Configuring political relationships to navigate host-country institutional complexity: Insights from Anglophone sub-Saharan Africa

Abstract

We examine how ties with multiple host-country political institutions contribute to MNE subsidiary performance in countries with weak formal institutions. We suggest that forging relationships between subsidiaries and host-country government actors, local chieftains, and religious leaders generates regulative, normative, and cultural-cognitive political resources. We integrate institutional and configuration theories to argue that similarity to an ideal configuration of the three political resources contributes to MNE subsidiary performance, and that the more dysfunctional host-country institutions, the greater the impact on performance. We test our hypotheses using primary and archival data from 604 MNE subsidiaries in 23 Anglophone sub-Saharan African countries and find support for our hypotheses. In our conclusion, we discuss the wider theoretical, managerial, and public-policy implications of our findings.

INTRODUCTION

Multinational enterprise (MNE) research has long considered non-market strategy to be an important determinant of internationalization (e.g., Funk & Hirschman, 2017; Zhao & Lu, 2016) and performance (e.g., Amore & Corina, 2021; Boddewyn & Doh, 2011; Sun, Doh, Rajwani, & Siegel, 2021). An MNE’s non-market strategy is its “concerted pattern of actions to improve its performance by managing the institutional or societal context of economic competition” (Mellahi, Frynas, Sun, & Siegel, 2016: 144). The construct takes multiple forms, including corporate social activity (Husted & Allen, 2009), corporate political activity (Mellahi et al., 2016), legal and public sensitization strategies (Baron, 1997), and self-categorization strategies (Curchod, Patriotta, & Wright, 2020). However, Sun et al. (2021) argue that corporate social activity (CSA) and corporate political activity (CPA) are the two principal non-market strategies. MNEs engage in CSA with multiple societal stakeholders to improve social and environmental practices (Husted & Allen, 2006; Teegen, Doh, & Vachani, 2004), and in CPA to deal with home-country and host-country political actors (Boddewyn & Doh, 2011; Cui, Hu, Li, & Meyer, 2018). Our study focuses on the CPA of MNE subsidiaries in host countries as a non-market strategy to navigate institutional complexity (Marquis & Raynard, 2015; Sheng, Zhou, & Li, 2011).

Prior research has examined under what conditions MNEs engage in CPA (e.g., Cui et al., 2018; Sun et al., 2021), however, extant studies have emphasized ties between MNEs and national government political institutions (see reviews by Mellahi et al., 2016; Sun et al., 2021). Some earlier studies have drawn on multiple theoretical perspectives such as a firm-specific or country-specific advantage, the resource-based view, or combinations of these, and argue that such ties help MNEs minimize host-country political risk and operating and transactional costs, and thereby enhance the performance of their subsidiaries (e.g., Holburn & Vanden Bergh, 2014; Zhao & Lu, 2016). While our understanding of how CPA can improve subsidiary performance has grown, empirical studies of the relationship between CPA and performance suffer from two major limitations (see Sun et al., 2021, pp. 1827–1832; Mellahi et al., 2016). First, while they have identified a wide variety of host-country political institutions, some formal, such as national governments, and some informal, for example, local chieftaincies and religious bodies (Amankwah-Amoah, Boso, & Kutsoati, 2022; Hoang, 2018), we still lack knowledge of how ties with them can be configured to best enhance MNE subsidiary performance (Amankwah-Amoah et al., 2022; Sun et al., 2021). Second, while some studies have suggested that the performance effect of CPA may be contingent upon an array of host-country institutional and market environment conditions (Bucheli & Salvaj, 2018), there has been limited theoretical specification and little empirical examination of these boundary conditions. Thus, we address an under-researched question: How can MNE subsidiaries achieve an ideal configuration of relational political resources from subsidiary ties with a multiplicity of host-country political institutions in poorly functioning host-country markets? We draw on Scott (1995, 2014) to develop a conceptual model to explain how the ties subsidiaries form with political institutions can provide the political resources needed to navigate host-country institutional complexity. In addition, we adopt the systems perspective of configuration theory to argue that the closer an MNE subsidiary comes to an ideal configuration of political resources, the better its performance.

We advance the multinational non-market strategy literature in two main ways. First, by integrating the Scottian view of institutions with configuration theory, we show how the capability to achieve similarity to an ideal configuration (SIC) of political resources contributes to performance. We apply Scott’s (1995, 2014) view of institutions as regulative, normative, and cultural-cognitive to the socio-cultural institutional complexity of sub-Saharan Africa (SSA), examining ties with national government actors, local chieftains, and religious leaders. The three CPAs generate separate but interrelated relational political resources that constitute firm-specific advantages. Ties with national government actors can generate regulatory and informational resources in the form of stronger enforcement of laws and regulations and favorable access to information on public policies, including changes in them. Ties with local chieftains can generate both regulatory and normative political capital and help reduce encroachment on intellectual property rights while enhancing local market legitimacy. Finally, ties with religious leaders can generate normative and cultural-cognitive political capital, which confers greater legitimacy on a firm and its brand. We contend that while each type of tie has its benefits, they may also complement each other or serve as substitutes. To gauge the performance outcomes of the interconnectivity between the three types of political resources, we use the systems perspective of configuration theory. We argue that getting close to an ideal configuration of the three political resources increases protection of MNE subsidiary assets and operations in weak host-country institutional environments. It minimizes political risk and operating and transaction costs, lowers vulnerability and volatility of sales revenue, and facilitates the creation and effective functioning of host-country markets and hence MNE subsidiary performance.

Second, prior CPA research has relied mostly on empirical evidence from developed markets. As there are few studies on emerging countries, we lack knowledge on the effect of CPA in less developed markets. Our study of MNE subsidiaries in sub-Saharan Africa provides a unique context to theorize about the benefits of CPA in institutionally weak host countries. A major strength of our study is that it integrates rich qualitative case studies with survey and archival data from multiple sources, over time, and across 23 Anglophone sub-Saharan African countries. This allows us to theorize about how MNE subsidiaries build ties with multiple political institutions to generate multiple political resources to enhance performance. Thus, we go beyond the already widely researched effect on performance of ties with national government political institutions. We show how achieving an ideal configuration of regulative, normative, and cultural-cognitive political resources, obtained from ties with host-country national government actors, local chieftains, and religious leaders, can result in sales and profit growth in a dysfunctional environment (Figure 1).

Figure 1.

Research model.

THEORETICAL BACKGROUND

The Multinational Enterprise and Non-Market Strategy

International business and strategy research has increasingly paid attention to differences in the internationalization behaviors and sources of competitive advantages of not only developed-country MNEs (DMNEs), but also emerging-market MNEs (EMNEs) (see Giuliani, Gorgoni, Günther, & Rabellotti, 2014; Hennart, 2012; Li, Quan, Stoian, & Azar, 2018; Luo & Tung, 2018; Steinberg, Urbig, Procher, & Volkmann, 2021). Importantly, while DMNEs have been shown to internationalize after first establishing a solid foundation in the home country (Johanson & Vahlne, 1977), EMNEs tend to internationalize at an earlier stage because of institutional constraints in home-country markets (Luo & Tung, 2007). Whereas DMNEs achieve competitive advantages in global markets by accumulating large stocks of resources and capabilities such as strong brands, advanced technologies, and access to well-developed financial infrastructure (Ramamurti & Williamson, 2019), EMNEs typically achieve competitive advantages by developing and leveraging institutional strategies that allow them to navigate weak home-country regulatory structures, under-developed capital markets, and rapid and turbulent changes (Marquis & Raynard, 2015).

Efforts to explain the unique challenges MNEs face in navigating cross-border institutional weakness have led to a stream of research on non-market strategies. Kostova and Zaheer (1999) emphasize the multiplicity of country institutional environments within which MNEs operate. Ramamurti’s (2001) two-tiered model explains how MNEs bargain with home-country and host-country governments and multilateral institutions to enhance competitive advantage. The institutional strategies of MNEs have often been studied within the context of multinational non-market strategy, which captures sustained actions to influence the regulatory, political, and social environment of the home country and host countries (Baron, 1997; Mellahi et al., 2016; Sun et al., 2021). Sun et al. (2021) see corporate social activities (CSA) (Husted & Allen, 2006) and corporate political activities (CPA) (Mellahi et al., 2016) as the two most studied non-market strategies. Earlier research has looked at other types of non-market strategies, including legal and public sensitization strategy (Baron, 1997), and self-categorization strategy (Curchod et al., 2020).

Corporate social activity seeks “to advance some social good that allows a firm to enhance organizational performance, regardless of motive” (Mellahi et al., 2016, p. 144). Some have assumed this strategy to be a self-regulating business model, that is, that firms are naturally socially accountable to the wider society (Husted & Allen, 2006; Lins, Servaes, & Tamayo, 2017; Teegen et al., 2004). Baron (1997) shows how Kodak used a non-market legal strategy against Fujifilm to influence international trade policy to facilitate its own market entry into Japan, to get access to broad distribution networks, and to increase its profits while reducing those of Fujifilm. Curchod et al. (2020) suggest that MNEs use both defensive and proactive self-categorization strategies such as rhetoric, labels, and narratives to achieve strategic fit. For example, eBay adopted this strategy to enter the French market (Curchod et al., 2020). Finally, Lucea and Doh (2012) argue that MNEs can build coalitions and share information with host-country institutions to achieve their strategic goals (see also Doh et al., 2012).

Our study focuses on host-country CPAs by which MNEs seek to influence regulatory, political, and social environments by leveraging ties with formal political institutions, including elected and unelected government officials and members of regulatory bodies ( Nell, Puck, & Heidenreich, 2015; Ge, Stanley, Eddleston, & Kellermanns, 2017), as well as informal political institutions such as the local community, civil society, and religious organizations (Sun et al., 2021; Webb, Khoury, & Hitt, 2020). Previous studies have found that CPA may prove effective in navigating “voids that occur in both formal and informal institutions” (Webb et al., 2020: 505). Ties with formal political institutions are forged to improve enforcement of favorable laws and regulations (Liou, Brown, & Hasija, 2021). MNEs are also known to make donations (e.g., monetary, and non-monetary contributions) to formal political institutions (Funk & Hirschman, 2017; Holburn & Vanden Bergh, 2014). In addition, they directly partner with formal political institutions charged with enforcing behaviors that facilitate efficient functioning of host-country markets. MNE subsidiaries provide financial support and partner with informal political institutions represented by local chieftains, religious leaders, and other non-governmental organizational actors attempting to improve societal wellbeing (Amankwah-Amoah et al., 2022).

Through CPA, MNE subsidiaries in host countries with weak formal institutions can contribute to institutional development in ways that are mutually beneficial to them and host countries (Sheng et al., 2011). Mair, Marti and Ventresca (2012) show that MNE subsidiaries can partner with political institutions to reduce unlawful market behavior (Sheng et al., 2011), such as the distribution and consumption of counterfeit or pirated goods (Amankwah-Amoah et al., 2022). In addition, they can use relational, infrastructure-building, and socio-cultural bridging political strategies to enhance the effectiveness of host-country institutions and assure the profitability and long-term survival of MNE subsidiaries (Marquis & Raynard, 2015). Still other studies have argued that MNEs use market actions, e.g., innovation, together with specific political activities, e.g., political mobilization, to remedy deficiencies and weaknesses in host-country political institutions (e.g., Funk & Hirschman, 2017; Sheng et al., 2011).

