Analysis of Solidarity Mechanisms Affecting the Performance of Ethnic Minority Business Groups in Africa

Abstract

Business groups comprise independently owned firms based on different types of owner solidarity, such as kinship, ethnicity, religion, or political identity. However, research has been slow to account for how the adverse effects of ethnic solidarity influence BG-affiliate firm performance. We investigate the interplay of owner ethnicity and their firms’ innovation and export performance. We find variations in affiliates’ performance based on their self-identified ethnicities by analyzing data from the World Bank’s Enterprise Surveys (WBES) across 20 sub-Saharan African countries. Notably, long-established migrant communities, including Indian, Middle Eastern, and European entrepreneurs, experienced waning performance within the BG structure. In contrast, group-affiliated firms led by Chinese entrepreneurs show significant outperformance compared to their African counterparts and minority group affiliates. This study contributes to a novel understanding of the heterogeneous relationship between ethnic solidarity and BG-affiliated firms’ performance across sub-Saharan Africa.

1. Introduction

Scholars have long celebrated ethnic solidarity among entrepreneurial communities as a valuable resource fostering cooperation and trust (Bonacich, 1973; Portes & Sensenbrenner, 1993; Landa, 1994; Chand & Ghorbani, 2011). In the context of business groups (BGs), Granovetter (2010) defines solidarity as an identity built on shared social bonds among group-affiliated firms and their personnel. Previous research has recognized that this shared identity is instrumental in establishing resource mobilization among group affiliates (Christensen-Salem et al., 2021; Khayesi et al., 2014; Lee & Park, 2017), which, in turn, underpins the competitive advantage and superior member firm performance (Ramachandran & Shah, 1999). However, much of this research accentuates the theorized benefits of co-ethnic solidarity while overlooking potential adverse effects (Portes & Sensenbrenner, 1993). While ethnic solidarity fosters cooperation and trust, it can also impose constraints that have received less attention in the literature. Strong intra-group ties may lead to the exclusion of outsiders, reinforce market segmentation, and create rigidities in decision-making. Such embeddedness can limit access to broader business networks, reducing firms’ ability to adapt and compete in diverse markets. In business groups, excessive ethnic homogeneity may also increase risks of nepotism, over-reliance on informal contracts, and reduced innovation due to limited external knowledge flows. By examining both the benefits and drawbacks of ethnic solidarity, this study contributes to a more nuanced understanding of its role in shaping firm performance, particularly in environments characterized by institutional voids.

Scholars identify multiple forms of solidarity prevalent among business groups in various global regions, including family (Masulis et al., 2011), kinship ties (Khanna & Rivkin, 2006), and other axes of solidarity, including caste, religion, regional origin, political affiliation, and even school attendance (Granovetter, 2010). Moreover, much business group research is conducted in a relatively small number of countries, particularly Japan, Korea, Taiwan, and China (Holmes et al., 2018), predominantly ethnically homogeneous nations where ethnic identity is unlikely to be a basis for solidarity. However, in ethnically heterogeneous nations, minority status may be a prominent source of solidarity that engenders a dominant homogeneous class of entrepreneurs (Davis et al., 2000). For example, in Indonesia and Thailand, ethnic Chinese represent 2.5% and 11% of the population and own 70% and 90% of private sector assets, respectively (Carney & Gedajlovic, 2002). Contrarily, in some contexts, ethnic minority entrepreneurs may encounter discrimination and exclusion.

Research in the 1990s on post-colonial sub-Saharan African (SSA) countries acknowledged that entrepreneurial classes segment along ethnic lines because ethnicity is a primary source of identity among BG affiliates and a source of competitive advantage for ethnic minorities (Kennedy, 1988; R. J. Fisman, 2003; Fafchamps, 2004; Biggs et al., 2002). However, recent research on contemporary SSA BGs is sparse (Hearn et al., 2018, 2023; Tajeddin & Carney, 2017, 2021), and interest in minority ethnic performance has diminished. Accordingly, considering both the positive and negative performance attributes of ethnically homogenous BG affiliates, we address their performance consequences in an understudied research context.

A further motivation for the study arises from an influential editorial claiming that the most significant issue for many African firms is the persistence of missing institutions “understood as the absence of market–supporting institutions… and contract enforcement mechanisms” (George et al., 2016, p. 377). In particular, George et al. (2016) call for research to prioritize how African firms (i) handle institutional voids and (ii) develop competitive capabilities. Appropriately, the BG literature states that the prevalence of institutional voids is a significant stimulus for affiliating with a BG.

We address this research agenda by drawing upon complementary insights from transaction costs theory (Khanna & Rivkin, 2006) and the economic sociology concept of social embeddedness (Granovetter, 1985) concerning ethnic inclusion and exclusion (Portes & Sensenbrenner, 1993; Waldinger, 1995; Koning & Verver, 2013) about facilitating exchange in a context of institutional voids in market support (Khanna & Rivkin, 2006). Therefore, a key question guiding this research is whether contemporary business groups (BGs) in sub-Saharan Africa (SSA), comprising ethnic minority affiliates can achieve a competitive advantage in the context of extensive institutional voids.

We test three hypotheses, beginning with a baseline transaction costs hypothesis (Khanna & Rivkin, 2006), proposing that BGs will enjoy a competitive advantage regardless of ethnicity compared with unaffiliated firms. We base our second hypothesis on the economic sociology concept of social embeddedness (Granovetter, 1985), specifically, minority ethnic embeddedness (Portes & Sensenbrenner, 1993). This hypothesis contrasts with the established view that ethnically homogenous minority BGs (Landa, 1994; Davis et al., 2000) enjoy a competitive advantage relative to indigenous African BGs (Kennedy, 1988; Fafchamps, 2004; Biggs & Shah, 2006). To do so, we offer a non-necessity hypothesis (Khanna & Rivkin, 2006) or null hypothesis that ethnically homogenous minority BGs hold neither competitive advantage nor disadvantage relative to their indigenous African-based counterparts.

Third, we develop a ‘China exceptionalism’ hypothesis based on China’s 2006 major foreign policy initiative on overseas trade and cooperation, assisting domestic firms to invest abroad while also developing China’s soft power (Bräutigam & Tang, 2014). The group affiliation process is based upon the state’s selection of private entrepreneurs with whom state-owned lead firms coordinate. We argue that combining China’s foreign policy and a novel, hybrid BG form may be an exceptional basis for competitive advantage. We support our three hypotheses based on WBES data on more than 1800 firms from a multi-country sample from some 20 sub-Saharan African (SSA) countries.

We make three contributions to the BG literature: First, we contribute to the sparse African business group literature with a unique multi-country study of the competitive advantage of group-affiliated entrepreneurs based on self-assessed ethnic identity. Secondly, we offer a complementary explanation to the prevalent institutional void perspective with an economic sociology concept of social embeddedness. This perspective suggests that established entrepreneurs affiliated with ethnically homogenous BGs may encounter the advantages and disadvantages of their embeddedness. However, our Chinese exceptionalism hypothesis suggests that state coordination and entrepreneurial ethnic identity co-determine the affiliation of recent arrivals, promoting a novel BG structure into sub-Saharan Africa. We begin by outlining our research context, theoretical foundations, and hypotheses.

