups dominate the Indian economy. The TATA group has remained the largest business group in India since 1939, despite the changes in regimes and governments. The importance of family ties in India is highlighted as regulations were placed restricting existing companies becoming bigger, as a result family members took control over other firms and in this regard these firms belonged to the family, yet not through equity. It is far more difficult to regulate business groups when they operate in this way (Khanna and Palepu, 2000) it also makes it more difficult to empirically analyze these groups. While business groups have seen more interest, their innovative capacity and effect on their host nation has not seen a lot of interest. On the topic of
innovation there is little information when researching India Paksitan and Bangladesh, there is in particular a dearth of information when research
Bangladesh and Pakistan. India is moving from imitation to innovation (Kale and Little, 2007) however, in which way they are becoming more innovative is still uncertain. Bangladesh faces difficulties with its innovative capacity and suffers from a lack of institutions restricting its innovative capacity. Pakistan too sees a lack of innovation. In Latin America the results show that through economic transition group affiliate firms become more innovative when compared to stand alone firms. In combination with the new data set provided by WBES new
research can be done on innovation and group affiliated firms in Asia.
Hypotheses
hypotheses used will be the financial institutions, legal institutions and labor market institutions. According to the Entry Barrier Thesis these institutions will be more beneficial, meaning they will increase the innovative propensity of GAFs more than they will increase the innovative propensity of SAFs. The reasoning behind each individual institution is given below, however, it is also important to note that these institutions are the typical institutions used when analyzing business groups (Botero et al., Khanna and Yafeh, 2007, Carney et al., 2011). The WBES dataset also provides better data when studying for these three
institutions. According to Herstatt et al., (2006) labor markets are important for innovation. According to Tiwari et al., (2008) another major impediment to innovation is corruption stating that improved law institutions will improve the innovation. Khanna and Palepu (2000) find that good financial markets are required for business groups to innovate. Currie et al., 1999 find that property rights are important for innovation.
Financial Institutions
When financial institutions become stronger or more efficient they can provide better financing for firms. As they become more efficient they can offer an expanding array of methods in which firms can finance their businesses. According to Lazonick (2007) when firms finance activity through debt it can spur innovation. GAFs have more equity to leverage and thus stand to benefit more from stronger financial institutions than SAFs.
this by reduced risk taking, less commitment to long term projects. If this is mitigated though stronger institutions GAFs stand to gain more than SAFs.
H1a: GAFs are more innovative than SAFs in stronger financial institutions. During economic transition the institutions of the nation will improve. When institutions improve the initial advantages of GAFs will be eroded away. The ability to pool financial resources to finance innovative investments when the financial institutions are not capable of offering them is an important source of innovation for GAFs. As the financial institutions improve, this advantage will erode and so too will the advantages of GAFs erode when the financial
institutions improve because then SAFs also have the credit and capital required for innovative investments.
H1b: GAFs are no more innovative than SAFs in stronger financial institutions.
Legal institutions
When legal institutions become stronger patent protection becomes stronger, as GAFs are more likely to have patents to protect they stand more to gain from this. GAFs are also often market leaders and according to Aghion et al., (2005), they have a larger incentive to invest in innovation to make sure that they remain a market leader. Another reason is that the agency issues which stem forth from the pyramid structures will be mitigated; the agency issues arise when there is a mismatch of agendas within the stakeholders of the firm. When legal institutions become stronger the votes of the stakeholders are better represented and
balanced, reducing this agency issue.
H2a: GAFs are more innovative than SAFs in stronger legal institutions.
H2b: GAFs are no more innovative than SAFs in stronger legal institutions.
Labor market
Weak labor market institutions make it difficult for firms to hire qualified labor. As the labor markets strengthen the GAFs stand more to gain from the higher quality labor force through the acquisition of a higher skilled workforce and the removal of the less skilled workers. As GAFs are large firms, they have more workers and therefor stand to gain more from the stronger institutions, as skilled labor increases the innovativeness of a firm.