Despite the potential benefits of CPA for MNEs and for host countries, prior studies drawing on insights from sociology (e.g., Granovetter, 1985; Portes & Sensenbrenner, 1993) and strategy (e.g., Adler & Kwon, 2002; Villena et al., 2011), have laid bare the “dark side” of CPA (e.g., Dieleman & Sachs, 2008; Lawton, Rajwani, & Doh, 2013; Liedong, 2021; Liedong, Aghanya, & Rajwani, 2020), notably when it swings the regulatory and enforcement pendulum to the advantage of MNEs at the expense of consumers and the wider society. We contend that this is likely to impede efficient functioning of the market. Indeed, Sheng et al. (2011) argue that political ties may not prove to be an effective mechanism for developing long-term relationships between MNEs and host governments due to innate goal divergence, with managers seeking to maximize economic returns for their firms, and political actors prioritizing their political careers. Ties with political actors may also be short-lived given high turnover in politics. Siegel (2007) argues that political ties may in fact prove to be a liability in that MNE operations can be put at risk when there is a change in political regimes. This is perhaps why Tajeddin and Carney (2019) find that African firms have instead established business groups to improve their access to skilled employees, formal credit, and technology.

Corporate Political Activity and MNE Subsidiary Performance

Researchers have drawn on a variety of theoretical perspectives to explain how, and under what conditions, CPA can contribute to MNE subsidiary performance. They have used Dunning’s eclectic paradigm, resource-based theory, agency theory, social capital theory, and institutional theory, individually and in combination. For instance, Holburn and Zelner (2010) integrate the resource-based theory and transaction cost economics to explain how political resources facilitate risky foreign direct investment in unstable host-country markets. Furthermore, Cuervo-Cazurra, Inkpen, Musacchio and Ramaswamy (2014) draw on agency theory, transaction costs economics, and the resource-based view to examine how government ownership of state-owned multinational companies influences their internationalization.

Institutional theory provides a valuable perspective to explain how various types of political resources affect MNE performance (Amankwah-Amoah et al., 2022; Rizopoulos & Sergakis, 2010). The CPA literature has focused rather narrowly on MNE ties with national governmental political institutions (De Villa, Rajwani, Lawton, & Mellahi, 2019; Sun et al., 2021) to the neglect of the broader spectrum of ties and partnerships that MNEs can build with host-country institutional actors. Yet, we know that MNEs establish ties with multiple host-country political stakeholders, including national, subnational, and grassroots-level political institutions (Amankwah-Amoah et al., 2022; Bucheli & Salvaj, 2018; Hoang, 2018; Luo & Chung, 2013). Following the arguments of the non-market strategy literature (e.g., Mellahi et al., 2016; Sun et al., 2021), of institutional theory (see exemplary studies in the Appendix A), and of research on the complexity of the institutional setting of sub-Saharan countries, we identify three major political institutions, thereby offering a more complete view of potential ties between MNE subsidiaries and host-country political institutions in countries with weak formal institutions.

Corporate Political Activity and Institutional Conditions in Sub-Saharan Africa

Scott (1995, 2014) has argued that a society’s institutional structure goes beyond the regulative institutions that have been studied in the multinational non-market strategy literature (Lawton et al., 2013; Mellahi et al., 2016; Sun et al., 2021). A major contention in Scott’s institutional theory is that institutions can be organized along a continuum ranging from regulative, to normative, to cultural-cognitive. Regulative institutional structures relate to society’s capacity to establish rules and enforce them (Scott, 2014). While the normative institutional pillar captures the norms and values of the society, the cultural-cognitive pillar relates to “the shared conceptions that constitute the nature of social reality and create the frames through which meaning is made” (Scott, 2014: 67). Recent studies have contended that MNE subsidiaries may cultivate regulatory ties with national political actors and members of regulatory bodies, normative ones with, for example, local chieftains and kings, and cultural-cognitive ones with local religious leaders so as to accumulate a range of relational political resources (Amankwah-Amoah et al., 2022). At the normative level subsidiaries can collaborate with other firms, industry stakeholders, and government actors to establish and enforce market norms, with long-term benefits to both MNE subsidiaries and market participants. Despite the call of Sun et al. (2021) for research on MNEs’ dynamic interactions with multiple host-country political institutional stakeholder constituencies and how such interactions can be structured to generate unique relational political resources (Amankwah-Amoah et al., 2022; Dorobantu, Kaul, & Zelner, 2017), the extant literature is limited with respect to how different combinations of these relational political resources explain heterogeneity in MNE subsidiaries’ competitive positions and performance in fragile institutional environments.

Importantly, Scott’s regulative, normative, and cultural-cognitive continuum provides a strong theoretical basis for understanding institutional complexity in sub-Saharan Africa. Importantly, socio-cultural institutional structures in many SSA societies confer absolute authority on institutional leaders whose legitimacy is derived from customs, traditions, and spirituality (Honig, 2019; Mbalyohere & Lawton, 2018; Tieleman & Uitermark, 2019). Hence, leaders of both formal and informal political institutions can greatly influence the behavior of individuals and organizations and exercise unfettered power over resource allocation. In addition, the SSA institutional environment includes other stakeholders such as pressure groups and non-governmental actors. The hybridity of SSA political institutions explains why it makes sense for subsidiaries to form close regulative, normative, and cultural-cognitive ties, with national government, local chieftaincy, and religious political institutions respectively, to acquire political (Amankwah-Amoah et al., 2022) and social resources (Acquaah, 2012).

The complexity of SSA institutions and how it drives multiplicity of CPAs is well argued in the extant literature. For example, anthropological and sociological studies have highlighted the complex configurations of cultural features that characterize the “African modes of livelihood, beliefs, attitudes, behaviors, even in languages, and artistic expression” (Awedoba, 2002: 21). Africa is characterized by a socio-cultural institutional structure that prioritizes conformity to established social norms, respect and loyalty to elders, family members, and religious leaders (Greenfield, Keller, Fulgni, & Maynard, 2003). This structure values interdependence, collaboration, association, and solidarity (Greenfield et al., 2003; Omobowale & Omobowale, 2019) and has given rise to multiple concurrent political institutions, national governments, local chieftaincies, and religious organizations (Gyekye, 2002). This institutional complexity is succinctly captured by Acquaah (2007: 1240): “while the power and authority in the formal political system are exercised by government officials, traditional political authority and leadership are in the hands of kings and chieftains of ethnic groups, cities, towns, and villages”. The fact that in SSA religion is deeply “built into the culture of the people and so is a way of life” (Gyekye, 2002: 17), gives religious leaders as well as traditional kings and chieftains a normative and cultural-cognitive authority.

Additionally, local chieftains legitimize specific behaviors and have the power to sanction, sometimes expel, individuals and organizations from the community for unacceptable ones (Gyekye, 2002). Cheeseman, Kanyinga and Lynch (2020) highlight the central role of traditional chiefs in Kenya’s state bureaucracy by virtue of their positions in provincial administrations, allowing them to be deeply involved in (1) implementing national policies and enforcing law and order, (2) mobilizing political support, and (3) linking the government with its citizens. Similarly, there is strong adherence to the norms of religious organizations, the dominant ones being traditional African religions, Christianity, and Islam, with religious norms and teachings influencing the attitudes, thoughts, and actions of individuals and organizations (Awedoba, 2002; Honig, 2019).

Local chieftains and religious leaders retain the authority to create, maintain, and enforce social norms and values within communities, including the power to ban sales of specific products and services that are not in keeping with local norms, values, and customs. For example, traditional tribal and religious leaders in Burundi have the authority to enforce social norms and values, and act as peacemakers during conflicts (Kasmi & Khan, 2021). Additionally, the constitutions of Malawi (Chiweza, 2007), South Africa (Ntsebeza, 2005; Oomen, 2005), and Ghana stipulate that local customs, traditions, and beliefs should be upheld. Tieleman and Uitermark (2019: 707) find that local chiefs in Ghana exercise “greater power as managers of the land and gatekeepers of the state bureaucracy, […and] continue to play the role of informal administrators and points of liaison between local communities and state institutions”. Evidence from Uganda also suggests that both traditional chiefs and religious organizations, e.g., Pentecostal-charismatic churches, play prominent roles in “influencing and shaping public policies, politics, and public spaces, like newspapers, that discuss and address public morality and decency in the country” (Bompani & Brown, 2015: 110). MNE subsidiaries can strengthen their ability to connect with these informal political institutions by appointing chieftains, kings and other royals, and religious leaders or their representatives, to their executive boards. Some subsidiaries employ liaison officers, often locals with network and kinship ties with chieftains and religious leaders, to facilitate such relationships (cf. Bompani & Brown, 2015; Mkodzongi, 2016; Tieleman & Uitermark, 2019).

Thus, it can be argued that MNE subsidiary performance in SSA may be a function of the complexity of the institutional environment and of the non-market strategy applied. We argue, therefore, that while different types of political resources can be generated from ties with different political institutions in SSA (Acquaah, 2012), the impact on MNE subsidiary performance depends on how close the configuration of political resources obtained is to the ideal one.

HYPOTHESIS DEVELOPMENT

Similarity to an Ideal Configuration of Political Resources and MNE Subsidiary Performance

Prior research that links MNE subsidiary performance to specific corporate political activities such as forging ties with national government (Parente, Rong, Geleilate, & Misati, 2019; Sheng et al., 2011), assumes that the resources generated exist in isolation. Yet, scholars suggest that MNE ties with host-country political institutions may be complex (Parente et al., 2019; Sun et al., 2021) and potentially interlinked (Amankwah-Amoah et al., 2022). Our study contributes to the non-market strategy literature by drawing insights from the conceptualization of Sun et al. (2021) and exploratory work of Parente et al. (2019) to theorize and empirically test the MNE subsidiary performance outcomes of a capability to orchestrate multiple interconnected political resources obtained from various corporate political activities. We rely on the systems perspective of configuration theory (Engelen, Gupta, Strenger, & Brettel, 2015) to build our argument about how a capability to configure multiple political resources explains variation in MNE subsidiary performance. The configuration theory view is a good fit because our research context exhibits a large number of possible configurations, and hence we can expect heterogeneity in outcomes (Drazin & Van de Ven, 1985; Doty, Glick, & Huber, 1993). Past studies indicate that this approach to modelling and testing offers a more holistic understanding of the performance effect of multiple configurations than reductionist approaches such as moderation analysis (e.g., Engelen et al., 2015; De Clercq et al., 2012).

Responding to the call of Sun et al. (2021) for additional research on the complexity of MNE corporate political activities, we hypothesize that similarity to an ideal configuration of political resources originating from ties with national government actors, local chieftains, and religious leaders, explains variation in MNE subsidiary performance. We define a similarity to an ideal configuration of political resources as the extent to which firms attain the highest level of ties with each of the three types of political institutions (Doty et al., 1993). Our hypothesis is premised on the observation that MNE subsidiaries in SSA engage in multiple political ties with different types of political institutions in an effort to extract different political resources (Amankwah-Amoah et al., 2022; Bucheli & Salvaj, 2018). Because each type of tie entails unique value-creating potential and risk (Adler & Kwon, 2002; Wu, 2008), the political resources obtained from them can be complementary or substitutive so that their ideal configuration leads to maximum subsidiary growth and profitability (Amankwah-Amoah et al., 2022; Carey, Lawson, & Krause, 2011).

The high collectivistic and informal cultures that characterize many SSA countries (Acquaah, 2007; Amankwah-Amoah et al., 2022) result in strong and complex interdependence between actors in formal political institutions and between them and actors in informal ones. This means that a connection with one political institution implies the possibility of one with another. In view of this dynamic, a high degree of connectivity is observed between government actors and local chieftains. For example, Gyampo (2009) finds that local chiefs successfully influence the outcome of both national and local elections because they have considerable authority in local jurisdictions. Similarly, elections in SSA are influenced by religious and ethical issues (Dovlo, 2006). For example, the relationship between national government officials and religious leaders is heightened during national and local political campaigns where one observes Christian Councils, the National Catholic secretariats, and Chief Imams playing a direct role in influencing national and local political narratives. The high degree of connectivity between major political institutions is highlighted in the 2020 Afrobarometer report on Ghana, which concludes that “religious and traditional leaders could be an important asset [… as] they enjoy greater popular trust and more contact with citizens than most other [political] leaders. A broader consultation with traditional and religious leaders who have close interaction with the people they lead might be an effective way to court public cooperation in the implementation of anti-COVID-19 measures” (Sanny & Asiamah, 2020: 1). In addition, Honig (2019) argues that besides having great local influence, traditional chiefs function as development intermediaries between national governments and local communities. Thus, connecting with one type of political actor means connecting with others.

