INTRODUCTION
Empirical research in finance and strategy continues to debate whether business groups are paragons or parasites (Khanna & Palepu, Reference Khanna and Palepu2000a, Reference Khanna and Palepu2000b; Yiu, Lu, Bruton, & Hoskisson, Reference Yiu, Lu, Bruton and Hoskisson2007). A substantial amount of studies, often in finance literature, indicate the effect of tunnelling, which leads to value expropriation of minority stockholders (Bae, Kang, & Kim, Reference Bae, Kang and Kim2002; Bertrand, Mehta, & Mullainathan, Reference Bertrand, Mehta and Mullainathan2002). Similarly, strategy scholars highlight the negative impact of business group (henceforth, BG) affiliation, due to organizational inertia (Kumar, Gaur, & Pattnaik, Reference Kumar, Gaur and Pattnaik2012). However, a rich body of empirical studies show that affiliated firms outperform stand-alone firms (Khanna & Palepu, Reference Khanna and Palepu2000a; Reference Khanna and Palepu2000b; Manikandan & Ramachandran, Reference Manikandan and Ramachandran2015), which leads us to question whether the value-constraining perspective dominates the value-enabling perspective or vice versa. Since mergers and acquisition are important and challenging corporate strategy actions, thus it is pertinent to use these contesting theoretical lenses of BG affiliation to theorize and empirically examine the acquisition performance of affiliate acquirers relative to stand-alone firms.
Defined as a network of legally separate and independent firms, BGs are connected through vertical ties of ownership control and horizontal ties of cross-holding and interlocking directors, in addition to family and social ties (Encarnation, Reference Encarnation1989; Khanna & Rivkin, Reference Khanna and Rivkin2001). In firms in both developed economies (e.g., Japan) and emerging markets (EMs) such as India, South Korea, China, and Latin American countries, BGs control significant ownership (Khanna & Rivkin, Reference Khanna and Rivkin2001; Leff, Reference Leff1978). Extant literature has extensively studied their effects on affiliate performance (Bertrand et al., Reference Bertrand, Mehta and Mullainathan2002; Khanna & Rivkin, Reference Khanna and Rivkin2001), innovation (Lee, Lee, & Gaur, Reference Lee, Lee and Gaur2016), internationalization (Elango & Pattnaik, Reference Elango and Pattnaik2007), and other firm-level outcomes (for a review, see Carney, Gedajlovic, Heugens, Van Essen, & Van Oosterhaut, Reference Carney, Gedajlovic, Heugens, Van Essen and Van Oosterhout2011). In support of the value-enabling perspective of BG affiliation, most of these studies have adopted an institutional economics lens, beginning with the premise that BG affiliation substitutes for market failures in emerging economies (Khanna & Rivkin, Reference Khanna and Rivkin2001).
In contrast, other studies highlight the dark or constraining implications of BGs. The first such implication is the principal–principal agency conflict (PP conflict), which deals with the expropriation of a firm's wealth by its controlling shareholders (Dalton, Hitt, Certo, & Dalton, Reference Dalton, Hitt, Certo and Dalton2007; Dharwadkar, George, & Brandes, Reference Dharwadkar, George and Brandes2000; Gilson, Reference Gilson2006), through subtle activities such as transfer pricing manipulations, misallocation of capital, and non-prudent ventures, as well as outright siphoning of resources (Johnson, La Porta, Lopez-de-Silanes, & Shleifer, Reference Johnson, La Porta, Lopez-de-Silanes and Shleifer2000). The severity of PP conflict is contingent on regulatory discipline – in particular, investor protection laws and the functioning of equity markets (La Porta, Lopez-de-Silanes, Shleifer, & Vishny, Reference La Porta, Lopez-de-Silanes, Shleifer and Vishny2000). In turn, such issues may be accentuated in EMs, where institutional voids are prevalent (Su, Xu, & Phan, Reference Su, Xu and Phan2008). A second constraining implication stems from organizational inertia (Hannan & Freeman, Reference Hannan and Freeman1984), which manifests in the institutionalized routines of all the affiliates, which may constrain their ability to adapt (Yiu et al., Reference Yiu, Lu, Bruton and Hoskisson2007).
In the context of post-acquisition performance, we frame our enquiry to develop two contrasting sets of arguments based on these value-constraining and value-enabling perspectives of BGs. Using a data set of 440 deals comprising both domestic and cross-border acquisitions by Indian firms during the period 2002–2013, we test our theoretical models. Our results show that relative to stand-alone firms, post-acquisition performance of affiliate acquirers, measured by Tobin's Q (Huang, Zhu, & Brass, Reference Huang, Zhu and Brass2017) and buy-hold abnormal return methodology (BHAR), is better than that of stand-alone firms. The results confirm the value-enabling role of BG affiliation in the context of post-acquisition performance. Furthermore, we investigate the role of intergroup heterogeneity among BGs, in terms of the effect of a BG's product diversification on the post-acquisition performance of its affiliate firm (Denrell, Fang, & Winter, Reference Denrell, Fang and Winter2003). Here again, we develop arguments based on competing theoretical perspectives to examine the role of the multidomain network structure of BGs on their post-acquisition performance. Our results are similar to our baseline hypothesis that post-acquisition performance is better for firms affiliated with BGs with extensive product diversification.
Our explanation and methods for testing it in turn make two distinct contributions to the current literature. Our post-acquisition performance results refute any causal link between BG affiliation and presence of any constraining logic; rather, they corroborate the value-enabling logic of group affiliation in an EM context. Because we also control for institutional development, this study confirms that, despite developments in the institutional landscape, the positive impact of BGs in an EM persists (Chittoor, Kale, & Puranam, Reference Chittoor, Kale and Puranam2015). Although extant literature addresses value-constraining and value-enabling logics of BGs separately, little research investigates these dual perspectives together. Additionally, in our post-hoc tests using cumulative abnormal returns (CAR), the results show that relative to stand-alone firms, performance of affiliate acquirers is negative and significant. These opposite and significant results further suggest that the market might be using negative heuristic biases for deals affiliated with BGs in accordance with the value-constraining logic. Although announcement effects measured as CAR do reflect future performance, we surmise that external stakeholders may not have enough information to assess the performance consequences (Lyon, Barber, & Tsai, Reference Lyon, Barber and Tsai1999). To the best of our knowledge, in EM context, this study is amongst the first to compare the deal performance measured as the announcement effects with the performance of a deal, post-completion (Meyer, Reference Meyer2015).
