Business groups and entrepreneurship in developing countries after reforms

JBR-08253; No of Pages 8 Journal of Business Research xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

Journal of Business Research

Business groups and entrepreneurship in developing countries after reforms Murali D.R. Chari a,⁎, Jaya Dixit b a b

Lally School of Management, Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY 12180, United States Indian School of Business, Hyderabad, India

a r t i c l e

i n f o

Article history: Received 25 October 2013 Received in revised form 12 December 2014 Accepted 13 December 2014 Available online xxxx Keywords: Business groups Reforms Entrepreneurship Developing countries Market development India

a b s t r a c t Drawing from the literature on markets and market development we develop hypotheses on the startup and growth of formal business by business groups relative to independent entrepreneurs after reforms. We then test the hypotheses in a large sample of firms that were started up after reforms in one large developing country—India. Findings show that (a) the likelihood of formal business startup by business groups, relative to independent entrepreneurs, declines with market development following reforms, (b) the likelihood of formal business startup by business groups, relative to independent entrepreneurs, is greater in industries privatized by reforms and in industries with greater foreign firm presence, and (c) formal businesses started by business groups experience greater growth than formal businesses started by independent entrepreneurs. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Starting and growing formal businesses, defined as privately owned or publicly owned firms that draw or seek to draw significantly on external resources, were very difficult in developing countries because of their underdeveloped markets before these countries enacted economic reforms (Leff, 1978, 1979). Business groups (BGs), which are collections of firms with common ownership and control by a family, were particularly successful in starting and growing formal businesses in many of these countries because they leveraged internal markets to supplement and overcome the constraints of underdeveloped markets (Amsden & Hikino, 1994; Leff, 1976). BGs, as a result, dominated formal business and economic activities in a number of developing countries including Argentina, Brazil, Chile, India, Malaysia, Mexico, Nicaragua, Pakistan, South Korea, South Africa, Taiwan, and Turkey (Amsden & Hikino, 1994; Guillen, 2000; Khanna & Palepu, 2000b; Leff, 1976, 1978, 1979). In recent years, a number of developing countries have enacted promarket economic reforms (hereafter called ‘reforms’) aimed at developing their markets to promote entrepreneurship and private enterprise (Hoskisson, Eden, Lau, & Wright, 2000). As a result of these reforms, these countries are becoming major economic forces in the world, and entrepreneurship (including the startup and growth of formal businesses) has been credited with playing ‘a key role in this development’ (Bruton, Ahlstrom, & Obloj, 2008: 1). Noting the growing importance of entrepreneurship in developing countries, and citing a paucity of research on the topic, scholars have called for more research into ⁎ Corresponding author. Tel.: +1 518 669 9164. E-mail addresses: [email protected] (M.D.R. Chari), [email protected] (J. Dixit).

entrepreneurship in developing countries (Bruton et al., 2008). Despite their widely acknowledged dominance in the startup and growth of formal businesses before reforms, there is virtually no research to our knowledge on the startup and growth of formal businesses by BGs in developing countries after reforms. Who starts businesses and who succeeds in growing their businesses are important questions for scholars, entrepreneurs, and policy makers (Shane, 2003). Relative to independent entrepreneurs, how likely are BGs to start and grow formal businesses after reforms? We examine the startup and growth of formal businesses by BGs relative to independent entrepreneurs in developing countries after reforms. We draw from the literature on markets and market development to advance hypotheses on the likelihood of formal business startup by BGs relative to independent entrepreneurs following reforms, and the relative success of such startups. We then test the hypotheses on a large sample of formal businesses started up in one large developing country that has enacted reforms, India. Our study makes important contributions as detailed in the Discussion section. Briefly, the study provides, to our knowledge, the first set of empirical findings on startup and growth of formal businesses by BGs in a developing country after reforms. The study also contributes to the literature on BGs by identifying some sources of BG advantage that persists after the reforms. 2. Background: entrepreneurial constraints and BGs in developing countries The institutional context, made up of a matrix of formal rules such as laws and regulations, informal norms, and the enforcement of rules,

http://dx.doi.org/10.1016/j.jbusres.2014.12.006 0148-2963/© 2014 Elsevier Inc. All rights reserved.