Labor market institutions hypothesis:
H2a: GAFs are more innovative than SAFs in stronger labor institutions
In inefficient labor markets firms will face difficulties hiring qualified workers because of regulations and a weak supply. Group affiliation can allow more easy access to qualified labor from within the group. Group affiliation has more of an incentive to train internally which compensates for the lack of strong labor market institutions.
H2b: GAFs are no more innovative than SAFs in stronger legal institutions
Dataset and parameters
difficulties and the nature of some issues can’t be properly dealt with through surveys.
This paper will focus on the newest data set from 2013-2014. The most
advantageous facet of this dataset is the provision on the ownership information which is vital determining which firms belong to GAFs and which firms do not. This distinguishes this data set from others. For example with the regulations in India families acquired firms through different measures than through equity making it difficult to discern if a firm belongs to a group or not
This paper will focus on the following nations: Bangladesh, India and Pakistan. From each of these nations’ only data on the manufacturing sector are taken. Because technological innovation is measured only manufacturing firms can be sampled. Bangladesh India and Pakistan are neighboring countries. Although the majority of the sample is from India, this should not hamper the results as the analysis is on individual firms and their individual assessment of the institutions. The data from Bangladesh and Pakistan was included to add variety for
robustness. This study focuses on 8614 firms of which 1144 firms are located in Bangladesh, 469 are from Pakistan and 7001 firms are from India.
Variables
To test the hypotheses, the following variables were chosen from the survey. Group-affiliated firms (GAFs): A dummy variable indicating if a firm belongs to a business group or is a standalone firm. Table 1 shows that in this sample approximately 20% of the firms are group affiliated firms.
Innovation (inno): A dummy variable obtained from question H.1 which indicates if a firm has introduced new or improved products and services in the last three years. This is used as an indicator if the firm has undertaken
technological innovation; this R&D helps the firm innovate as manufacturing firms introduce new products or production methods. It also helps firms imitate from more developed firms allowing them to innovate more (Cohen and
Firmsize (firmsize): a categorical indicator of the size of the firm obtained from question A.6. The possible values being: 1(micro) if the firm has less than 5 employees; 2(small) if the firm has between 5 and 19 employees; 3(medium) if it has between 20 and 99 employees; 4(large) if the firm has more than 100
employees. Table 1 shows that in this sample the average firm size lies between micro (between 0 and 5 employees) and small (between 5 and 19 employees). Firm age (agefirm): Obtained from question B.5, in what year the establishment began operations. In this sample the average age of the firm is 21 years, with a range of 0 to 150 meaning the oldest firm began operations 150 years ago. Iso-quality (isoquality): Obtained from question B.8, if the establishment has an internationally recognized quality certification. In this sample approximately 28% of the firms have reported to have internationally recognized quality certification.
Citysize (citysize): Obtained from question A.3, the size of the locality. The values range from 2 to 5. The possible values being: 2 a city with population over 1 million; 3 a population over 250,000 to 1 million: 4 a population between 50,000 and 250,000; 5 a population less than 50,000. In this sample the average population is around 3(a population over 250,000 to 1 million).
Product diversification (divers): Obtained from question D.1a3, what percentage the main activity or product represents of the total sales. In this sample approximately 92% of the total sales are derived from the main activity or product.
Obstacle to finance (obst_fin): Obtained from question k.30, variable indicating to what extent the access to finance represents an obstacle for the firm. This is a categorical variable ranging from 0 (no obstacle) to 4(very severe obstacle). This variable has been multiplied by -1; this is done to make the results more intuitive as now the higher the value the better the institution is. In this sample the
average perception of the obstacle of law is of a minor obstacle.
as now the higher the value the better the institution is. In this sample the average perception of the obstacle of law is of a minor obstacle.
Obstacle to law (obst_law): Obtained from question J.30c, variable indicating if business licensing and permits is an obstacle to the firm. This is a categorical variable ranging from 0 (no obstacle) and 4 (very severe obstacle). This variable has been multiplied by -1; this is done to make the results more intuitive as now the higher the value the better the institution is. In this sample the average perception of the obstacle of law is of a minor obstacle.