We thus begin by providing a background of literature on post-acquisition performance. Next, we discuss research into BGs and their role in an EM context. We then develop our theoretical arguments using value-constraining and value-enabling perspectives of BG to derive contrasting hypotheses about the post-acquisition performance of BG affiliate acquirers. Similarly, we develop contrasting hypotheses of positive or negative impacts of diversification of BG on the acquisition performance of affiliated firms. After, we elaborate the methodology and variables used, we discuss the empirical findings of our analysis. The final section contains conclusion, limitations, and some scope for further research.
THEORETICAL BACKGROUND AND HYPOTHESES
Business Groups in Emerging Markets
Despite valuable contributions from an impressive body of empirical research, studies on post-acquisition performance have resulted in inconclusive evidence (King, Dalton, Daily, & Covin, Reference King, Dalton, Daily and Covin2004; Moeller & Schlingemann, Reference Moeller and Schlingemann2005). Nevertheless, acquisitions continue to be a popular lever for inorganic growth strategy, and in response, a recent stream of research relies on various theoretical lenses to explain the potential antecedents and contingencies of post-acquisition performance. It acknowledges the role of concentrated ownership, in the form of BGs, which reduces principal–agent conflict and thereby enhances wealth creation for the acquirer (Bhaumik & Selarka, Reference Bhaumik and Selarka2012). In places such as India, South Korea, and Latin America, BGs tend to be controlled by families; in contrast, the government remains the dominant equity owner in Chinese BGs (Yang, Ru, & Ren, Reference Yang, Ru and Ren2015). Considering the critical importance of BGs in EMs, we believe it is pertinent to examine their impacts on major strategic decisions about mergers and acquisitions (M&As).
In EMs, BGs represent a network form of organization in which member firms are interconnected by social relationships, such as family ties, kinship, or ethnic or friendship ties (Encarnation, Reference Encarnation1989). A rich body of literature has examined BGs’ influences on various outcomes, such as growth strategies (Manikandan & Ramachandran, Reference Manikandan and Ramachandran2015), entrepreneurship (Chari & Dixit, Reference Chari and Dixit2015), financial performance (Kim, Kim, & Hoskisson, Reference Kim, Kim and Hoskisson2010), degree of internationalization (Lamin, Reference Lamin2013; Yiu, Reference Yiu2011), and research and development investments. Unlike multidivisional governance structures, in which divisional heads report to a firm's corporate headquarters, the affiliate firm in a BG is a separate legal entity with its own governance bodies, shareholders, directors, and auditors. These structural characteristics alleviate the resource allocation concerns typical in M-form organizations (Manikandan & Ramachandran, Reference Manikandan and Ramachandran2015). As a result, BG-affiliated firms enjoy limited autonomy while receiving benevolent interventions from group headquarters (Kumar et al., Reference Kumar, Gaur and Pattnaik2012). By filling the voids in the economic institutions of imperfect factor markets for labour, capital, and technology, BGs play a critical role (Guillén, Reference Guillén2000; Khanna & Rivkin, Reference Khanna and Rivkin2001). Nevertheless, literature in both strategy and finance domains has adopted two divergent theoretical perspectives to highlight the impact of BGs on their affiliate firms: (1) the value-constraining logic that cites PP conflict (Dharwadkar et al., Reference Dharwadkar, George and Brandes2000) and organizational inertia (Hannan & Freeman, Reference Hannan and Freeman1984) and (2) the value-enabling logic that refers to how BGs fill institutional voids (Khanna & Palepu, Reference Khanna and Palepu2000a, Reference Khanna and Palepu2000b).
Value-Constraining Logic of BGs
Agency theorists argue that though concentrated ownership in the hands of company directors may ameliorate manager–owner agency conflict, it might create PP conflict or type-II agency conflict (Bhaumik & Seralaka, Reference Bhaumik and Selarka2012). In contrast with the traditional agency problem (Fama & Jensen, Reference Fama and Jensen1983; Jensen & Meckling, Reference Jensen and Meckling1976), type-II agency conflict signifies the potential for value expropriation to occur, such that an opportunistic block equity holder deprives minority holders of the returns on their investments (Cho, Reference Cho1999; Dharwadkar et al., Reference Dharwadkar, George and Brandes2000). Non-performance can also result from various forms of tunnelling (Bertrand et al., Reference Bertrand, Mehta and Mullainathan2002; Gao & Kling, Reference Gao and Kling2008; Johnson et al., Reference Johnson, La Porta, Lopez-de-Silanes and Shleifer2000), such as misallocation of capital, transfer-pricing manipulations by buying supplies at above-market prices and selling at lower prices, suboptimal investment decisions, and so forth (Claessens, Djankov, & Lang, Reference Claessens, Djankov and Lang2000; Morck, Wolfenzon, & Yeung, Reference Morck, Wolfenzon and Yeung2005; Young, Peng, Ahlstrom, Bruton, & Jiang, Reference Young, Peng, Ahlstrom, Bruton and Jiang2008). In the context of Korea, Bae at el. (Reference Bae, Kang and Kim2002) find support for tunnelling hypothesis such that in acquisition announcements by a group affiliated firm, minority shareholders suffer. They find that mean cumulative abnormal return of large chaebol bidders with controlling ownership is −5.77 percent which is significant at one percent level. These findings are also echoed by studies of Ferris, Kim, and Kitsabunnarat (Reference Ferris, Kim and Kitsabunnarat2003) and Kim et al. (Reference Kim, Lyn, Park and Zychowicz2005). We submit that the severity of this conflict is amplified in contexts such as EMs, where regulatory regimes are weak and the chances of block holder opportunism to occur are higher (La Porta et al., Reference La Porta, Lopez-de-Silanes, Shleifer and Vishny2000). In such environments, firms with concentrated equity holding have lower firm value and reduced growth (Li & Qian, Reference Li and Qian2013; Lins, Reference Lins2003; Morck et al., Reference Morck, Wolfenzon and Yeung2005). The key manifestations of PP conflict results in cross-subsidizing the poorly performing firms at the cost of better firms (Chittoor et al., Reference Chittoor, Kale and Puranam2015) and the allocation of resources toward suboptimal projects, leading to value expropriation (Kim, Lyn, Park, & Zychowicz, Reference Kim, Lyn, Park and Zychowicz2005). For example, controlling shareholders may do M&A deals just to pursue private benefits, rather than for economic reasons (Bae et al., Reference Bae, Kang and Kim2002). Examining Korean capital investment decisions, Kim et al. (Reference Kim, Lyn, Park and Zychowicz2005: 969) find that ‘BG-affiliated firms make investment decisions that diverge from the wealth maximization principle, whereas non-chaebols exhibit behaviour that seems to be more consistent with the maximization of shareholder wealth’. Such expansion tends to benefit controlling shareholders rather than maximizing the wealth of the whole firm.