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

2

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx

significantly impacts economic activity in a country (North, 1990). Restrictive laws and regulations in developing countries were not supportive of markets, and the resulting deficiencies in “factor and product markets” made startup and growth of formal businesses very difficult (Leff, 1978, 1979: 46). Underdeveloped factor markets made the mobilization of capital, labor, and other production factors required to start formal businesses difficult (Leibenstein, 1968). Venture capitalists who mobilize capital to fund formal business startups were nonexistent. Paucity of educational institutions left a void in the labor market for trained managers and professionals (Khanna & Palepu, 2010). Absence of product market intermediaries made it difficult to obtain information to make entrepreneurial decisions (Khanna & Palepu, 2010). Restrictive regulations, in addition, introduced bureaucratic barriers to start formal businesses. These conditions led to pessimism about the prospects for startup and growth of formal businesses in developing countries (Leff, 1979). BGs were particularly successful under these conditions because they were able to supplement and overcome the deficient product and factor markets. Specifically, founders of BGs mobilized capital initially from their extended families and ethnic communities to start and operate multiple formal businesses (Leff, 1976). They subsequently leveraged internal markets, created by pooling the resources of their various businesses, to start and grow additional formal businesses (Amsden & Hikino, 1994; Leff, 1978, 1979). To overcome talent market voids BGs often ran their own management development programs (Khanna & Palepu, 2010). When faced with a lack of intermediary products, BGs started new businesses to produce these products. Operating businesses in multiple industries generated internal information flows which helped reduce uncertainties regarding production and investment decisions (Leff, 1979: 53). BGs adapted technologies imported from developed countries to overcome deficiencies in domestic technology sources (Amsden & Hikino, 1994). BGs were also better at dealing with bureaucratic hurdles as their multiple businesses provided economies of scope in lobbying (Ghemawat & Khanna, 1998; Leff, 1978: 668). Finally, repeated entry into new businesses provided BGs extensive experience in the entrepreneurial process (Amsden & Hikino, 1994; Guillen, 2000). BGs, thus, were credited with overcoming the entrepreneurial constraints of the pre-reform era. The success of BGs has spawned a large body of literature examining their effects on firm performance, risk sharing, innovation, and socio-economic welfare (Colpan, Hikino, & Lincoln, 2010; Khanna & Palepu, 2000b; Singla & George, 2013). Recognizing the importance of well-developed markets for economic growth, a number of developing countries have enacted reforms to transform their institutional contexts in support of market development and private enterprise (Hoskisson et al., 2000). What effect does this have on the startup and growth of formal businesses by BGs? 3. Hypotheses We draw from the literature on markets and market development for our hypotheses. Our main arguments are as follows. Markets evolve gradually following reforms and hence the startup and growth of formal businesses by BGs after reforms are best observed as markets develop over time (North, 1990). The development of markets eases prereform era constraints by increasing the availability of resources through markets, and this would lower the advantages of BGs in starting formal businesses relative to independent entrepreneurs. Accessing resources from markets however is not universally advantageous because markets are not always efficient. Specifically, when measurement problems are high, arriving at price and terms of contract are difficult for market participants, and therefore markets will be inefficient (Akerlof, 1970; Alchian & Demsetz, 1972; Coase, 1937; Williamson, 1979). Markets are also inefficient when the reliability of supply is important (Bernanke & Gertler, 1987), and when the need for secrecy prevents sufficient disclosure to aid measurement and thereby the

determination of price and terms of contract (Coffee, 1984). Because of market inefficiency, access to internal markets would provide BGs an advantage in starting formal businesses in certain areas of the economy, and in growing formal business startups. We develop these arguments below. 3.1. Market development following reforms and the startup of formal businesses by BGs Reforms are efforts to change the institutional context in support of market development and private enterprise (Hoskisson et al., 2000; North, 1990). Reforms typically include measures that strengthen shareholder protection, deregulate labor restrictions, and liberalize entry for product and factor market intermediaries (Chari & David, 2012; Cuervo-Cazurra & Dau, 2009; Khanna & Palepu, 2000a). Greater protection for shareholders contributes to equity market development by increasing investor confidence, which in turn increases access to and lowers the cost of capital (La Porta, Lopez-De-Silanes, Shleifer, & Vishny, 1997). Development of equity markets also enables entrepreneurs to defray startup risks. Entry liberalization for venture capitalists enables these intermediaries to raise capital and fund startups. Entry liberalization for educational institutions increases the availability of trained managers and professionals. Entry and growth in the number of product and factor market intermediaries such as market research firms, consultants, and rating agencies increase the amount of information and intermediary services available through the market for entrepreneurial decision making. The development of markets following reforms thus increases the availability of resources required to start formal businesses. Market development, however, occurs gradually over time, rather than immediately, after reforms (Ghemawat & Khanna, 1998). This is because institutional change, including those intended to support market development, is overwhelmingly incremental (North, 1990). While changes in formal rules can, at least in theory, be enacted quickly, changes in the other two components of the institutional matrix are necessarily incremental (North, 1990). Informal norms that extend and complement the new formal rules only evolve gradually over time through repeated exchange by organizations within the new policy framework (North, 1990). Supervisory institutions such as regulatory agencies that enforce the new rules also evolve only over time through a process of learning by doing (Panagariya, 2008). Even the new formal rules often evolve as regulatory agencies discover changes necessary to make the rules enforceable. The availability of resources required to start businesses therefore can be expected to increase as markets develop over time following reforms rather than immediately after reform initiation. An increase in resources available through markets will favor formal business startup by independent entrepreneurs who can now more easily marshal the necessary resources through the markets. In addition, as the availability of resources through markets increases, BGs' advantages in starting formal businesses that stem from their ability to use internal markets to overcome or supplement the underdeveloped markets would decline. The above arguments suggest a decline in the relative likelihood of formal business startup by BGs as markets develop following reforms. Hypothesis 1. The likelihood of formal business startup by BGs, relative to independent entrepreneurs, declines with market development following reforms.

3.2. Market inefficiency and the startup of formal businesses by BGs As noted by Coase (1937), Alchian and Demsetz (1972), and Williamson (1979), even within well-developed market economies there are contexts where measurement problems render markets inefficient. Thus, while the development of markets suggests a decline in