We contend that SIC of political resources from the three constituencies will boost MNE subsidiary performance. First, ties with multiple political institutions may serve as a buffering mechanism to generate diverse and redundant political resources (Amankwah-Amoah et al., 2022; Sidki Darendeli & Hill, 2016) which can also provide the necessary flexibility to navigate complex institutional environments (Amankwah-Amoah et al., 2022; Parente et al., 2019). Second, as each institutional actor controls specific rent-generating political resources, concurrently pursuing ties with all of them provides access to diverse political resources. For example, ties with national government actors enable MNE subsidiaries to obtain stronger enforcement of laws and regulations and access to information on existing and future public policies, while ties with local chieftains can help protect MNEs from encroachment on intellectual property rights, as local chieftains have the authority to prescribe acceptable behaviors in their localities (Tieleman & Uitermark, 2019). Third, ties with local chieftains can also provide local legitimacy for MNE brands as local chieftains can help communicate brand messages to local consumers and secure market space for them. This is documented by Omobowale and Omobowale’s (2019) study of informal markets in Ibadan (Nigeria), which shows how access and retention of market space is facilitated by the support of local chiefs and queens. Additionally, ties with both local chieftains and religious leaders can generate normative and cultural-cognitive political capital because they are entrusted with protecting the morality, customs, traditions, and spirituality of society and hence their recommendation can provide legitimacy to the subsidiary’s brands (Honig, 2019; Kavas, Jarzabkowski, & Nigam, 2020). Thus, these multiple connections among regulatory, normative, and cultural-cognitive political resources may constitute a causally ambiguous firm-specific advantage which competitors may find difficult to imitate because it is socially complex. Having ties with multiple institutions may also reduce the MNE’s power asymmetry and hence improve performance. Accordingly, we argue that:

Hypothesis 1:

Similarity to an ideal configuration of relational political resources from ties with national government actors, local chieftains, and religious leaders is positively associated with MNE subsidiary performance in sub-Saharan African countries.

Moderating Effect of Dysfunctional Market Conditions

Beyond the positive effect on MNE subsidiary performance of similarity to an ideal configuration of multiple political resources, an unanswered research question is which host-country environment conditions are likely to maximize this effect? We further extend the non-market strategy literature (particularly Gao, Zuzul, Jones, & Khanna, 2017; Parente et al., 2019) by arguing that this effect will be greater in dysfunctional host markets.

Researchers have shown that the positive impact of CPA on subsidiary performance is heightened when MNEs operate in less-developed countries. Weak institutions have been shown to inhibit creation and effective functioning of markets (Narooz & Child, 2017; Webb et al., 2020). They cause bureaucratic bottlenecks, poor disclosure, weak governance systems, and dysfunctional markets because of ineffective legal enforcement of laws (Tracey & Phillips, 2011). Scholars have contended that under conditions of institutional failure, a void (or a vacuum) is created in the institutional environment (Gao et al., 2017), marked by widespread illicit trading activities and growing violation of intellectual property (Amankwah-Amoah et al., 2022). This has a negative impact on operating and transactional costs, market value of assets, and sales (Julian & Ofori‐Dankwa, 2013). We argue that MNE subsidiaries respond to these institutional weaknesses by leveraging political resources from multiple political institutions (Bucheli & Salvaj, 2018; Sidki Darendeli & Hill, 2016).

We advance several arguments to explain how political resources from national governmental, local chieftain, and religious political institutions help lower the negative impact of institutional weaknesses on MNE subsidiary performance. First, as the political risk and operating and transaction costs associated with host-country dysfunctional market conditions increase, the value of regulatory resources in the form of stronger enforcement of laws and regulations and preferential access to information on regulatory changes increases; second, ties with regulatory bodies allow firms to obtain more effective enforcement of intellectual property laws and standards so as to reduce threats to brands and sales revenue; third, the value of the protection and legitimacy given by local chieftains to a subsidiary’s operations and brands in local markets increases; and fourth, because religious followers see their leaders as custodians of moral values and norms, the influence those leaders have on the purchasing behavior of consumers becomes more valuable. Thus, we predict that as markets become more dysfunctional, the value of political resources from ties with national governmental, local chieftaincy, and religious political institutions increases. Accordingly, we argue that:

Hypothesis 2:

The positive effect on MNE subsidiary performance of similarity to an ideal configuration of relational political resources from ties with national government actors, local chieftains, and religious leaders is strengthened under greater host-country dysfunctional market conditions in sub-Sahara African countries.

RESEARCH METHODS

Research Setting

While much of the international business and strategy research has largely focused on MNE operations in industrialized countries and BRIC emerging markets, there are several reasons why sub-Saharan Africa offers a suitable context to theorize about the performance effects of non-market strategy. First, although global trading activities in sub-Saharan Africa have traditionally been dominated by European (and to some extent North American) MNEs, the recent emergence of developing-economy MNEs is reshaping many African markets (Kaplinsky & Morris, 2009). For example, China is now estimated to be Africa’s largest trading partner, with Sino-African trade topping approximately US$200 billion per year by 2020 (Shepard, 2019; Siu, 2019), while an Economist report estimates that Chinese firms represent one-eighth of the continent’s industrial output (The Economist Special Report, 2022). A joint report by the Africa Development Bank and the OECD estimates that “Africa’s economic outlook continues to brighten. Its real GDP growth, estimated at 3.4% for 2019, is projected to accelerate to 3.9% in 2020 and to 4.1% in 2021” (Africa Economic Outlook, 2020). It has also been estimated that the African Continental Free Trade Area (AfCFTA) could increase the value of intra-African trade by approximately 15 to 25% by 2040 and boost economic output by US$29 trillion by 2050 (Signé & Gurib-Fakim, 2019; Songwe, 2019), and this is expected to be powered by a young population rising to 2.5 billion in 2050. This evolving economic, social, and demographic landscape, therefore, makes SSA a fruitful location for our study (Oke et al., 2022).

Second, despite its exciting potential, socio-political instability, civil unrests, and ethnic clashes make sub-Saharan Africa highly unpredictable (Fon, Filippaios, Stoian, & Lee, 2021; Luiz, Stringfellow, & Jefthas, 2017). Unfavorable policies on procurement, taxation, asset transfer, and appropriation of profit (De Villa, Rajwani, & Lawton, 2015; Mbalyohere & Lawton, 2018) undermine the resilience and performance of MNE subsidiaries (Acquaah, 2012; Parente et al., 2019). Faced with such conditions, MNE subsidiaries vie to develop ties with different classes of political actors (Amankwah-Amoah et al., 2022; Sidki Darendeli & Hill, 2016).

Third, the highly collectivistic and informal relationship culture that characterizes many SSA societies justifies the use by MNE subsidiaries of multifaceted non-market political strategies (Acquaah, 2007; Amankwah-Amoah et al., 2022). Thus, these contextual nuances make SSA a suitable setting for our study. Given the paucity of research on this topic, we designed two complementary empirical studies: an exploratory one to generate initial insights into the types of political relationships in which MNE subsidiaries engage, and a large-scale survey and archival study to test a model of how CPA helps MNEs compete in sub-Saharan African markets.

The exploratory study

The purpose of the exploratory study was to investigate the extent of MNE subsidiary CPA in sub-Saharan African countries. To unearth the commonalities and differences in MNE subsidiary CPAs, and allow for greater coverage and contextual relevance, we used a multiple case study approach. We used the following criteria to select our cases: (1) subsidiary longevity of operations; (2) experience of institutional failures and (3) willingness of its senior executives to provide information. In line with best practice, we identified relevant informants using a range of approaches which have been found to be generally effective when conducting research in weak institutional environments. These included industry contacts, research networks, direct referral, and snowballing (see Bartholomew & Smith, 2006; Tayeb, 1994). We used these methods to identify senior executives of MNE subsidiaries in the beverage, electronic appliance, and pharmaceutical industries. The networks and contacts of these senior executives were then used to access national government politicians, local chieftains, and religious leaders. Ultimately, 23 senior executives from six MNE subsidiaries participated in the exploratory study: two in beverages, two in electronic and electrical appliances, and two in pharmaceuticals. In addition, we interviewed five national government officials directly involved in detecting trade in counterfeit and pirated goods and enforcing laws against it. We also interviewed five local chieftains and four religious leaders in jurisdictions where there has been significant illicit trading. One team of co-authors transcribed the interviews while another team double-checked the transcriptions to ensure that all statements and sentiments were accurately captured. A summary of the profile of the interviewees used in the exploratory study is available in Appendix B.

Qualitative data analysis

The qualitative data obtained from the exploratory study were analyzed to ensure replicability and robustness. We followed the analytical approach advanced by Gioia, Corley and Hamilton (2013), which structures data into three components: first-order, second-order, and aggregated themes. The approaches we adopted are reflected in our use of initial coding via an open coding process, i.e., first-order, where the coding units captured a range of political initiatives and activities, political resource development, and strategies. Then, using axial coding (Strauss & Corbin, 1990), we integrated the first-order into second-order themes, with greater focus and insights (see Fleming, Collins, Bardwell, Fowler, Boyd, Small, & McNeil, 2020). These steps enabled us to gain insights about major underlying issues from interviews and the archival data collected. Subsequently, we developed aggregated themes to shed light on the different types of political activities developed by the subsidiaries (Gioia, 2021; Gioia et al., 2013). For example, we combined interviewee comments that “institutional conditions relating to high transaction costs and risks to investments in SSA elevated the concerns among subsidiary managers” and “lack of certainty about law enforcement mechanisms coupled with weak protection for the business operations and strategic assets” and “weakening institutional conditions in SSA are creating a hostile environment for businesses” into the theme; “institutional weaknesses inhibits smooth functioning of markets, and undermine subsidiary ability to create and sustain economic value in SSA”. Additional information on the themes that were extracted from the data is available in Appendix C.

The Survey and Archival Studies

From the themes, constructs, and relationships that emerged from the exploratory study, we developed a multi-time and multiple source research design by combining survey and archival data from MNE subsidiaries operating in 23 Anglophone sub-Saharan countries. We studied the relationships between four constructs: subsidiary experience with dysfunctional market conditions; subsidiary CPA; subsidiary sales growth; and subsidiary profit growth. Following Wiklund and Shepherd (2003) we minimized potential reverse causality (Van der Stede, 2014) by setting a 3-year time lag between the independent and moderator variables – corporate political activities and dysfunctional market conditions – and the dependent variables – subsidiary sales growth and subsidiary profit growth. Additionally, the time lag was also important because the performance effects of the CPAs studied might take time to materialize. Furthermore, having multiple respondents per firm helped reduce singularity bias and common method variance concerns (Krause, Luzzini, & Lawson, 2018; Montabon, Daugherty, & Chen, 2018).

We were not able to find a comprehensive database of MNE subsidiaries in sub-Saharan Africa. We relied instead on five sources to create a list of MNE subsidiaries in the 23 Anglophone sub-Saharan countries: (1) the African Report which publishes data on the top 500 companies in sub-Saharan Africa; (2) commercial sections of embassies and high commissions; (3) chambers of commerce; (4) industry associations; and (5) publicly available databases where available (e.g., MNC monitor in South Africa). We focused on Anglophone sub-Saharan Africa to ensure linguistic equivalence of responses. We relied on country co-investigators and research associates to generate a list of subsidiary firms in each sub-Saharan African country (see Appendix D for information on the roles played by the country co-investigators). We treated each MNE subsidiary as a unit of analysis. We ended up with a sampling frame of 3300 subsidiaries with full contact details for senior executives such as country director, CEO, managing director, finance director, chief operations officer, and stakeholder liaison manager. With endorsement letters from a local university and the public relations and marketing directorate of national investment promotion authorities, the country co-investigators and the authors contacted the senior executives of all 3300 subsidiaries (by telephone, e-mail, and in-person visit). A total of 970 subsidiaries agreed to participate in the study.