The second part of constraining logic is that BGs could suffer from legacy issues and organizational inertia (Carney, Reference Carney2008; Hannan & Freeman, Reference Hannan and Freeman1984). Here, the constraining aspect implies that, relative to stand-alone firms, the embeddedness of BGs makes them more resistant to change and could lead to suboptimal decisions and practices (Yiu et al., Reference Yiu, Lu, Bruton and Hoskisson2007). These inertial forces in groups arise due to the ‘proliferation of rules, routines and internal organizational arrangements’ (Guillén, Reference Guillén2002: 511) and can create institutionalized routines and increased bureaucracy (Kumar et al., Reference Kumar, Gaur and Pattnaik2012). Owing to these rigidities, the affiliated firm may apply the routines which may not be suitable in integrating the target firm, leading to impaired performance (Levitt & March, Reference Levitt and March1988). Carney and Gedajlovic (Reference Carney and Gedajlovic2003) note that administrative heritage and powerful path dependence lead to structural inertia in Asian business groups and hinder the development of new skills. We submit that acquisitions are complex strategy actions as target firms are rooted in different rules and institutions. Hence, organizational inertia associated with a group could hinder the amalgamation of the affiliate firm and the target firm and even lead to chaos affecting the performance outcomes for the acquirer firm. On the other hand, non-affiliated firms may be more flexible and open to structural transformation in order to extract the synergistic benefits of acquisitions. Thus, we hypothesize:
Hypothesis 1a: In EMs, firms affiliated with BGs will have poor post-acquisition performance relative to unaffiliated firms.
Value-Enabling Logic of BGs
In EMs, advantages of group affiliation go beyond the ‘substitute for market failures’ premise and include sharing of resources, knowledge, and brand equity (Khanna & Palepu, Reference Khanna and Palepu2000a). We suggest that in the context of M&As, these advantages of BG affiliation would lead to benevolent impact on the post-acquisition performance. We now discuss that relative to standalone firms, how BG affiliation would create a value-enabling logic in both ex-ante and ex-post phase of an acquisition process.
In ex-ante phase of an M&A, the focal acquirer firm must deal with high uncertainty due to information asymmetry, which can lead to a host of challenges in the ex-ante deal-making phase. A business group, with its social capital, could help mitigate the concerns of information asymmetry in target screening and selection, negotiation, and due diligence, which are accentuated in emerging economies (Akerlof, Reference Akerlof1970; Wright, Filatotchev, Hoskisson, & Peng, Reference Wright, Filatotchev, Hoskisson and Peng2005). The role of the BG's management cadre also is critical, in that the central management team delivers managerial services (Khanna & Palepu, Reference Khanna and Palepu1997). In addition, the risk of an ‘empire-building’ motivation is mitigated in the case of a BG affiliated acquirer, because the acquirer firm must exhibit unbiased due diligence, executed and monitored by a central decision-making body (Jensen, Reference Jensen1986). Therefore, the concerns of type-I agency issues and their manifestations in an M&A setup become mitigated (Singh & Delios, Reference Singh and Delios2017).
In a similar vein, BG affiliation could help maximize ex-post benefits from the acquisition too. Because BGs have repertoires of unique resource and capabilities, this structure potentially provides an ideal setting for an extended resource-based view, in which a focal firm can access its network's resources and capabilities (Arya & Lin, Reference Arya and Lin2007; Dyer & Singh, Reference Dyer and Singh1998; Lavie, Reference Lavie2006). An affiliated firm is connected to other member firms vertically in the form of equity ties, through horizontal linkages in the form of director interlocks, or both. Thus, after an M&A deal closes, these network resources can help the focal firm exploit opportunities that may remain elusive to other firms (Guillén, Reference Guillén2000). Attesting the greater growth opportunity for affiliated firms, Manikandan and Ramachandran (Reference Manikandan and Ramachandran2015) find that in measures of Tobin's Q, around 60 percent of affiliated firms exceeded their comparable set of non-affiliated firms. In most deals, wealth addition occurs when cost and revenue synergies are realized (Scherer & Ross, Reference Scherer and Ross1990). Thus, BG-affiliated firms might utilize the best operational practices employed across the network of the affiliate firms to realize cost synergies. Although cost synergies are typically easily realized, revenue synergies are challenging and take time to materialize (Rappaport & Sirower, Reference Rappaport and Sirower1999). For a group-affiliated firm, it is possible to increase the chances of realizing revenue synergies, usually through cross-selling. Moreover, the realization of revenue synergies is a function of the focal firm's ability to create new product-market domains by recombining resources (Cullinan, Le Roux, & Weddigen, Reference Cullinan, Le Roux and Weddigen2004). Access to group wide scientific talent and technology act as catalyst for product or service innovation too (Kim & Lui, Reference Kim and Lui2015). Similarly, the BG's financial controls facilitate affiliate firms’ innovativeness (White, Hoskisson, Yiu, & Bruton, Reference White, Hoskisson, Yiu and Bruton2008). The BG's reputational capital can be instrumental in legitimizing new product or service innovations, resulting in increased probability of greater revenue synergies (Belenzon & Berkovitz, Reference Belenzon and Berkovitz2010; Choi, Lee, & Williams, Reference Choi, Lee and Williams2011). Finally, political capital helps reduce external dependencies and secure approvals and support from the government (Faccio, Reference Faccio2006; Zheng, Singh, & Mitchell, Reference Zheng, Singh and Mitchell2015). On the basis of these arguments, we propose that:
Hypothesis 1b: In EMs, firms affiliated with BGs will have better post-acquisition performance relative to unaffiliated firms.
