This paper investigates the so far unexplored theoretical tension associated with foreign investor valuations of emerging market firms (EMFs) that tap into global capital markets. One line of argument expands the liability of foreignness perspective into the capital market domain where a foreign firm faces legitimacy and unfamiliarity issues. In contrast, there is recognition of the benefits to global investors in EMFs associated with geographic portfolio diversification. Our core premise is that EMFs face a cost–benefit trade-off associated with the extent of institutional distance when venturing abroad in search of financial resources. Based on a theoretical model grounded in institutional theory, we find a curvilinear effect of institutional distance on foreign equity valuations. Furthermore, we show that information spillovers through a focal firm’s domestic capital market and international product market strategies moderate this relationship.
Hinweise
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
1 Introduction
Need to access financial resources to support growth aspiration and increasing level of financial market integration motivate the emerging market firms (EMFs) to expand into foreign capital market (Payne et al., 2013). As some of the emerging markets such as India and China, have large and growing corporate sectors, there is an opportunity for EMFs to choose from multiple equity markets, including those outside their home countries (Fernandes, 2011). Equity offerings from EMFs also create significant investment portfolio diversification opportunity for global investors especially the ones from developed economies (Chakrabarti et al., 2008). Therefore, selecting among non-domestic capital markets aiming to obtain higher valuations represents a critical but largely unexplored strategic decision by EMFs, which is in the focus of this paper.
Though institutional distance between home and host locations commonly treated as a source of cost for getting appropriate valuation in the foreign equity market (Bell et al., 2012a, 2012b), there is gradual recognition in the international economics and finance fields of the benefits of institutional differences between markets to global investors. The benefits stem, for example, from unsystematic risks reduction through geographic portfolio diversification (Bansal & Clelland, 2004; Chang et al., 2016; Hagelin & Pramborg, 2004). Also, researchers suggest that the greater institutional distance between the firm’s home and host markets indicates lesser synchronization of financial trends in the two locations. In turn, as systematic risks are reduced, demand among foreign investors may increase, and the focal firm profits from reduced costs of capital (Amit & Livnat, 1988; Gu et al., 2019).
Anzeige
In contrast, international business (IB) research has long emphasized that firms aiming to expand into foreign markets face liabilities of foreignness (LoF), or “additional costs a firm operating in a market overseas incurs that a local firm would not incur” (Zaheer, 1995: 343). The concept of LoF has been developed in IB literature within two interrelated streams, namely internalization (Buckley & Casson, 1976; Buckley et al., 2002) and the internationalization process of firms (Johanson & Vahlne, 1977). Internalization argument indicates boundaries of the firm are set on the basis of intermediate product flows and based on the LoF created through transaction costs. Internationalization process theory explores gradual learning as a mechanism to overcome LoF and takes relevance to reduce institutional/cultural distance and/or market uncertainty. Integrating internalization and internationalization process view of the firm, (Casson, 2000; Gulamhussen, 2009) show how sequential expansion emerges as a special case of internalization. Prior studies provide evidence of LoF in different strategic contexts, such as a foreign market entry (e.g. Meyer et al., 2009; Wu & Salomon, 2016) or the performance and survival of firms entering foreign markets (e.g. Mata & Alves, 2018). LoF is also expanded into two related but contextually different dimensions namely the liability of country foreignness and the liability of regional foreignness (Qian et al., 2013). In our context we anchor our arguments mainly based on internationalization process perspectives. Recent research expands the LoF perspective into the capital markets domain where foreign firms may face legitimacy and unfamiliarity issues (Bell et al., 2012a, 2012b). This is specially more applicable for EMFs as they operate under different domestic institutional norms compared to what is typically experienced by the developed market investors (Khanna & Palepu, 1997). Therefore, a significant body of literature indicates home bias among investors against EMFs from dissimilar institutional environments and the benefits of tapping into (potentially vast) foreign capital markets are negatively affected by institutional differences between countries (Cornaggia et al., 2020). A key conclusion of these studies grounded in institutional theory is that the higher are differences between institutional settings in the firm’s home and host countries, the lower will be the valuation in the foreign equity market (or greater cost of capital).
Combined, these theoretical arguments strongly point to a cost–benefit trade-off associated with EMFs tapping into foreign equity markets (Purkayastha & Kumar, 2021). Much of the literature analyzing cross-border listing tends to focus on how firms from a multitude of home countries compare when they list in the same host country (Doidge et al., 2004; Payne et al., 2013). However, at present, there is limited research in the IB fields on the cost–benefit trade-offs associated with host capital market choices of firms from a specific emerging market when they issue equity abroad. Our paper aims to fill these theoretical and conceptual gaps.
Theoretically, we build on the institutional theory perspective which is focused “on the different degree to which the existing institutions in a given country support effective economic activity and coordination between economic actors” (Kostova et al., 2020: 470). Although prior studies have identified various operationalizations of differences in the firm’s macro-institutional environments when arguing that information costs, unfamiliarity costs, and cultural differences are major sources of capital market liability of foreignness or CMLOF (Bell et al., 2012a, 2012b), we focus on institutional distance and build on theoretical mechanisms from both organizational institutionalism and institutional economics (Berry et al., 2010; Kostova et al., 2020). Our decision is driven by Kostova et al., (2020: 468) assertion that “institutional distance provides a broader view of national contexts, encompassing not only cultural but also regulatory and cognitive elements… also allows the capturing of the dynamic aspects of context, reflecting important institutional changes in countries throughout the world”. Therefore, we use comprehensive set of mechanisms including issuing firm’s legitimacy, liability of foreignness, adaptation, and the home and host countries’ institutional quality to develop our arguments. Since investors are usually taking such a broad view of national context when investing into foreign equity (Bris et al., 2012; Siegel, 2009), our focus on institutional distance is appropriate. Therefore, we summarize our first research questions as: How does institutional distance between the emerging market firm’s home and host capital market influence its valuation in the host equity market?
By emphasizing the importance of institutional drivers of the EMF’s liability of foreignness and investors’ portfolio diversification, however, we do not assume that the cost–benefit trade-offs associated with institutional distance are equivalent for all firms from a particular country. Therefore, we also look at firm level factors as moderators, in line with multi-level research grounded in the institutional theory (Bell et al., ). An established line of argument in strategic management and IB suggests that EMFs can mitigate LoF associated with institutional distance by selecting an appropriate entry mode (e.g. Meyer et al., 2009) or imitating strategic actions of local firms (e.g. Wu & Salomon, 2016). We extend these arguments further by shifting focus on contingency effects of the EMF’s strategic actions in other markets, including capital and product markets. Building on prior research on information spillovers (Joe & Oh, 2018), we propose that foreign investors may draw information from the firm’s other financial market activities as well as from the focal firm’s international product market strategies. Surprisingly, questions of how information spillovers from firm’s operating market interact with the cost–benefit trade-offs from equity offerings in a specific foreign market remain largely unexplored. Hence, our second research question is: How do contingency factors associated with information spillovers from other capital markets and international product market affect the relationship between institutional distance and valuation in the host equity market?
Anzeige
Our empirical analysis shows that the relationship between institutional distance and focal firm’s valuation at the foreign equity market is inverted U-shape. We also find evidence of spillover effects from listing in the domestic equity market (Healy & Palepu, 2001) and from foreign expansion through product market internationalization (Meyer, 2004) as two possible strategies, which flatten the base, non-linear relationship. Fixed effect panel regression analysis performed on a manually created unique panel dataset of 2553 firm-years spanned over 17 years from 190 Indian firms listed in foreign equity markets supports the hypotheses.
We contribute to the IB literatures in three ways. First, the possibility that institutional “differences may provide opportunities for exploration, learning, and growth” (Stahl et al., 2016: 622) is missing in previous narrative specially in the context of foreign listing as dominant view was to look at institutional distance only as source of informational frictions. Our findings indicate that, while facing CMLOF, equity offerings of EMFs from institutionally distant markets can provide benefits related to portfolio diversification (Markowitz, 1991). Therefore, our finding enriches our understanding of the implications of institutional distance on strategic outcomes in the context of capital raising decisions in EMFs. Second, extending the theoretical argument from Bell et al., (2012a, 2012b), we offer a contingency model, where information spillovers from home market listing reduce both the degree of uncertainty and benefits of portfolio diversification surrounding security issued by EMFs (Aguilera et al., 2015). We relax the implicit assumption in prior theory that firm-related information flows are embedded into and/or originate from the host capital market’s institutional setting and hence we derive that there could be an information spillover effect between institutionally distant capital market segments (Hasan et al., 2011). Third, extending earlier literature on the distribution of positive and negative spillover from product market internationalization among stakeholders (Meyer, 2004), we find that spillovers from product market internationalization of EMFs not only reduce CMLOF, but also reduce the beneficial aspect of investment portfolio diversification by embedding investment seeking firm into the host market’s institutional settings. Our findings add to the growing work on internationalization- related information spillovers from one market to others (Benveniste et al., 2003; Joe & Oh, 2018). By bringing together information flows associated with the firm’s product and factor market, we offer a more comprehensive theoretical model of the EMFs’ internationalization.