2. Context, Theory, and Hypotheses

2.1. Context

BGs have a long history in sub-Saharan Africa, a region with pervasive institutional voids. A product of Africa’s colonial past, European BGs were a core component of the African economy for over a century. Described as merchant-multinationals (Jones, 2000), investment groups (Chapman, 1985), as well as BGs (Jones & Wale, 1998), the Colonial–era BGs members affiliated with a European-based parent that coordinated a two–way trade between Africa and Europe. These groups, comprising large but loosely connected, or ‘freestanding companies’ (Wilkins, 1988), such as Jardine-Matheson, the United Africa Company, Anglo–American, and De Beers, shared strategic resources, including information and finance. They cooperated to achieve mutual goals, such as government influence and a favorable trade policy. In the post-colonial era, several of these groups repatriated their capital to friendlier jurisdictions, but others continued to operate with different organizational structures and product–market strategies (Goldstein, 2010; Jones, 2000).

In the process of decolonization, communities of migrant traders, a product of the colonial era, emerged as thriving enterprises in sub-Saharan Africa. But they attracted little attention from the colonial authorities. These communities operated as middleman minorities (Bonacich, 1973; Johnson, 2017) in the retail, import–export, and transportation sectors (Kennedy, 1988). The middlemen businesses were typically owned and operated by migrant entrepreneurs from South Asia, Middle Eastern, and European settler communities whose prevalence across the region is well documented (Biggs & Shah, 2006; Fafchamps, 2004; Isaac, 1974). For example, Chua (1998, p. 21) notes, “India’s Gujarati have been prominent or predominant in business enterprises from Fiji to virtually the entire eastern coast of the African continent”. A Lebanese diaspora formed ethnic-based BGs in the West African states of Benin, Gambia, Ghana, Liberia, and Sierra Leone (Davis et al., 2000). In southern Africa, including Mozambique, South Africa, and Zimbabwe, the minority business elite comprised the European diaspora. While some exported their capital following the coming of post-colonial states, most minorities chose to stay on as residents after independence.

Unlike the large, multinational European BGs, these domestic groups have little political influence and affiliate into ethnically homogenous networks. Indeed, when describing entrepreneurs from ethnic migrant communities, Fafchamps (2004) uses the terms networks and business groups interchangeably. Compared with loosely linked European BGs, networks have strong horizontal linkages based on relational contracting (Williamson, 1985). Independent firms establish partnerships by granting trade credit, producing intermediate products, and performing other functions such as warehousing and distribution (Biggs et al., 2002; Biggs & Shah, 2006). Most importantly, network groups informally enforce contracts in a context where legal contract enforcement costs are high (R. Fisman, 2001; R. J. Fisman, 2003). More recently, significant numbers of mainland China-based BGs have entered sub-Saharan Africa (Bräutigam, 2003; Bräutigam & Tang, 2014).

2.2. Business Groups and Competitive Advantage

We begin with a baseline transaction cost-based hypothesis stating the received view of the competitive advantage of emerging market BGs with particular reference to the arguments about African BGs’ competitive advantage. That BGs improve their affiliates’ international competitive advantage is a well-established principle in the literature (Tajeddin & Carney, 2019; Eduardsen et al., 2022). This view suggests institutional voids in factor markets and contract enforcement present entrepreneurs with resource assembly and obstacles to market transactions (Hoskisson et al., 2013).

First, the most acknowledged way BG solidarity can enhance affiliates’ competitive advantage is by reducing the cost of contracting with group members and facilitating resource assembly (Chang & Hong, 2000; Khanna & Palepu, 2010). Research worldwide suggests BGs can form internal markets for three types of resources: credit and equity, skilled executives and scarce skilled labor, and sharing intangible assets such as reputation and information (Carney et al., 2011; Masulis et al., 2011). Consistent with the BG competitive advantage principle, developmental theorists in Africa suggest that credit-based market exchange is costly due to weak public institutions protecting property rights and contracts. African firms avoid contractual default by making either spot market transactions or contracting within BGs (Biggs et al., 2002; Fafchamps, 2000).

Secondly, Khanna and Palepu (2010) suggest that BGs can compensate for the lack of skilled human resources when there are gaps in the labor force training and education. They propose that affiliated BGs deploy the needed personnel from their collective pool of talented workers as necessary. Scholarly research suggests that this is a particularly problematic issue in sub-Saharan African countries; firms encounter shortages of adequately skilled employees due to the limitations of the local educational and vocational training institutions (Kiggundu, 2002).

Third, firms from emerging economies face the additional challenge of ‘catching up’ to the productivity frontier occupied by mature economies. Domestic firms approach this goal by utilizing ‘intermediary technology’ to build their competencies, enabling participation in international activities (Hobday, 1995). In the African context, we expect BGs will progressively improve their ability to provide their affiliates with access to productivity-enhancing mid-range technologies. For example, one study found that BGs promote the adoption of information and communications technologies among SME African group affiliates (Tajeddin & Carney, 2019). Hence, by enabling credit and skilled human resource exchanges and disseminating better technologies and management practices, BGs can elevate their affiliates’ competitive performance (Ramachandran & Shah, 1999). These arguments suggest our baseline hypothesis. From a transaction cost perspective, business groups act as a substitute for missing market institutions by reducing the cost of transactions, enhancing resource assembly, and mitigating risks associated with weak formal institutions (Shin et al., 2022). From a hierarchical governance perspective, business groups create internal markets that reduce uncertainty and opportunism, leading to more efficient operations. This logic directly supports our hypothesis that business group affiliates will have a competitive advantage over unaffiliated firms due to their ability to circumvent market inefficiencies and leverage group-based resources.

Hypothesis 1.

BG affiliates will outperform unaffiliated firms due to their competitive advantages.

2.3. Social Embeddedness in Ethnically Homogeneous Minority BGs

To develop this hypothesis, we contrast the positive view of ethnically homogenous BG (Landa, 1994; Davis et al., 2000) with a nuanced view that identifies potential adverse effects on BG affiliates’ performance. Research in the 1990s established that ethnic minority entrepreneurs in informal networks start larger and grow faster due to their location in information and financial networks (Ramachandran & Shah, 1999). Biggs and Shah (2006) found that minority entrepreneurs establish long-term business relationships in tight ethnically based networks to solve market failures and lack formal institutions. Advantages arise because group member lenders typically have better information about borrowers’ opportunities and can make better credit risk appraisals (Biggs et al., 2002). Trade credit from suppliers within an entrepreneur’s ethnic community is a vital source of financing (R. J. Fisman, 2003). Trade credit promotes firm efficiency as it is negatively related to inventory shortages and positively associated with capacity utilization (R. Fisman, 2001). These groups can exclude affiliates who default on credit payments from future within-group transactions. Thus, affiliates are incentivized to honor their debts (Fafchamps, 2000, 2004). While social embeddedness within ethnically homogeneous business groups fosters trust, reciprocity, and contract enforcement, it can also impose constraints. Portes and Sensenbrenner (1993) highlight that strong in-group solidarity may limit external opportunities, discourage innovation, and create exclusionary barriers that hinder competitive adaptability. Additionally, these groups may experience pressures for conformity and resource-sharing obligations that reduce individual firm autonomy and strategic flexibility (Yenkey, 2018). Over time, such factors may limit growth and make ethnic minority BGs less adaptable to broader market changes, suggesting that their competitive performance may not systematically exceed that of indigenous African BGs.