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx

the relative likelihood of formal business startup by BGs overall (as argued above), there would be pockets of the economy where, due to measurement problems, markets are inefficient and therefore internal markets provide BGs an advantage in starting formal businesses. We argue that starting formal businesses in response to two new opportunities accompanying reforms represents such pockets. In the pre-reform period, developing country governments reserved certain industries for state owned enterprises because these industries were considered critical to the economy, and because these industries involved complex operations and large investments which the governments felt were beyond the capacity of private enterprises. With reforms, many of these industries are privatized (i.e., de-reserved) because of the governments' revised assessment that private firms are better able to unleash the growth potential of these industries (GOI, 1991). While privatized industries are not new industries, they are new to private enterprise. The lack or rarity of relevant predecessors (i.e., private enterprises), and the consequent dearth of existing strategies to observe and evaluate, would therefore render envisioning viable strategies for these industries more difficult (Carroll & Delacroix, 1982). BGs would have an advantage in envisioning strategies in these industries because their prior startup experience would provide a frame of reference and the absorptive capacity to better interpret new information and make entrepreneurial conjectures in these new domains (Shane, 2003). Experience in the entrepreneurial (i.e., startup) process is tacit, intertwined with the organizational context where it arose, and therefore difficult to measure, price, and access through the market (Kogut & Zander, 1993). BGs can access this experience through their internal markets by moving employees with the relevant experience to the new startup (Guillen, 2000). The market for other resources such as capital is also likely to be inefficient in this context. The rarity or lack of relevant predecessor firms would make returns from the proposed venture difficult for market participants to measure and evaluate, and therefore arrive at price and terms of contract (Akerlof, 1970; Alchian & Demsetz, 1972). Access to internal markets would provide BGs an advantage in this context both because they provide an internal source for the resources and because their existing resources (i.e., the pooled assets, track records, and relationships) can be leveraged to gain better access to the markets. BGs, for example, can use their demonstrated success in prior ventures to signal credibility to resource providers. BGs have existing networks of social ties with resource providers for their other businesses which they can leverage. BGs also have the advantage of using guarantees from existing businesses to assuage resource provider skepticism. Finally, to the extent bureaucratic hurdles remain in these industries, their economies of scope in lobbying will provide BGs an advantage in overcoming these hurdles. For all of these reasons, we expect that BGs would have an advantage in starting formal businesses in industries privatized by reforms. Hypothesis 2. The likelihood of formal business startup by BGs, relative to independent entrepreneurs, is greater in industries privatized by reforms. Another source of opportunities accompanying reforms involves new markets and market growth created by foreign firms entering or scaling-up their presence in developing countries. A number of foreign firms enter or scale up their presence in developing countries following reforms (Bhattacharya & Michael, 2008). Foreign firms introduce new and technologically superior products to developing countries, which create new markets and expand potential market size (Khanna & Palepu, 2006). Because of their inability or unwillingness to modify their products and business models to developing countries, foreign firms limit themselves to upper income segments and large cities where affordability of their higher priced products is greatest (Khanna

3

& Palepu, 2006). Awareness of the new products, however, creates a large unmet demand for these new products among lower income segments and markets outside large cities, which can be effectively addressed only at much lower price points (Khanna & Palepu, 2006). This unmet demand offers significant opportunities for BGs and others to enter these industries. Entering these markets will require new business models to produce and sell the new and technologically superior products at much lower prices. The newness of these markets and the consequent absence of predecessors would make it difficult for entrepreneurs to envision suitable strategies. BGs have extensive experience in adapting imported technologies and products to local market needs from the pre-reform era. This prior experience will provide BGs a frame of reference and the absorptive capacity to interpret new information and make entrepreneurial conjectures and envision strategies for these new markets (Shane, 2003). As argued earlier, experience is difficult to measure, price, and hence access through the markets. Internal markets on the other hand enable BGs to draw on this experience. Due to the paucity of indigenous technology sources, producing the technologically superior products at much lower prices would likely require the acquisition and adaptation of imported technologies. Given uncertainties over enforcement, technology access through markets is risky—particularly in developing countries. BGs would have an advantage in accessing technology in this context because their experience with importing technology in the pre-reform era and their ability to stake the reputation of the entire group of businesses would help allay technology provider concerns (Khanna & Palepu, 2000b). Reputation and trust are tacit assets generated over time, and therefore not easily accessed by independent entrepreneurs through the market. The newness of these markets and the consequent absence or rarity of predecessors will also make it difficult for the providers of other relevant resources to measure and evaluate the venture and hence arrive at the price and terms of the contract (Akerlof, 1970; Alchian & Demsetz, 1972). As a result the market will be inefficient for these resources as well. As elaborated in developing the previous hypothesis, internal markets will provide BGs an advantage in this context, both by providing an internal source for the resources and by enabling better access to the market for these resources. For all these reasons, BGs would have an advantage in starting formal businesses in industries where foreign firms are successful and hence have a greater presence. Hypothesis 3. The likelihood of formal business startup by BGs, relative to independent entrepreneurs, is greater in industries with a greater foreign firm presence.

3.3. Growing the formal businesses The factors that lead to formal business startups are not necessarily the ones that determine their growth. The very act of starting a business causes information about the entrepreneurial opportunity to diffuse and invites competition (Shane, 2003). Successful growth of formal business startups therefore depends not only on the ability to exploit opportunities, but also on protecting entrepreneurial gains from competition (Ireland, Hitt, & Sirmon, 2003). Entrepreneurial gains can be protected from competition when opportunities are exploited by leveraging resources in a manner that is hard for competitors to imitate (Barney, 1991). As we argue below, seeking resources while limiting information leakage to prevent competition makes markets inefficient. Growing formal business startups also requires reliable access to resources, and markets can be unreliable. Access to internal markets provides BGs advantages in overcoming these inefficiencies, and consequently in growing formal business startups. First, the conflict between the resource providers' need for disclosure of relevant information to arrive at price and terms, and the entrepreneurs' need to limit leakage of sensitive information would mean