2 Theoretical Background
Rational models of portfolio choice in the finance literature suggest that investors hold diversified portfolios to reduce non-compensated risk and that diversification among politically risky countries improves the risk-return characteristics of optimal portfolios (Levy & Sarnat, 1970). Extending this argument to the context of international equity markets, portfolio diversification into foreign securities has long been advocated as one important way of enhancing average returns while reducing portfolio risk. There are multiple pathways through which equity offerings from foreign firms provide value to the foreign investors.
Equity offering from an institutionally distant firm serves as a signal of the underlying high value of the security to investors on an ongoing basis. This is because the issuing firm is ready to face market driven valuation of the intrinsic quality of their offering in a foreign setup for a longer term. Hence, it signals the firm’s attributes and quality and the associated portfolio diversification benefits to the foreign investors, in line with the “bonding hypothesis” (Doidge et al., 2004). Equity seeking foreign companies from institutionally distant location overcome information gaps in the host market environment by carefully opting for the market where investors and analysts have an understanding and proven expertise in these industries. For example, a foreign mining firm might list in the London Stock Exchange or a foreign digital platform firm might list in the NASDAQ. Better analyst coverage and knowledgeable investors of such industries reduce unsystematic risks and generate greater interest among host market investors about the institutional distant foreign firms and hence enhances the benefits of portfolio diversification through investing in these companies. Systematic risks refer to the degree to which the firm’s performance covaries with the economy as a whole. More recent research produced evidence that international portfolio diversification benefits are dependent on the variation of risk profile of the issuing country (Driessen & Laeven, 2007; Montgomery & Singh, 1984). Since emerging markets in general are embedded in different institutional settings compared to developed market institutions, investors may be motivated to hold equities that are linked to emerging markets (or based in distant economies) to reduce systematic risks in their portfolio. Though global markets have tended to become more integrated as a result of greater liberalization and deregulation, studies indicate persisting low correlations between different markets, which creates arbitrage opportunities for both short-term and long-term investors (Gilmore & McManus, 2002).
The gradual lowering of institutional barriers over time has enabled more frequent cross-border equity capital flows (Bell et al., 2012a, 2012b). However, prior literature consistently found that, as compared to a local firm, a foreign firm faces steeper challenges when accessing capital in host capital markets, such as higher cost of capital, higher listing fees and audit fees, greater risk of lawsuits, lower liquidity, and less analyst coverage (Bhattacharya et al., 2007). IB literature distinguishes the country dimension of liability of country foreignness (LoF) from the regional liability of regional foreignness (LoF). The liability of country foreignness refers to the costs of doing business across countries which includes the structural, relational and institutional costs in addition to the costs directly associated with spatial distance. By contrast, the liability of regional foreignness is related to policy-driven regional integration within regional groups (such as EU and ASEAN). In other words, it is based on government bias, that is, unfavorable government policies towards firms from other regions. When considering the listing location choice based on liability of country foreignness and liability of regional foreignness, researchers observe the important phenomenon that most firms will consider capital markets outside of their own region rather than within their own (Bailey et al., 2006). The situation frequently happens for Asian firms as well. For example, most of Chinese firms hope to be listed in New York rather than in Hong Kong or Singapore though the costs (also more stringent market scrutiny and reporting requirements) involved in the former are much higher than in the latter. Though relevance of country and regional separation of LoF is empirically supported (Qian et al., 2013), there is a great deal of integration among capital markets. Hence, CMLOF is an integrated phenomena as which does not different valuation based on a region.
We argue that both perceived and actual differences in governance standards contribute to CMLOF. Investor protection is a key determinant for the investor’s decision. Hence, the (perceived) differences between the investor protection in the host market and the one in the issuing firm’s home market determines investors’ perceptions of the corporate governance followed by the issuing firm (Payne et al., 2013). Despite the potential gains from investing internationally, investors in both developed and developing markets strongly prefer to invest in domestic firms (a phenomenon termed as home bias) rather than foreign firms operating in the same capital markets (Tesar & Werner, 1995). From an institutional perspective, differences in accounting systems, corporate governance, and investment regulations create asymmetries of information between investor and issuing firm. Furthermore, due to the lack of tax harmonization, foreign investors often find it difficult to get refunds or credits for taxes paid abroad.
We also use theoretical arguments grounded in the literature on information spillovers which suggest that the firm’s activity in and exposure to other capital markets and international product markets may be important contingencies that moderate the base effect.
Domestic capital market participation increases market scrutiny and reporting by equity analysts for the focal firm. Domestic capital market participation also requires greater disclosure to the investor community specially when governance standard of the economy is improving (Bruton et al., 2010; Moore et al., 2012). Overall, it acts as a conduit for the information spillover and therefore, minimizes managerial entrenchment and owner opportunism that are often prevalent in the emerging market firms (Chittoor et al., 2015). Prior empirical studies also indicate that an increase in the quality of information spillover positively influences the investors and reduces the cost of capital (Bertomeu et al., 2011; Lambert et al., 2007). As the second contingent factor, there is a gradual recognition of the significant knowledge outflows back from host country organizations to foreign multinationals (Singh, 2007). The prior literature argues that foreign direct investment and subsidiary creation might be a vehicle that facilitates global diffusion of information spillover between distant markets (Pagano et al., 2002). The view of emerging market subsidiaries are acting as ‘listening posts’ to facilitate information spillover from the host country is reflected in Luo and Tung's (2018) ‘springboard’ arguments where emerging market firms engage in risky foreign acquisition to acquire tacit knowledge of the foreign market firms. The similar pattern of information spillover through product market internationalization is empirically supported for 17 emerging market economies as well (Gorodnichenko et al., 2014). To summarize, prior research indicates that institutional distance between the firm’s home and host markets creates a cost–benefit effect in terms of investor valuations of the firm’s equity. Further, this cost benefit trade-off is affected by contingency factors associated with information spill-overs between product and factor markets. In the following section we translate this over-arching theoretical perspective into testable hypotheses.
3 Hypotheses Development
As we indicated in the previous section, there are clear benefits in terms of the association between institutional distance between issuing country and host location and the host market equity valuation for the following reasons. First, cross-listing of EMFs into another market (especially when host market is in developed country) generally requires more disclosure of organizational health and potential growth trajectory to conform to the institutional norms of the host market. Hence, when EMFs offer equity opportunity in a foreign equity market, it signals firm’s ability to navigate through the foreign listing process and lower unsystematic risks. This is because the issuing firm is ready to face market driven valuation of the intrinsic quality of their offering in a foreign setup for a longer term. It also reduces subjective perception and uncertainty about the issuing firm’s credibility. Thus, EMFs listed in the foreign market provide portfolio diversification opportunity for the local investors. Moreover, legal requirements and tax implications make investment into emerging market’s American Depository Receipts (ADRs) or Global Depository Receipts (GDRs) more convenient for foreign institutional and individual investors to invest (Hansda & Ray, 2003) compared to directly investing into emerging market’s equity market. Thus, the combined effect of a positive signaling of the quality/novelty and convenience of investment makes foreign investor interested to invest into equity offerings from EMFs leads to higher valuation.
Second, systematic risks refer to the degree to which the firm's performance covaries with the economy as a whole and investors with geographically diversified portfolios have an opportunity to hedge against economic cycle swings in a particular country (Olibe et al., 2008). For example, emerging markets have weathered relatively better the negative impact of the 2008 global financial crisis compared to developed economies. Specifically, recent research indicated that none of the developed as well as most of the emerging countries are integrated with Indian stock market in the long run (Mukherjee et al., 2005). Though there are some evidence of unidirectional influence of Nasdaq on two major Indian stock markets namely Bombay Stock Exchange (BSE) or National Stock Exchange (NSE) (Hansda & Ray, 2003), international portfolio diversification is beneficial in our research context due to low correlation between different financial markets and the differences of the risk perceptions of the market (Fabozzi et al., 2002). This is because systematic risks are always dependent on the uncertainty associated with the country specific economic forces. Also, investment in the firms from institutionally distant country that are driven by different set of economic priorities provides opportunity to avoid certain country specific systematic risks. Due to this low systematic risk, equity offerings from EMFs offer hedging opportunity to the foreign investors and generates greater valuation at the host locations.
However, prior literature consistently found that, may be negative valuation effects of CMLOF along with increasing institutional distance that are relevant in our research context, one in which an EMF raises capital in a foreign market through ADRs or GDRs. First, prior research found that better protection of minority shareholders leads to higher valuation of firms (Porta et al., 2002). When laws are protective of the outside investors and well enforced, investors are ready to invest and willing to pay more for the financial assets. The quality of investor protection also influences funding supply and thus profoundly impacts capital structure. As perceived and actual differences in governance standards contribute to CMLOF, investor protection in the context of emerging market is a key determinant for the investor’s decision. Historically emerging markets including India are identified with low investor protection and weak legal enforcement (Leuz et al., 2003). As greater institutional distance increases the difference between governance standards between locations, we argue that the differences in the extent of investor protection (typically driven by governance standards) in emerging markets compared to developed market reduces investor’s positive sentiments towards financial instruments offered by the foreign firms.
Second, research on home bias indicates that investors appear to invest predominantly in their home country often ignoring foreign opportunities (Coval & Moskowitz, 1999). Though there is an increase in cross-border information flow, investors do not choose to learn what others know as specializing in what they already know is a more profitable strategy. In addition, while choosing what to learn, investors seek to make their information set as different as possible from the average investor’s. Thus, investor’s focus is confined to a small set of familiar stocks. Consequently, investors are more likely to evaluate more positively the stocks of firms that are located close to the investor compared to the firms from distant economy. We expect that along with the increasing level of the institutional distance, there will be an upward movement of the home bias (Ahearne et al., 2004).