Once having secured agreement from the participating subsidiaries, a two-page survey instrument capturing the independent, moderator, and control variables was administered to the senior executives in phase one (2016), and a one-page survey capturing the sales growth, profit growth, and other financial performance indicators was administered to finance managers and senior accountants in phase two (2019). We asked the senior executives who participated in the phase one survey to provide us with the contact details of finance managers or senior accountants in their respective firms for the phase two survey. The finance managers were asked to provide both perceptual indicators (i.e., the extent to which financial indicators compared with the subsidiaries’ performance objectives for the 2017 and 2018 financial years), and objective financial figures from internal archival records for those years.

The country co-investigators administered in-person the two survey instruments to all 970 MNE subsidiaries. To enhance participation, each subsidiary was promised an executive summary of the study and a 1-day virtual workshop on corporate political network management. Eventually, 637 valid responses were received for the phase one survey, representing a 66% response rate, which is consistent with previous studies that relied on in-person surveys in developing-economy countries (e.g., Acquaah (2007) = 53%, Amoako-Gyampah, Boakye, Famiyeh and Adaku (2020) = 59.6%, and Zhu, Krikke and Caniëls (2018) = 81.8%). Additionally, we received 604 matched valid responses for the phase two survey. To establish causality and minimize common method bias, we used the 2018 objective financial performance data provided by the finance managers to create the dependent variables, and the 2016 data provided by the senior executives for the independent and moderator variables. Table 1 shows the countries where the subsidiaries on which we obtained data are located. It also shows that the country distribution of our sample is representative of its economic size as measured by its gross domestic products (World Bank, 2021). Table 2 shows that 30% of the MNE subsidiaries operate in the computer industry (including software design, computer hardware, networking infrastructure, component manufacturing and information technology services), 28% in the pharmaceutical industry, 26% in fast-moving consumer goods (FMCGs), and 16% in hospitality industry.

Table 1.

Number of subsidiaries studied by country

No.	Country	Number of subsidiaries	Percent	Country GDP (thousands of USD) = 2020
1.	Nigeria	119	19.7	440,776,971.54
2.	South Africa	84	13.9	419,946,428.13
3.	Kenya	67	11.1	110,347,079.52
4.	Ethiopia	62	10.3	111,271,112.33
5.	Ghana	54	8.9	77,594,279.05
6.	Rwanda	44	7.3	11,070,356.52
7.	Seychelles	17	2.8	1,320,053.79
8.	Tanzania	17	2.8	67,775,101.79
9.	Lesotho	14	2.3	2,518,468.89
10.	Gambia	13	2.2	2,078,070.68
11.	Mauritius	13	2.2	11,156,657.77
12.	Sierra Leone	13	2.2	4,200,380.12
13.	Uganda	12	2.0	40,434,701.52
14.	Botswana	11	1.8	17,613,846.47
15.	Saint Helena	10	1.7	46,319,308.00
16.	Eswatini (Swaziland)	9	1.5	4,941,373.18
17.	Zambia	8	1.3	21,203,059.08
18.	Zimbabwe	8	1.3	26,217,726.72
19.	Liberia	6	1.0	3,486,741.37
20.	Namibia	6	1.0	12,236,250.78
21.	South Sudan	6	1.0	11,997,800.76
22.	Cameroon	6	1.0	45,238,613.48
23.	Malawi	5	0.8	12,626,718.07
Total		604	100.0

Table 2.

Industry breakdown of subsidiaries

Industry of operation	Percentage (%)
Computer	30
Pharmaceutical	28
FMCGs	26
Hospitality	16
Total	100

Measure Development

Predictor variable: Similarity to ideal configuration of corporate political ties

It emerged from the interviews that subsidiary managers engage in extensive networking with major political leaders and that most subsidiaries have a formal political networking and lobbying unit to facilitate political relationships. Accordingly, we drew on the interviews and the existing literature (e.g., Acquaah, 2007) to develop three-dimensional CPA construct to capture the extensiveness and depth of their activities. We asked subsidiary managers to rate the extent to which the questionnaire items accurately described their firms’ activities along three dimensions: (a) ties with national government actors; (b) ties with local chieftains; and (c) ties with religious leaders in sub-Saharan African markets. As shown in Table 3, Cronbach’s alpha for the three CPA dimensions ranged between 0.726 and 0.947. In line with the inter-organizational networking literature (e.g., Kale & Singh, 2007), we validated our CPA constructs by estimating their correlation with a subsidiary manager’s political networking experience (r = 0.440, p = 0.001), and their employment in a division working on political networking (r = 0.420, p < 0.001).

Table 3.

Validity and reliability results

Constructs/indicators, composite reliability (CR), average variance extracted (AVE), Cronbach’s alpha (CA)	Standardized factor loadings	t-values
Environmental dynamism (CR = 0.925; AVE = 0.860; CA = 0.924)
In this industry, technology has changed rapidly over the last 3 years	0.911	Fixed
The industry sector in which our company operates is technology intensivea
In this industry, technology advancements occurred at a fast pace during the last 3 years	0.944	12.948
Business leadership ties (CR = 0.900; AVE = 0.751; CA = 0.898)
Over the past 3 years
Our managers have built many relationships with people in ‘businesses’ (e.g., local businesses, local managers).	0.940	Fixed
Our relationships with people in ‘businesses’ (e.g., local businesses, local managers) are deep.	0.830	26.756
We interacted with our contacts in ‘businesses’ (e.g., local businesses, local managers) very frequently.	0.825	26.365
Institutional void (CR = 0.844; AVE = 0.643; CA = 0.843)
Property rights	0.825	Fixed
Ethics and corruption	0.806	19.052
Influence peddling	0.773	18.687
Host-country dysfunctional market condition (CR = 0.947; AVE = 0.900; CA = 0.947)
Our host sub-Saharan African market is characterized by conditions of
Illegal copying of new products and services by other firms	0.943	Fixed
Counterfeiting of products and trademarks by other firms	0.954	26.516
Consumption of counterfeit and unbranded products and servicesa
National government leadership ties (CR = 0.729; AVE = 0.574; CA = 0.726)
Over the past 3 years
Our managers have built relationships with many people in national governments (e.g., with national government officials, law makers)	0.718	Fixed
Our relationships with people in national governments (e.g., with national government officials, law makers) have been deep	0.795	11.785
We frequently interacted with our contacts in national governmentsa
Local chieftain leadership ties (CR = 0.783; AVE = 0.644; CA = 0.781)
Over the past 3 years
Our managers have built relationships with many local chieftain leaders (e.g., paramount kings, traditional chiefs, sub-chiefs)	0.842	Fixed
Our relationships with local chieftain leaders (e.g., paramount kings, traditional chiefs, sub-chiefs) have been deep	0.761	13.937
We frequently interacted with our contacts in traditional councilsa
Religious leadership ties (CR = 0.947; AVE = 0.900; CA = 0.947)
Over the past 3 years
Our managers have built relationships with many religious leaders (e.g., bishops, pastors, imams, traditional priests)	0.945	Fixed
Our relationships with religious leaders (e.g., bishops, pastors, imams, traditional priests) have been deep	0.952	21.809
We frequently interacted with our contacts in major religious bodies (e.g., churches, mosques)a

^aDeleted due to poor loading.

We constructed our similarity to ideal configuration variable as the degree to which a firm simultaneously attains highest values of ties with national political institutional leadership actors (NPL), with local chieftaincy political institutional leadership actors (LCL), and with religious political institutional leadership actors (REL) (Doty et al., 1993). To do this, we first identified and validated sets of indicators for each corporate political activity. Next, we averaged each set of indicators to capture their theoretical constructs (e.g., De Clercq, Dimov, & Thongpapanl, 2010; Engelen et al., 2015). We then calculated for each subsidiary in the sample the Euclidean distance from the ideal configuration, which is the one where NPL, LCL, and REL take concurrently the highest values. All indicators were rated on a seven-point scale where 7.0 is the highest value (De Clercq et al., 2010). Since large Euclidean distance values from the ideal configuration indicates ‘misfit’, and because we hypothesize that firms that approach the ideal configuration should perform better, particularly in dysfunctional institutional environments, we multiplied the Euclidean distance values by –1.0 to facilitate interpretation (De Clercq et al., 2010).

$$\text{Similarity}\text{to}\text{ideal}\text{configuration}=-\sqrt{{\left(X_{\mathit{ij}}-X_{\mathit{mj}}\right)}^2}$$

where X_ij represents the value of attribute j (NPL, LCL, REL) for firm I and X_mj represents the maximum value for that attribute (i.e., 7.0).

Moderator variable: Dysfunctional market condition

Based on the insights gained from our interviews and on the literature on dysfunctional market conditions and institutional voids (e.g., Cuervo-Cazurra & Genc, 2008; Parente et al., 2019; Li & Atuahene-Gima, 2001; Li & Zhang, 2007; Yaprak, Yosun, & Cetindamar, 2018), we measured dysfunctional market conditions as the extent to which a sub-Saharan African host market is characterized by: (1) illegal copying of new products and services, (2) counterfeiting of products and trademarks by other firms, and (3) consumption of copied and counterfeited products and services.

Outcome variable: MNE subsidiary performance

We followed previous research (e.g., Acquaah, 2012; Julian & Ofori-Dankwa, 2013) and used two objective indicators of subsidiary financial performance: percentage growth in sales and percentage growth in profits. We obtained this archival financial information from the databases of local investment promotion centers, subsidiary annual reports (where available), and internal archival sources (where access was granted in the phase two of the survey studies). For example, all MNE subsidiaries in Ghana are required by law to register with the Ghana Investment Promotion Center and to provide it annually with key financial information. There are similar investment promotion centers in other African countries, for example the Kenya Investment Authority, the Nigerian Investment Promotion Commission, and the Rwanda Investment Promotion Agency, where we obtained data on subsidiary performance for 2017 and 2018.

Control variables

Following the existing literature on MNE subsidiaries (e.g., Nell et al., 2015; Nguyen & Rugman, 2015), we controlled for external environment, industry, and firm-related covariates. Specifically, we controlled for environmental turbulence as it can affect subsidiary performance (Cui, Griffith, & Cavusgil, 2005). Previous research also indicates that institutional void can also influence subsidiary performance (Shi, Sun, Yan, & Zhu, 2017), so we controlled for its effects by asking managers to rate the quality of institutions in the host sub-Saharan African market on a scale of 1 = very poor to 7 = very high: property rights, ethics and corruption, and influence peddling. Prior research also shows that firm performance depends on the industrial sector in which it operates (Grazzi, Jacoby, & Treibich, 2016), as well as on firm size and age (or experience) (Ling, Zhao, & Baron, 2007). The nationality of a subsidiary’s parent firm can also have an impact on a subsidiary’s performance (Jiang, Jung, & Makino, 2020), and so can its total expenditure on political activity, including lobbying (Rajwani & Liedong, 2015). Additionally, some scholars have argued that having a dedicated unit to manage a firm’s inter-organizational relationships (including its political networks), as well as a subsidiary parent’s experience in handling political relationships, increases effectiveness in managing such relationships (Wu, Choi, & Rungtusanatham, 2010). Studies have also shown that ties with other businesses in a host-country market can impact on a subsidiary’s performance (Sheng et al., 2011; Siegel, 2007); hence we controlled for business leadership ties. By controlling for all these covariates, we account for any potential omitted variables that might undermine the stability of our findings.