BG affiliation can also ease access to formal finance, such as banks, which favor granting credit to an affiliate of a reputable group (Fafchamps, 2004).

While some ethnic minority entrepreneurs may encounter discrimination in the broader community (Landa, 1994), ethnic entrepreneurship research identifies several mechanisms that compensate for such marginalization. First, ethnic communities may share certain value imperatives resulting from socialization processes that inculcate an underlying moral order that shapes behavior among co-ethnics, described as value introjection (Portes & Sensenbrenner, 1993, p. 1323). A second mechanism concerns the development of reciprocity among ethnic communities interacting through frequent and recurring exchanges, fostering collaboration to make in-group economic resources available to one another. Doing so enables flexibility through reduced transaction costs. Such groups have communal monitoring and self-enforcing and credible sanctioning capacity described as ‘enforceable trust’ (Portes & Sensenbrenner, 1993).

However, despite the body of research highlighting the advantages of closed ethnic business groups and networks, Portes and Sensenbrenner (1993) stress the adverse effects of within-group solidarity, including member conformity and conservatism, potentially unlimited claims on member goodwill, and levelling, which are group pressures to limit upward social mobility. Such adverse effects can make such groups prone to stagnation due to their exclusion of outsiders (Waldinger, 1995; Yenkey, 2018).

Paradoxically, exclusion costs can arise from entrepreneurial efforts to reduce transaction costs of searching, screening, and verifying trading partners’ reliability. To the extent that members of a particular minority group confine their transactions to one another, they tend to develop a closed network with many redundant links (Burt, 2000). Entrepreneurs in closed groups may readily share information in their networks but ignore or devalue information from outside sources (Yenkey, 2015). Such closed communities can produce several inertial effects on business innovation. One study finds that when new entrants to stable networks are scarce, stable business networks will have ‘lock-in’ effects that “reinforce static patterns of business exchange” (Biggs & Shah, 2006, p. 306). Hence, information about opportunities and better business practices from individuals outside the ethnic group may remain undiscovered (Yenkey, 2015), reducing the flow of stimulus for innovation (Min et al., 2022).

However, recent research on ethnic entrepreneurship suggests that later-generation minority entrepreneurs rely less on the exclusivity of ethnic resources (Koning, 2013; Koning & Verver, 2013; Verver et al., 2020; Verver & Koning, 2024). Due to generational change, more complex and layered identities emerge, enabling entrepreneurs to operate at the intersections of minority and majority ethnic groups, resulting in ‘mixed embeddedness’ (Koning, 2007; Koning & Verver, 2013).

In contrast with the closed network view of minority BG competitive advantage, our second hypothesis is mixed embeddedness (Koning & Verver, 2013). Mixed embeddedness theory challenges the received view that minority entrepreneurs generate competitive advantage from membership in closed networks (Koning & Verver, 2013; Högberg & Mitchell, 2023). This mixed embeddedness perspective suggests that minority entrepreneurship becomes increasingly embedded in ethnic communities and mainstream society. In this view, ethnic entrepreneurs tend to “rely on connections within and beyond their ethnic group … when hiring employees or targeting consumers” (Verver et al., 2020, p. 3). Increasingly, ethnic entrepreneurs in sub-Saharan Africa identify business opportunities and realize new venture creation at the intersection of diverse groups that counter existing intra-group segregation norms (Griffin-EL & Olabisi, 2018). Thus, in the mixed embeddedness view, traditional ethnic identities have become less relevant over the generations, finding greater heterogeneity among entrepreneurs’ identities.

Due to these social processes, we suggest that ethnically homogeneous BGs will unbundle and disappear as constituent members form more diverse relationships in the broader community. Thus, the visible ethnic characteristics of BG affiliation have become less relevant. Therefore, we propose a non-necessity hypothesis (Khanna & Rivkin, 2006), which is the null hypothesis that group-affiliated ethnic minority entrepreneurs will be no different than their indigenous African counterparts. Consistent with the mixed embeddedness perspective, we expect both minority and African entrepreneurs will likely form multiplex and mixed ties based on family, direct or indirect ownership, interlocking directorates, ethnicity, language, religion, and school attendance, but none of these ties are a necessary basis of group affiliation. Khanna and Rivkin (2006) argue that social ties observed by researchers may not form the basis of group affiliation or define the group’s boundaries. Indeed, social ties that enable entrepreneurial cooperation may be invisible to researchers (Cruz & Centeno-Caffarena, 2023). Accordingly, we propose the following:

Hypothesis 2.

Ethnic minority-owned group-affiliated firms will neither underperform nor outperform their indigenous African group affiliates.

2.4. China’s BGs in Africa: Newcomer Exceptionalism

Since 2006, mainland China’s firms have become important foreign direct investors (FDI) in SSA, and in 2013, surpassed the US as SSA’s largest foreign direct investor. Other Asian investment in SSA is minimal. There is some FDI from overseas Chinese communities in Mauritius, and there is fleeting investment from Taiwanese in automobile parts in Nigeria (Bräutigam, 2003). Moreover, Japanese and Korean investment in SSA is limited mainly to financial aid (Cornelissen, 2016; Kwon et al., 2021). Thus, the vast majority of Asian FDI is from mainland China. Asian investment in sub-Saharan Africa has changed substantially with China’s 2006 overseas trade and cooperation policy, assisting domestic firms to invest in Africa while also developing China’s soft power (Bräutigam & Tang, 2014). Investment has intensified following China’s Belt and Road Initiative (Li et al., 2022).

With this hypothesis, we argue that China’s BG affiliates’ competitive advantage will differ from those described in hypothesis two above. BGs controlled by Chinese owners in Africa are newcomers to the region. We reason that their competitive advantage will continue to benefit from the group principle described in hypothesis one. China’s business groups are a novel state-led hybrid comprising state and private-owned firms, and the latter have enjoyed substantial state assistance in internationalizing their operations (Deng, 2009). This hybrid model aligns with the broader concept of social embeddedness, where firms leverage state-backed institutional advantages and ethnic-based solidarity to navigate foreign markets. Unlike traditional BGs in SSA, which are often deeply rooted in local economies, Chinese business groups benefit from state-facilitated market entry while maintaining intra-group coordination through strong ethnic ties. This dual embeddedness enables them to circumvent local institutional voids while reinforcing internal resource-sharing mechanisms. Relying on the ‘ethnic enclave’ perspective, newcomer minority entrepreneurs typically prefer to work with their co-ethnic population to gain socio-economic returns despite antagonism from the indigenous population (Zhang et al., 2016).