Diversification of BGs: Value Constraining or Value Enabling?
Most prior research has made the implicit assumption that BGs are diversified conglomerates (Majumdar & Bhattacharjee, Reference Majumdar and Bhattacharjee2014). However, all BGs are not alike, and substantial differences mark their diversification strategies (Kedia, Mukherjee, & Lahiri, Reference Kedia, Mukherjee and Lahiri2006). For example, many renowned groups in India (e.g., TATA, Reliance, Aditya Birla group, Mahindra) span various industries, but other groups (e.g., Hero MotoCorp, Amtek Auto) have a very narrow focus. Therefore, the differential impacts on post-acquisition performance due to a BG's degree of diversification represents a pertinent question (Carney et al., Reference Carney, Gedajlovic, Heugens, Van Essen and Van Oosterhout2011). In this hypothesis, we advance two reasons which could potentially lead to a negative impact of diversification of the BG on the post-acquisition performance of the affiliate firm.
Since the chances of resources shifting from one-member firm to another are greater in a group with greater diversification (George & Kabir, Reference George and Kabir2008; Young et al., Reference Young, Peng, Ahlstrom, Bruton and Jiang2008), it could lead to more chances for value expropriation. In such largely diversified groups, BGs serve to cross-subsidize poor-performing firms at the cost of other firms (Chittoor et al., Reference Chittoor, Kale and Puranam2015). In the context of emerging markets, a large number of such intra-group transactions are not transparent to outside investors or analysts, hence outside investors demand risk premia (Carney, Reference Carney2008; Dewenter, Novaes, & Pettway, Reference Dewenter, Novaes and Pettway2001; Gomes, Reference Gomes2000). This lack of transparency is also reflected in higher analyst forecast error and dispersion for BG firms (Pattnaik, Chang, & Shin, Reference Pattnaik, Chang and Shin2013). Hence, we suggest that manifestation of these heightened PP conflicts could lead to higher cost of capital, therefore greater financing cost might reduce the acquisition performance.
Second, we submit that a highly diversified group is controlled and coordinated by a central organization, through a corporate development unit (Khanna & Rivkin, Reference Khanna and Rivkin2001; Leff, Reference Leff1978). This body typically maintains strict monitoring and control over the investment decisions of its member firms, and it shoulders the responsibility of managing processes and routines to share knowledge and capabilities across firms (Keister, Reference Keister1998; White et al., Reference White, Hoskisson, Yiu and Bruton2008). This control may manifest in organizational routines and result in a semiautomatic pattern of behaviours and actions (Cyert & March, Reference Cyert and March1963; Zollo & Winter, Reference Zollo and Winter2002). Literature has acknowledged the positive impact of strategic agility to realize better post-acquisition performance (Junni, Sarala, Tarba, & Weber, Reference Junni, Sarala, Tarba and Weber2015). Thus, as discussed previously, such situations could lead to greater organizational inertia (Hannan & Freeman, Reference Hannan and Freeman1984) and render the dominant logic less effective for complex actions of mergers and acquisitions (Nadolska & Barkema, Reference Nadolska and Barkema2007). In summary, this constraining logic could prevent the BG affiliate acquirer from making the necessary adjustments to routines and processes to reflect the new post-acquisition scenario. Conversely, firms affiliated to groups with low product diversification would have less commitments to established patterns of behaviours and institutionalized practices and hence could be more efficient in adapting to the target firm post-acquisition. Thus, we propose that:
Hypothesis 2a: In EMs, among BG-affiliated firms, group-level product diversification will relate negatively to post-acquisition performance of the affiliated firm.
Conversely, we argue that a firm's growth opportunities are a function of its organizational governance structure and framework (Penrose, Reference Penrose1959). Accordingly, a high degree of BG diversification could create more opportunities for its member firms (Khanna & Palepu, Reference Khanna and Palepu2000a). Greater BG diversification relates positively to the growth opportunities for its member firms, because it provides affiliate firms with access to a comprehensive portfolio of resources and capabilities that could be critical in a post-acquisition scenario (Denrell et al., Reference Denrell, Fang and Winter2003; Lamin, Reference Lamin2013). In line with the institutional voids hypothesis, Ramaswamy, Purkayastha, and Petitt (Reference Ramaswamy, Purkayastha and Petitt2017) suggest that compared with a less diversified group, a diversified BG can address institutional voids more effectively. They also argue that, in contrast with narrowly focussed groups, groups with greater diversification possess a larger set of competencies that can help them create opportunities associated with diverse skills. Byun, Choi, Hwang, and Kim (Reference Byun, Choi, Hwang and Kim2013) find that greater diversification in a group generates a co-insurance effect such that cost of capital for BG-affiliated firm is 0.48% lower relative to stand-alone firms, thereby substantially reducing the financing cost for the acquirer firm.
A highly diversified BG also possesses routines and processes to manage the complexity associated with its multiunit, multimarket organization (Mayer, Stadler, & Hautz, Reference Mayer, Stadler and Hautz2015). Such capabilities may help affiliate firms to ease the complexities associated with extracting synergies in an M&A deal. Manikandan and Ramachandran (Reference Manikandan and Ramachandran2015) find that greater portfolio diversification provides affiliate firms with better chances for growth. This diversification could also help an acquirer firm realize revenue synergies, which otherwise remain very challenging. Beyond extracting cost and revenue synergies, firms from EMs conduct many deals to acquire strategic resources and assimilate knowledge and technological capabilities (Luo & Tung, Reference Luo and Tung2007). As discussed previously, vertical and horizontal linkages of a BG may help it share the best practices developed in the group, which aid the focal acquirer's efforts to absorb new knowledge (Frenz & Letto-Gillies, Reference Frenz and Letto-Gillies2009; White et al., Reference White, Hoskisson, Yiu and Bruton2008). This experience is richer and offers broader knowledge across multiple context when the BG diversification is higher. Put together, we thus propose that greater product diversification of the affiliate group can help the affiliate improve its post-acquisition performance. Thus, we hypothesize:
Hypothesis 2b: In EMs, among BG-affiliated firms, group-level product diversification will relate positively to post-acquisition performance of the affiliated firm.