SURVEY AND ARCHIVAL DATA ANALYSES AND RESULTS

Tests of Reliability and Validity

We followed the procedure recommended by Bagozzi and Yi (2012) to conduct confirmatory factor analysis for all the constructs measured with multiple indicators. We used the maximum likelihood estimation method in Mplus 7.4 to analyze the data. Exact model fit was evaluated using the conventional chi-square (χ²) difference tests. Following the suggestion of Hu and Bentler (1999), and Bagozzi and Yi (2012), we also used a combination of approximate fit heuristics to provide a broader perspective on model fit. Specifically, Bagozzi and Yi (2012) recommend that non-centrality-based measures such as Root Mean Square Error of Approximation (RMSEA), relative fit indices including Non-Normed Fit Index (NNFI), Comparative Fit Index (CFI), and absolute fit index such as Standardized Root Mean Squared Residual (SRMR) should be reported. The psychometric literature suggests that normed chi-square (i.e., χ²/df) should be ideally less than 2.00; RMSEA ≤ 0.07; NNFI ≥ 0.90; CFI ≥ 0.95; and SRMR ≤ 0.07 (Bagozzi & Yi, 2012). Based on these criteria, we obtained excellent fit: χ² = 135.560; df = 83; χ²/df = 1.633; RMSEA = 0.032; NNFI = 0.991; CFI = 0.987; and SRMR = 0.019.

Next, we subjected all constructs to reliability, convergent validity, and discriminant validity tests. The standardized factor loadings for each item are significant at the 1% confidence level, providing support for convergent validity (see Table 3). Further, Table 3 shows that the composite reliability (CR) and average variance extracted (AVE) values for each construct exceed the required benchmarks of 0.70 and 0.60, respectively, confirming the internal consistency and convergent validity of the measures (Bagozzi & Yi, 2012). Using Fornell and Larcker’s (1981) test, we assessed the discriminant validity of the measures to determine whether the AVE for each construct exceeded the highest shared variance of each pair of constructs. This was the case, thus showing that we have discriminant validity.

Hypothesis Testing and Results

Following studies of the activities of MNE subsidiaries in sub-Saharan Africa (e.g., Julian & Ofori-Dankwa, 2013), we assessed the distribution of our variables to determine whether there were significant departures from normality. We found that our archival subsidiary performance variables (sales growth and profit growth), subsidiary size (total full-time employees), subsidiary experience in Africa, percentage of sales spent on political activities, number of full-time subsidiary staff involved in political activities, and length of time executives spent on CPA either peaked or were negatively skewed. Accordingly, we used a natural logarithmic transformation procedure to correct for non-normality. The descriptive statistics and zero-order correlations for the study variables are shown in Table 4. The correlation results indicate that the three relational political resources are related: there is a tendency for MNE subsidiaries to simultaneously network with national government actors and local chieftains (r = 0.466, p = 0.001) (a complementary relationship) while substituting any of these two political activities with networking with religious leaders (r = − 0.256, p = 0.001; r = − 0.308, p = 0.001; respectively) (a substituting relationship).

Table 4.

Descriptive statistics and correlation results

Variables		1	2	3	4	5	6	7	8	9	10
1.	Sales growth	1
2.	Profit growth	0.780	1
3.	Similarity to ideal configuration	0.086	0.030	1
4.	NPL	− 0.049	− 0.075	0.631	1
5.	LCL	− 0.008	− 0.061	0.600	0.466	1
6.	REL	0.163	0.165	0.326	− 0.256	− 0.308	1
7.	Dysfunctional market condition	− 0.118	− 0.083	0.183	0.371	0.380	− 0.349	1
8.	MNE nationalitya	0.048	0.043	− 0.052	− 0.170	− 0.190	0.241	− 0.191	1
9.	Computer	− 0.021	0.000	− 0.055	− 0.076	0.045	− 0.039	− 0.079	0.189	1
10.	Pharmaceutical	− 0.061	− 0.048	0.044	0.117	0.073	− 0.097	0.040	− 0.228	− 0.244	1
11.	FMCGs	0.007	0.035	0.006	− 0.003	− 0.054	0.051	− 0.025	0.035	− 0.078	− 0.053
12.	Subsidiary years in Africa	− 0.105	− 0.081	− 0.037	− 0.102	− 0.082	0.073	− 0.033	0.062	− 0.019	0.058
13.	Subsidiary size	− 0.070	− 0.069	− 0.097	− 0.185	− 0.194	0.155	− 0.072	0.135	− 0.039	− 0.076
14.	Environmental turbulence	0.152	0.144	− 0.089	− 0.097	− 0.173	0.092	− 0.031	0.103	− 0.128	0.009
15.	Hours spent on political activities	0.048	0.065	0.017	− 0.054	0.014	0.058	− 0.057	− 0.017	0.004	− 0.030
16.	Percentage of total sales spent on political activities	0.193	0.263	0.001	− 0.078	− 0.036	0.078	0.036	− 0.034	− 0.129	0.128
17.	Number of people for political activities	0.030	− 0.012	− 0.044	− 0.148	− 0.132	0.155	− 0.124	0.165	0.000	− 0.108
18.	Institutional void	− 0.051	− 0.047	− 0.021	0.005	0.039	− 0.067	0.079	0.087	0.024	− 0.060
19.	Business leadership ties	0.047	0.023	− 0.004	− 0.008	− 0.125	0.124	0.023	− 0.077	− 0.155	0.077
Mean		19.613	18.628	− 4.955	0.000	0.000	0.000	3.894	0.465	0.263	0.142
Standard deviation		22.564	21.684	1.495	1.449	1.500	1.531	1.522	0.499	0.441	0.350

Variables		11	12	13	14	15	16	17	18	19
1.	Sales growth
2.	Profit growth
3.	Similarity to ideal configuration
4.	NPL
5.	LCL
6.	REL
7.	Dysfunctional market condition
8.	MNE nationalitya
9.	Computer
10.	Pharmaceutical
11.	FMCGs	1
12.	Subsidiary years in Africa	0.008	1
13.	Subsidiary size	0.036	0.375	1
14.	Environmental turbulence	0.041	0.096	0.016	1
15.	Hours spent on political activities	− 0.024	0.222	0.271	0.033	1
16.	Percentage of total sales spent on political activities	− 0.010	− 0.065	− 0.063	0.166	0.033	1
17.	Number of people for political activities	0.015	0.277	0.636	− 0.010	0.198	0.003	1
18.	Institutional void	0.021	− 0.102	− 0.044	− 0.181	0.006	0.019	− 0.038	1
19.	Business leadership ties	0.038	0.074	0.077	0.178	0.062	0.086	0.075	− 0.103	1
Mean		0.017	2.957	4.537	4.236	1.710	2.311	2.453	2.618	4.436
Standard deviation		0.128	0.822	2.079	1.482	1.075	1.129	1.649	0.939	1.251

NPLN national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties.

^aDummy variable (MNE from emerging economy = 1, MNE from developed economy = 0).

Correlations with an absolute value larger than 0.08 are significant at the 0.05 level.

We mean-centered the similarity to ideal configuration (SIC) and dysfunctional market environment variables, created an interaction term as a product of the two mean-centered variables, and then evaluated the hypotheses using a moderated regression model that includes the mean-centered variables and their interaction term alongside the control variables (Aguinis, Edwards, & Bradley, 2017). Since our dependent variable, MNE subsidiary performance, is operationalized at two unique performance dimensions, sales growth and profit growth, we estimated the following two models:

$$\begin{array}{cc}& \text{Subsidiary}\text{sales}\text{growth}(\%)={\beta }_0+{\beta }_1\text{MN}+{\beta }_2\text{A}+{\beta }_3\text{B}+{\beta }_4\text{C}\hfill \\ \hfill & +{\beta }_5\text{E}+{\beta }_6\text{S}+{\beta }_7\text{H}+{\beta }_8\text{P}+{\beta }_9\text{N}\hfill \\ \hfill & +{\beta }_{10}\text{T}+{\beta }_{11}\text{U}+{\beta }_{12}\text{V}+{\beta }_{13}\text{SIC}\hfill \\ \hfill & +{\beta }_{14}\text{DMC}+{\beta }_{15}\text{SIC}\ast \text{DMC}+{\epsilon }_1\hfill \\ \hfill & \text{Subsidiary}\text{profit}\text{growth}(\%)={\beta }_0+{\beta }_1\text{MN}+{\beta }_2\text{A}+{\beta }_3\text{B}+{\beta }_4\text{C}\hfill \\ \hfill & +{\beta }_5\text{E}+{\beta }_6\text{S}+{\beta }_7\text{H}+{\beta }_8\text{P}+{\beta }_9\text{N}\hfill \\ \hfill & +{\beta }_{10}\text{T}+{\beta }_{11}\text{U}+{\beta }_{12}\text{V}+{\beta }_{13}\text{SIC}\hfill \\ \hfill & +{\beta }_{14}\text{DMC}+{\beta }_{15}\text{SIC}\ast \text{DMC}+{\epsilon }_1\hfill \\ \hfill \end{array}$$

where β₀ is the regression constant; β_1–15 are the regression coefficients; ε₁ is the error term ; MN = MNE nationality; A = computer industry; B = pharmaceutical industry; C = FMGCs industry; E = subsidiary experience (in years) in Africa; S = subsidiary size; H = hours spent on political activities; P = percentage of total sales spent on political activities; N = number of people involved in political activities; T = environmental turbulence; U = business ties; V = institutional voids; DMC = dysfunctional market condition (mean-centered); SIC = SIC of NPL, LCL, and REL (mean-centered); SIC*DMC = product of the mean-centered forms of DMC and SIC.

The results for the two models are presented in Table 5. They indicate that SIC of NPL, LCL, and REL has a significant positive effect on sales growth (β = 1.719, p = 0.011) but not profit growth (β = 1.025, p = 0.116), in partial support of Hypothesis 1. Additional results reveal that the interaction between SIC and dysfunctional market condition is positively associated with sales growth (β = 1.039, p = 0.004) and profit growth (β = 0.879, p = 0.012), which provides support for Hypothesis 2.

Table 5.

Main results – effects of SIC of NPL, LCL, and REL on subsidiary performance under varying levels of dysfunctional market conditions (DMC)

Independent variables:	Hypothesis	Dependent variables		VIF
		Sales growth	Profit growth
		β (SE) [p value]	β (SE) [p value]
Control effects
MNE nationalitya		0.070 (2.044) [0.973]	− 0.881 (1.980) [0.656]	1.167
Computer		− 0.031 (2.300) [0.989]	1.033 (2.228) [0.643]	1.156
Pharmaceutical		− 3.328 (2.978) [0.264]	− 2.570 (2.885) [0.373]	1.152
FMCGs		− 0.824 (6.964) [0.906]	4.974 (6.746) [0.461]	1.020
Subsidiary years in Africa		− 3.230 (1.244) [0.010]	− 2.284 (1.205) [0.059]	1.244
Subsidiary size		− 1.770 (0.640) [0.006]	− 1.538 (0.620) [0.013]	1.882
Hours spent on political activities		1.388 (0.922) [0.133]	1.875 (0.893) [0.036]	1.114
Percentage of total sales spent on political activities		3.074 (0.876) [0.001]	4.567 (0.848) [0.000]	1.089
Number of people involved in political activities		1.774 (0.755) [0.019]	0.870 (0.732) [0.235]	1.735
Environmental turbulence		1.723 (0.709) [0.015]	1.326 (0.687) [0.054]	1.180
Institutional void		− 0.732 (1.059) [0.490]	− 0.847 (1.026) [0.410]	1.070
Business leadership ties		0.227 (0.814) [0.781]	− 0.294 (0.789) [0.709]	1.100
Main effects
Similarity to ideal configuration (SIC)	H1: +	1.719 (0.672) [0.011]	1.025 (0.651) [0.116]	1.125
Dysfunctional market condition (DMC)		− 1.775 (0.668) [0.008]	− 1.375 (0.647) [0.034]	1.115
Moderating effect:
SIC × DMC	H2: +	1.039 (0.359) [0.004]	0.879 (0.348) [0.012]	1.095
R2		0.123	0.129
F		4.721	5.006
p value		0.001	0.001

NPL national governmental leadership ties, LCL local chieftain political ties, REL religious leadership ties.