While much of the Chinese investment targets the acquisition of commodities, there is also substantial potential for export-oriented manufacturing, promising potential for structural transformation of African manufacturing. Recent research has found that African Chinese-owned firms are increasing their exports to other African international markets (Sun, 2017). Some high-profile state investments are in special economic zones (SEZs), in which African governments provide land and Chinese BGs build transportation, telecommunications, and other infrastructure, in a process described as ‘going global in groups’ (Bräutigam & Tang, 2014). This view recounts the deliberate creation of clusters of export-oriented firms working closely with other group members, which has boosted many African regions’ export capacities. Chinese entrepreneurs’ internationalization process reflects a ‘follow the leader’ strategy, where private entrepreneurs follow in the footsteps of a state-sponsored group leader. This approach further strengthens the role of ethnic solidarity in business group dynamics. Rather than integrating into local economies over time, these firms rely on well-established transnational networks for financial capital, labor, and business intelligence (Zhou et al., 2023). While this strategy ensures high operational efficiency and risk mitigation, it also fosters economic insularity, limiting their ability to cultivate deeper market relationships with non-Chinese firms (Wood & Cooke, 2023).

China’s BGs have established seven special economic zones in Africa since 2006 (Bräutigam & Tang, 2014), and the number is increasing (Feng & Pilling, 2019). For example, The Guandong New South Group operates the Ogun state SEZ in Nigeria. After seven years of operation, the zone has some 50 registered companies, mainly export-oriented businesses, manufacturing products such as ceramics, plastics, furniture, and footwear for African markets. Chinese investment in Africa is not limited to prominent state and private business groups (Feng & Pilling, 2019). The McKinsey management consultancy estimates more than 90% of the 10,000 Chinese businesses in Africa are privately owned (Feng & Pilling, 2019). Some observers believe that independent Chinese entrepreneurs are the driving force behind Africa’s manufacturing restructuring because of their willingness to take risks that Chinese state-owned enterprises will not (Sun, 2017). However, state and private group firms are well resourced. The lead Chinese firm in an African SEZ is required to have a minimum turnover requirement of USD 2 billion. Moreover, the Chinese Ministry of Foreign Trade carefully selects qualified private firms to relocate to the zones and provides subsidies and long-term loans. Some provinces offer additional incentives for local firms to relocate to the zones (Bräutigam & Tang, 2014).

Compared with African-based firms, Chinese BG affiliates typically have better-skilled (imported) labor, plant and equipment, and technical know-how. Indeed, a frequent criticism of Chinese firms is that they bring their skilled labor and capital goods and form few linkages to the local economy (Morrisey, 2012). The selection process for access to these resources suggests that these affiliates will be technically competent and high-performing entities even by the standards set by China’s highly competitive environment (Feng & Pilling, 2019). Newcomer ethnic minorities typically retain their home country citizenship, which enables trade facilitation across borders by taking advantage of border regulations (Welter et al., 2018). However, this strategy also contributes to self-segregation, as Chinese firms often prioritize business dealings within their ethnic networks rather than establishing broader commercial partnerships with indigenous African firms. Unlike Indian and European BGs, which have gradually embedded themselves through interethnic collaborations, new-generation Chinese entrepreneurs maintain a more insular approach, reinforcing their reliance on intra-group trade (Zhou et al., 2023). While this strategy enhances short-term efficiency, it raises questions about long-term sustainability and economic integration within SSA markets (Wood & Cooke, 2023).

Finally, BG-affiliated firms are likely to benefit from non-market advantages. Because the Chinese government supports these ventures, they are likely to exercise influence with senior African decision-makers in their host environments. Hence, our China exceptionalism hypothesis states the following:

Hypothesis 3.

Chinese-owned BG affiliates will outperform their African-based counterparts.

We summarize our theoretical framework in Figure 1.

3. Methods

3.1. Data Source

We obtained the data used in this study from the World Bank Enterprise Survey (WBES, 2015), which employs face-to-face structured interviews with firm representatives, typically top managers or functional managers. The WBES dataset undergoes standardized cleaning and validation processes by the World Bank, ensuring its reliability and reducing the risk of data inconsistencies or biases before it is publicly released.

The WBES comprehensively covers less developed and emerging economies, including firm-level data from 125,000 firms across 139 countries. The World Bank collects enterprise survey data to assess the investment climate and gain insight into firm behavior and performance in these settings. Local World Bank staff administer the survey in person, as one-on-one interviews with firm representatives, usually top managers or functional managers who know their firm’s overall operations.

Using local staff to administer the survey suggests that the interviewer will be familiar with the local language and culture. Given its rigorous approach and resulting reliability, the WBES data are used extensively in economics (Mitton, 2016), Research Policy (Cirera & Muzi, 2020), international business (Cuervo-Cazurra, 2016; Lewandowski et al., 2022), and studies of BG affiliation (Castellacci, 2015; Tajeddin & Carney, 2019; Commander et al., 2024). The data provide information on firm ownership, the owner’s nationality, group affiliation, annual export revenues, technology sophistication, credit access, resource management practices, and information for our control variables. These data allow us to distinguish between ethnic minority and indigenous African ownership.

Our research uses the latest World Bank Enterprise Surveys (WBES, 2015), incorporating firm owner’s ethnicity data. The sample spans from 2006 to 2015 and includes 17,150 firms across 41 sub-Saharan African countries. We examine firms affiliated with BGs to facilitate a meaningful comparison with non-affiliated firms. Consequently, we excluded firms with majority foreign ownership from our analysis. As a result, we reduced our sample size to 8016 firms across 20 countries in sub-Saharan Africa.

Finally, we utilized propensity score matching (PSM) to ensure comparability between BG-affiliated and independent or unaffiliated firms, mitigating potential selection bias among the two groups (Rosenbaum & Rubin, 1983). The PSM approach reduces endogeneity concerns (Dehejia & Wahba, 2002) and obtains consistent results. This approach of using matched sample analysis enabled us to choose a control group (firms affiliated with business groups) that was very similar to the treatment group (unaffiliated firms). It helped us remove any observations from the control and treatment groups that were not comparable, thus ensuring our analysis focused solely on like cases.

To examine the propensity score, we used a Probit model that considers various known factors defining the characteristics of firms affiliated with a business group, such as the number of employees, firm age, and industry. To address this issue, we employed the “psmatch2” STATA-18 command with the Kernel function, optimizing bandwidth to balance bias and efficiency. The Kernel method for PSM balances the treatment (business group-affiliated) and control (non-affiliated) firms while retaining more observations compared to nearest-neighbor matching. Unlike nearest-neighbor matching, Kernel matching uses a weighted average of multiple control firms, reducing variance and improving precision. Additionally, the Kernel method ensures a smoother distribution match, minimizing extreme weight distortions. Thus, this allowed us to compute the propensity score for every firm and indicate the closest matches within the standard support range for unaffiliated firms. The final matched sample consisted of 1803 firms, with 907 unaffiliated firms and 896 firms affiliated with a business group.

We conducted an additional analysis using the “pstest” command in Stata-18 to verify that our sample was well-balanced. The “pstest” function enabled us to perform a T-test to ensure no significant mean differences between groups across all covariates post-matching. A T-test result that is not statistically significant (p > 0.05) indicated a balanced sample. The T-test results were insignificant (p > 0.5), indicating a good balance. Additionally, “pstest” assessed standardized biases, which should be minimal (a bias below 10% is acceptable) for all covariates after matching. Our analysis showed that the bias rates for all covariates were below 10% (approximately 3.9%), confirming that our sample is indeed well balanced.