These theoretical arguments combined strongly point to the presence of cost–benefit trade-offs associated with institutional distance for the EMFs looking for equity market participation abroad. Thus, we hypothesize:
Hypothesis 1 (H1): The relationship between institutional distance (between an emerging market firm’s home market and the host equity market) and the firm’s valuation at the host equity market is an inverted U-shape.
By emphasizing the importance of institutional drivers of investors’ portfolio diversification and the firm’s liability of foreignness we, however, do not assume that the cost–benefit trade-offs associated with institutional distance are equivalent for all firms from a particular host country. Our theoretical arguments grounded in the literature on information spillovers suggest that the firm’s activity in and exposure to other capital markets and international product markets may be important contingencies that moderate the baseline non-linear effect. As our first moderating mechanism, we consider that spillover effects from the focal EMF’s participation in the home capital market. Specifically, we argue that listing in the domestic stock market influences the signaling effect of foreign equity offering. Listing in the home market indicates the firm’s commitment to disclose financial information compared to unlisted peers and lower unsystematic risks. In addition, emerging markets are improving the expected level of corporate governance practices and undertaking regulatory transformation. In such environments, which demand higher transparency in business transactions and appropriate disclosure of audited financial results, a longer duration of the domestic listing of an EMFs indicates better availability of firm specific information over a longer period of time. Thus, the spillover (through domestic listing) of the available information, risks, and benefits (or intangible firm-specific assets) reduces the signaling effect and dilutes benefits associated with the opportunity to invest into a firm’s ADR/GDR offering to the foreign investors. Second, the spillover through domestic listing reduces the effect of hedging against systematic risks across the countries. Large emerging markets including India’s financial market are becoming closely connected with the global economy. Such monetary integration leads to stronger stock market synchronization. The underlying mechanisms for the increasing level of synchronization are through the elimination of exchange rate volatility, adoption of common monetary policy, and the convergence of inflation expectations. When a firm is listed for longer duration in its domestic market, global capital market synchronization allows spillovers of the historic tacit knowledge (or firm-specific advantages) about the firm to the foreign equity market. To illustrate, Tata Steel, an Indian steel company (incorporated in 1907) is included in Nify-50 index (represents about 66.8% of the free float market capitalization of the stocks listed on NSE in 2019) since its inception in 1996 in National Stock Exchange (NSE) whereas United Spirits (incorporated in 1999) is included only in 2014. Thus, a hedging opportunity for a foreign investor is less by investing in Tata steel compared to United Spirits as risk exposure of Tata Steel is more comparable to the similar firm in the foreign market due to synchronization effect.
On the other hand, domestic equity market listing duration reduces the cost of institutional distance as well. First, as emerging markets are undergoing institutional transformation, there is an improvement of the protection of minority shareholders or reduction of principal-principal agency problem (Dharwadkar et al., 2000). Thus, longer domestic listing duration increases the confidence of the foreign investors as the issuing firm is maintaining its home market listing status for longer period of time under the environment of improving institutional standards. Hence, we expect in reduction in some part of the negative effect on the foreign valuation. Second, spillover of capital market practices through longer domestic listing reduces some of the home bias. A recent study identifies that home bias persists not because investors cannot learn what locals know, nor because such information is expensive, but because investors do not choose to learn what others know (Van Nieuwerburgh & Veldkamp, 2009). This close linkage between information and the investment choices becomes weaker due to reduction of asymmetry of information as longer duration requires higher quality of disclosure of information and indicates adherence to local governance practices. Thus, we hypothesize:
Hypothesis 2 (H2): The inverted U-shaped relationship between institutional distance and the emerging market firm’s valuation at the host equity market will be flattened by the duration of the firm’s home capital market listing.
As the second contingent factor, we consider the moderating effect of the focal firm’s internationalization. The prior literature argues that foreign direct investment and subsidiary creation might be a vehicle that facilitates global diffusion of information spillover between distant markets (Pagano et al., 2002). Building on this research, we argue that spillover effect from the EMF’s product market internationalization reduces the benefits of institutional distance on equity valuation in the host market for the following reasons. First, product market internationalization requires firm to invest strategic assets such as capital and knowledge within the host market (Grant, 1996). To ensure success from the investment, the firm (more so when it is from emerging market) discloses many relevant information through multiple channels and tries to increase transparency to the foreign stakeholders such as governments, partners and consumers. This spillover of the information about the firm’s intrinsic values (or internalized assets) has a two- fold effect. It reduces the signaling effect of the quality of the equity offering and also it dilutes novelty of the offering through foreign listing. Thus, we argue that increasing level of product market internationalization reduces positive valuation of foreign listing. Second, product market internationalization links the issuing firm’s home market and equity host market. Increasing level of product market internationalization makes the systematic risks between countries correlated and increases the issuing EMF’s embeddedness into the host market. The reduction of systematic risks as well diminishes the attractiveness of investing into locally offered equity from the foreign firm.
In a similar vein, the contingent effect of product market internationalization is visible with regard to the cost of institutional distance on equity valuation in the host market. First, product market internationalization indicates commitment of the foreign firm to align its corporate governance practices with the host market institutional norms. Though increasing level of institutional distance increases concerns associated with poor investor’s protection, product market internationalization alleviates some of those concerns as it indicates legitimacy of the firm among foreign stakeholders. Thus, it weakens the effect of liability of foreignness. Second, product market internationalization also reduces some of the sources of home bias such as information asymmetry. Regulatory norms in most of the market, and especially in the developed markets, force the EMF to adhere to host market’s higher level of disclosure standard. This spillover from factor market reduces some of the opaqueness of the equity issuing firm and thus reduces the effect of home bias on the foreign equity valuation. Finally, as suggested in the eclectic paradigm (Dunning, 1980), another possibility is that firms expanding internationally are generally those that are successful in their domestic operations. In other words, the firm needs to be successful in its home-country market before it can go international successfully. Therefore, continued and rapid international involvement of a domestic firm will provide foreign investors with a positive signal and, therefore, mitigate the negative impact of high institutional distance between home and host countries. In summary, we hypothesize:
Hypothesis 3 (H3): The inverted U-shaped relationship between institutional distance and the emerging market firm’s valuation at the host equity market will be flattened by the firm’s extent of product market internationalization.
4 Methods
4.1 Sample Selection and Data
India is an ideal setting for research on expansion of EMFs from a single home country into foreign capital market as many Indian firms are active on both domestic and global product and capital markets. To illustrate this, we gather relevant data from publicly available sources on two Indian firms that are listed in the foreign equity market through ADR/GDR. First illustration is Tata Steel which is part of Tata Group and was established in India as Asia’s first integrated private steel company in 1907. It is listed in the Indian stock market – NSE since 1994 and launched its GDR in the London Stock Exchange in 2009. The current market cap (as on September, 2022) of Tata Steel GDRs is 1.4 billion pounds. Other than India, Tata Steel operates in 25 countries with key international operations in Netherlands and the United Kingdom, and employs around 80,500 people. As per Reuters report,1 the listing of Tata Steel GDRs was the largest ever from an Indian firm as it has raised $500 million (304.2 million pounds) on the London Stock Exchange. Second illustration is Infosys Limited (Infosys) which is an Indian multinational information technology (IT) company. Infosys is the second-largest Indian IT company and the 602nd largest public company in the world (as per Forbes Global 2000 ranking). Infosys has 100 + development centers across the world (as of March 2018), with major presence in India, United States, China, Australia, Japan, Middle East and Europe. Infosys has listed in India equity market in 1993, offered ADRs in the United States in 1999. It is the first Indian company to be listed on Nasdaq and current market cap (as on September, 2022) of its ADR offering is US$72 billion.
Further, there are a set of institutional factors that makes India as a suitable research context. First, institutional distance is a significant factor for Indian firms venturing into foreign equity markets. Though there is a gradual transformation of Indian institutional settings more towards developed market condition, still India is very different in multiple ways from other markets as institutional transformation is slow process and that takes its own unique shape (Stucchi et al., 2015). Second, since the 1991 economic liberalization, many Indian firms have made considerable expansions into multiple different global product and capital markets (Chittoor et al., 2009). Therefore, Indian firms provide a critical sample for empirically verifying the role of institutional distance in the context of foreign equity valuation in multiple host markets and contingent effect of product market internationalization. Third, India has a relatively well-developed domestic equity market, which gives Indian firms the opportunity to raise capital in their home market. In sum, firms in our research context show substantial variance with regards to international expansion, home and host capital market usage while their institutional distance with the foreign equity market in developed economies remains considerably high.
Data about firms and their financials are collected from the Centre for Monitoring Indian Economy (CMIE) Prowess database, which has been used extensively by previous emerging market researchers (Kumar et al., 2020). We first identified all Indian firms (215 in total) that have raised capital in the foreign equity market since 1995. We excluded 17 firms from the financial sector as the capitalization strategies of firms in this sector are not comparable in other sectors of the economy (Khanna & Rivkin, 2001). We also removed six firms owned by a foreign entity and two state owned firms resulting in a final sample of 190 firms. We take 2001 as the starting year of our analysis because India has undergone a significant institutional transformation process in the first decade after the start of the economic liberalization in 1991 (Ahluwalia, 2002). Thus, starting at 2001 helps us to avoid the spurious effect of economic disruption on the foreign equity valuation.