Table 1: summary statistics
Mean Standard Deviation min max
The econometric model
The goal of the econometric analysis is to estimate how the financial, legal, and labor markets influence the relationship between innovation (inno) and group affiliated firms (gafs).
The data has an inherent issue because of a self-selection bias (Khanna 2000). This means that some characteristics of firms can affect the probability of it belonging to a group and it can affect the firm’s propensity to innovate. This means some variables, measured or unobserved will impact the likelihood of a firm belonging to a group and also its innovativeness. The more successful a firm is the more likely it will be acquired by a group. As a result when we measure group affiliated firms their innovativeness, there is a bias as the affiliated firms belong to the groups are chosen precisely because they are more innovative. There is a self-selection into groups between GAFs and SAFs, the stronger firms are more likely to be selected into becoming GAFs. This means that GAFs will always appear to be more innovative, this affects the estimations of the innovativeness of GAFs.
To adjust for the self-selection bias a two-equation model is required, a model which takes into account the probability that a firm belongs to a group and the firm’s innovative propensity. The two equations form a recursive model because the dependent variable group affiliation (GAF) is also an explanatory variable in estimating innovation (inno). The dependent variables: innovation and gafs are dummy variables; a recursive bivariate probit model can model the probability of each of the dependent variables. Having dichotomous dependent variables
reduces the validity of many tests; the bivariate probit model is applicable with two dichotomous dependent variables of which the results are between 0-1. The first equation estimates the factors which may determine why a firm is acquired into a group and the second equation measure the factors for that firm’s
innovativeness. This way the self-selection bias is reduced as it is also included in the calculations for the innovative propensity. The model is specified as such: Equation 1: GAFs = α1 + β1firmsizei + γ1agefirmi + δ1isoi + ζ1diversi + εı1
Equation 2: inno= ω2GAFsi + β2 firmsizei+ γ2agefirmi+ δ2isoi+
The dependent variable of the first equation is gafs, and the explanatory
variables are: firmsize, agefirm, iso, divers. The dependent variable in equation two is innovation and the explanatory variables are: gafs, firmsize, agefirm, isoquality, divers, obst_law, obst_fin and obst_lab. The variable gafs in equation 2 is expected to be positive; the literature indicates that gafs should be more innovative propensity than stand-alone firms. The control variables firmsize agefirm iso and drivers are expected to be positive, meaning that they are expected to be positively related to the innovative propensity. The indicators for institutions such as obst_law obst_fin and obst_lab also expected to be positive, in line with the entry barrier thesis, meaning the better the institution the more innovative GAFs are vis-à-vis SAFs.
Because there is a self-selection issue, the firm-specific characteristics can affect the likelihood of a firm belonging to a group and that firm’s innovative capacity. This increases the likelihood that the variable GAFs is correlated with the error term. It can also overestimate the effect group affiliation on innovation. The two equations are estimated through a recursive bivariate probit model, where both equations are calculated simultaneously. The dependent variable of equation 1 is placed as an independent variable in equation 2, meaning that equation 1 is placed on the right hand side of the second equation. Because the firm-specific features According to Greene (2003: 715-716) with this method the self-selection bias can be circumvented this way.
Results
Table 4 bivariate probit model
Basic Model H1 H2 H3
Dependant variable inno inno inno inno
GAFs 0.0422 0.0422 0.0422 0.0422 (0.13) (0.13) (0.13) (0.13) firmsize 0.173*** 0.173*** 0.173*** 0.173*** (3.30) (3.30) (3.30) (3.30) agefirm 0.00307** 0.00307** 0.00307** 0.00307** (3.21) (3.21) (3.21) (3.21) isoquality 0.371*** 0.371*** 0.371*** 0.371*** (11.92) (11.92) (11.92) (11.92) obstacle_to_finance -0.0198 -0.0198 (-1.55) (-1.55) obstacle_to_law -0.0183 -0.0183 (-1.29) (-1.29) obstacle_to_labor -0.0376* -0.0376* (-2.53) (-2.53) Constant -0.844*** -0.844*** -0.844*** -0.844*** (-14.97) (-14.97) (-14.97) (-14.97) Observations 8614 8614 8614 8614 t statistics in parentheses * p < 0.05, ** p < 0.01, *** p < 0.001
The majority of the control variables are significant and positive. Firms are more likely to be innovative when they are large. The results show that with each step the size of the firm (firmsize) increases the innovative propensity by 17.7% .The age of the firm (agefirm) increases the innovative propensity by 0.3% and is significant, meaning that every additional year adds 0.3% to the innovative propensity. If the firm has an isoquality (iso) certification this adds 37% to the innovative propensity of the firm and is significant.
econometric estimation cannot conclude that they are more innovative than SAFs.