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

4

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx

that markets will be inefficient in providing the resources necessary for growth. Internal markets within BGs can help overcome this inefficiency both by providing an internal source for the required resources and, as explained earlier, by facilitating access to markets at better terms. For example, BGs can use guarantees from their existing businesses to negotiate access to capital without disclosing sensitive information. Second, growth depends on the appropriate timing of investments (McGrath, 1996). Investing too much too soon risks greater losses if uncertainties do not resolve favorably, and investing late risks losing early mover advantages. Appropriate timing of investments therefore requires that investments are made as critical uncertainties are resolved (McGrath, 1996). Markets are subject to cycles, and funds through markets can become unavailable when startups need them for investment (Bernanke & Gertler, 1987; Rajan, 1994). Internal markets within BGs can help overcome this limitation by providing funds to facilitate investment timing that is dictated by the resolution of critical uncertainties rather than by the vagaries of market cycles. Finally, knowledge of how to manage and grow startups is tacit and hence is best learned by experience (Shane, 2003). Supporting this idea, research on serial and other habitual entrepreneurs shows that prior entrepreneurial experience is an important predictor of startup growth and success (Eesley & Roberts, 2006). BGs, having grown by repeatedly starting businesses, are the quintessential habitual entrepreneurs in developing countries. Their greater experience with the entrepreneurial process would therefore provide BGs an advantage over independent entrepreneurs in successfully growing formal business startups. As argued earlier, experience is tacit and difficult to measure and therefore inefficient to access through markets. BGs can access this resource though their internal markets by moving people with the relevant experience to the focal startup. For all the above reasons, BGs would have an advantage in growing formal business startups. Hypothesis 4. Formal businesses started by BGs experience greater growth than formal businesses started by independent entrepreneurs.

4. The Indian context India followed a socialist economic policy from its Independence in 1947. The policy involved extensive restrictions on firms, including requiring entrepreneurs to obtain licenses from the government to start businesses, expand capacity, and even change product range. Foreign firm entry and operation were also restricted. In addition, several industries were reserved exclusively for state owned enterprises. These practices, often referred to as the “License Raj,” severely constrained the development and functioning of markets and dampened entrepreneurship and private enterprise (Chari & Banalieva, 2014). In 1991, India initiated pro-market economic reforms to develop markets and promote private enterprise (Chari & David, 2012). The need for licenses to start businesses was abolished except for a short list of industries with safety and security implications; the requirement of government permission to expand capacity and change product range was repealed; restrictions on foreign firm entry and operations were liberalized; all but a few industries previously reserved for the State were opened to private enterprise; shareholder protections were strengthened, and supervisory institutions were created (Panagariya, 2008). These changes have transformed the economy substantially from the restrictive “License Raj” to a more market oriented economy contributing to the development of markets and the strengthening of private enterprise and entrepreneurship (Panagariya, 2008). To be sure, some aspects of the economy are not fully open and bureaucratic hurdles remain despite reforms. Nevertheless, reforms have substantially freed-up the economy from the shackles of government control, and have helped propel India from a laggard to one of the leading emerging economies in the world (Panagariya, 2008). India, therefore, is an appropriate and interesting context to test our hypotheses.

5. Methods, data, and measures 5.1. Methods We estimate a logistic regression model of the following general form to test Hypotheses 1 through 3. The model is estimated on a sample of formal business startups in India from 1992, the first year after the country enacted reforms, to 2008 (Chari & David, 2012). BG affiliate ¼ b0 þ b1 market development þ b2 de‐reserved industry þ b3 foreign firm presence þ b4−n controls þ e: BG affiliate in the model is a binary variable with value 1 if the startup is by a BG and 0 if the startup is by an independent entrepreneur. Market development (defined in the Measures section) represents the level of market development in the country in the year of the startup. De-reserved industry is an indicator variable with value 1 if the startup is in an industry that was removed from the list of industries reserved for the State as part of the reforms, and 0 otherwise. Foreign firm presence represents the extent of foreign firm sales in the industry in the year of the startup. Controls in the model represent other factors that may influence the dependent variable. Specifically, BGs may have a prior propensity to compete in certain industries. We control for this by using the ratio of BG firms to total number of firms in the industry in the year before reform initiation. We also control for industry sector and the year of the startup. We control for industry sector rather than more finely defined industries because of the latter's correlation with the de-reserved industry variable. Hypothesis 1 expects the likelihood of formal business startup by BGs relative to independent entrepreneurs to decline with market development following reforms. A negative and significant coefficient b1 in the model will support the hypothesis. Hypothesis 2 expects that the likelihood of formal business startup by BGs relative to independent entrepreneurs is greater in dereserved industries. A positive and significant b2 in the model will support the hypothesis. Hypothesis 3 expects that the likelihood of formal business startup by BGs relative to independent entrepreneurs is greater in industries with a greater foreign firm presence. A positive and significant coefficient b3 in the model will support the hypothesis. To test Hypothesis 4 we estimate a regression model of the following general form. Growth ¼ b0 þ b1 BG affiliate þ b2−n controls þ e: Growth, in the above model, represents the growth of the startup in the first three years following the commencement of business. BG affiliate is a binary variable with value 1 if the startup is by a BG and 0 if the startup is by an independent entrepreneur. Other variables that may also have an impact on the startup's growth, namely total assets of the startup in the first year of operation, industry of the startup, the level of market development in the startup year, and year dummies to capture unobserved time period effects, are included as controls. Hypothesis 4 expects that formal businesses started by BGs experience greater growth than formal businesses started by independent entrepreneurs. A positive and significant coefficient b1 in the model will support the hypothesis. Since some of the startups may not have data to compute growth because they went out of business, or were acquired by other businesses, or were started after 2005 and our dataset ends in 2008, there is the possibility of regression results being influenced by this selection/survivor bias. We use a Heckman type estimation to control for this bias. 5.2. Data The Center for Monitoring the Indian Economy (CMIE) is the primary source of our data. CMIE is a reputed source, and has been used extensively in prior research (Elango & Pattnaik, 2013). Our interest in this

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx

5

Table 1a Formal business startups by industry sector and year.