Our initial sample is an unbalanced panel dataset of 2804 firm-years (some firms started operating after 2001) over 17-year (2001–2017) period. Subsequently, we further removed 259 observations where a firm’s return on assets was more than four standard deviations away from the sample mean (Khanna & Rivkin, 2001), debt-to-equity ratio was more than 100, foreign sales to total sales ratio was more than 100, as these values are very likely to be mistakes or misrepresentations. Finally, we lagged our explanatory variables by 1 year to minimize problems of reverse causality. Our final sample, thus, includes 2545 observations pertaining to 190 firms spanning over 17 years. The year-wise distribution of the number of firms, which raised equity from foreign markets is described in Table 1. The location-wise distribution of the number of firms, which raised equity from foreign markets are detailed in Tables 2, 3 and 4 respectively.
Table 1
Year wise distribution of the number of firms raised equity from one or two or three foreign market
Year
Number of firms raised equity from one foreign market
Number of firms raised equity from two foreign markets
Number of firms raised equity from three foreign markets
2001
1
2002
2
2003
2
2004
3
2005
5
2006
5
2007
6
2008
7
2009
32
1
2010
84
7
1
2011
108
10
1
2012
134
11
1
2013
142
12
1
2014
139
12
1
2015
135
12
1
2016
34
7
1
Total
839
72
7
Table 2
Location wise distribution of the number of firms, which raised equity from a primary foreign market
Host Location
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
Luxembourg
1
1
1
2
2
3
4
25
64
82
102
105
99
98
20
609
Singapore
1
6
7
7
9
11
10
51
UAE
1
1
1
1
1
1
1
1
1
9
UK
3
7
10
15
18
19
18
6
96
USA
1
1
1
2
2
2
2
2
2
6
8
9
9
10
9
8
74
Total
1
2
2
3
5
5
6
7
32
84
108
134
142
139
135
34
839
Table 3
Location wise distribution of the number of firms, which raised equity from a secondary foreign market
Host Location
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
Luxembourg
2
3
3
4
4
4
4
24
UK
1
1
1
1
1
1
6
USA
1
4
6
7
7
7
7
3
42
Total
1
7
10
11
12
12
12
7
72
Table 4
Location wise distribution of the number of firms, which raised equity from a tertiary foreign market
Host Location
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
Total
UK
1
1
1
1
1
1
1
7
Total
1
1
1
1
1
1
1
7
4.2 Key Variables
4.2.1 Dependent Variable
Most prior research on foreign equity market participation is based on Initial Public Offering (IPO). IPO research either focuses on immediate return and valuation of the offerings from multiple home markets in one host market (Bruton et al., 2010) or compared valuation and listing location between two locations (Moore et al., 2012). In this paper, we aim at determining the factors that influence the valuation of equity offerings from one home market in multiple host markets. Therefore, we use depository receipts (DRs) – one of the most common ways of creating equity offerings in foreign capital market. DRs are negotiable certificates, which represent ownership of shares in a company registered in a country other than the host country (Jayaraman et al., 1993). DRs fall into the category of financial products designed to facilitate international investment and provide an alternative to the direct purchase of ordinary shares in an overseas stock exchange. Individual investors can benefit from DRs by avoiding the complications and additional costs associated with international investment. DR-issuing firms benefit from increased demand for their shares and a broadened shareholder base. DRs represent one or more shares of foreign-company stock held by the bank in the local market. ADRs (which is one of the commonly traded DRs) offer U.S. investors a means to gain investment exposure to non-U.S. stocks without the complexities of dealing in foreign stock markets. In addition to ADRs, GDRs give issuers exposure to the global markets outside their home market. GDRs are offered to investors in two or more markets and are most commonly used to raise capital in Europe and the United States. Both ADRs and GDRs are usually denominated in U.S. dollars, but may also be denominated in Euros. Both ADRs and GDRs are tradeable in the host market stock exchanges, pay dividends in the local currency and trade like regular stock. To illustrate, the Indian ICICI Bank Ltd. has an ADR issued by Deutsche Bank that trades on the NYSE. As per U.S. Security and Exchange Commission (SEC), some foreign companies list their securities in multiple markets, which may include U.S. markets. However, most foreign stocks trade in the U.S. markets as ADRs only.
Following Doidge et al. (2004), we use Tobin’s Q to measure valuation at foreign equity market (FEqMktVal) of the firm’s ADRs and GDRs. Tobin’s Q is calculated using the following formula:
$$\text{Tobin's Q}=\frac{\text{Book value of total assets }-\text{ Book value of equity }+\text{ Market value of equity}}{\text{Book value of total assets}}.$$
This is one of the commonly used measurements of equity valuation that is applicable to both domestic and ADRs/GDRs based foreign equity listing (Black et al., 2006; Sarkissian & Schill, 2008). First, we identify the number of equity shares issued against ADRs and GDRs and total number of outstanding stocks. Then, we calculate the ‘domestic stock price’ in the context of India as the average (as the stock prices of firms listed in both BSE and NSE exchanges show marginal differences) of yearly weighted average prices from the BSE and NSE stock exchanges (when the firm is listed both in BSE and NSE) or yearly weighted average price from either BSE or NSE based on the listing location. Subsequently, we calculate book value and market value of ADRs and GRDs to measure Tobin’s Q.
4.2.2 Independent Variable
We adopted a multi-stage process based on Berry et al. (2010) and Beugelsdijk et al. (2018) to measure Institutional distance (InsDist). We first manually searched the listing locations (e.g. USA, UK, Luxembourg, Singapore, and UAE in our sample) of the firm from annual reports. Second, we capture (a) economic distance, (b) financial distance, (c) political distance, and (d) administrative distance between India (the home market) and the foreign listing location. We capture (a) economic distance on the basis of (i) difference in income level, (ii) prevailing inflation rates, and intensity of trade ((iii) import and (iv) export) with the rest of the world and (b) financial distance on the basis of (i) difference in credit to private sectors, (ii) stock market capitalization, and (iii) number of listed companies based on data collected from World Development Indicators (WDI) from World Bank website (http://data.worldbank.org/data-catalog/world-development-indicators). We capture (c) political distance on the basis of (i) Policy-making uncertainty from The Political Constraint Index (POLCON) Dataset (https://mgmt.wharton.upenn.edu/faculty/heniszpolcon/polcondataset/), (ii) democratic character from Freedom House (https://freedomhouse.org/report/freedom-world), (iii) Size of the state from WDI, (iv) WTO membership from World Trade Organization (WTO) dataset (https://timeseries.wto.org/), and (v) Regional trade agreement from WTO. We capture (d) administrative distance on the basis of (i) Colonizer–colonized link from CIA Factbook (https://www.cia.gov/library/publications/the-world-factbook/), (ii) Common language from CIA Factbook, (iii) Common religion from CIA Factbook, and (iv) Legal system from various sources in internet. Finally, we aggregate the four dimensions to form the institutional distance construct using the Mahalanobis technique (Mahalanobis et al., 1937). The Mahalanobis index is relevant in our context as we have included a mix of highly and lowly correlated indicators. (Beugelsdijk et al., 2018).
4.2.3 Moderating Variables
Duration of participation in domestic capital market (MktParPeriod) is operationalized as the difference between the year of listing at BSE or NSE and the year of analysis. In case of dual listing at BSE and NSE, we computed the sum of the capital market participation at BSE and NSE. To illustrate, if a firm is listed in BSE since 1990 and in NSE since 2000, we calculate 14 years ((2002 – 1990 = 12) + (2002 – 2000 = 2)) as domestic capital market participation period in year 2002. Our approach is appropriate because we want to capture the extent of information spillovers from the home capital market to the foreign market. To this end, participation in two home market stock exchanges generates much more information about the firm than participation in just one market and the summation of the values makes sense.
We used two ratios to compute a firm’s extent of product market internationalization (PMI). First, we compile the ratio of foreign assets (or long-term investment outside India) to total assets (FATA) in each year. Second, we calculate the ratio of firm’s number of overseas subsidiaries, irrespective of entry mode, to total number of subsidiaries (OSTS) in each year. FATA captures spillover effect from access to foreign resources and OSTS indicates institutional and cultural spillover from foreign market (Johanson & Vahlne, 1977). Next, we integrate these two measures into a composite measure by adding up the two ratios. Hence, our measure takes on values ranging from 0 to 2, with 2 is indicating the highest level of product market internationalization.