To ensure a representative sample across different firm sizes, we employed the World Bank’s firm size classification based on the number of permanent workers: small firms (fewer than 50 employees), medium firms (50–100 employees), and large firms (more than 100 employees). Our sample comprises 26% large, 42% medium, and 32% small firms. Notably, small- and medium-sized firms are predominantly in the retail sector, whereas large firms are concentrated in manufacturing. The sample spans 24 industrial sectors, with retail industry firms representing 27% of the sample, food manufacturing 13%, other manufacturing 9%, and hospitality 7%. We used the standard WBES survey questionnaire, which includes structured questions on firm characteristics, ownership, performance, and external constraints.

3.2. Variables and Measurement

Dependent Variable: To assess the competitive advantage of firms, we utilized two indicators: export intensity and product innovation. Much research considers exporting as a leading indicator of a firm’s competitive advantage (Porter, 1990; Crick et al., 2011). We measured competitive advantage using export intensity as a dependent variable. The basis of our export intensity variable is the answer to the WBES question Qd3c: ‘What percent of sales were direct exports?’. The second measure of competitive advantage we considered is product innovation, which is used frequently to indicate a firm’s competitive advantage (Friar, 1995; Kuncoro & Suriani, 2018; Siqueira & Cosh, 2008). The WBES assesses product innovation by asking respondents: “During the last three years, has this establishment introduced new or significantly improved products or services?”. Product innovation is a dummy variable, where Yes is coded as 1 and No as 0 (see

Table 1. Variable definition.

	Variable	Definition	Source
DV.	Export Intensity	Sales exported directly as percentage of total sales.	WBES
	Product Innovation	During the last three years, has this establishment introduced new or significantly improved products or services?	WBES
INV.	Business Group Affiliation	Dummy indicating whether firms are part of larger enterprise.	Calculated from WBES
Mod. V.	Ethnicity	The ethnic identity of the current owner.	WBES
Control Variables	Firm size	Number of permanent workers: small = less than 50, medium = 50–100, large = more than 100 employees.	WBES
	Firm Age	The number of years between the firm’s founding year and the year of its interview.	WBES
	Foreign Own.	Ownership of private foreign individuals, companies or organizations.	WBES
	Managerial Experience	How many years of experience working in this sector does the top manager have?	WBES
	Training	Formal training programs for permanent, FT employees ran in last fiscal.	WBES
	GDP/Export	Exports as a percentage of GDP.	The Global Competitiveness Report (GCR)
	DBR_ (DTF)	Doing Business distant to frontier score.	Ease of Doing Business (World Bank Group, 2015)
	Fragile Index	Fragile States Index: 0–120_higher more fragile.	Fund for Peace Fragile States Index (2015)
	Labor Regulation	Percent of firms choosing labor regulations as their biggest obstacle.	Calculated from WBES
	Political Instability	Percent of firms choosing political instability as their biggest obstacle.	Calculated from WBES

).

Independent Variable: The most common indicator of BG affiliation is a public firm listed on a public stock exchange that is also partly owned by another firm at a specified threshold (Carney et al., 2011), making cross-national BG comparisons difficult. The WBES survey data are ideal for assessing BG affiliation, as they apply a standard definition of group affiliation. These data also meet the criteria for group affiliation identified in the literature, suggesting that affiliates are legally independent companies that maintain a stable relationship with another firm (Castellacci, 2015).

The WBES identifies affiliated firms by the following conditions: enterprises (i) must be registered legally for tax purposes, (ii) are required to make financial decisions independently and produce financial statements for their firm, separate from the groups, (iii) must manage and control their payroll, and (iv), must be owned privately by domestic individuals, companies or organizations. Affiliated firms self-identify as not a ‘firm on its own’ but connected to a larger enterprise. We noted above the conventional use of the term network to describe Africa’s business groups (Fafchamps, 2004). However, the WBES definition closely corresponds to Kennedy’s description of African merchant and trading groups as a “distinctive type of African business venture based on a series of semi-independent branch firms yet linked to a parent enterprise” (Kennedy, 1988, p. 187). By this definition, BG affiliation is somewhat prevalent among African businesses, with some 15% of privately owned sub-Saharan African firms reporting a group affiliation (WBES, 2015).

Moderating Variables. To measure the ethnicity of firm owners, we used the question: ‘What is the nationality of origin of the current largest owner?’ (see question AFB.4a. in WBES for sub-Saharan Africa region), asking informants to select from indigenous African, Indian, Middle Eastern, Asian, European, and Other. In our sample, indigenous African-owned firms comprise 63.1% of our sample, with the remainder being minority entrepreneurs: 19.1% European, 8.1% Indian, 3.2% Middle Eastern, and 6.4% Asian1.

Control Variables: As indicators of the competitive advantage of firms (export and innovation performance), we chose five firm-level control variables that previous studies have identified as related. We use firm age (years since founding) and firm size (total number of permanent workers) to capture demographic characteristics that are predictive of export intensity and product innovation (Chen et al., 2016). Cerrato and Piva (2012) also observed how export intensity is typically higher in firms with greater foreign ownership, so we include a control for foreign ownership. Chen et al. (2016) also highlighted in their research that a higher level of foreign ownership positively affects innovation performance.

Research shows that managerial experience is a vital source in enhancing the direct export performance of firms (Cieślik et al., 2022) while negatively impacting innovation (Daveri & Parisi, 2015). Therefore, we control for accumulated learning/knowledge acquired by managers during their working years in a managerial position. Training employees gives them access to advanced knowledge, which boosts a company’s ability to innovate (Dostie, 2018) and impacts firm export performance (Mubarik et al., 2020).

Furthermore, as factors that influence firms’ competitive advantage, we selected five country-level control variables previously identified by studies as relevant. We expected an elevated export intensity in countries whose exports account for a more significant percentage of GDP. Thus, we controlled for this variable (Cumming et al., 2014). To control for the impact of ‘business regulations and their enforcement’ across 21 economies in our sample on the competitive advantages of firms, we used the average score provided in the World Bank’s Doing Business report.

Moreover, the conditions that threaten the stability of nations may affect a firm’s competitive advantage, so we controlled the Fragile States Index (0–120; higher scores are more fragile) of countries produced by the Fund for Peace to access countries based on social, economic, and political indicators such as cohesion, economic decline, political corruption, and security. We also controlled for political instability and the degree of difficulty in navigating labor regulations in countries, as detailed in the World Bank report. We also control for year and industry-fixed effects.