METHODS
Empirical Context
We draw on evidence from India, an EM where the quality of corporate governance has been weak (Estrin & Prevezer, Reference Estrin and Prevezer2011) and equity concentration in the form of family-owned BGs is ubiquitous (Cuervo-Cazurra, Reference Cuervo-Cazurra2006). The dominance of family-owned BGs and cross-ownership patterns amongst the affiliated firms suggests that this context offers an appealing and unique setting in which to test the value-constraining versus value-enabling perspectives of BG affiliation. In contrast with state- and bank-owned BGs in China and Japan, respectively, Indian BGs are independent of any state intervention and solely responsible for their strategic actions (Ramaswamy et al., Reference Ramaswamy, Purkayastha and Petitt2017).
Our choice of the time period for this study (2002–2013) was influenced by the stage wise institutional reforms adopted in India. Although product market reforms in India began in 1991, capital market reforms started taking place only in the early half of the past decade (Majumdar & Bhattacharjee, Reference Majumdar and Bhattacharjee2014). These financial market reforms served as a catalyst for surges in domestic and cross-border acquisitions by Indian firms (Popli & Sinha, Reference Popli and Sinha2014).
Data
For data related to the M&A deals, such as their announcement date and the Standard Industrial Classification codes of the acquirer and target firms and other deal characteristics, we relied on the Thomson Financial SDC Platinum database, a commonly used database for M&A studies in developed and emerging economy contexts (Gaur, Malhotra, & Zhu, Reference Gaur, Malhotra and Zhu2013). We collected data on company financials and stock prices from the Centre for Monitoring Indian Economy's (CMIE) database, called Prowess.
We considered data on M&A deals completed by listed Indian acquirers from 2002 to 2013. Starting with the full population of deals, we considered majority stake deals (equity stake bought >= 50%) closed by publicly listed Indian acquirers. Next, we filtered out buybacks, deals by financial firms, and deals in which the acquirer's identity was not clear. To test H2a and H2b, we considered only deals carried out by BG-affiliated firms. The sample distribution is given in Table 1.
Industry and year wise deal numbers
Performance Measurement
Dependent variable
We employed Tobin's Q as a measure to capture post-acquisition performance. Following extant studies, we measured the post-acquisition performance of an acquirer using Tobin's Q one year after the focal deal (Hauser, Reference Hauser2018). We measured Tobin's Q by calculating the ratio of market capitalisation plus total assets minus net worth to total assets.
Explanatory variables
For H1a and H1b, we use a dichotomous variable, namely business group, where 1 represents a group-affiliated firm and 0 represents a stand-alone firm. For H2a and H2b, we measure group diversification by taking a count measure of distinct industries of the BG (Elango & Pattnaik, Reference Elango and Pattnaik2007). Table 2 summarizes the variables, measures, and sources of data.
Variable definition and data sources
Control variables
Following existing studies examining post-acquisition performance, we controlled for several explanatory variables that represent potential confounding effects in our regression models. We control for promoter ownership, a key component in the ownership and governance structure of a firm, representing the percentage equity held by the Indian promoter(s) (Ayyagari, Dau, & Spencer, Reference Ayyagari, Dau and Spencer2015). To control for the prior performance of the acquirer, we take a profitability ratio, namely prior profit margin. Because a target's public status accounts for the amount of information available to the acquirer and thus could affect performance (Capron & Shen, Reference Capron and Shen2007), we include a binary variable to indicate this public status (target listing status = 1 for public firms and 0 otherwise). Following Gubbi, Aulakh, Ray, Sarkar, and Chittoor (Reference Gubbi, Aulakh, Ray, Sarkar and Chittoor2010), we control for the potential slack of the acquiring firm, measured through debt-to-equity ratio. Extant literature investigating the impact of relatedness on M&A performance (King et al., Reference King, Dalton, Daily and Covin2004) suggests that the degree of similarity between acquirer and acquired firms improves post-acquisition performance. Therefore, we include a binary variable namely relatedness to code related deals (relatedness = 1 for deals in related industries and 0 otherwise). We control for the percentage stake (Stake Acquired) acquired by the acquirer in the focal deal. Because R&D intensity also influences post-acquisition performance (Le, Park, & Kroll, Reference Le, Park and Kroll2014), we control for it. To capture the state of institutional reforms, we also include a variable called reforms index that uses the Index of Economic Freedom published by the Heritage Foundation and The Wall Street Journal, as used frequently in studies in EM contexts (Kane, Holmes, & O'Grady, Reference Kane, Holmes and O'Grady2007). Furthermore, we segregate cross-border deals because they are complex and risky – even more so for EM acquirers, due to their liabilities of foreignness and emergingness (Madhok & Keyhani, Reference Madhok and Keyhani2012; Zaheer, Reference Zaheer1995). We employ a binary variable to this end (cross-border deal = 1 and 0 otherwise). Since high technology also increases the complexity in valuation of the target firms (Capron & She, Reference Capron and Shen2007), we also control for the type of economic activity of the target firm (OECD, 2011) and classify it as high-tech or low-tech using a binary variable (High-tech Target = 1 for high-tech targets and 0 otherwise). Finally, we also use a binary variable, called, recession year, to account for any fixed effects due to global economic downturn. We have winsorized all variables at 1% level.
RESULTS
Main Findings
We report the descriptive statistics and partial correlation coefficients of the full sample in Table 3 and for BG-affiliated firms in Table 4. The correlations between the independent variables are consistently moderate to low. Nevertheless, to mitigate multicollinearity concerns, we measure the variance inflation factor; its value ranged from 1.01 to 1.63, confirming that multicollinearity was not an issue in our empirical models.
Descriptive statistics and correlations
Notes: * p < 0.1, ** p < 0.05, *** p < 0.01
Descriptive statistics and correlations for BG affiliated firms
Notes: * p < 0.1, ** p < 0.05, *** p < 0.01
In Table 5, we present the ordinary least squares (OLS) regression results. Model 1 contains only control variables with Tobin's Q as the dependent variable. Models 2 and 3 test for the influence of our hypothesized relationships on the dependent variable. The error terms in the OLS regression models are clustered per BGs.
Ordinary least square regression of performance using Tobin's Q as the dependent variable
Notes: Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01
In Model 2 of Table 5, we find a positive and significant coefficient for BG affiliation (b = 0.106) with a p value of 0.013. This finding implies that BG affiliation is associated with a 10.6% increase in acquisition performance, measured through Tobin's Q, one year after deal completion. That is, acquiring firms which are affiliated to a business group are likely to exhibit 10.6% higher acquisition performance as compared to non-affiliated acquirers. This finding provides support for H1, and thus attests to the value-enabling perspective of BG affiliation.