4.2.4 Control Variables
In line with previous literature on foreign capital market investments, we use a number of control variables identified as important in prior studies including firm-specific factors that generate differences in performance that result from strategic decisions. Business group affiliation is one of the dominant forms of organizational structure in EMFs and a group often functions as an internal capital market (Khanna & Rivkin, 2001). We control for such access to the group’s internal capital market by computing the natural logarithm of the total sales at group level (BGS). We control for investment into innovation (RDInt) because research-intensive firms may have better valuation abroad. We measure RDInt as R&D expenses per sales. We also include marketing intensity (MktInt; measured as the ration of total marketing expenses (including advertising, sales promotions, and sales and distribution expenditures to total sales) as it captures the possible degree to which information about the focal firm is pushed in the (foreign) markets. We include firm size (FirmSize) measured as the natural logarithm of total sales) to control for significant resources available to the relatively larger firms, which may help to address LoF. As a high performing firm has a greater possibility to receive higher valuation, we control for the performance of the firm (ROA) and measure it as profit before interest and tax divided by total assets. We also include the firm’s leverage (debt-to-equity ratio- DtE) because it indicates how much leverage a company is using and captures firm’s financial condition. As the presence of independent directors indicates lower levels of agency problems and higher legitimacy among foreign investors, we control for the percentage of independent directors in the board (BoDI) (Bell et al., 2014). We include foreign debt (IFB)- foreign currency borrowings to control existing presence in foreign financial market (Bell et al., 2012a, 2012b) and import of raw materials (IRM)- ratio of import of raw materials and sales to control existing legitimacy in foreign product market (Chittoor et al., 2009). We use the presence of one of the big four auditor firms (Big4) as a dummy to control for legitimacy effects that these highly reputed auditors might generate (Aguilera et al., 2015). To control, macro-economic condition at the home market (as the domestic firm is less likely to go public listing in the foreign capital market when its domestic one becomes more developed as the general economy grows), we include home market economic growth rate (EGR_Home) as the annual percentage growth rate in %. Next, the foreign country locations namely USA, UK, Luxembourg, Singapore, and UAE in our sample have different time zones with respect to India (even different equity markets in USA has multiple time zones). Hence, there could be exogeneous effect of the time zone differences of foreign equity market on the dependent variable (valuation at foreign equity market). Theoretically, the time zone differences do not influence equity valuation as any such differences will be absorbed in the equity market pricing mechanism. Otherwise, there would be an arbitrage opportunity. To address any such possibility econometrically, we add a control variable to capture time zone differences across different equity (TimeZoneDiff). We measure TimeZoneDiff as the maximum time difference between equity market locations where firm has offered ADRs or GDRs. Finally, to control for the region effect (Region_Effect), we include a dummy variable with ‘0’ indicating “go public listing” within the firm’s own (geographic) region and ‘1’ beyond its region. To ensure that we separate out all the unobserved year effects, including macroeconomic and environmental effects, we also use 16 annual dummies to control for period effects pertaining to the study period. Table 5 contains an overview of our measures.
Table 5
Variables and measurement
Variable
Measurement
Dependent variable
Valuation at foreign equity market (FEqMktVal)
Tobin Q = \(\frac{\text{Book value of total assets }-\text{ Book value of equity }+\text{ Market value of equity}}{\text{Book value of total assets}}\)
Independent variable
Institutional distance (InsDist)
• Economic distance = Mahalanobis index (Income level, Prevailing inflation rates, Intensity of import, Intensity of export)a
– Income level: GDP per capita- constant 2010 US$
– Inflation rates: GDP deflator- annual %
– Intensity of import: Imports of goods and services- % of GDP
– Intensity of export: Exports of goods and services- % of GDP
• Financial distance = Mahalanobis index (Private credit, Stock market capitalization, Number of listed companies)b
– Private credit: Domestic credit to private sector as % of GDP
– Stock market capitalization: Market capitalization of listed domestic companies as % of GDP
– Number of listed companies: Total number of listed domestic companies compared to total number of domestic companies
– Political distance = Mahalanobis index (Policy-making uncertainty, Democratic character, Size of the state, WTO membership, Regional trade agreement)
– Policy-making uncertainty: political stability measured by considering independent institutional actors with veto power from The Political Constraint Index (POLCON) Datasetc
– Democratic character: democracy score from Freedom Housed
– Size of the state: Government consumption as % GDP from WDI
– WTO membership: membership in WTO from World Trade Organization (WTO) datasete
– Regional trade agreement: dyadic membership in the same trade bloc from WTO
• Administrative distance = Mahalanobis index (Colonizer–colonized link, Common language, Common religion, Legal system)
– Colonizer–colonized link: whether dyad shares a colonial tie from CIA Factbookf
– Common language: % population that speak the same language in the dyad from CIA Factbook
– Common religion: % population that share the same religion in the dyad from CIA Factbook
– Legal system: whether dyad shares the same legal system from various sources in internet
InstDist = Mahalanobis index (Economic distance, Financial distance, Political distance, Administrative distance)
Mahalanobis index (xA, xB) = [(xB – xA)T * C -1 * (xB – xA)]0.5 where xA and xB is a pair of objects, and C is the sample covariance matrix
Moderating variables
Duration of participation in domestic capital market (MktParPeriod)
(i) If the firm is listed at either BSE or NSE, then we calculate the difference between year of listing in BSE or NSE and year of analysis
(ii) In the firm is listed at both BSE and NSE, then we calculate the difference between year of listing in BSE and NSE and year of analysis separately and take the sum of the individual differences
Extent of product market internationalization (PMI)
The sum of (i) the ratio of foreign assets to total assets and (ii) the ratio of firm’s number of overseas subsidiaries to total subsidiaries
Control variables
Group level slack resources (BGS)
Natural logarithm of the sales aggregated at business group level
Investment into innovation (RDInt)
R&D expenses per sales
Investment into marketing (MktInt)
The ratio of the percentage of total annual marketing expenses (including advertising, sales promotions, and sales and distribution expenditures) to sales
Firm size (FirmSize)
Natural logarithm of firm level sales
Performance of the firm (ROA)
Profit before interest and tax divided by total assets
Capital structure (DtE)
Debt to equity ratio
Presence of independent director (BoDI)
Percentage of independent directors in the board
Foreign debt (IFB)
Foreign currency borrowings
Import of raw materials (IRM)
Ratio of import of raw materials and sales
Presence of one of the reputed auditor firms (Big4)
Binary variable; Take ‘1’ in case of presence of one of the big 4 auditing firms, else ‘0’
Economic growth rate in the home market (EGR_Home)
The annual percentage growth rate in %
Time Zone Difference (TimeZoneDiff)
The maximum time difference between equity market locations where firm has offered ADRs or GDRs
Region effect (Region_Effect)
A dummy variable with 0 indicating "go public listing" within the firm's own (geographic) region and 1 beyond its region
We test our models using panel regressions. Panel estimation procedures allow to control for unobserved firm-level heterogeneity and reduce the possibility of biased parameter estimates and thereby of spurious results (Greene, 2003). Considering that we have created a panel dataset, we have the option of either apply fixed effect or random effect regression model. Theoretically, the variance we seek to explain exists at two levels: within- and between-firm. As explained by Certo et al. (2017), within-firm variance represents change occurring within firms over time and between-firm variance denotes change that occurs between firms over time. Our independent variable (Institutional Distance) varies exogenously for a specific firm over period of time based on institutional changes both in home country (India in our case) and host countries (one or more than one foreign location based on the number of foreign equity offerings). But, the changes between firms are endogenous as it is a function of two different firms’ internal decision to decide on the foreign equity location. Therefore, fixed effect regression model suits best for our research context. Empirically, Hausman test also indicates that fixed effect generalized least-squares (GLS) panel regression (Kennedy, 1998) is more suitable for our sample (χ2 = 41.7, df = 17, p < 0.001) than random effects estimations.
5 Results
Table 6 provides descriptive statistics and correlations. We observe that all three variables of interest– institutional distance (r = 0.19; p < 0.001), duration of home capital market participation (r = 0.08; p < 0.001), and extent of product market internationalization (r = 0.14; p < 0.001) are statistically significantly correlated with the dependent variable (valuation at foreign equity market). We checked variance inflation factors (VIF). The VIF values range from 1.010 to a maximum of 4.041 (mean = 1.666), which indicates that multicollinearity is not an issue (Hair et al., 2006).