3.3. Analysis

To test hypothesis 1 (H1), we conducted estimations for Models 1a and 1b, where firms’ export intensity and product innovation were regressed on BGA and various control variables. Given that the dependent variable for product innovation is a dummy variable, we utilized a Probit regression to assess Model 1b. We conducted analyses using Models 2b and 3b to test hypotheses 2 (H2) and 3 (H3). We aimed to explore the moderating effect of minority identity on the relationships between BG affiliation and export intensity and between BG affiliation and product innovation. We estimate the following six equations:

Export intensity = α + β BGA + β Control Variables + industry and time controls + ε	(1)
Product Innovation = α + β BGA + β Control Variables + industry and time controls	(2)
+ ε
Export intensity = α + β BGA + β African + β Indian + β Middle Eastern + β European	(3)
+ β Asian + β Control Variables + industry and time controls + ε
Export intensity = α + β BGA + β African + β Indian + β Middle Eastern + β European	(4)
+ β Asian + β BGA * African +/− β BGA * Indian +/− β BGA * Middle Eastern +/− β
BGA * European + β BGA * Asian + β Control Variables + industry and time controls
+ ε
Product Innovation = α + β BGA + β African + β Indian + β Middle Eastern + β	(5)
European + β Asian + β Control Variables + industry and time controls + ε
Product Innovation = α + β BGA + β African + β Indian + β Middle Eastern + β	(6)
European + β Asian + β BGA * African +/− β BGA * Indian +/− β BGA * Middle
Eastern +/− β BGA * European + β BGA * Asian + β Control Variables + industry and
time controls + ε

where α is the constant, β is the coefficient vector, and ε is the error term. To evaluate the propositions of H1, we used Ordinary Least Squares (OLS) regression for export intensity (Equation (1)) and Probit regression for product innovation (Equation (2)). OLS regression is appropriate for export intensity, a continuous variable, as it provides unbiased and efficient estimates while allowing straightforward interpretation of coefficients. It ensures a clear assessment of how business group (BG) affiliation influences export intensity. We use probit regression for product innovation, a binary outcome (1 = introduced a new product/service, 0 = no innovation). Unlike OLS, Probit accounts for non-linearity and ensures predicted probabilities remain within the 0–1 range. It better captures the threshold-based nature of innovation decisions, where firms innovate only when certain conditions are met. This approach ensures statistical robustness, aligning with the dependent variables’ nature and improving our estimates’ accuracy. To examine the potential impact of our moderator on firms’ export intensity (Equation (4)) and product innovation (Equation (6)) concerning H2 and H3, we utilized OLS regression for export intensity and Probit regression for product innovation. We test our hypotheses as follows: H1 suggests that firms affiliated with business groups in Africa have a competitive advantage over non-affiliated firms, as evidenced by higher export intensity and more significant product innovation.

The equations for H2 indicate that ethnic minority-owned firms affiliated with a BG do not show a competitive advantage in export intensity and product innovation compared to their African-based counterparts. Conversely, H3’s associated equations suggest that Chinese-owned firms affiliated with a BG possess a better competitive advantage in export intensity and product innovation than their African-owned counterparts.

3.4. Results

We report all variable definitions in Table 1 and the sources of all the variables used in our regressions. We present the descriptive statistics in

Table 2. Means and correlation.

Variables	Obs.	Mean	Std. Dev.	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)	(11)	(12)	(13)	(14)
(1) Export	1794	3.629	14.61	1.000
(2) Innovation	1803	0.413	0.492	0.036	1.000
(3) BG	1803	0.497	0.5	−0.013	0.265	1.000
(4) Ethnicity	1803	2.101	1.622	0.073	−0.168	0.033	1.000
(5) Firm Age	1803	19.53	16.91	0.094	−0.052	−0.162	0.139	1.000
(6) Firm Size	1803	1.95	0.757	0.169	−0.091	−0.295	0.107	0.179	1.000
(7) Foreign Own	1780	4.113	14.36	0.050	−0.039	−0.147	0.089	0.039	0.191	1.000
(8) Mang. Exper.	1757	15.60	10.17	0.083	−0.059	−0.197	0.110	0.330	0.152	0.059	1.000
(9) Training	1678	0.399	0.49	0.098	−0.042	−0.073	0.071	0.091	0.256	0.093	0.093	1.000
(10) GDP/Expo.	1803	31.38	15.11	0.018	0.048	−0.079	0.028	0.020	−0.004	0.093	0.056	0.033	1.000
(11) DBR_DFT	1803	54.30	10.46	0.070	−0.223	−0.068	0.367	0.145	0.083	0.021	0.158	0.170	−0.026	1.000
(12) Fragile	1803	86.53	15.33	−0.063	0.179	0.091	−0.361	−0.164	−0.070	−0.058	−0.168	−0.146	−0.329	−0.842	1.000
(13) Labor Regulation	1793	3.828	6.514	0.042	0.156	−0.070	−0.094	0.042	0.006	0.051	0.039	−0.003	0.253	−0.329	0.227	1.000
(14) Political Instability	1793	4.026	4.016	0.024	−0.240	−0.117	0.189	0.083	0.124	0.156	0.115	0.166	0.300	0.610	−0.765	−0.279	1.000

, which contains the means, standard deviations, and correlation coefficients: export intensity at the top and political instability at the bottom. There are some possible collinearity issues in the correlation matrix. The Doing Business index of countries and political instability are closely correlated (above 0.2) with innovation performance, firm size is highly correlated (above 0.2) with BGA, and the Doing Business index of countries is highly correlated (above 0.3) with minority identity. Our multicollinearity statistics address this correlation issue; the variance inflation factor (VIF) is less than three for all variables, relieving concerns about multicollinearity.

To test our hypotheses, we estimate the two equations on the export intensity (Model 1a) and product innovation performance (Model 1b) reported in

Table 3. Effect of BGA and interaction effect of BGA and ethnic identity.

	Hypothesis 1		Hypotheses 2 and 3
Variable	Model 1 (OLS) Export	Model 2 (Probit) Innov.	Model 3 (OLS) Export	Model 4 (OLS) Export	Model 5 (Probit) Innov.	Model 6 (Probit) Innov.
Firm Age	0.04 *	0	0.038	0.036	0.001	0.001
Firm Size	2.309 ***	0.037	2.246 ***	2.27 ***	0.025	0.038
Foreign Own	−0.015	0.002	−0.023	−0.016	0.003	0.002
Mang. Exper.	0.045	0.001	0.042	0.047	0	0
Training	1.447 *	−0.064	1.426 *	1.32 *	−0.026	−0.046
GDP/Expo.	0.061	0.006 *	0.071 *	0.064 *	0.006 *	0.007 *
DBR_DFT	0.176 *	0.024 **	0.189 **	0.174 *	0.026	0.026 ***
Fragile	0.15 *	−0.002	0.192 **	0.150 *	0.004 ***	0.001
Labor Regulation	0.119	−0.052 ***	0.204	0.126	−0.044 **	−0.051 ***
Political Instability	−0.056	0.01	−0.08	−0.078	0.007	0.007
BGA	1.285 *	0.927 ***	1.238 *	0.567	0.884 ***	0.915 ***
African			-	-	-	-
Indian			0.465	−0.66	0.401 **	0.277
Middle Eastern			4.74 **	3.20	−0.0170	0.090
European			1.82 *	1.179	−0.116	0.142
Asian			0.058	−5.64 *	0.678 ***	−0.115
BG * African				-		-
BG * Indian				1.620		0.071
BG * Mid. Eastern				3.075		−0.259
BG * European				0.790		−0.597 **
BG * Asian				7.56 **		1.10 ***
Constant	−26.33 *	−2.507 *	−30.41 **	−25.85 *	−3.198 **	−2.948 **
Industry control	Yes	Yes	Yes	Yes	Yes	Yes
Year control	Yes	Yes	Yes	Yes	Yes	Yes
Observations	1597	1584	1597	1597	1584	1584
Adjusted/Pseudo R2	0.113	0.294	0.115	0.141	0.306	0.316
F	6.088		5.69	5.68
Chi-square		630.135	-		656.00	673.51

* p < 0.05, ** p < 0.01, *** p < 0.001.