H1a hypothesized that stronger financial institutions will increase the
innovativeness of GAFs vis-à-vis SAFs. The results show that a categorical rise in the perceived obstacle of law will increase the innovative propensity of GAFs by 1.98%. However, the result is not significant and therefor the hypothesis cannot be verified. The results are unable to conclude verify that GAFs are more
innovative than SAFs in stronger financial institutions.
H2a hypothesized that legal institutions will increase the innovativeness of GAFs Vis-à-vis SAFs. The results show that that a categorical rise in the perceived obstacle of law will increase the innovative propensity of GAFs by 1.86%. However, the result is insignificant and the hypothesis cannot be verified. The results are unable to conclude that stronger institutions of law will increase the innovativeness of GAFs.
H3a hypothesized that stronger labor institutions will increase the
innovativeness of GAFs vis-à-vis SAFs. The results show that a reduction in the perceived obstacle of labor will increase the innovative propensity by 3.76%. The result is significant at the 10% significance level. At the 10% significance level each categorical rise in the quality of labor institution increases the
innovativeness of GAFs vis-à-vis SAFs by 3.76%. H3A can only be accepted at the 10% interval level.
The results are inconclusive as the hypotheses can’t be verified. Further measures need to be taken to properly analyze to test for the hypotheses. The bivariate probit model despite trying to account for the self-selection bias is not enough to overcome this issue. While the variables have the correct sign they are not significant. The paper fails to show that GAFs are more innovative than SAFs and it fails to show that stronger institutions make GAFs more innovative vis-à-vis SAFs.
Conclusion and discussion
Bangladesh, India and Pakistan. This paper compared the differences in
innovativeness between group affiliated firms and non-group affiliated firms and how the labor market, financial and legal institutions affect the innovativeness of these group affiliated firms. The empirical analysis used the WBES dataset from 2013 -2014 and analyzed 8614 manufacturing firms.
The results of the empirical model fail to provide evidence for the Entry Barrier Thesis. Further empirical research is necessary to prove the entry barrier thesis. The bivariate probit model did not overcome the self-selection bias. It is also possible that a different empirical model is required to aptly test the Entry Barrier Thesis. Further research using differing or improved econometric models is required to verify the literature and the hypotheses of this paper and the entry barrier thesis.
The literature provides a strong theoretical foundation for the Entry Barrier Thesis but the empirical analysis fails to verify the literature. There are several reasons why the model failed to verify the theory. The first reason is that the relationship between business groups and their informal ties with the
government is far more influential than expected. The literature acknowledges this but it is almost impossible to add this as data to the model due to the informal nature of the relationship (La Porta, Lopez-de-Silanes, and Schleifer. 1999; Rajan & Zingales 2003). This would increase the error margin of the model and help explain why the results are insignificant. The tenacity of the TATA group is emphasized in how they dealt with reforms by splitting equity up amongst family members.
The second reason is that the model is incorrect, that despite attempts to remedy the issues of self-selection more should be used than a bivariate probit model. Castellacci(2015) uses propensity score matching combined with a bivariate probit model with interaction variables and finds significant results.
References
Abegaz, Berhanu. "The diversified business group as an innovative organisational model for large state-enterprise reform in China and Vietnam."International Journal of Entrepreneurship and Innovation Management 5, no. 5 (2005): 379-400.
Aghion, Philippe, Nicholas Bloom, Richard Blundell, Rachel Griffith, and Peter Howitt. Competition and innovation: An inverted U relationship. No. w9269. National Bureau of Economic Research, 2002.