Agriculture, fishing & mining Manufacturing Power & construction Retail & wholesale Financial services Non-financial services Total

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

Total

0 6 0 0 0 2 8

3 74 5 9 10 7 108

15 170 11 27 70 25 318

38 290 18 55 220 59 680

15 132 22 42 111 52 374

4 69 7 32 68 29 209

7 111 18 41 59 42 278

12 189 19 41 71 93 425

6 158 18 40 47 108 377

6 105 22 41 88 92 354

8 103 37 52 115 103 418

35 274 34 192 343 331 1209

21 168 31 106 295 228 849

11 111 29 50 113 130 444

15 75 33 41 91 132 387

3 21 19 10 29 110 192

0 8 12 7 18 44 89

199 2064 335 786 1748 1587 6719

paper is on formal business startups. A distinguishing characteristic of formal businesses is that, because they draw or seek to draw significantly on external resources, they prepare and disclose their audited financial information to fulfill their statutory obligations and/or to inform and attract external resource providers. While it does not cover all business in the economy, CMIE is among the more comprehensive sources of data on private and public firms that prepare and disclose their audited financial statements with over 19,000 privately owned and publicly traded firms in the database. The CMIE data therefore is well suited for our study. We begin with the population of all firms listed in CMIE between 1989 and 2008. We then select firms for which CMIE lists the year of incorporation as 1992 or later (7807 firms)—India enacted reforms in 1991. We then cross check the year of incorporation listed in CMIE manually using company sources and their filings with the government. In the vast majority of cases, the year of incorporation reported in CMIE corresponded with the year of incorporation or the year of commencement of business found through our manual check. In a small fraction of cases (about 7%) we found that the firm had been in operation before the year of incorporation reported in CMIE. We dropped these firms from our sample. The remaining startups included 532 that were owned by foreign firms or the State. Since our focus is on private enterprise by domestic firms, we drop these 532 startups to arrive at our final sample of 6719 formal business startups. Entrepreneurial startups have a high failure rate, and a large fraction of startups may not survive beyond their 3rd year (Shane, 2009). Not all firms that cease to exist can be considered to have failed, however, since some may have been successful but acquired by another firm. Since we require data for four years to compute our growth variable, our sample for testing Hypothesis 4 is a subset of the 6719 formal business startups (specifically, 3128 formal business startups) for which data on 4 years of operation are available. To get a sense for the fate of the rest of the firms, we tracked information from public sources for a randomly chosen sample of 1554 firms. The vast majority of the firms (about 79%) appear to have ceased operations. About 3% of the firms were acquired or merged into another firm, and we could not find any information on the rest (about 17%) of the firms. 5.3. Measures CMIE classifies firms as affiliates of BGs, or non-BG firms, the accuracy of which has been independently verified (Khanna & Palepu, 2000b). No firm is affiliated to more than one BG. We measure BG affiliation of the startup (BG affiliate in our models and tables) using an indicator variable with value 1 if the startup firm is classified by CMIE as affiliated to a BG and 0 otherwise. Since our sample excludes startups by others in the economy besides BGs and independent entrepreneurs, a value of 0 on our measure of BG affiliate represents startups by independent entrepreneurs. We track seven indicators of market development, namely the ratio of stock market capitalization to GDP, the number of portfolio managers, the number of credit rating agencies, the number of venture capital firms, the number of colleges for general education, the number of colleges for professional education, and the number of firms engaged in providing information and other business services, each year, to measure Market development. These indicators measure the extent of development in capital, labor, and product markets, and are similar to ones

used in prior research (Khanna & Palepu, 2000a). Appendix A lists the indicators and data sources. The seven indicators are highly correlated and we therefore take the first principal component of the seven indicators as our measure of market development. The first principal component accounts for a large proportion of the variance (78%) in the data indicating that it adequately captures the seven indicators. Consistent with our arguments in this paper, and the observation of other scholars (Khanna & Palepu, 2000a), the measure of market development shows a statistically significant upward slope with time following reforms. A number of industries were reserved for state owned enterprises before the reform initiation in 1991. All but three of these have been de-reserved as part of the reforms. Some of the industries were dereserved in 1991 at the outset of reforms, and others were dereserved in subsequent years as part of the reforms. We obtain information on de-reservation from relevant agencies of the Indian government, and construct an indicator variable to measure de-reserved industry with value 1 if the startup was in an industry that was de-reserved by reforms and 0 otherwise. Foreign firm presence is measured as sales of foreign firms in the industry divided by total sales in the industry, in the year of the startup. CMIE identifies foreign firms or subsidiaries of foreign firms, which we use to identify foreign firms in the industry (Chari & David, 2012). Dummy variables, one for each sector, namely agriculture, fishing and mining, manufacturing, power and construction, financial services, and non-financial services, are used to measure the industry sector. We control for unobserved time period effects using year dummies. Prior propensity of BGs to compete in industries is measured as the ratio of the number of BG affiliated firms in each industry to the total number of firms in the industry in 1990—the year before the reform initiation. With respect to variables in our model to test Hypothesis 4, Growth of the startup is measured as the average annual change in total assets of the startup in the three years following the firms' first year of operation. We winsorize the variable at the 1st and 99th percentiles to avoid undue influence from extreme values. BG affiliation of the startup is measured as described earlier. The growth of the startup may be influenced by industry. We control for industry using dummy variables, one for each four digit industry. Growth could be influenced by the initial size of entry. We control for this using total assets of the startup in its first year of operation (initial asset size). Economic conditions at the time of entry may also influence growth. We therefore use the level of market development in the startup year, and year dummies as additional controls. 6. Results Table 1a shows the distribution of formal business startups in the sample across industry sectors and years. Table 1b shows the proportion of formal business startups in each industry sector started by BGs and independent entrepreneurs. Table 2 shows means, standard deviations, and correlations between variables. The correlations between independent variables are low. Multicolinearity, therefore, is unlikely to bias our results. Table 3 shows test results for hypotheses 1–3. Model 1 includes just the control variables. In models 2, 3, and 4, we add each independent variable by itself to test Hypotheses 1, 2 and 3 respectively. Model 5 is

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

6

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx

Table 1b Share of formal business startups in each industry sector by business groups (BGs) and independent entrepreneurs (IEs).