^aDummy variable (MNE from emerging economy = 1, MNE from developed economy = 0).

β = Unstandardized estimates.

SE = Standard error.

Additional and Robustness Analyses

We undertook several additional analyses to determine the specific level of interaction between SIC and dysfunctional market condition and its impact on sales and profit growth. First, we used the Johnson–Neyman (J–N) technique in Hayes’ PROCESS to explore the magnitude and direction of associations between the predictor and the outcome variables for each level of market dysfunctionality (Hayes, 2018). This was necessary as the moderated regression analysis reported earlier does not help to determine the precise level of dysfunctionality at which SIC affects sales and profit growth. The J–N analysis included all study control variables. The results are graphed in Figure 2a and b. With respect to the magnitude of effect, the results show that the paths from SIC to both sales growth and profit growth are strengthened as levels of market dysfunctionality increase in magnitude. Regarding direction of effect, results show that when markets are mildly dysfunctional, SIC tends to have negative effects on both sales and profit growth, and inversely in the case of highly dysfunctional market conditions. Specifically, for the seven-point rating scale used to capture dysfunctional market condition, results show that SIC has an increasingly significant positive effect on sales growth for levels of dysfunctional market condition greater than or equal to 3.7, while for levels of dysfunctional market condition less than or equal to 1.6 and greater than or equal to 4.25, SIC has increasingly negative and positive effects on profit growth, respectively.

Figure 2.

a Effect of SIC of NPL, LCL, and REL on sales growth under varying levels of DMC. Note: 1. NPL national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties. 2. The regression coefficient and associated confidence intervals (95%) are generated using the Johnson–Neyman technique in PROCESS for SPSS. 3. For x ≥ 3.7, all effects are positive and statistically significant. b Effect of SIC of NPL, LCL, and REL on profit growth under varying levels of DMC. Note: 1. NPL national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties. 2. The regression coefficient and associated confidence intervals (95%) are generated using the Johnson–Neyman technique in PROCESS for SPSS. 3. For x ≥ 4.25, all effects are positive and statistically significant.

Second, given the weak and non-significant direct effects of SIC on both sales and profit growth, we estimated a quadratic regression model to explore the extent to which varying levels of SIC are associated with both performance outcomes. Contrary to our prediction of linear relationships, the results shown in Table 6 and Figure 3 indicate that the impact of SIC on profit growth is non-linear and resembles a U-shape (p = 0.100). Thus, both low and high SIC values result in approximately 13.5% and 16% profit growth, respectively, whereas a moderate level of SIC generates about 12.5% profit growth. Although the quadratic results obtained for profit growth are nonsignificant (p = 0.159), the plotted relationship in Figure 3 indicates more of a J-shaped relationship.

Table 6.

Additional analyses results – curvilinear effects of SIC of NPL, LCL, and REL

Independent variables	Dependent variables:		VIF
	Sales growth	Profit growth
Control effects	β (SE) [p value]	β (SE) [p value]
MNE nationalitya	0.584 (2.054) [0.776]	− 0.412 (1.984) [0.836]	1.164
Computer	0.039 (2.314) [0.986]	1.093 (2.236) [0.625]	1.156
Pharmaceutical	− 3.090 (2.996) [0.303]	− 2.412 (2.894) [0.405]	1.151
FMCGs	0.078 (7.017) [0.991]	5.887 (6.778) [0.386]	1.022
Subsidiary years in Africa	− 2.912 (1.249) [0.020]	− 2.000 (1.207) [0.098]	1.238
Subsidiary size	− 1.765 (0.644) [0.006]	− 1.528 (0.622) [0.014]	1.882
Hours spent on political activities	1.466 (0.928) [0.115]	1.951 (0.896) [0.030]	1.115
Percentage of total sales spent on political activities	3.239 (0.879) [0.001]	4.708 (0.850) [0.001]	1.085
Number of people involved in political activities	1.802 (0.760) [0.018]	0.894 (0.734) [0.224]	1.734
Environmental turbulence	1.878 (0.712) [0.009]	1.423 (0.688) [0.039]	1.176
Institutional void	− 0.851 (1.065) [0.425]	− 0.930 (1.029) [0.366]	1.075
Business ties	0.378 (0.821) [0.646]	− 0.143 (0.793) [0.857]	1.105
Dysfunctional market condition	− 1.651 (0.683) [0.016]	− 1.223 (0.660) [0.065]	1.151
Similarity to ideal configuration (SIC)	1.237 (0.681) [0.070]	0.565 (0.658) [0.391]	1.141
SIC squared	0.428 (0.303) [0.159]	0.483 (0.293) [0.100]	1.090
R2	0.112	0.123
F	4.245	4.729
p value	0.001	0.001

NPL national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties.

^aDummy variable (MNE from emerging economy = 1, MNE from developed economy = 0)

β = Unstandardized estimates.

SE = Standard error.

Figure 3.

Curvilinear effects of SIC of NPL, LCL, and REL on MNE subsidiary performance. Note: NPL national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties.

The SIC perspective suggests that subsidiaries that approach the highest levels of each of the three types of relational political resources attain superior performance (Doty et al., 1993). However, this configuration perspective does not reveal the performance effects of other configurations of the predictor variables. To assess the robustness of our results testing Hypothesis 1, we estimated a three-way interaction model to determine how configurations with low (i.e., one standard deviation below the mean), and high (one standard deviation above the mean) levels of the three corporate political activities affect performance. The results shown in Table 7 indicate that the three-way interaction term is positively related to both sales (β = 0.512, p = 0.022) and profit growth (β = 0.429, p = 0.045). Figure 4a and b show the surface plots for the two three-way interaction models. Compared to the other configurations, high levels of all three CPAs are associated with high sales and profit growth.

Table 7.

Robustness results – three-way effects of NPL, LCL, and REL

Independent variables	Dependent variables		VIF
	Sales growth	Profit growth
	β (SE) [p value]	β (SE) [p value]
Main effects
MNE nationalitya	0.119 (2.119) [0.955]	− 1.590 (2.038) [0.436]	1.241
Computer	0.243 (2.325) [0.917]	1.710 (2.236) [0.445]	1.170
Pharmaceutical	− 2.268 (3.009) [0.451]	− 1.781 (2.894) [0.539]	1.165
FMCGs	0.118 (7.018) [0.987]	5.086 (6.749) [0.451]	1.026
Subsidiary years in Africa	− 3.034 (1.253) [0.016]	− 2.042 (1.206) [0.091]	1.250
Subsidiary size	− 1.829 (0.650) [0.005]	− 1.627 (0.625) [0.009]	1.919
Hours spent on political activities	1.466 (0.932) [0.116]	2.026 (0.896) [0.024]	1.128
Percentage of total sales spent on political activities	3.078 (0.887) [0.001]	4.603 (0.853) [0.001]	1.108
Number of people involved in political activities	1.719 (0.763) [0.025]	0.763 (0.734) [0.299]	1.755
Environmental turbulence	2.042 (0.714) [0.004]	1.508 (0.687) [0.029]	1.186
Institutional void	− 0.98291.070) [0.359]	− 0.886 (1.029) [0.389]	1.088
Business leadership ties	0.289 (0.830) [0.728]	− 0.448 (0.799) [0.575]	1.134
Dysfunctional market condition	− 1.260 (0.763) [0.099]	− 0.727 (0.733) [0.322]	1.438
National political leadership ties (NPL)	0.242 (0.829) [0.771]	0.624 (0.797) [0.434]	1.565
Local chieftain leadership ties (LCL)	1.411 (0.845) [0.096]	0.252 (0.813) [0.757]	1.684
Religious leadership ties (REL)	0.940 (0.823) [0.254]	1.341 (0.791) [0.091]	1.606
Interaction effects
NPL × LCL	0.047 (0.476) [0.922]	0.235 (0.458) [0.608]	1.432
NPL × REL	− 0.06390.493) [0.899]	0.169 (0.474) [0.721]	1.875
LCL × REL	− 0.106 (0.470) [0.822]	− 0.55790.452) [0.219]	1.707
NPL × LCL × REL	0.512 (0.222) [0.022]	0.429 (0.214) [0.045]	1.805
R2	0.123	0.142
F	3.516	4.140
p value	0.001	0.001

NPL national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties.

^aDummy variable (MNE from emerging economy = 1, MNE from developed economy = 0).

β = Unstandardized estimates.

SE = Standard error.

Figure 4.

a Three-way interaction effects of non-market relational political resources on sales growth. Notes: 1. NPL national governmental leadership ties, LCL local chieftain leadership ties, REL religious leadership ties. 2. “Low” and “High” values correspond to − 1 standard deviation and + 1 standard deviation values. b Three-way interaction effects of non-market relational political resources on profit growth. Notes: 1. NPL national governmental leadership ties; LCL local chieftain leadership ties; REL religious leadership ties. 2. “Low” and “High” values correspond to − 1 standard deviation and + 1 standard deviation values.

DISCUSSION

Our study draws on insights from Scott’s (1995, 2014) approach to institutions and on the systems perspective of configuration theory to explore the extent to which similarity to an ideal configuration of relational political resources affects MNE growth in sales and profits in dysfunctional market conditions. We tested our hypotheses on both survey and archival data from 604 MNE subsidiaries in 23 Anglophone sub-Saharan African countries where subsidiaries seek to form ties with national government actors, local chieftains, and religious leaders. Our findings show that similarity to an ideal configuration of the three corporate political activity (CPA) dimensions is indeed associated with growth in sales and profit. We further find that the impact on performance of similarity to an ideal configuration on performance is amplified in dysfunctional market conditions. We next outline the theoretical contributions we make, after which we will present the managerial and policy implications of our findings.

Theoretical Contributions

Our findings advance the multinational non-market strategy literature in several ways. First, prior international business and strategy research suggests that non-market strategies take many forms (Bucheli & Salvaj, 2018), and are more valuable in some contexts than in others (Hillman & Wan, 2005). However, there is limited conceptualization of the mechanisms linking relational political resources stemming from CPA to MNE subsidiary performance in weak institutional environments. Notable exceptions are Parente et al. (2019) that look at firm-stakeholder relationships in Central Africa, and Amankwah-Amoah et al. (2022) that consider political networking in sub-Saharan Africa. Still, more research is needed if we are to understand the multiple forms of relational political resources obtained from CPA and their impact on subsidiary performance.

We contribute to the non-market strategy stream (e.g., De Villa et al., 2019; Lawton et al., 2013; Mellahi et al., 2016) with a study drawing on insights from Scott’s (1995, 2014) perspective on institutions to conceptualize three forms of CPA that generate multiple relational political resources. Scott categorizes institutions into regulative, normative, and cultural-cognitive which are argued to be distinct “pillars”. The regulative pillar has to do with the degree of conformity to laws and regulations, the normative one to social norms and values, and the cultural-cognitive to the extent of shared concepts and frameworks. We adopt this perspective and extend non-market strategy research by arguing that to generate a multiplicity of relational political resources the subsidiaries of MNEs must cultivate regulative, normative, and cultural-cognitive ties with national government actors, local chieftains, and religious leaders respectively. We propose, and find empirical support for, the notion that achieving similarity to an ideal configuration of those three relational political resources explains heterogeneity in performance in the weak institutional environment of sub-Saharan Africa (SSA).