Table 6
Means, standard deviations and correlations (n = 190)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1. Valuation at foreign equity market
2. Institutional Distance
0.19***
3. Duration of home capital market participation
0.08***
0.35***
4. Extent of product market internationalization
0.14***
0.16***
0.14***
5. Group level slack resources
0.09***
0.08***
0.03
0.09***
6. Investment into innovation
0.17***
0.11***
0.08***
0.08***
− 0.06**
7. Investment into marketing
0.05*
− 0.02
− 0.06**
0.08***
− 0.01
0.16***
8. Firm size
− 0.02
0.24***
0.59***
0.05**
0.11***
0.07***
− 0.13***
9. Performance of the firm
− 0.07***
− 0.02
0.05*
− 0.06**
− 0.07***
0.01
− 0.08***
0.21***
10. Capital structure
− 0.04
− 0.03
− 0.02
− 0.03
− 0.06**
− 0.02
0.02
0.05*
− 0.03
11. Presence of independent director
0.10***
0.05**
0.26***
0.09***
0.13***
0.06**
− 0.02
0.25***
− 0.01
− 0.03
12. Foreign debt
− 0.01
0.11***
0.17***
0.03
0.03
0.09***
0.04
0.20***
0.01
0.05**
0.13***
13. Import of raw materials
0
0.01
0.02
− 0.01
0.04
0
0.01
0.04
− 0.01
0.01
− 0.02
0.03
14. Presence of one of the reputed auditor firms
0.01
0.15***
0.21***
0.04
− 0.06**
0.08***
0.06**
0.20***
0.04*
− 0.03
0.03
0.12***
0.02
15. Economic Growth Rate (Host)
− 0.10***
− 0.14***
− 0.11***
− 0.13***
− 0.04
− 0.02
0
− 0.06**
0.11***
− 0.03
0.04*
0.04
0.02
− 0.02
16. Time Zone Difference
0.27***
0.36***
0.30***
0.26***
0.15***
0.05**
0.01
0.15***
− 0.20***
0.03
0.14***
0
− 0.01
0.05**
− 0.48***
17. Region Effect
0.02
0.44***
0.04*
0
0.04*
− 0.03
− 0.03
0.04*
− 0.03
− 0.01
0.04*
0.11***
0.03
− 0.04
− 0.07***
0.15***
Mean
2.87
1.11
24.30
0.08
7.20
0.00
0.03
4.14
0.10
1.36
0.15
0.04
0.10
0.10
9.51
0.93
0.02
Std. Div
8.51
1.66
20.10
0.20
2.35
0.01
0.07
2.49
0.13
3.66
0.11
0.08
0.58
0.30
8.69
1.55
0.14
Max
140.66
19.44
135.00
1.73
11.14
0.33
1.48
11.13
1.59
87.82
0.57
0.61
25.34
1.00
27.47
3.50
1.00
Min
0.00
0.52
0.00
0.00
− 3.55
0.00
0.00
− 5.71
− 2.66
0.00
0.00
0.00
0.00
0.00
− 2.90
0.00
0.00
*p < 0.05, **p < 0.01, ***p < 0.001
Table 7 presents regressions results with beta, standard error, and p-value. In the first step, we entered only control variables (Model 1). The variables for the linear and quadratic terms of institutional distance are added in Model 2. Wald test chi-square statistics confirm that the inclusion of the institutional distance linear term and then of the quadratic term improves model fit (model 2: χ2 = 18.2, df = 1, p = 0.000). Cohen's f2 also indicates large effect size of our model. The linear term of InsDist is positive (β = 1.501) and significant (p = 0.000). Following Lind and Mehlum (2010), we confirm the inverted U-shaped relationship in three ways. First, the squared term of institutional distance has a statistically significant negative association with the foreign equity market valuation, which indicates that high levels of institutional distance are detrimental to the firm’s valuation (Model 2: β = − 0.085, p = 0.000). Second, we find 8.843 as turning point. This value is well within the range of [0.52; 19.44] for institutional distance in our sample and between 75 and 100 percentile of the data range. This also indicates that until institutional distance is lower than 8.843, one unit increase in institutional distance improves valuation at foreign equity market by 1.5% whereas valuation at foreign equity market starts reducing at 0.085% rate/unit of institutional distance once institutional distance passes the mark of 8.843. We further test to confirm that the slope is positive (0.85) at the minimum and negative (− 1.42) at the maximum value of the data range. Overall, these results provide strong evidence of inverted U-shape relationship between institutional distance and foreign equity market valuation, thus supporting Hypothesis 1.
Table 7
Results of fixed effect panel data regression analyses
Fixed effect models
Model 1 (Control)
Model 2 (H1)
Model 3 (H2)
Model 4 (H3)
Dependent variable
Valuation at foreign equity market
Valuation at foreign equity market
Valuation at foreign equity market
Valuation at foreign equity market
Control variables
β
Standard error
p-value
β
Standard error
p-value
β
Standard error
p-value
β
Standard error
p-value
Group level slack resources
1.042
0.352
0.003
1.078
0.351
0.002
1.172
0.350
0.001
1.118
0.352
0.002
Investment into innovation
10.978
14.341
0.444
7.489
14.289
0.600
5.430
14.204
0.702
6.922
14.236
0.627
Investment into marketing
− 0.959
2.682
0.721
− 0.836
2.669
0.754
− 1.395
2.654
0.599
− 0.904
2.662
0.734
Firm size
− 0.379
0.203
0.061
− 0.399
0.202
0.048
− 0.429
0.201
0.033
− 0.401
0.201
0.046
Performance of the firm
1.564
1.327
0.239
1.455
1.321
0.271
1.907
1.317
0.148
1.693
1.317
0.199
Capital structure
− 0.031
0.043
0.469
− 0.031
0.043
0.468
− 0.024
0.043
0.574
− 0.025
0.043
0.556
Presence of independent director
4.772
1.901
0.012
4.846
1.894
0.011
4.573
1.885
0.015
4.704
1.886
0.013
Foreign debt
4.402
2.514
0.080
3.642
2.506
0.146
3.710
2.500
0.138
3.847
2.498
0.124
Import of raw materials
0.050
0.278
0.857
0.015
0.276
0.958
0.033
0.275
0.903
0.053
0.275
0.847
Presence of one of the reputed auditor firms
− 0.425
0.869
0.625
− 0.686
0.876
0.434
− 0.484
0.875
0.581
− 0.698
0.875
0.425
Economic growth rate (home)
− 0.177
0.345
0.607
− 0.181
0.343
0.598
− 0.187
0.342
0.584
− 0.167
0.343
0.625
Time zone difference
0.377
0.509
0.458
0.189
0.508
0.711
− 0.138
0.600
0.819
0.052
0.598
0.931
Region effect
− 1.890
1.310
0.149
− 5.010
1.478
0.001
− 6.149
1.541
0.000
− 5.386
1.490
0.000
Independent variable
Institutional distance
1.501
0.301
0.000
3.180
0.536
0.000
1.861
0.338
0.000
Institutional distance2
− 0.085
0.020
0.000
− 0.178
0.047
0.000
− 0.102
0.026
0.000
Moderating variables
Duration of home capital market participation
0.047
0.055
0.401
− 0.016
0.053
0.770
Product market internationalization
4.005
0.969
0.000
7.462
1.407
0.000
Institutional distance * duration of home capital market participation
− 0.040
0.010
0.000
Institutional distance2 * duration of home capital market participation
We add interaction terms to test our two contingencies (Model 3 and 4). Again, Wald test chi-square statistics confirm that the inclusion of the additional terms improves model fit in both Model 3 (χ2 = 13.2, df = 1, p = 0.000) and 4 (χ2 = 4.1, df = 1, p = 0.043). In Model 3, the squared term of institutional distance is negative and significant (β = − 0.178, p = 0.000). Furthermore, the interaction is estimated as statistically significant and positive (β = 0.002, p = 0.000), which lends support to our H2. Similarly, in Model 4 and as expected, the coefficient of the squared term of institutional distance (β = − 0.102, p = 0.000) is negative and significant. The statistically significant positive coefficient (β = 0.121, p = 0.043) of the interaction term supports H3. Thus, the inverted U-shaped relationship between institutional distance and valuation at foreign equity market becomes flatter in the presence of longer duration of domestic listing and greater extent of product market internationalization.
In order to gain more insight on exactly how domestic capital market participation period and extent of product market internationalization influence the direct effect, we plot polynomial graphs displaying the interaction results through split sample analysis (Aiken et al., 1991). First, we split the sample into two parts- less than mean of MktParPeriod and more than mean of MktParPeriod. Then we run the direct effect model separately for these two samples. Similarly, we split the sample into two parts- less than mean of PMI and more than mean of PMI. Then we run the direct effect model separately for these two samples. Split sample results are shown in Table 8. Both the positive (model 5: 5.705 vs. model 6: 1.547) and negative (model 5: − 0.773 vs. model 6: − 0.085) slopes for the sample with less than mean of MktParPeriod are steeper compared to the sample with more than mean of MktParPeriod. Similarly, both the positive (model 7: 2.223 vs. model 8: 1.445) and negative (model 7: − 0.124 vs. model 8: − 0.082) slopes for the sample with less than mean of PMI are steeper compared to the sample with more than mean of PMI. Figure 1 shows that the non-linear relationship between institutional distance and valuation at foreign equity market becomes flatter in case of above mean of domestic capital market participation period compared to lower than mean of domestic capital market participation period. Thus, these findings are largely in line with the relationships proposed in H2. Figure 2 shows that the relationship between institutional distance and valuation at foreign equity market becomes flatter in presence of product market internationalization and supports H3.