To analyse the differential impacts arising out of heterogeneity in BG characteristics, we turn to the subsample of only group-affiliated firms. In Table 5, Model 3 shows the performance implication of group diversification. The results are similar to our previous finding, that is, there is a positive and significant co-efficient of group diversification (b = 0.00676) with a p value of 0.019. The result implies that a one-unit increase in group diversification would lead to a 0.67% increase in acquisition performance measured by Tobin's Q over one year. Therefore, acquiring firms affiliated to a more diverse business group are likely to exhibit higher acquisition performance. In other words, an addition of a new industry by a business group in its portfolio of companies is likely to increase the acquisition performance of its affiliate companies by 0.67%. This finding lends support to H2b and implies that amongst BG-affiliated firms, diversification is helpful in attaining better acquisition performance. Overall, the results present a coherent picture that the value-constraining perspective of BGs is misplaced, and group affiliation provides value-enabling opportunities in a post-acquisition setting for its member firms.
We must acknowledge that M&A performance is likely to be affected by other factors such as payment method (Carow, Heron, & Saxton, Reference Carow, Heron and Saxton2004), and relative size of the acquisition (Fuller, Netter, & Stegemoller, Reference Fuller, Netter and Stegemoller2002). Due to data-availability issues, we could run an empirical test for a sub-sample of firms using the above-mentioned two additional control variables, here too the results were qualitatively same as our primary tests.[Footnote 1]
Robustness Checks
To test the robustness of our main findings, we employed alternate measures of post-acquisition performance. First, we employed two variants of Tobin's Q such that we calculate value of Tobin's Q two and three years after the focal deal, while controlling for value of acquirer's Tobin's Q one year prior to the deal (Huang et al., Reference Huang, Zhu and Brass2017). In Table 6, the coefficient of variable business group in Model 1 is positive and significant (b = 0.114) with a p value of 0.030. This implies that a business group affiliated acquirer will exhibit 11.4% higher performance as compared to a non-affiliated acquirer. Similarly, in Model 3 of Table 6, where the dependent variable is Tobin's Q measured three years after the focal deal, the coefficient of variable business group is positive and significant (b = 0.217) with a p value of 0.000, validating the results obtained in the primary tests. Therefore, the robustness tests on the different versions of Tobin's Q reconfirm the results of our primary tests that if an acquiring company is affiliated to a business group then it is likely to show a higher acquisition performance.
Robustness check of performance measures
Notes: Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01
Similarly, the coefficients of variable group diversification in Model 2 (b = 0.00863, p = 0.009) and Model 4 (b = 0.00683, p = 0.030) of Table 6 are positive and significant, thus, reconfirming the results obtained earlier.
In the second set of robustness tests, we employed buy and hold abnormal return (BHAR) methodology to estimate the post-acquisition performance. The BHAR measure captures long-term shareholder wealth that results from an M&A deal (Chakrabarti, Gupta-Mukherjee, & Jayaraman, Reference Chakrabarti, Gupta-Mukherjee and Jayaraman2009; Lyon et al., Reference Lyon, Barber and Tsai1999) and its main advantage is ‘to yield an abnormal return measure that more accurately represents investor experience’ (Lyon et al., Reference Lyon, Barber and Tsai1999: 198). For estimating this measure, we consider a time period of 12 months, measured from the completion of a deal. Essentially, we compare the 12-month return of the focal firm with a characteristic based reference portfolio for the same period (Lyon et al., Reference Lyon, Barber and Tsai1999; Mitchell & Stafford, Reference Mitchell and Stafford2000). The procedure to estimate BHAR consists of three steps.
First, we constructed 150 reference portfolios of firms listed on the Bombay Stock Exchange at the end of September every year, based on similarity in their size, market-to-book ratio, and prior performance (Barai & Mohanty, Reference Barai and Mohanty2014). To ensure that the returns of reference portfolios are not affected by M&A events, we excluded event firms from our data while forming such portfolios. We also excluded firms with missing values on required firm characteristics such as market value of equity or the market-to-book ratio. Furthermore, we excluded firms that experience infrequent trading in their equity stocks and included only those firms with monthly return data for at least 8 months in a given year. We considered the market value of equity as a proxy for size and one-year stock returns prior to the portfolio formation day as a proxy for prior performance (Basuil & Datta, Reference Basuil and Datta2015; Lyon et al., Reference Lyon, Barber and Tsai1999). We divided all non-event firms into deciles, according to their market value of equity, and constructed 10 size-based portfolios. Then, we divided each of the size-based portfolios into quintiles according to market-to-book ratio of the component firms. Last, we divided each of the 50 resulting portfolios into three portfolios, according to the prior stock performance of the component firms. This exercise led to formation of 150 reference portfolios. We repeated this exercise in September of every year.
Second, we assigned each acquiring firm to a matching portfolio according to size, market-to-book ratio, and prior performance. Third, we calculated the BHAR of the acquiring firm by measuring the difference between the buy-and-hold return of the event firm and the buy-and-hold return of the matching reference portfolio, calculated as the average monthly returns of portfolio firms compounded over 12 months:
where RB is the return on matching reference portfolio, Rit is the monthly return for stock i in month t, T is the holding period of 12 months, and nt is the number of firms in a portfolio. Then, we calculated 12 months of buy-and-hold returns by compounding monthly returns for the acquirer firm starting from the month after the deal completion. The difference between the buy-and-hold return of the acquiring firm and the buy-and-hold return of the matching reference portfolio is the 12-month BHAR:
where BHARi is the BHAR of event firm i over the 12-month period (T), Ri is the monthly return of an event firm i, t is a given month, and T is the holding period.
As shown in Model 5 in Table 6, the variable business group has a positive and significant relationship with the dependent variable (b = 0.178) with a p value of 0.000. These results suggest that, if acquisition performance is measured as buy and hold abnormal return after deal completion, then business group affiliated acquirers are likely to exhibit 17.8% higher acquisition returns as compared to non-affiliated acquirers. This difference is statistically significant and highlights the immense advantage held by affiliated acquirers. Similarly, as shown in Model 6 of Table 6, group diversification has a positive and significant coefficient too (b = 0.0046, p = 0.075). If an acquirer is affiliated with a business group which has larger diversification, then for every increase in a distinct industry in the group's portfolio, the former is likely to exhibit 0.46% higher acquisition performance.