Our empirical study of MNE subsidiaries across Anglophone sub-Saharan Africa shows that specific relational political resources may be generated from ties with national government actors, local chieftains, and religious leaders. We find that (1) ties with national government actors can give MNE subsidiaries regulatory resources such as access to information on the likelihood of legal and regulatory changes, can improve the effectiveness of public policy and institutions, and can promote partnerships with regulatory bodies to enforce laws and regulations; (2) ties with local chieftains can lead to preferential access to land resources, confer legitimacy, assure protection of property rights, and bolster enforcement of informal societal norms and values that facilitate the creation and effective functioning of local markets; and (3) ties with religious leaders who by virtue of their moral authority influence producers and consumers, can encourage moral behavior that also supports efficient functioning of local markets.

Further, our findings show that MNE subsidiaries that have ties with national government actors also have ties with local chieftains, a complementary relationship, but in that case, they do not emphasize ties with religious leaders. We find that the closer a subsidiary comes to an ideal configuration of the three political resources obtained from CPA, the higher sales growth (see Table 5). This suggests that consumption is highly influenced by religious convictions. However, we find that the path from SIC to subsidiary profit growth is more complex. Table 5 shows that SIC has a weak association with profit growth at mean levels. We observe a U-shaped relationship at one standard deviation above the mean value of SIC (see Table 6 and Panel B in Figure 3), suggesting that both low and high levels of SIC are associated with profit growth.

One way to explain this curvilinearity is that initially there are costs associated with building political relationships, but subsequently the relationship pays off in the form of less political risk and lower operating and transaction costs, resulting in sales growth, and presumably in profit growth. This finding extends previous research that shows a linear association between corporate political ties and financial performance (e.g., Frynas, Mellahi, & Pigman, 2006; Peng & Luo, 2000; Sun, Mellahi, & Thun, 2010). Our finding of a curvilinear effect of SIC on profit growth supports the argument of Sheng et al. (2011) that ties with national governmental political institutions only affect performance when markets are dysfunctional and demand uncertain. Furthermore, our results clarify Acquaah’s (2012) equivocal findings that ties with elected government officials reduce performance, while those with bureaucrats increase it.

We further extend the multinational non-market strategy literature, adding to previous studies that have looked at MNE subsidiaries in emerging markets (e.g., Puck, Rogers, & Mohr, 2013; Sheng et al., 2011) by showing that the performance benefits of relational political resources are amplified when market conditions are dysfunctional (see Figure 2a and b). This confirms the findings of Sheng et al. (2011) regarding the performance benefits of ties with national government actors in uncertain markets such as China, and extends the existing literature by demonstrating how similarity to an ideal configuration of multiple relational political resources from a broad spectrum of actors enables MNE subsidiaries to grow sales and profits in dysfunctional market conditions.

Managerial Implications

The few non-market strategy studies that include SSA have emphasized the importance of ties as a major determinant of performance (Acquaah, 2012; Parente et al., 2019). We provide guidelines for managers of subsidiaries seeking to configure political strategies to navigate complexity in SSA. We suggest that managers may leverage ties with national government actors to access regulatory resources and build national-level partnerships to cope with institutional weaknesses. It is important to note the presence of strong connections between national government and local political institutions in these countries. Formal national government efficacy depends on the support of local chieftains and religious leaders. Hence, MNE subsidiaries should develop ties with all three political institutions.

For example, our study finds that local chieftains wield significant authority in many African societies and hence are able to greatly influence allocation of natural resources. The authority of the king of Asanteman in the Ashanti region of Ghana, the king of Eswatini in Swaziland, and the traditional ruler of Ugbo Kingdom in Nigeria in some ways surpasses that of elected national government officials. Thus, although ties with national governmental actors generate national-level regulatory resources, traditional local chieftains enforce norms, values, and practices at the local level, and they can be enlisted to strengthen the legitimacy of a subsidiary and provide protection for its investments.

In addition, the close connections between national government, local chieftaincy, and religious institutions in SSA may also explain the need to obtain relational resource from the three political institutions. Importantly, local chieftains play dual roles as traditional political heads of their kingdoms and chiefdoms as well as chief priests of traditional religions. In their priestly role, they are believed to be empowered by their ancestors to protect their people and property (Ndemanu, 2018). Consequently, national government officials often consult them as well as religious leaders when enacting national laws and policies. This is why an ideal configuration of political resources features ties to all three political institutions in this context.

Public-Policy Implications

National and local government policymakers continue to enact policies aimed at encouraging domestic firms to internationalize and MNEs to increase investment inflows to boost domestic economic growth and new job creation (Chen, Paik, & Park, 2010). They offer juicy incentives in the form of tax breaks, lower interest rates, friendly trade policies, deregulation, and grants, loans, and subsidies (Guimón, Chaminade, Maggi, & Salazar-Elena, 2018). Our study shows that an alternative strategy is to encourage subsidiaries to establish ties with the national government itself, local chieftaincies, and religious institutions in order to remedy institutional weaknesses. Research shows that beyond the vacuum that may exist in formal national governmental institutions, such as courts, weaknesses may also exist in informal institutions (Mair et al., 2012; Webb et al., 2020). Mair et al., (2012: 819) show that “voids can result from conflict and contradictions among institutional bits and pieces from local political, community, and religious spheres”. Webb et al., (2020: 505) argue that an “inability of norms, values, and beliefs and their localized representations to facilitate stable, efficient, and effective transactions” can increase operating and transaction costs, and prevent the creation and effective functioning of markets. MNE subsidiaries have relationships with all three types of political institutional actors, albeit with different degrees of closeness, and hence they have the opportunity to partner with these political actors to strengthen their capacity to support the creation and functioning of markets in SSA.

Limitations and Directions for Future Research

Our study’s conclusion is subject to limitations that provide avenues for further research. First, while the multinational non-market strategy literature includes several different strategies, such as CSA and CPA as well as legal and public sensitization strategies, and the potential of these strategies to interact as complements or substitutes (Reimann, Ehrgott, Kaufmann, & Carter, 2012; Sun et al., 2021), our study focuses exclusively on CPA to explain heterogeneity in MNE subsidiary performance. Additional research may draw insights from the conceptual work of Boddewyn and Doh (2011), Doh, Rodrigues, Saka-Helmhout and Makhija (2017), and Webb et al. (2020) to examine the extent to which CPA and other non-market strategies are complementary or substitutive responses to institutional weaknesses. Future studies may also examine the extent to which business group affiliation is a complement or a substitute for non-market strategies in weak institutional environments (Tajeddin & Carney, 2019).

Second, we do not directly observe how dysfunctional markets influence MNE subsidiary sales growth and profit growth. Although senior executives named weaknesses in the institutional environment as a major challenge citing ineffectual laws and listing consequences such as trademark infringement, we are not able to see how illegal copying of new products and services and their counterfeiting affects subsidiary sales and profit growth. Our reliance on the perceptions of senior executives, while a valid approach to capturing such phenomena, is open to bias. A possible solution to this limitation may be for future research to rely on more objective measures of market dysfunction, for example, established international indexes of institutional quality and regulatory quality.

Third, our context is 23 Anglophone SSA countries out of the 46 sub-Saharan countries listed by the United Nations Development Program. Although socio-economic conditions in all 46 are similar, there are important differences between Anglophone and Francophone countries that need to be considered when generalizing our findings. In addition to language, there are differences in the role of public authorities, the concepts and methods used to formulate and implement public policies, and traditional political institutions. Furthermore, inter-country institutional conditions also vary within the 23 countries studied. Studies that take a more in-depth look at inter-country differences across SSA may help broaden our understanding of how ties with a variety of political actors might mitigate specific dysfunctional market conditions.

Fourth, despite the setting of our empirical study in Anglophone sub-Saharan Africa, we believe that our findings may be applicable to other locations having similar socio-political institutions, particularly ones where there are strong connections between formal regulative political institutions and informal normative and cultural-cognitive institutions. One possibility may be Latin America’s Hispanic societies where the Catholic Church has considerable influence, including in the selection of national political and business leaders (Daudelin, 1995). Another possibility is East Asia, particularly mainland China, where the Guanxi system governs ties between businesses and both formal and informal socio-political institutions, and a third is South Asia, especially India, where there is a fusion of business, kinship, and religious ties.

Fifth, although African nations have sought to create a common regulatory framework for businesses operating within the African Union and the recently created African Continental Free Trade Area, the effort is still in its infancy compared to other trade blocs, such as the European Union. Accordingly, African countries still differ in terms of how they regulate competition, stimulate investment, and control lobbying. Consequently, it may be difficult to generalize our findings to Francophone Africa and beyond. Future studies may seek to rectify this shortcoming.

Biographies

Nathaniel Boso

is O.R. Tambo Africa Research Chair of Entrepreneurship and Employability and Professor of International Marketing and Strategy at Kwame Nkrumah University of Science and Technology, Ghana, and Extraordinary Professor at University of Pretoria’s Gordon Institute of Business Science, South Africa. His research interests span international entrepreneurship, marketing, and supply chain strategies, which have appeared in such journals as Journal of World Business, Journal of Business Venturing, Journal of Product Innovation Management, Journal of International Marketing, etc. He sits on the editorial review board of International Marketing Review and Africa Journal of Management as an Associate Editor. Nathaniel earned a PhD from Loughborough University (United Kingdom), and is a citizen of Ghana and the United Kingdom.

Joseph Amankwah-Amoah

is Professor of International Business at the University of Kent, United Kingdom. He has published in journals such as Journal of World Business, British Journal of Management, Industrial and Corporate Change, Business History, International Business Review, Journal of International Management, Management International Review, and IEEE Transactions on Engineering Management. He has published over 150 articles in leading journals. He earned a PhD from Swansea University (United Kingdom), and is a citizen of the United Kingdom.

Dominic Essuman

is a Lecturer (Assistant Professor) in Sustainable Management at the University of Sheffield, United Kingdom. Dominic’s research interests revolve around supply chain strategy, resilience, and sustainability. He has published in major supply chain journals including International Journal of Production Economics, Supply Chain Management: An International Journal, and Journal of Purchasing and Supply Management. Dominic holds a PhD in Logistics and Supply Chain Management from Kwame Nkrumah University of Science and Technology, Ghana, and is a citizen of Ghana.

Oluwaseun E. Olabode

is a Lecturer (Assistant Professor) in Marketing at University of Bradford, United Kingdom. Her research interests span international marketing, disruption innovation, and organizational strategy, which have appeared in leading international scholarly journals such as International Marketing Review and Journal of Business Research. Oluwaseun obtained a PhD in Marketing from University of Leeds (United Kingdom), and is a citizen of Nigeria.

Patience Bruce

is a doctoral researcher at Kwame Nkrumah University of Science and Technology, Ghana. Her research interests revolve around organizational leadership, resilience, and high-performance work systems in low-resource contexts. Her scholarly works have been published in journals such as International Journal of Production Economics and several book chapters. Patience is a citizen of Ghana and the United Kingdom.

Magnus Hultman

is an Associate Professor of Marketing and International Business at Goodman School of Business, Brock University, Canada. He was previously an Associate Professor of Marketing at Leeds University Business School, University of Leeds. His research interests revolve around international marketing strategy, branding, and various aspects of tourism marketing, and his work has appeared in journals such as Industrial Marketing Management, International Business Review, International Marketing Review, Journal of Business Research, Journal of International Marketing, and Tourism Management among others. Magnus earned a PhD in industrial marketing from Luleå University of Technology in Sweden, and is a citizen of Sweden.

James Kofi Kutsoati

is a freelance consultant to several multinational enterprises in the Cocoa sector. He was previously Deputy Chief Executive of Operations at Ghana COCOBOD in Ghana. James holds a PhD in operations strategy from Kwame Nkrumah University of Science and Technology, and is a citizen of Ghana.