Table 8
Results of split sample analyses
Fixed effect models
Model 5
(Below average duration of home capital market participation)
Model 6
(Above average duration of home capital market participation)
Model 7
(Below average extent of product market internationalization)
Model 8
(Above average extent of product market internationalization)
β
Standard Error
p-value
β
Standard Error
p-value
β
Standard Error
p-value
β
Standard Error
p-value
Control variables
Group level slack resources
0.337
0.346
0.330
0.037
0.504
0.942
0.347
0.272
0.203
2.518
1.038
0.016
Investment into innovation
− 11.364
13.776
0.410
111.217
39.342
0.005
4.404
18.524
0.812
45.867
34.890
0.189
Investment into marketing
0.239
2.661
0.928
− 20.730
6.601
0.002
1.136
2.564
0.658
− 13.082
6.673
0.051
Firm size
− 0.086
0.208
0.680
0.803
0.515
0.119
0.092
0.188
0.623
− 0.021
1.004
0.983
Performance of the firm
2.388
1.584
0.132
1.883
2.060
0.361
− 0.052
1.203
0.965
0.049
3.685
0.989
Capital structure
− 0.048
0.049
0.327
0.009
0.070
0.898
− 0.023
0.035
0.515
− 0.054
0.458
0.907
Presence of independent director
3.858
2.029
0.057
− 1.455
2.730
0.594
1.264
1.506
0.402
0.929
5.674
0.870
Foreign debt
5.564
2.832
0.050
1.134
4.052
0.780
3.188
2.157
0.140
− 4.088
9.318
0.661
Import of raw materials
0.071
0.269
0.792
0.009
0.632
0.989
0.026
0.219
0.905
− 3.212
6.696
0.632
Presence of one of the reputed auditor firms
− 0.444
1.472
0.763
0.694
1.175
0.555
0.242
0.807
0.764
− 2.903
3.091
0.348
Economic Growth Rate (Home)
− 0.011
0.021
0.596
0.018
0.029
0.533
− 0.002
0.016
0.924
0.086
0.061
0.159
Time zone difference
1.135
0.183
0.000
0.393
0.186
0.035
0.835
0.134
0.000
0.569
0.358
0.113
Region Effect
− 7.832
2.002
0.000
− 2.121
2.242
0.344
− 5.787
1.429
0.000
− 1.782
6.019
0.767
Independent variable
Institutional distance
5.705
0.728
0.000
1.547
0.305
0.000
2.223
0.280
0.000
1.445
0.619
0.020
Institutional distance2
− 0.773
0.120
0.000
− 0.085
0.020
0.000
− 0.124
0.021
0.000
− 0.082
0.042
0.050
FE model indices
Goodness of estimation
F = 17.1598
[0.000]
F = 7.5532
[0.000]
F = 20.6505
[0.000]
F = 2.5068
[0.001]
# of observations
1462
1083
2001
544
Number of firms
159
121
188
101
Fig. 1
Flattening moderating effect of domestic capital market participation period on the relationship between institutional distance and valuation at foreign equity market
Fig. 2
Flattening moderating effect of product market internationalization on inverted U-shaped relationship between institutional distance and valuation at foreign equity market
×
×
5.1 Endogeneity Tests: Reversed Causality and Sample Selection Bias
There is a possibility of endogeneity issue due to presence of unobserved factors (e.g. firm’s capability to overcome LoF) in the error term that are correlated with the explanatory variables (institutional distance in our case). To further assess and alleviate the risk of omitted variable endogeneity due to self-selection bias, we use a multi-stage process suggested by Semadeni et al. (2014). First, we need strong and exogenous instruments, and subsequently we use those instruments in two-stage models. We identify export intensity (measured as the ration of export to total sales) and foreign ownership (measured as the sum of (i) proportion of shares held by foreign promoters, (ii) proportion of shares held by foreign institutional investors as non-promoters, and (iii) proportion of shares held by foreign venture capital investors as non-promoters) as two potential instrumental variables. Prior research emphasizes that export (Salomon & Jin, 2010) and foreign ownership (Bruton et al., 2010) help firm to learn about the foreign market. We argue that level of export and foreign ownership might influence the level of institutional distance measured between home and host locations and therefore, these instruments are theoretically exogenous. Empirically (Table 9), these instruments proved to be statistically significant in the first stage (export intensity: p = 0.093, foreign ownership: p = 0.073) and they are strongly (F = 2.3313, p = 0.010) correlated to the endogenous variable. Having identified strong instruments, we used two-stage least squares regressions and excluded the instruments from the second stage. To determine whether our specification is appropriate, we first check the strength of instruments (model 10: F = 14.414, p = 0.000) (typical rule-of-thumb value is 10 or more to avoid weak instruments (Stock & Yogo, 2005)) and exogenous (overidentifying restrictions) as they produce a non-significant Sargan test result (model 10: p = 0.850). We used the Durbin–Wu–Hausman test on both models and the result (model 10: F = 11.337, p = 0.001) suggests that there is a potential of endogeneity in our models. We observe similar pattern on instruments strength, consistency, and overidentifying restrictions in our results in model 10, 11, and 12. Therefore, we tested endogeneity corrected models and results (model 10 (H1): InstDist2 − β = − 0.562, p = 0.000; model 11 (H2): InstDist2 * MktParPeriod- β = 0.030, p = 0.039; model 12 (H3): InstDist2 * PMI − β = 2.132, p = 0.020) support hypothesized relationships.
Table 9
Endogeneity test to check reverse causality using instrumental variable regression analyses
Models
Model 9
(Instrument relevance)
Model 10
(H1)
Model 11
(H2)
Model 12
(H3)
Dependent variable
Institutional distance
Valuation at foreign equity market
Valuation at foreign equity market
Valuation at foreign equity market
Control variables
β
Standard Error
p-value
β
Standard Error
p-value
β
Standard Error
p-value
β
Standard Error
p-value
Intercept
0.111
0.108
0.305
− 3.425
1.202
0.004
− 17.522
8.182
0.032
− 8.789
3.617
0.015
Group level slack resources
0.014
0.012
0.253
0.175
0.082
0.033
0.110
0.120
0.360
0.224
0.108
0.037
Investment into innovation
9.260
1.848
0.000
20.245
22.530
0.369
30.621
27.547
0.266
1.567
37.292
0.966
Investment into marketing
− 0.178
0.411
0.665
2.358
2.835
0.406
1.053
3.819
0.783
6.310
4.227
0.136
Firm size
0.104
0.013
0.000
− 0.725
0.142
0.000
− 0.812
0.228
0.000
− 0.956
0.267
0.000
Performance of the firm
0.096
0.218
0.659
− 0.536
1.510
0.723
2.974
2.414
0.218
1.562
2.015
0.438
Capital structure
− 0.015
0.007
0.042
− 0.005
0.054
0.927
0.005
0.074
0.948
− 0.007
0.070
0.926
Presence of independent director
− 1.000
0.259
0.000
7.273
1.838
0.000
1.667
3.068
0.587
7.367
2.495
0.003
Foreign debt
0.311
0.351
0.375
− 1.184
2.407
0.623
5.262
4.689
0.262
3.116
3.802
0.413
Import of raw materials
− 0.033
0.047
0.490
0.273
0.327
0.404
0.624
0.506
0.218
0.337
0.436
0.440
Presence of one of the reputed auditor firms
0.619
0.094
0.000
− 0.836
0.671
0.213
− 1.283
1.000
0.199
− 0.654
0.871
0.453
Economic Growth Rate (Home)
0.009
0.004
0.013
− 0.044
0.029
0.129
− 0.055
0.043
0.205
− 0.084
0.050
0.095
Time Zone Difference
0.324
0.021
0.000
− 0.855
0.635
0.178
− 1.286
1.095
0.240
− 2.236
1.426
0.117
Region Effect
4.824
0.200
0.000
− 20.783
5.331
0.000
− 27.682
10.922
0.011
− 29.699
10.834
0.006
Independent variable
Institutional distance
9.768
2.638
0.000
32.092
14.641
0.028
19.839
7.931
0.012
Institutional distance2
− 0.562
0.156
0.000
− 2.076
0.968
0.032
− 1.278
0.528
0.016
Moderating variables
Duration of home capital market participation
0.430
0.201
0.033
− 0.031
0.025
0.211
Product Market Internationalization
− 2.867
2.680
0.285
33.480
12.901
0.010
Institutional Distance * Duration of home capital market participation
− 0.517
0.249
0.038
Institutional Distance2 * Duration of home capital market participation
Instruments Strength for Export Intensity and Foreign Ownership
F = 2.3313
[0.010]
Instruments Strength for Institutional Distance (F-test)
F = 14.414
[0.000]
F = 1703.000
[0.000]
F = 356.2
[0.000]
Instruments Strength for Institutional Distance * Duration of home capital market participation (F-test)
F = 3.820E31
[0.000]
Instruments Strength for Institutional Distance * Product Market Internationalization (F-test)
F = 1.215E32
[0.000]
Consistency of instruments (Durbin-Wu-Hausman Test)
F = 11.337
[0.001]
F = 9.698
[0.002]
F = 11.05
[0.001]
Overidentifying Restrictions (Sargan-Hansen Test)
F = 0.036
[0.850]
F = 0.057
[0.811]
F = 0.017
[0.896]
# of observations
2545
2545
2545
2545
Number of firms
190
190
190
190
Unobserved factors (e.g. firm’s ability to manage post listing process to receive favorable valuation) may influence the probability of entering the sample (e.g. firms that are listed in foreign equity market) and the dependent variable (e.g. valuation in foreign equity market) and therefore, introduce sample-selection bias (Certo et al., 2016). We use two-stage Heckman process where first stage estimates (through probit model) the probability of an observation’s entering the sample based on exclusion restrictions, and the second stage uses a selection parameter, the inverse Mills ratio to account for potential sample-selection bias (Heckman, 1979). We argue that (i) net cash flow from operating activities and (ii) total liabilities are potential exclusion restrictions (or exogenous variables) that predict whether or not an observation appears in our sample. Both net cash flow from operating activities (β = − 0.0007, p = 0.000) and total liabilities (β = 0.001, p = 0.000) are significant in the probit model. In the second stage, inverse Mills ratio (lambda) is not significant (β = 6.289, p = 0.114; ρ = 0.452). As sample selection bias requires both a significant exclusions restrictions in the first stage and a significant lambda (Certo et al., 2016), we rule out the potential sample-selection bias.2
5.2 Robustness Tests
Our results are robust against modifications of our measurements and estimation methods. First, regarding the measurement of institutional distance, we explored if our results hold when we aggregate our dimensions of distance differently. Using the Kogut and Singh’s (1988) aggregation method, we found that our results are qualitatively similar: The squared term of institutional distance is significantly negatively associated the valuation at the foreign equity market (β = − 0.001, p = 0.083), which supports H1. Both interaction effects (β = 0.0001, p = 0.049 for the coefficient of term InstDist2*MktParPeriod; β = 0.007, p = 0.005 for the coefficient of the term InstDist2*PMI) are also supporting H2 and H3.