Belenzon, Sharon, and Tomer Berkovitz. "Innovation in business groups." Management Science 56, no. 3 (2010): 519-535.
Belloc, Filippo. "Corporate governance and innovation: A survey." Journal of Economic Surveys26, no. 5 (2012): 835-864.
Botero, Juan Carlos, Simeon Djankov, Rafael La Porta, Florencio López de Silanes, and Andrei Shleifer. "The regulation of labor." NBER working paper w9756
(2003).
Carney, Michael, Eric R. Gedajlovic, Pursey PMAR Heugens, Marc Van Essen, and J. Hans Van Oosterhout. "Business group affiliation, performance, context, and strategy: A meta-analysis." Academy of Management Journal 54, no. 3 (2011): 437-460.
Castellacci, Fulvio. Institutional Voids or Entry Barriers? Business Groups,
Innovation and Market Development in Emerging Economies. No. 2013-05. SPRU-Science and Technology Policy Research, University of Sussex, 2013.
Castellacci, Fulvio. "Institutional Voids or Organizational Resilience? Business Groups, Innovation, and Market Development in Latin America." World
Development 70 (2015): 43-58.
business group affiliation promote firm innovation? A tale of two emerging economies." Organization Science 17, no. 5 (2006): 637-656.
Chari, Murali DR, and Parthiban David. "Sustaining superior performance in an emerging economy: An empirical test in the Indian context." Strategic
Management Journal 33, no. 2 (2012): 217-229.
Chakrabarti, Abhirup, Elena Vidal, and Will Mitchell. "Business transformation in heterogeneous environments: The impact of market development and firm strength on retrenchment and growth reconfiguration." Global Strategy Journal 1, no. 1‐2 (2011): 6-26.
Chung, Chi-Nien, Ishtiaq Mahmood, and Will Mitchell.Political connections and business strategy: The impact of types and destinations of political ties on business diversification in closed and open political economic contexts. Working paper, National University of Singapore, Singapore, 2007.
Cohen, Wesley M., and Daniel A. Levinthal. "Absorptive capacity: a new
perspective on learning and innovation." Administrative science quarterly(1990): 128-152.
Claessens, Stijn, Joseph PH Fan, and Larry HP Lang. "The benefits and costs of group affiliation: Evidence from East Asia." Emerging Markets Review 7, no. 1 (2006): 1-26.
Colpan, Asli M., Takashi Hikino, and James R. Lincoln, eds. The Oxford handbook of business groups. Oxford Handbooks Online, 2010.
Cuervo-Cazurra, Alvaro, and Luis Alfonso Dau. "Promarket reforms and firm profitability in developing countries." Academy of Management Journal 52, no. 6 (2009): 1348-1368.
Currie, David, Paul Levine, Joseph Pearlman, and Michael Chui. "Phases of imitation and innovation in a North-South endogenous growth model." Oxford Economic Papers 51, no. 1 (1999): 60-88.
Fortune, Annetta, and Will Mitchell. "Unpacking firm exit at the firm and industry levels: The adaptation and selection of firm capabilities." Strategic Management Journal 33, no. 7 (2012): 794-819.
Ghemawat, Pankaj, and Tarun Khanna. "The nature of diversified business groups: A research design and two case studies." The Journal of Industrial Economics 46, no. 1 (1998): 35-61.
Granovetter, Mark. "Coase revisited: Business groups in the modern economy." Industrial and corporate change 4, no. 1 (1995): 93-130.
Greene, William H. "Econometric analysis, 5th." Ed.. Upper Saddle River, NJ (2003). Hazari, Rabindra Kishen, and A. N. Oza. The structure of the corporate private sector: A study of concentration, ownership and control. Asia Pub. House, 1966. Herstatt, Cornelius, Rajnish Tiwari, Stephan Buse, and Dieter Ernst. "India's National Innovation System: Key elements and corporate perspectives."East-West Center Working Paper (2008).
Himmelberg, Charles P., and Bruce C. Petersen. "R & D and internal finance: A panel study of small firms in high-tech industries." The Review of Economics and Statistics (1994): 38-51.