Post-Hoc Analysis
As a post-hoc analysis, we also estimated abnormal returns around deal announcements by employing the standard event study methodology (Brown & Warner, Reference Brown and Warner1985; Kothari & Warner, Reference Kothari and Warner1997; MacKinlay, Reference MacKinlay1997). We estimated the acquisition performance by observing the cumulative abnormal return (CAR) for the focal event. We calculate the expected return for an event firm using the following market model:
where E (Rit) is the expected return for event firm i for a given day t, Rmt is the return on market index (m) for day t, and the parameters α and β are estimated by regressing daily share returns over the market returns. The estimation window is the historical 250 trading days, starting 30 days (–280 to –31 days) before the focal deal announcement. We calculate the abnormal return for a given day t as the difference between the expected and actual returns for an event firm i:
Because we are interested in capturing the abnormal returns over an event window, we add the abnormal returns for each day in the event window. Following extant research in EM contexts (Chi, Sun, & Young, Reference Chi, Sun and Young2011; Gaur et al., Reference Gaur, Malhotra and Zhu2013), we calculated the CAR for a five-day window (t = –2 to + 2), for each event firm i:
In Table 7, we present the ordinary least squares (OLS) regression results with CAR as the proxy of post-acquisition performance. Model 1 of Table 7 tests for the influence of our hypothesized relationship between BG affiliation and acquisition performance measured as CAR observed over a five-day window (t= −2 to + 2). The error terms in these regression models are clustered per BGs. In Model 1, we find a negative and significant coefficient for BG affiliation (b = –0.0095) with a p value of 0.027, which implies that acquisition performance as measured through CAR around deal announcements is approximately 1% lower for BG-affiliated acquiring firms. To assess the robustness of the results with respect to the choice of different event windows, we considered two alternate measures of the dependent variable that captured announcement effects over 7-day and 11-day event windows (Gubbi & Elango, Reference Gubbi and Elango2016). We observed that the empirical results are qualitatively same when we consider a longer event window for CARs. As shown in Models 2 and 3 of Table 7, the BG affiliation variable has a negative relationship with announcement returns of a 7-day window (t = –3 to +3) (b = –0.012) with a p value of 0.012, as well as in an 11-day window (t = –5 to +5) (b = –0.011) with a p value of 0.089. Thus, our findings suggest that on deal announcements, relative to standalone firms, BG affiliation is associated with 1.2% and 1.1% lower abnormal returns over a 7-day and an 11-day event window, respectively. These results do not attest to the value enabling perspective of BG affiliation, whereas, in the earlier tests using Tobin's Q and BHAR measures, our results attest to the value-enabling perspective of BG affiliation.
Post-hoc analysis using cumulative abnormal return
Notes: Standard errors in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01
In Table 7, Model 4 shows the implications of group diversification on announcement returns. We observe that contrary to main results, BG diversification has a negative impact on M&A performance observed over an event window of 5 days (b = −0.000963, p = 0.002). The results imply that a one-unit increase in group diversification would lead to approximately 0.1% fall in abnormal returns around deal announcements. For the alternate measures of CAR as well, the relationship between group diversification and announcement effects is negative and significant for both the 7-day (b = –0.001, p = 0.001) and the 11-day window (b = –0.001, p = 0.001) in Models 5 and 6, respectively. Thus, even if we consider a different event window (7-day or 11-day), our findings suggest that for deal announcements, a one-unit increase in group diversification would lead to a 0.1% fall in CARs.
The above findings imply that amongst BG-affiliated firms, a greater degree of diversification is perceived as negative for acquisition performance observed over different event windows around deal announcements. These results are contrary to our main findings using Tobin's Q or BHAR as a measure of performance and confirm to the value-constraining hypotheses for BG affiliation and group diversification.
Because acquisitions are complex, research in a variety of disciplines (e.g., strategy, behavioural finance, economic sociology, social psychology) empirically investigates acquisition performance (King et al., Reference King, Dalton, Daily and Covin2004). These studies provide ample evidence that markets might not be perfectly efficient (Fama & French, Reference Fama and French2007; Schijven & Hitt, Reference Schijven and Hitt2012); therefore, it is unreasonable to assume that the market completely subsumes the intricacies of deals in an EM context. Zajac and Westphal (Reference Zajac and Westphal2004) note that market value could be socially constructed by investors’ beliefs. Reinforcing this possibility, Young (Reference Young2016) notes that Chinese acquirers’ poor performance in global M&A waves may be attributed to the inefficiency of Chinese stock markets. Although prior literature notes some strengths of short-horizon methods, a growing body of empirical research suggests that investors may under- or overreact to corporate actions (Fama & French, Reference Fama and French2007; Zajac & Westphal, Reference Zajac and Westphal2004), and thus, the change in stock prices may not completely incorporate the effect of the information conveyed through such actions (Oler, Harrison, & Allen, Reference Oler, Harrison and Allen2008). Therefore, we suggest that the negative relationship between BG affiliation and performance, and similarly between group-diversification and performance, could be on account of bounded rationality on part of investors (March & Simon, Reference March and Simon1958), or the usage of heuristics in judging the potential value creation in a deal (Campbell, Sirmon, & Schijven, Reference Campbell, Sirmon and Schijven2016; Fiske & Taylor, Reference Fiske and Taylor1991).
DISCUSSION
In this study, we theorize and empirically examine whether BG affiliation constrains or enables value creation in the context of the post-acquisition performance of BG-affiliated firms. With a data set of Indian acquirers, comprising both domestic and cross-border acquisition deals during 2002–2013, the empirical results show that, relative to stand-alone firms, group affiliation confers value-enabling benefits to affiliate firms in realizing better post-acquisition performance.