. Starting with hypothesis 1, the principle of group competitive advantage suggests a positive effect. Thus, the results in Table 3 support hypothesis 1, showing that affiliation with a business group significantly enhances the competitive advantage of affiliated firms over independent ones (DV: export intensity, β = 1.28, p < 0.1; DV: product innovation, β = 0.927, p < 0.001).

We test the remaining hypotheses on the impact of BGA and minority identity on the export intensity Equation (4) and product innovation Equation (6) reported in Table 3. Hypothesis 2 predicts that the positive relationship between BG affiliation and export intensity will show no difference from African BG affiliates when moderated by minority ethnic groups (European, Indian, and Middle Eastern) embedded in African society. At the same time, the signs on the coefficients of the minority groups, including Indian (β = 1.62, p > 0.1), Middle Eastern (β = 3.075, p > 0.1), and European (β = 0.79, p > 0.1) in Equation (4) are positive. All the coefficients are not significant, which supports hypothesis two by indicating that minority ethnic groups do not exhibit any performance difference in terms of export intensity compared to their African-based counterparts. Regarding innovation performance, another measure of competitive advantage, the results indicate no significant differences in innovation between minority-affiliated firms from India (β = 0.071, p > 0.1) and the Middle East (β = − 0.259, p > 0.1) and their African-based counterparts. However, minority-affiliated European firms (β = −0.597, p < 0.05) demonstrate lower product innovation performance than their African-based counterparts. Therefore, our findings support hypothesis 2 for firms affiliated with India and the Middle East and offer partial support for European-affiliated firms (indicating no difference in export intensity between European-affiliated firms and their African-based counterparts, yet revealing underperformance in product innovation).

Hypothesis 3 posits that firms with Asian affiliations will have a more significant competitive advantage (higher export intensity and product innovation) than their African-based counterparts. The results in Table 3 provide robust support for this hypothesis; in Model 2b, the coefficient on the interaction between Asian ethnicity and BG affiliation (Asian BG) is positive and significant (DV: export intensity, β = 7.56, p < 0.05) as well as in Model 3b (DV: product innovation, β = 1.10, p < 0.05). Figure 2 and Figure 3 show the interaction relationship, where Chinese group affiliations show greater export intensity and product innovation performance than their African-based counterparts, respectively.

4. Discussion

Based upon the transactions cost theory BG competitive advantage in an environment of institutional voids (Khanna & Rivkin, 2006), our baseline hypothesis finds support. This finding is novel: a meta-analysis of business group performance (Carney et al., 2011) found that BG had a competitive advantage, as reflected in affiliates’ financial performance, in 17 jurisdictions and negative financial performance in 12 jurisdictions, including Nigeria and South Africa. Thus, our study covering 20 SSA jurisdictions indicates affiliate competitive advantage for BGs across the region. However, we measure performance (competitive advantage) by export intensity and product innovation, and we find a significant difference between affiliates self-identified as particular ethnicities. The following discusses the potential sources of ethnic entrepreneurs’ heterogeneous export and innovation performance levels.

Approximately two decades ago, a series of studies using data collected in the early 1990s by the World Bank-sponsored Regional Programme of Enterprise Development (RPED) found that minority-owned European, Indian, and Middle Eastern entrepreneurs organized into BG networks enjoyed a series of advantages over enterprises owned by their African-based counterparts. These advantages included better access to supplier credit (Biggs & Shah, 2006; Fafchamps, 2000), larger size and faster growth (Ramachandran & Shah, 2007), and greater productivity and capacity utilization (R. Fisman, 2001).

The underlying logic to explain these group advantages is a transaction costs theory of institutional voids that increase transaction costs to independent firms (Fafchamps, 2004). However, using the most recent WBES data (2015), 15 to 20 years later, we found no support for an ethnic minority competitive advantage. Our theoretical perspective suggests that BG’s affiliation has both advantageous and potentially adverse effects on affiliates’ performance. We reason that minority-owned group affiliates enjoyed the theorized advantages of BGs sampled in the 1990s. Still, the BG sample of two decades later suggests that adverse effects of group affiliation may have come to the fore in the form of group closure and competitive stagnation, which has diminished their competitive advantage.

Building on Koning and Verver’s (2013) findings on mixed embeddedness among ethnic Chinese entrepreneurs in Southeast Asia, we propose that similar dynamics may emerge in Africa. However, given the distinct historical, economic, and institutional conditions in Africa, this remains an open question for future research, requiring further empirical investigation to assess whether Chinese entrepreneurs in Africa are transitioning toward greater intersectionality and integration. Such a social change may produce ethnically heterogeneous business networks and business groups. This change is consistent with Bonacich’s (1973) germinal paper on middleman minorities, where she suggests that “Ethnic enclaves may develop among settlers out of convenience of common language, but the community tie tends to be much weaker and more likely to dissipate over time” (p. 586). From a management practice perspective, it matters little what form of solidarity underpins BG functioning so long as it facilitates interfirm trust and group continuity and promotes affiliate competitive advantage. However, dependence on a single axis of solidarity can be a source of vulnerability, and BG affiliates should display caution in doing so. On the contrary, group members relying on multiple forms of solidarity are likely to be more robust among late-generation ethnic entrepreneurs (Cruz & Centeno-Caffarena, 2023). Thus, we describe hypothesis 2 as a ‘non-necessity’ phenomenon (Khanna & Rivkin, 2006), suggesting that visible entrepreneurial characteristics, such as ethnicity or kinship, may no longer form a basis of group affiliation.

Accordingly, we propose a double movement suggesting that contemporary indigenous African entrepreneurs (McDade & Spring, 2005), assisted by improving institutions, have begun neutralizing minority BGs’ former advantages. Secondly, closed minority groups may stagnate or open up to a broader range of ethnic members. Thus, we propose a null hypothesis suggesting that minority BGs will neither under nor overperform unaffiliated indigenous or independent minority firms.

Third, we find strong support for our Chinese BG exceptionalism hypothesis. Chinese owners of group-affiliated firms significantly outperform independent firms and other BG affiliated with non-Chinese owners regarding export intensity and product innovation performance. Thus, these groups’ export performance is consistent with the transaction cost theory of group closure and internal contracting. Although anecdotal evidence suggests that these businesses’ Chinese owners have strong connections to the political class, other evidence suggests tension between these owners and indigenous African entrepreneurs and employees (Chu & Fafchamps, 2022). Overall, Chinese entrepreneurs in Africa tend to function within closely knit ethnic networks, which provide strong transnational support but also limit their integration into local markets. As Wood and Cooke (2023) highlight, their economic engagement in South Africa remains relatively shallow, constrained by regulatory uncertainties, language barriers, and crime-related risks. Despite a significant influx since the late 1990s, many operate on the economy’s periphery, focusing on cost-sensitive market segments with limited capital access. Unlike Indian and European business groups, which have built deep-rooted commercial networks over time, new-generation Chinese entrepreneurs primarily rely on “Primary Groups”, tight ethnic-based connections facilitating business entry and resource access (Zhou et al., 2023). These ethnic solidarity networks provide strategic advantages by offering financial, labor, and business linkages, distinguishing them from African and other minority entrepreneurs who often depend more on institutional support (Zhou et al., 2023). However, this insularity reinforces self-segregation, further complicating long-term local integration (Zhou et al., 2023). While these strategies ensure initial market success, their impact on long-term sustainability and broader economic contribution remains uncertain. Future research should explore whether evolving socio-economic conditions might compel Chinese entrepreneurs to establish deeper local ties and engage in cross-ethnic business collaborations.