Ogechi Adeola

is an Associate Professor of Marketing at Lagos Business School, Pan-Atlantic University, Nigeria, and a Board Member of University of Kigali, Rwanda. Her research interests revolve around marketing, entrepreneurship, tourism and gender in sub-Saharan Africa context. Her research has appeared in journals such as Annals of Tourism Research, Tourism Management, Journal of Business Research, Industrial Marketing Management, International Marketing Review, and Psychology & Marketing. Ogechi earned a Doctor of Business Administration from the University of Manchester (United Kingdom), and is a citizen of Nigeria.

Appendix A

See Table 8.

Table 8.

Studies that have examined non-market strategies using institutional theory arguments

Authors (year)	Data/methodology	MNE non-market strategy construct	Outcome variables	Theoretical perspectives	Key argument	Key findings
Cui and Jiang (2012)	132 Foreign direct investment entries made by Chinese firms during 2000–2006	State ownership	Foreign market ownership structure	Institutional theory	State ownership in a Chinese firm has an impact on FDI ownership decisions.	The effects of home-regulatory, host-regulatory, and host-normative pressures on a firm to choose a joint-ownership structure are stronger when the state-ownership level is high.
Wang, Hong, Kafouros and Wright (2012)	626 Chinese manufacturing firms that invested in 1,390 overseas projects from 2006 to 2007	Degree of state ownership; level of government affiliation	Overseas investment level, overseas investment location, overseas investment type	Institutional theory and resource-based view (RBV)	Government involvement comes with institutional pressures that influence resource allocation, and overcome information, transaction, and resource constraints.	Government involvement has an impact on the level, location (developing versus developed economies), and type of investment undertaken (resource versus market-seeking).
Cuervo-Cazurra et al. (2014)	Conceptual paper	Government ownership	Internationalization and foreign investment	Integration of agency, transaction cost economics, RBV, resource dependence; neo-institutional theories	While overseas investments made by state-owned multinational companies (SOMNCs) may be outside the protection of home governments, some SOMNCs invest in foreign markets to achieve political objectives rather than economic or profitability objectives.	SOMNCs are more likely to select host countries in which their state ownership is perceived to be more legitimate. SOMNCs are more likely to engage in legitimacy-building actions in host countries than private MNCs.
Henisz and Zelner (2005)	Conceptual paper	Political actors, powerful local interest groups	Possibility of institutions to be overturned, altered, or reinterpreted by political actors	Neo-institutional theory and bargaining-power perspective	Policy-making is an ongoing bargaining process that comprises interactions among investors, interest groups, citizens, and political actors who have different interests.	Emergent institutions are more likely to be overturned, altered, and reinterpreted by political actors when institutions become highly divergent from legitimate reference points.
Hillman and Wan (2005)	169 wholly owned MNE subsidiaries and archival data from Western European subsidiaries of US MNEs	Information political strategy; incentive political strategy; constituency-building political strategy	Information political strategy; incentive political strategy; constituency-building political strategy	Institutional theory	MNE subsidiaries will decide on the type of political strategy to use based on the number of years of operation within a country, its size, the nature of the host country, and parent-firm factors such as level of international diversification, global integration, and use of financial incentives.	MNE subsidiaries are subject to institutional duality and need to attain internal and external legitimacy by utilizing three political strategies depending on the number of years of operation, degree of pluralism/corporatism in the host country, and parent-firm factors.
Rizopoulos and Sergakis (2010)	Conceptual	Political leverage in home country; nature and number of MNE links with other actors inside a policy network	Political leverage	Institutional theory	The MNE’s structural position within the policy network determines its home-country political leverage.	MNEs can have various levels of political leverage ranging from negotiated power (open policy network and strong MNE position) and dominant positions (closed policy network and strong MNE position) to autonomy (open policy network and weak MNE position) and predominance of other rival interests (closed policy network and weak MNE position).
Sun et al. (2010)	142 Interviews over 11 years (1980s to 2000s)	Political embeddedness of MNEs in host-country political networks	Output volumes, market performance	Institutional theory	The value of politically embedded relationships in emerging economies can change over time because of the level of change that occurs in these contexts. Political changes can render distribution networks and product markets obsolete, especially if they were controlled by political forces.	MNE’s political embeddedness can facilitate market-performance outcomes and provide competitive advantages in the host country. However, the deeper the level of MNE political embeddedness, the higher the likelihood of cost inefficiency through structural lock-in in liberalized markets.
Wang, Cui, Vu and Feng (2022)	Case study of Chinese state-owned enterprises in Africa	Ownership-based political capital: Political ties and political identity	N/A	Institutional theory	MNEs utilize political ties and political identity as a response mechanism to institutional voids (human capital and industry-standard voids) in host countries as these voids increase the transaction costs of Chinese state-owned enterprises (SOEs) when accessing human capital or implementing industry norms.	The Chinese SOEs leveraged political ties to overcome human-capital voids by securing large projects and high-end markets in Africa (allied fleet organizing strategy) and relying on institutional support from home governments (co-dependent alliance strategy).
Mbalyohere, Lawton, Boojihawon, & Viney, 2017	Case studies of five MNEs in the electricity generation sector	Primary and integrative political capabilities	Integrative political capabilities for advanced institutionalization	Institutional theory, Dunning’s OLI paradigm	MNEs need political capabilities to successfully navigate the changing political environment in emerging markets.	Primary and integrative political capabilities need to be reviewed regularly due to the changing and transformational nature of emerging markets.
Albino-Pimentel, Oetzel, Oh and Poggioli (2021)	FDI investments from US firms between 2003 and 2011	Political capabilities/ influence-based non-market capabilities. Environmental, social and governance capabilities	MNE foreign direct investment	Institutional theory	MNEs can leverage their political, environmental, social, and governance capabilities to enter conflict-affected countries successfully.	MNEs that have strong political capabilities and strong environmental and social ties to governments are not affected by peace agreements and changes in the political environment in countries affected by conflict.

Appendix B

See Table 9.

Table 9.

Profile of interviewees

Number of political institutional actors	Position	Roles pertain to:
National governmental political institutional leaders
NGPL-1	Senior government official	Detecting, regulating, and enforcing laws against counterfeit and pirated goods trading.
NGPL-2	Government official
NGPL-3	Government official and in local government
NGPL4	Senior government official
NGP-5	Senior government official
Local chieftain political institutional leaders
LC-1	Local chieftaincy	Being seen as influential leaders within the communities impacted by dysfunctional market practices.
LC-2	Local chieftaincy
LC-3	Local chieftaincy
LC-4	Local chieftaincy
LC-5	Local chieftaincy
Religious political institutional leaders
RL-1	Religious leader	Being seen as influential moral authority within the communities impacted by dysfunctional market practices.
RL-2	Religious leader
RL-3	Religious leader
RL-4	Religious leader

Number of MNE senior executives (SE)	Position in the firm	Industry work experience (years)	Country of origin and industry	MNE subsidiary ID
SE-1	Executive	30	European pharmaceutical firm in multiple SSA countries.	Company 1
SE-2	Executive	25
SE-3	Executive (Marketing)	30
SE-4	Executive/advisor	14
SE-5	Affairs manager	21
SE-6	Executive	10
SE-7	Executive	12
SE-8	Executive	33
SE-9	Executive	15	European pharmaceutical firm in multiple SSA countries.	Company 2
SE-10	Manager/executive	21
SE-11	Sales manager	29
SE-12	Executive officer	34
SE-13	Sales manager	16
SE-14	Regional manager	23
SE-15	Medicinal research chemist	11
SE-16	Associate affairs associate	12	United States pharmaceutical firm in 20 SSA countries.	Company 3
SE-17	Medical representative	12
SE-18	Sales manager	Not disclosed
SE-19	Senior manager	22	Asian electronic and electrical appliance firm in 10 SSA countries.	Company 4
SE-20	Senior executive	23
SE-21	Senior executive	Not disclosed	United States electronic and electrical appliance company in 20 SSA countries.	Company 5
SE-22	Senior executive	Not disclosed
SE-23	Sales manager	Not disclosed	European Electronics and appliance firm.	Company 6

Appendix C

See Table 10.

Table 10.

Themes that emerged from the data

First-order themes	Second-order themes and synthesis	Aggregated themes
Institutional conditions related to high transaction costs and subsidiaries’ investments in SSA elevated the concerns of subsidiary managers. Weakening institutional conditions in SAA are creating a hostile environment for some businesses. Lack of certainty about law enforcement mechanisms coupled with weak protection for the business operations and strategic assets.	Institutional infrastructure dysfunctions are impacting sub-Saharan African markets and organizational behaviors. Institutional infrastructure dysfunctions are creating additional hurdles for foreign firms.	Host-country institutional dysfunction exerts adverse effects on MNE subsidiary activities.
High legal arbitration costs and delays in adjudication of business disputes. Government red tape and weak applications of laws incentivizing poor/unethical business behavior. Dissimilarities in terms of the regulatory and normative institutions between the home and host country.	Institutional dysfunctions incentivize organizations to seek to connect with bureaucratic government officials and other political actors. Pressure to develop political resources. Inherent difficulties in bridging the institutional distance between home and host countries.	Triggers for CP strategies formulation to promote policy change and enhance MNE subsidiary performance.
Poor outlook of potential to alter the institutional constraints. Potential of policy lobbying and political contributions to alter the policy environment.	Neutralizing the negative aspects of the dysfunction seen as a strategic necessity.	Improving the performance of MNE subsidiary performance in host countries.
Recognizing the potential value that can be accrued from the use of traditional leadership in communities. Local chieftaincy leaders such as paramount kings and traditional chiefs seen to provide access to market opportunities and improve business competitiveness. Interactions with traditional institutions can benefit firms.	A network of traditional leaders can be beneficial to businesses in reaching other stakeholders such as new customers and governments. Grassroots traditional networking via traditional leaders.	Local chieftaincy leadership networking to fill institutional constraints.
Religious groups can provide businesses with access to community networks with greater levels of trust. Religious groupings seen as potential vehicles for mobilizing community and potential customers’ support. Religious leaders such as bishops and pastors’ endorsement can improve the standing of the business.	Cultivating multiple links to religious leaders improves access to explore and/or exploit market opportunities.	Religious leadership networking to fill institutional constraints.

Appendix D: Activities of Country Co-Investigators (CCIs)

Country Co-Investigators (CCIs) were research collaborators who participated in the research project from various countries in the 23 Anglophone SSA countries. CCIs played a significant role in the conduct of the on-site research activities such as data collection in their respective countries, participation in instrument design and data analysis, creation of reports and documents, as well as knowledge mobilization efforts in their countries. CCIs collecting data in a country were viewed as the primary researchers representing that country. We focused on MNE subsidiaries in the central, eastern, southern, and western SSA countries. Northern Africa was excluded because the focus of the study was SSA, and because Northern Africa tends to be classified as part of the Middle Eastern and Northern Africa region.

Specific responsibilities of the CCIs included:

• Review the survey instruments and provide feedback;

• Participate in virtual meetings with the principal investigator’s team to share ideas, raise questions and concerns, and receive updates;

• Find research participants in their country by using approved criteria and procedures for procuring research participants;

• Manage timely data collection in the countries they represent. In most cases, data were collected via in-person surveys;

• Help in the creation of the relevant databases regarding their countries;

• Participate in virtual meetings to exchange ideas on the results of surveys; and

• Provide interpretation of results that pertain to their specific country and to the overall meaningfulness of the research.

Research Associates/Assistants

The activities of the CCIs were supported locally by Research Associates/Assistants who were post-doctoral researchers and PhD students working under the direction of the CCIs. The researchers provided support in communicating with MNE subsidiary senior executives, assisted in overseeing fieldworkers, and supported local CCIs in data cleaning, coding, entry, and report writing. They also supported the CCIs in web-scraping of published articles and industry reports on MNE subsidiaries.