We conducted this study in an EM context, thus placing it at the confluence of BG literature and M&A performance. Scores of studies empirically examine the phenomenon of M&As in finance and strategy domains, and post-acquisition performance being the most widely researched question. In this article, we not only analyse post-acquisition performance in a unique contextual setting, marked by firm's BG governance structures in emerging economies, but we also apply contrasting theoretical lenses. First, leveraging the theoretical notions of PP conflict and organizational inertia, we highlight the possibility of a detrimental impact of BG affiliation on the corporate strategy of M&As (Bae et al., Reference Bae, Kang and Kim2002), which we term as ‘value-constraining’ perspective. Second, we develop a ‘value-enabling’ perspective of BGs, predicting a positive impact of BG affiliation on post-acquisition performance (Khanna & Palepu, Reference Khanna and Palepu2000a, Reference Khanna and Palepu2000b). We argue that BG affiliation could help affiliate firms not only overcome the ex-ante challenges of an M&A deal but also improve the likelihood of realizing the benefits from it, resulting in an improved post-acquisition performance.
In addition, we examine the impact of intergroup heterogeneity attributes in terms of product diversification, which could have a differential impact on the post-acquisition performance of BG-affiliated firms. All groups are not alike (Ramaswamy et al., Reference Ramaswamy, Purkayastha and Petitt2017); therefore, we consider contextual factors, to highlight heterogeneity in the positive or negative impacts of BG affiliation on the post-acquisition performance of member firms (Carney et al., Reference Carney, Gedajlovic, Heugens, Van Essen and Van Oosterhout2011). To that end, we predict the role of a BG's product diversification in its impact on post-acquisition performance amongst affiliated firms. The empirical results attest that BG affiliation and greater product scope of a BG lead to a better post-acquisition performance by an affiliated firm. This finding underscores the extended resource-based view, which holds that a member firm may exploit the resources, experience, and capabilities of other affiliate firms in the same BG (Arya & Lin, Reference Arya and Lin2007; Lavie, Reference Lavie2006). Empirically, we verified our results with two sets of robustness checks. First, we calculated variants of Tobin's Q – two and three years after the deal. Moreover, we also proxied for the performance of the acquirer firm using the buy hold and abnormal return methodology. All these additional tests verify our results in primary tests. The results of some of the control variables relative to the performance results match our expectations. Firms’ R&D intensity and potential slack seems to help them realise better performance through the acquisitions. However, we observe no significant difference in the performance of cross-border and domestic acquisitions.
Finally, we also did some post-hoc tests using cumulative abnormal return technique as a measure of post-acquisition performance. The short-window based results performance refute the theorization that BG affiliation and greater diversification of a BG positively impact the wealth creation of an affiliate acquirer. To verify the results, we also did checks with alternate event windows for capturing cumulative abnormal returns, but qualitatively results remained the same. The reversal of performance results in the announcement effects methodology suggest that in contrast with the notions of financial theory (Fama, Reference Fama1970), investors tend to apply negative heuristics to a BG affiliate acquirer deal and label it as a ‘bad deal’ (Campbell et al., Reference Campbell, Sirmon and Schijven2016; Oler et al., Reference Oler, Harrison and Allen2008; Schijven & Hitt, Reference Schijven and Hitt2012). However, there is some evidence that such might deals create some value when compared with a stand-alone firm.
This article thus makes several contributions. BGs constitute an important governance structure in EMs and examining the effect of BG affiliation on M&A performance can help extend both BG and acquisition performance literature. Contradicting the value-constraining perspective, our results indicate an overall positive impact of BG affiliation (Chittoor et al., Reference Chittoor, Kale and Puranam2015). It is possible that the advantages of BG affiliation neutralize the negative effects, if any, due to type-II agency concerns and organizational inertia, and as a result, the affiliation effect is positive for realizing the post-acquisition objectives. Recent studies in EM contexts argue that with institutional development, affiliation advantages may attenuate (Carney, Shapiro, & Tang Reference Carney, Shapiro and Tang2009; Lee, Peng, & Lee, Reference Lee, Peng and Lee2008; Zattoni, Pedersen, & Kumar, Reference Zattoni, Pedersen and Kumar2009), while others suggest that despite economic reforms, the positive impacts of BG affiliation persist (Manikandan & Ramachandran, Reference Manikandan and Ramachandran2015). Because we control for institutional development, the empirical results support our arguments that even in an improving institutional landscape, the value-enabling perspective of BG persists, because of the BG network's extended resources, which affiliated firms can exploit (Lavie, Reference Lavie2006). Although the value-enabling perspective and its erosion, if any, has received extensive attention in strategy literature, with this study we attempt to test both the value-constraining and value-enabling perspectives together to help decipher the role of BG affiliation in a post-acquisition setting (Kim, Haleblian, & Finkelstein, Reference Kim, Haleblian and Finkelstein2011). This study also provides one of the earliest attempts in an EM context to compare announcement effects with post-completion performance. Although behavioural finance and economic sociology researchers have extensively studied over- and under reactions in announcement effects, strategy literature has paid little attention to such phenomena (Preda, Reference Preda2007; Zuckerman, Reference Zuckerman2004), more so in the context of emerging markets where the problem could be further accentuated (Shapiro & Li, Reference Shapiro and Li2016; Young, Reference Young2016). Investor reaction is subject to bounded rationality (March & Simon, Reference March and Simon1958); thus, decision makers tend to rely on heuristics or mental shortcuts to arrive at judgements (Campbell et al., Reference Campbell, Sirmon and Schijven2016; Fiske & Taylor, Reference Fiske and Taylor1991). In other words, investors underestimate the performance benefits of BG affiliation and a negative heuristic bias dominates for deals affiliated with BGs, according to the value-constraining logic (Hendricks & Singhal, Reference Hendricks and Singhal2001; Shleifer & Vishny, Reference Shleifer and Vishny2003).
Limitations and Scope for Further Research
This study has some limitations, on which continued studies may build to advance BG literature. In addition to product diversification, a host of other variables may change the impact of PP conflict and thus performance. Therefore, we call for studies that examine BG heterogeneity in terms of corporate governance practices. Scholars have noted that EM firms historically tend to have notional boards, governed passively by rubber-stamping members (Colli & Colpan, Reference Colli and Colpan2015). To advance this stream of research, we suggest examining the impact of board independence and professionalization on M&A performance and other such corporate strategy actions. Moreover, research can also look at the impact of timing (early-movers or late-movers) on post-acquisition performance. Finally, a natural extension of this study would be to analyse and compare the effect of various types of BGs on the post-acquisition performance of their affiliate firms in other emerging economies.