This paper seeks to contribute to the literature on business groups and minority entrepreneurship. With a rare multi-country large-sample survey, we add to our understanding of the BG ownership heterogeneity in sub-Saharan Africa and we theorize potential sources of adverse effects of group affiliation that limit their competitive advantage. In particular, our findings concerning China’s state-owned groups and privately owned affiliates represent novel insights into a hybrid high-performing group structure.

The ethnic foundations of BG affiliation and its effects on international performance are not central in mainstream BG studies. In their literature review, Holmes et al. (2018) find that just four countries (Japan, Korea, Taiwan, and India) dominate the BG literature. Solidarity in the latter three countries is often based on kinship, while Japan’s BGs organize around regional banks. Moreover, ethnicity is relatively homogeneous in Japan, Korea, and Taiwan (Evans, 1995); therefore, ethnicity is unlikely to form the basis of solidarity. However, ethnicity forms the foundation for BG solidarity in various regions, including Southeast Asia, India, and Latin America (Landa, 1994; Davis et al., 2000). Thus, we suggest that research on this type of BG warrants scholarly attention in addition to family and kinship-based solidarity.

To the minority entrepreneurship literature, we contribute to understanding how social relations foster mutual advantage. We have identified research suggesting that closed minority business groups benefit their affiliates, but these advantages appear to decline with later generations. In particular, we emphasize the potentially adverse economic effects of intra-ethnic solidarity. Traditional segregated solidarity can underpin multiple forms of exclusion and market segmentation (Yenkey, 2015). As Fafchamps (2004, p. 305) puts it: “Finding ways of ensuring non-discriminatory markets is thus essential for sustained market–based economic development.” However, with the progression of the generations, the mixed embeddedness perspective points to the evolution of more layered and complex identities. There is also an emerging consensus that African ethnic identities are the subject of ongoing renegotiation (Lynch, 2018). Over the past two decades, Africa’s rapid urbanization has stimulated such a renegotiation. Urbanization is sometimes associated with the intensification of ethnic tensions (Abascal & Baldassarri, 2015). However, urbanization also leads to opportunities for integration, new understandings, and the emergence and amalgamation of new identities.

4.1. Limitations

One limitation of the study is that we rely on export intensity and product innovation to indicate competitive advantage rather than financial performance. Unfortunately, our data do not permit financial performance assessment, such as return on assets or net income. However, the limitation of WBES data is that it is exclusively firm-specific. While the data can show whether a firm is group-affiliated or independent, it does not show whether it belongs to a specific business group. Thus, while we can see whether group-affiliated firms have a competitive advantage, we do not know whether the group’s composition is homogeneous or heterogeneous. Therefore, the interpretation of the findings for hypothesis two relies on inferences based on theoretical insight and the de facto erosion of export and innovation performance. This is a task for future research. Additionally, the WBES for sub-Saharan Africa provides data on the principal owner’s origin only for specific periods, which constrained our data collection. This limitation restricts scholars from capturing recent dynamics in Africa’s entrepreneurial ecosystem, particularly among ethnic entrepreneurs and local business owners, especially during and after the COVID-19 pandemic.

4.2. Recommendations for Further Research and Policy Implications

Our findings highlight important avenues for future research and offer practical insights for policymakers and business leaders. While previous studies emphasized the competitive advantages of minority-owned business groups, our results suggest that these advantages may diminish over time. Future research should explore whether this trend continues, particularly as younger generations of minority entrepreneurs adapt to Africa’s evolving institutional and economic landscape. Understanding how ethnic-based business groups transition toward more diverse, cross-ethnic collaborations could provide deeper insights into changing business dynamics. Further studies could examine whether financial performance indicators, such as return on assets, align with our findings on export intensity and product innovation. Access to firm-level longitudinal data would help determine if the stagnation of minority business groups is a temporary phase or part of a broader structural shift. Institutional reforms in sub-Saharan Africa appear to be neutralizing some of the historical advantages of minority-owned business groups, and further research could explore how these changes influence group performance over time.

Our findings also emphasize the exceptional performance of Chinese business groups in Africa. However, the long-term sustainability of their success remains an open question. Future studies could investigate whether these firms move beyond enclave-based business models and integrate more deeply into local economies. Comparative research across regions, particularly examining ethnic business networks in Southeast Asia and Latin America, could provide broader insights into how hybrid business models evolve in different institutional settings.

From a policy perspective, fostering collaboration between diverse business groups and local enterprises could enhance economic integration. Initiatives that promote cross-group partnerships improve financial accessibility, and clarify regulatory frameworks may help maximize the benefits of ethnic solidarity while reducing market segmentation. Strengthening institutional support for minority and indigenous entrepreneurs could further level the playing field and encourage inclusive economic development. Strategic engagement with Chinese business groups should also be considered to ensure their competitive success translates into broader economic benefits.

For business leaders, adapting to changing institutional dynamics is key. Minority-owned business groups may need to expand beyond traditional ethnic networks by engaging in regional trade agreements and partnerships. Over-reliance on a single form of solidarity, such as ethnicity or kinship, may limit long-term stability. Entrepreneurs who embrace multi-ethnic affiliations and institutional support mechanisms could gain a competitive advantage. In particular, collaboration between African and Chinese business groups through joint ventures and technology-sharing agreements could create more inclusive business ecosystems and enhance long-term economic resilience. Our study underscores the shifting role of ethnic-based business groups in Africa and the growing influence of institutional reforms. Policy makers and entrepreneurs can contribute to more inclusive and competitive regional markets by fostering broader business networks and sustainable collaboration strategies.

5. Conclusions

BGs can cope with Africa’s pervasive institutional voids (George et al., 2016). While extensive literature suggests that ethnic ownership forms the basis of group solidarity in sub-Saharan Africa, we find that group affiliation heterogeneously affects their competitive advantage. Affiliates owned by Indian, Middle Eastern, and European entrepreneurs show no significant difference from their African-based counterparts regarding their competitive advantage. In contrast, we find that group-affiliated firms’ Chinese owners significantly outperform independent firms and other BG affiliated with non-Chinese owners. We suggest that these groups’ appearance and evident competitive advantage are a product of political and economic developments beyond Africa. Whether or not China’s continuing rise and the internationalization of firms from China will have comparable effects in other regions, such as Latin America and Central Asia, also appears to be a fruitful avenue for future research by international business scholars.