Agriculture, fishing & mining Manufacturing Power & construction Retail & wholesale Financial services Non-financial services

BG's share

IE's share

15% 21% 36% 20% 21% 24%

85% 79% 64% 80% 79% 76%

the full model with all the independent variables added together. The results are consistent across all models, and we interpret the full model. The Chi2 statistic is significant indicating a good model fit. With respect to control variables, prior propensity is not significant. The tests for joint significance shows that the year dummies as a group and industry sector dummies as a group are significant. With respect to the variables of interest, the coefficient of market development is significant and negative supporting Hypothesis 1. The coefficients of both de-reserved industry and foreign firm presence are significant and positive supporting Hypotheses 2 and 3 respectively. Table 4 shows test results for Hypothesis 4. We use a Heckman regression model to control for potential selection/survivor bias. The model first estimates the inverse mills ratio associated with the probability that a startup is part of our sample of 3128 that survived to be included for our growth analysis, based on all independent variables in the model and the ratio of debt to assets of the firm in its first year of operation. The model then incorporates this inverse mills ratio in estimating the regression model for growth. Model 1 includes only the control variables. In model 2 we add BG affiliate to test Hypothesis 4. The model has a significant Chi2 statistic indicating a good model fit. With respect to control variables, the level of market development exerts a significant positive influence and initial asset size exerts a significant negative influence on growth. The tests for joint significance shows that the year dummies as a group and the industry dummies as a group are also significant. With respect to the hypothesized variable, the coefficient of BG affiliate is significant and positive supporting Hypothesis 4. 7. Discussion Our results show the following. First, the likelihood of formal business startup by BGs relative to independent entrepreneurs declines as markets develop following reforms. Second, the likelihood of formal business startup by BGs relative to independent entrepreneurs is greater in industries privatized by reforms and in industries with a greater foreign firm presence. Third, formal businesses started by BGs experience greater growth than those started by independent entrepreneurs.

Since underdeveloped markets set the stage for BGs' advantages in the startup and growth of formal businesses in the pre-reform era, a naïve extension would suggest that, as markets develop after reforms, BGs' advantages in the startup and growth of formal businesses would decline. Our findings above, however, show a more complex dynamic, with BGs having advantages in starting formal businesses in some areas and having advantages in growing formal business startups, even though their share of formal business startups in the economy declines. As pointed out by Baumol (1993), a good theoretical analysis can provide insight into and explain outcomes that are not easily accessible with naïve analysis. Our analysis which recognizes both the beneficial role of market development in increasing resource availability, and the inefficiencies of markets under certain circumstances, provides an explanation for the more complex dynamic that we observe. On the one hand, the greater availability of resources through markets, as the naïve analysis would indicate, explains the reduction in the share of formal business startups by BGs as markets develop. On the other hand, market inefficiencies due to measurement problems explain why internal markets provide BGs with advantages in starting formal businesses in the two particular areas of the economy. Similarly, market inefficiencies triggered by measurement problems, the need to limit disclosure of sensitive information, and the need for reliable funds explain why internal markets provide BGs advantages in growing formal business startups. Our study contributes to the literature on entrepreneurship in developing countries and to the literature on BGs. Who starts businesses and to what effect are important questions in entrepreneurship. We provide the first set of answers on formal business startups by BGs relative to independent entrepreneurs, and the growth of such startups, in a developing country after reforms. The decline in the proportion of formal business startups by BGs following reforms that we find shows a broadening in the base of entrepreneurs, and underscores the importance of independent entrepreneurs for formal business startup after reforms. For policy makers, the broadening of the base of entrepreneurs suggests that their intent to stimulate entrepreneurship through reforms has been successful. Relatively little is known in the literature about independent entrepreneurs in developing countries. Our finding that the importance of independent entrepreneurs for formal business startup is not matched by the success of their startups, therefore, suggests an urgent need for scholars to study this population of entrepreneurs and the various obstacles they may face in growing their formal business startups. Our finding of BGs' advantages in starting formal businesses in the two areas of the economy also holds interesting implications. Specifically, as we argued, market inefficiencies make these opportunities more difficult for independent entrepreneurs to exploit. These opportunities, consequently, would be underexploited in the absence of BGs.

Table 2 Means, standard deviations, and correlations. Panel A: startup model variables Mean

Std. dev.

1

2

3

4

1 Business group affiliate 0.22 0.41 2 Market development 0.25 1.72 −0.07 3 De-reserved industry 0.08 0.27 0.10 −0.03 4 Foreign firm presence 0.08 0.10 0.05 −0.14 −0.07 5 Prior propensity 0.58 0.23 0.01 0.05 −0.10 −0.13 N = 6719. Correlations larger (smaller) than 0.03 (−0.03) are significant at p b 0.05. Industry sector and year dummies not included in the interest of brevity. Panel B: growth model variables

1 2 3 4

Growth Business group affiliate Initial asset size Market development

Mean

Std. dev.