Second, we alternatively measure institutional distance using The Mahalanobis index and five items ((1) business freedom, (2) trade freedom, (3) property rights, (4) investment freedom, and (5) financial freedom) of the economic freedom index developed by the Heritage Foundation (Meyer et al., 2009). Economic freedom index incorporates the World Bank Doing Business indices and provides information about an overall view of institutions, firms, and individuals in a country to pursue their business activities. Results support hypothesized models- the coefficient of the squared term of institutional distance being negative (β = − 0.143, p = 0.004), and both interaction terms being positive and significant (β = 0.011, p = 0.015 for the coefficient of InstDist2*MktParPeriod; β = 0.192, p = 0.038 for the coefficient of InstDist2*PMI). These results are closely aligned with the findings of our primary estimations.
Third, there are small set of cases in our sample where firm withdrew ADR/GDR offerings. To ensure that such occurrences do not influence our result, we run regression models on a restricted sample. The direct effect – InstDist2 (β = − 0.089, p = 0.000), first moderating term – InstDist2*MktParPeriod (β = 0.002, p = 0.000), and second moderating term – InstDist2*PMI (β = 0.120, p = 0.055) are as per hypothesized model. The models of the robustness tests are not shown but are of course available upon request.
Finally, we conducted additional analyses to examine whether the results hold good without the presence of (1) super large firms and (2) family owned and controlled firms in the sample. Therefore, using two different subsamples (1) without super large firms and (2) without family owned and controlled firms, we separately run all three models. First, we observe that after excluding top one percentile of the firms in our sample based on firm size, all three hypotheses hold good. This indicates that our results are not driven by the foreign equity issuance decisions of the super large firms. Second, we collected new data for promoter block holding which indicates the % of ownership of the original family members. After excluding all the observations where promoter block holding is more than 50%, we received support for all the three hypotheses.3
6 Discussion
Extending the LoF arguments as discussed extensively in the context of product market internationalization (Zaheer, 1995), Bell and colleagues (2012) started an interesting conversation on the costs (termed as CMLOF) and cost mitigation strategies of the firm that expands into foreign capital market. Though empirical evidence of CMLOF is visible in the literature (Moore et al., 2012), there is an increasing trend of firms accessing foreign capital market. As per SEC bulletin, more than 2000 ADRs from 70 countries are available in 2012 (Office of Investor Education & Advocacy, 2012). Popularity of investment in GDRs motivated Luxembourg stock exchange to create country specific GDR indices such as Lux GDRs India and the Lux GDRs Taiwan. As on 31st December, 2017, there are 496 NYSE-listed non-U.S. issuers from 46 countries. This is because integration of global financial market has created a positive environment of raising capital from foreign market (Peng & Su, 2014). Building on the costs-benefits analysis of EMFs that are listed in foreign equity market, we develop a theoretical framework that explains how portfolio diversification and CMLOF are both associated with the institutional distance and contingent effects of domestic equity market listing duration and extent of product market internationalization. Proposed model is supported with the help of fixed effect panel regression analysis of a dataset of 2553 firm-years from 190 Indian firms.
We find institutional distance between issuing firm’s home market and foreign equity location benefits emerging market firm’s ADR/GDR as opportunity associated with the portfolio diversification allows these firms to receive preferential valuation in the foreign equity market. But, eventually cost of CMLOF catches-up and we see a negative trend at the higher side of institutional distance. Though the implications of distance between home and host market is discussed extensively in the international business research (Evans & Mavondo, 2002), most of the time distance has been considered detrimental to achieve organization’s goal (Tihanyi et al., 2005). Building on the institutional distance research in organizational institutionalism and institutional economics (Kostova et al., 2020), our study indicates relevance of contextual factor such as portfolio diversification to develop a more comprehensive understanding of the implications of the institutional distance between home and host market. Inverted U-shaped finding also implies continuing relevance of portfolio diversification argument (contrary to complete market integration theory argument) while drawing a limit in such benefits due to increasing costs associated with CMLOF. In summary, it indicates importance of integrative approach to consider both costs and benefits to expand our understanding of capital market internationalization for EMFs (Purkayastha & Kumar, 2021; Stahl et al., 2016).
We find two organizational contingency factors that influence the effect of institutional distance in the foreign market valuations of the equity offerings from EMFs. First, informational spillovers across the distant equity markets reduce both positive implication of portfolio diversification and negative implication of CMLOF. This result indicates necessity of considering more encompassing sources of legitimate intermediaries (Bell et al., 2012a, 2012b). Second, product market internationalization plays an important role in impacting the effect of institutional distance on the foreign market valuation (Lindorfer et al., 2016). This finding reverses the spillover conversation based on the effect of foreign subsidiaries in emerging markets (Spencer, 2008). We find that spillover may happen from emerging market to developed market as well especially in the context of reducing asymmetry of information among foreign investors through product market internationalization. Therefore, it indicates need for more comprehensive discussion on direction of spillover in the context of global financial market.
Empirically, it is important to carefully consider whether the curve inflects over the relevant data range, meaning a sufficient number of cases are present in both the upward and downward facing parts of the curve (Shaver, 2007). In other words, positive linear term and negative quadratic term may also indicate the existence of a positive but diminishing effect of institutional distance on the foreign equity valuation (Meyer, 2009). In our case, the turning point (8.310) lies outside of two standard deviations above (4.437) and below (-2.219) the mean (1.109) for the independent variable. Therefore, instead of ‘typical’ inverted U-shape effect (Haans et al., 2016), it is indicating a positive diminishing effect rather than a complete negative turn of the relationship. Thus, theoretically, our results support a cost–benefit trade-offs based on institutional distance while the positive effect of portfolio diversification is more prominent compared to the negative effect of CMLOF.
Our paper provides some important managerial implications. As the relationship between institutional distance and foreign market valuation demonstrates inverted U-shape relationship, there is an optimum inflection point beyond which marginal benefits of portfolio diversification from institutional distance become negligible. Thus, firm specific analysis of suitable institutional distance between home and host equity market will allow firm to achieve preferential valuation in the foreign equity market. Managers in-charge of capital management process need to be aware of the contingent effect of duration of domestic listing and extent of product market internationalization. Though management may not be able to influence duration of domestic listing the way they can control extent of product market internationalization, better understanding of the available levers that influences foreign market valuation is an important managerial take-away.
Our research is not free from the limitations that provides future opportunities for further development. First, we confine our analysis to the non-linear effect of the institutional distance on the foreign equity market valuation. Similar analysis can be expanded into two other forms of foreign capital expansion- debt and private equity investment (Bell et al., 2012a, 2012b). Second, distance between home and host market manifested in multiple forms. A further expansion of our analysis can be done using Berry et al. (2010)’s work on cross-national distance. Third, we look at the foreign equity participation considering only India as home market. Our research design is motivated by our interest to understand the determinants of ADRs and GDRs valuation from one home country into multiple foreign equity markets. A multi-home country analysis using contrasting (large vs. small or emerging vs. developed) set of home countries will provide more comprehensive understanding of the phenomenon. Fourth, apart from our firm-level moderators, other inside factors related to intrinsic quality of a firm's offering may affect strategic decisions to search for foreign capital. To develop greater understanding on the factors that help Indian firms to tap into foreign equity markets, further research can be done to find firm-specific contingency attributes. Fifth, firm's specific characteristics (e.g., past strategic development and current status) could lead to a positive appraisal or valuation by the foreign equity markets. For example, the firm may possibly earn a premium for its share from the foreign equity market if it has greater capacity to develop new products/technologies and/or maintain stable cash flow through international market diversification. Therefore, the industry in which a firm operates and the strategy on which it relies could be critical to the realization of superior performance which lead to its share premium by the capital markets. Therefore, further research focusing on the influence of firm-specific factors on foreign equity valuations may be another promising avenue. Lastly, there is a possibility that collaboration in the form of joint venture or alliance with the firm from a developed economy may help an Indian firm to use externalities for foreign equity market expansion. Therefore, a further extension of the current research findings in the context of the foreign firm’s involvement in the Indian domestic market will be a valuable research endeavor.
7 Conclusion
Traditionally, IB and International Finance have considered internationalization strategy in product and capital markets as two separate research domains. By focusing on the impact of institutional distance on the firm’s valuations in foreign markets, our research indicates that institutional distance and information spillover effects between product and factor markets may be interlinked. In this study, we attempt to extend the ongoing conversation on the firm-level globalization strategies by offering more holistic, inter-disciplinary research framework that can advance further our understanding of internationalization processes and their outcomes, in particular in the context of emerging market firms.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
MANAGEMENT INTERNATIONAL REVIEW is a double-blind refereed journal that aims at the advancement and dissemination of research in the fields of International Management.
Due to space constraints, we do not include the details of robustness test in the paper. Tabulated results for all the robustness analyses are available from the authors upon request.