Kale, Dinar, and Steve Little. "From imitation to innovation: The evolution of R&D capabilities and learning processes in the Indian pharmaceutical
industry." Technology Analysis & Strategic Management 19, no. 5 (2007): 589-609. Khanna, Tarun. "Business groups and social welfare in emerging markets:
Existing evidence and unanswered questions." European Economic Review44, no. 4 (2000): 748-761.
Khanna, Tarun, and Krishna Palepu. "The future of business groups in emerging markets: Long-run evidence from Chile." Academy of Management journal 43, no. 3 (2000): 268-285.
Khanna, Tarun, and Yishay Yafeh. "Business Groups and Risk Sharing around the
World*." The Journal of Business 78, no. 1 (2005): 301-340..
Khanna, Tarun, and Yishay Yafeh. "Business groups in emerging markets: Paragons or parasites?." ECGI-Finance Working Paper 92 (2005).
Kim, Hicheon, Heechun Kim, and Robert E. Hoskisson. "Does market-oriented institutional change in an emerging economy make business-group-affiliated multinationals perform better? An institution-based view." Journal of International Business Studies 41, no. 7 (2010): 1141-1160.
Porta, Rafael, Florencio Lopez‐de‐Silanes, and Andrei Shleifer. "Corporate ownership around the world." The journal of finance 54, no. 2 (1999): 471-517. La Porta, Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny. "The quality of government." Journal of Law, Economics, and organization 15, no. 1 (1999): 222-279.
Lazonick, William. "The US stock market and the governance of innovative enterprise." Industrial and Corporate Change 16, no. 6 (2007): 983-1035. Leff, Nathaniel H. "Industrial organization and entrepreneurship in the
developing countries: The economic groups." Economic development and cultural change (1978): 661-675.
Mahmood, Ishtiaq P., and Chang-Yang Lee. "Business groups: entry barrier– innovation debate revisited." Journal of Economic Behavior & Organization 54, no.
4 (2004): 513-531.
Mahmood, Ishtiaq P., and Will Mitchell. "Two faces: Effects of business groups on innovation in emerging economies." Management Science 50, no. 10 (2004): 1348-1365.
Mahmood, Ishtiaq, Chi-Nien Chung, and Will Mitchell. "The evolving impact of combinatorial opportunities and exhaustion on innovation by business groups as market development increases: The case of Taiwan."Management Science 59, no. 5 (2013): 1142-1161.
4 (2004): 513-531.
Mahmood, Ishtiaq P., and Will Mitchell. "Two faces: Effects of business groups on innovation in emerging economies." Management Science 50, no. 10 (2004): 1348-1365.
Mahmood, Ishtiaq P., Hongjin Zhu, and Edward J. Zajac. "Where can capabilities come from? Network ties and capability acquisition in business groups."Strategic Management Journal 32, no. 8 (2011): 820-848.
Morck, Randall, Daniel Wolfenzon, and Bernard Yeung. Corporate governance, economic entrenchment and growth. No. w10692. National Bureau of Economic Research, 2004.
Myers, Stewart C., and Nicholas S. Majluf. "Corporate financing and investment decisions when firms have information that investors do not have."Journal of financial economics 13, no. 2 (1984): 187-221.
Nelson, Richard R. "Recent evolutionary theorizing about economic change." Journal of economic literature (1995): 48-90.
Rajan, Raghuram G., and Luigi Zingales. "The great reversals: the politics of
financial development in the twentieth century." Journal of financial economics69, no. 1 (2003): 5-50.
Reeb, David, Mariko Sakakibara, and Ishtiaq P. Mahmood. "From the editors: Endogeneity in international business research." Journal of International Business Studies 43, no. 3 (2012): 211-218.
Schneider, Ben Ross. "Economic liberalization and corporate governance: the resilience of business groups in Latin America." Comparative politics(2008): 379-397.
Stein, Jeremy C. "Information production and capital allocation: Decentralized versus hierarchical firms."The journal of finance 57, no. 5 (2002): 1891-1921. Strachan, Harry W. Family and other business groups in economic development: The case of Nicaragua. Praeger Publishers, 1976.