1

2

3

5.86 0.26 31.10 −0.42

38.39 0.44 169.54 1.19

0.15 −0.03 0.04

0.17 −0.03

0.00

N = 3128. Correlations larger (smaller) than 0.04 (-0.04) are significant at p b 0.05. Industry and year dummies not included in the interest of brevity.

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx Table 3 Results for Hypotheses 1–3: likelihood of formal business startup by BGs. Model 1

Model 2

Model 3

De-reserved industry

Industry sector dummies Year dummies Constant Chi2 Log likelihood

Model 5

1.65*** (4.65) −0.03 (0.20) Included Included −1.68*** (8.91) 242.4 −3426.4

−4.50** (3.16) 0.81*** (7.35) 2.20*** (6.05) −0.00 (0.03) Included Included 19.02** (2.90) 307.3 −3394

0.66*** (6.16)

Foreign firm presence Prior propensity

Model 4

−4.45** (3.13)

Market development

0.04 (0.25) Included Included −1.64*** (8.71) 221.2 −3437.1

0.040 (0.26) Included Included 18.93** (2.90) 234.5 −3430.4

0.08 (0.48) Included Included −1.70*** (9.00) 257.5 −3418.9

N = 6719. * p b 0.05, ** p b 0.01, *** p b 0.001; all two tailed tests. All Chi2 values are significant at p b 0.001. ‘t’ values in absolute numbers are in parentheses.

BGs therefore play a more limited but important role in starting formal businesses after reforms. Our study also contributes to the literature on BGs. Based on the observation that BGs are prevalent in countries with underdeveloped markets, research attributed BG advantages and existence to their ability to bridge voids in markets. As a corollary, as markets evolved following reforms, scholars expected the advantages of BGs to erode and the groups to become irrelevant (Khanna & Palepu, 1999: 126). Despite some evidence that the performance advantages of groups erode, there is little evidence for BGs' imminent demise and a number of authors have called for research to explain their continuing presence (e.g., Colpan et al., 2010). Our analysis shows that their extensive experience with the entrepreneurial process and their access to internal markets continue to confer advantages in starting formal businesses in certain areas of the economy and in growing formal business startups. Given the surge in entrepreneurship in developing countries after reforms (Bhattacharya & Michael, 2008), the continuing advantages that we identify contribute to explaining, at least partially, the presence and relevance of BGs after reforms. There are limitations to our study that can be addressed in future research. The availability of firm level data on a large sample of formal business startups makes India an ideal context to test our hypotheses. As in India, BGs used similar mechanisms to overcome entrepreneurial constraints and dominate the formal business landscape in many other developing countries. It is therefore reasonable to expect that

Table 4 Test results for Hypothesis 4. Effect of BG affiliation on formal business startup's growth. Model 1

Model 2

−0.01⁎ (2.36) 32.66⁎

9.65⁎⁎⁎ (3.96) −0.01⁎⁎ (3.19) 40.75⁎⁎

(2.00) Included Included Included −9.66 (0.47) 306.8 6719 3128

(2.68) Included Included Included −6.91 (0.34) 359.0 6719 3128

Business group affiliate Initial asset size Market development Industry dummies Year dummies Inverse mills ratio Constant Chi2 N n

All two tail tests. All Chi2 values are significant at p b 0.001. ‘t’ values in absolute numbers are in parentheses. Regression analyses with Heckman selection. N = sample size before selection, n = sample size after selection. ⁎ p b 0.05. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001.

7

our results may also hold in other developing countries that have enacted reforms. Further research using samples from other developing countries, however, will be necessary to confirm this expectation. 8. Conclusion Our study shows that the environment for entrepreneurship changes as markets develop after reforms. The proportion of formal business startups by BGs in the economy declines as a result. The BGs, however, have advantages in starting formal businesses in certain areas, and in growing their formal business startups. The findings enrich our understanding of factors influencing startup and growth of formal businesses in developing countries after reforms and shed light on some sources of BG advantage that persists after the reforms. Acknowledgments The authors thank Elitsa Banalieva, Associate Professor at North Eastern University, and Shyam Kumar, Associate Professor at Rensselaer Polytechnic Institute, for helpful comments on the earlier drafts of the manuscript. The authors also thank Sri Raghavan, CEO of BizSciences for helpful comments and suggestions. Appendix A. Items (indicators) measuring market development

Item

Aspect of market development

Source

Ratio of stock market capitalization to GDP Number of portfolio managers

Capital market

The World Bank

Capital market

Number of credit rating agencies

Capital market

Number of venture capital firms Number of colleges for general education

Capital market

Number of colleges for professional education

Labor market

Number of firms engaged in providing information and other business services including market research, product testing, and management consulting

Product market

Handbook of statistics on the Indian securities market, published in 2009 by the Securities and Exchange Board of India Handbook of statistics on the Indian securities market, published in 2009 by the Securities and Exchange Board of India Securities and Exchange Board of India Statistics of higher and technical education, published in the year 2011 by the Ministry of Human Resource Development, Government of India Statistics of higher and technical education, published in the year 2011 by the Ministry of Human Resource Development, Government of India The data base of companies in the Indian economy, compiled by the Center for Monitoring the Indian Economy

Labor market

References Akerlof, G. A. (1970). The market for “lemons”: quality uncertainty and the market mechanism. Quarterly Journal of Economics, 488–500. Alchian, A. A., & Demsetz, H. (1972). Production, information costs, and economic organization. American Economic Review, 62(5), 777–795. 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. Barney, J. (1991). Firm resources and sustained competitive advantage. Journal of Management, 17(1), 99–120. Baumol, W. J. (1993). Formal entrepreneurship theory in economics: existence and bounds. Journal of Business Venturing, 8(3), 197–210. Bernanke, B. S., & Gertler, M. (1987). Financial Fragility and Economic Performance. National Bureau of Economic Research Cambridge, Mass., USA.

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006

8

M.D.R. Chari, J. Dixit / Journal of Business Research xxx (2014) xxx–xxx

Bhattacharya, A. K., & Michael, D. C. (2008). How local companies keep multinationals at bay. Harvard Business Review, 86(3), 84–95. Bruton, G. D., Ahlstrom, D., & Obloj, K. (2008). Entrepreneurship in emerging economies: where are we today and where should the research go in the future. Entrepreneurship: Theory and Practice, 32(1), 1–14. Carroll, G. R., & Delacroix, J. (1982). Organizational mortality in the newspaper industries of Argentina and Ireland: An ecological approach. Administrative Science Quarterly, 169–198. Chari, M. D. R., & Banalieva, E. R. (2014). How do pro-market reforms impact firm profitability? The case of India under reform. Journal of World Business. http:// dx.doi.org/10.1016/j.jwb.2014.05.004. Chari, M. D. R., & David, P. (2012). Sustaining superior performance in an emerging economy: an empirical test in the Indian context. Strategic Management Journal, 33, 217–229. Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386–405. Coffee, J. C., Jr. (1984). Market failure and the economic case for a mandatory disclosure system. Virginia Law Review, 717–753. Colpan, A. M., Hikino, T., & Lincoln, J. R. (Eds.). (2010). The Oxford Handbook of Business Groups. New York: Oxford University Press. Cuervo-Cazurra, A., & Dau, L. (2009). Promarket reforms and firm profitability in developing countries. Academy of Management Journal, 52(6), 1348–1368. Eesley, C., & Roberts, E. (2006). The second time around?: Serial entrepreneurs. MIT Working Paper. Elango, B., & Pattnaik, C. (2013). Response strategies of local firms to import competition in emerging markets. Journal of Business Research, 66(12), 2460–2465. Ghemawat, P., & Khanna, T. (1998). The nature of diversified business groups: a research design and two case studies. The Journal of Industrial Economics, 46(1), 35–61. GOI (1991). Statement on Industrial Policy. Ministry of Industry, Government of India. Guillen, M. F. (2000). Business groups in emerging economies: a resource-based view. Academy of Management Journal, 43(3), 362–380. Hoskisson, R. E., Eden, L., Lau, C. M., & Wright, M. (2000). Strategy in emerging economies. Academy of Management Journal, 43(3), 249–267. Ireland, R. D., Hitt, M. A., & Sirmon, D. G. (2003). A model of Strategic Entrepreneurship: The Construct and Its Dimensions. Journal of Management, 29(6), 963–989. Khanna, T., & Palepu, K. (1999). The right way to restructure conglomerates in emerging markets. Harvard Business Review, 77, 125–135. Khanna, T., & Palepu, K. (2000a). 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. (2000b). Is group affiliation profitable in emerging markets? An analysis of diversified Indian business groups. The Journal of Finance, 55(2), 867–891. Khanna, T., & Palepu, K. (2006). Emerging giants: building world-class companies in developing countries. Harvard Business Review, 84(10), 60–69. Khanna, T., & Palepu, K. (2010). Winning in Emerging Markets: A Road Map for Strategy and Execution. Boston, MA: Harvard Business Press. Kogut, B., & Zander, U. (1993). Knowledge of the firm and the evolutionary theory of the multinational corporation. Journal of International Business Studies, 24(4), 625–645. La Porta, R., Lopez-De-Silanes, F., Shleifer, A., & Vishny, R. W. (1997). Legal determinants of external finance. The Journal of Finance, 52(3), 1131–1150. Leff, N. H. (1976). Capital markets in the less developed countries: the group principle. In R. McKinnon (Ed.), Money and Finance in Economic Growth and Development (pp. 97–122). New York: Marcel Dekker. Leff, N. H. (1978). Industrial organization and entrepreneurship in the developing countries: the economic groups. Economic Development and Cultural Change, 26(4), 661–675. Leff, N. H. (1979). Entrepreneurship and economic development: the problem revisited. Journal of Economic Literature, 17, 46–64. Leibenstein, H. (1968). Entrepreneurship and development. American Economic Review, 58(2), 72–83. McGrath, R. G. (1996). Options and the Entrepreneur: Toward A Strategic Theory of Entrepreneurial Behavior. Snider Entrepreneurial Center, Wharton School. North, D. C. (1990). Institutions, Institutional Change, and Economic Performance. Cambridge, MA: Harvard University Press. Panagariya, A. (2008). India: The Emerging Giant. New York: Oxford University Press. Rajan, R. G. (1994). Why bank credit policies fluctuate: a theory and some evidence. Quarterly Journal of Economics, 109(2), 399–441. Shane, S. (2003). A General Theory of Entrepreneurship: The Individual–Opportunity Nexus. Northampton, MA: Edward Elgar. Shane, S. (2009). Failure is a constant in entrepreneurship. The New York Times. Retrieved from http://boss.blogs.nytimes.com/2009/07/15/failure-is-a-constant-inentrepreneurship/. Singla, C., & George, R. (2013). Internationalization and performance: a contextual analysis of Indian firms. Journal of Business Research, 66, 2500–2506. Williamson, O. E. (1979). Transaction-cost economics: the governance of contractual relations. Journal of Law and Economics, 22(2), 233–261.

Please cite this article as: Chari, M.D.R., & Dixit, J., Business groups and entrepreneurship in developing countries after reforms, Journal of Business Research (2014), http://dx.doi.org/10.1016/j.jbusres.2014.12.006