Franchise’s locational concentration in Brazil: a contextual perspective Open Access

The Entrepreneurial Ecosystem (EE) approach has emerged and gained traction over the last decade because of its capacity to contextualize entrepreneurial activity within a structural view that emphasizes the contextual features in which entrepreneurs and other agents are embedded (Wurth, Stam, & Spigel, 2022). This concept seeks to understand how entrepreneurial activity occurs, placing greater emphasis on local dynamics than on individuals per se, granting particular attention to Institutional Arrangements and Resource Endowments (Stam & Van de Ven, 2021). Following this rationale, individual characteristics lose centrality in favor of an analytical shift towards interactions between agents and institutions in fostering entrepreneurial activity (Stam, 2015). This perspective is relevant to understanding why certain locations have a greater propensity for business success and growth than others (Brown & Mason, 2017). However, a critical view of the topic discusses the multiple configuration possibilities in terms of ecosystems, leading to similar results and highlighting the absence of a single pattern to be followed for fostering entrepreneurial activity (Schrijvers, Stam, & Bosma, 2023; Alves, Fischer, & Vonortas, 2021).

However, EE studies have suffered from what Donaldson (2021) classifies as “entrepreneurial myopia” that is, an excessive focus on high-tech startups in specific industries. This generates a neglect of other cases of productive entrepreneurship, as defined in Baumol (1990), a situation in which a broader notion of the interplay between EE elements and businesses with socioeconomic relevance remains largely uncharted (Bennett, Vedula, & Araki, 2025). In this vein, recent literature has increasingly called for a broader view of entrepreneurship when assessing EE dynamics (O’Connor & Audretsch, 2023). Following these arguments, our entrepreneurial focus in this study is oriented towards franchises by considering their economic relevance and innovative potential (Kaufmann & Dant, 1999). The understanding of franchising as an entrepreneurial activity has been increasing recently (Watson, Dada, López-Fernández, & Perrigot, 2020). This arrangement materializes in the relationship between two types of entrepreneurs: franchisors – those who develop an idea and a model – and franchisees – those who locally exploit the developed idea.

Although the adoption of franchise business model indicates a high degree of standardization and replicability of processes and products (Fadairo, Kaswengi, Lanchimba, & Bitti, 2020), the replication of business models in different sites involves high complexity, especially when considering environmental influences on locational strategies. In this sense, the high geographical dispersion of network outlets is characteristic of franchising in Brazil. Although socio-economic factors are more visible, strong differences in institutional aspects and the local business environment are also recognized (Melo, Delgado, Corrêa, & Borini, 2020), which impact decisions and the results of spatially dispersed operations such as franchise networks.

Beyond the scarcity of studies adopting an EE perspective in a franchising context, this study addresses a key gap: understanding how local contextual factors influence the performance of franchise networks, particularly in emerging economies (Bui, Jambulingam, & Amin, 2022). In such contexts, external dynamics are often more volatile than those in mature economies, affecting the operation of EEs operate (Alves et al., 2021; Cao & Shi, 2021). Based on these considerations, this study aims to answer the following research question: How do the vectors of local EEs influence the locational concentration of franchise units in Brazilian cities? Accordingly, our research objective was to explore the relationship between the vectors of local Entrepreneurial Ecosystems and the locational concentration of these units in Brazilian cities.

This study proposes a longitudinal perspective. We analyzed the temporal scope within the context of the economic turbulence experienced by Brazil over the last 10 years. During this period, the recession observed between 2014 and 2016 stood out, resulting in two years of GDP decline (almost 7% in the period), as well as a decline of over 4% observed in 2020 due to the new coronavirus pandemic. Brazilian franchising was affected by this turbulence. Data from the Brazilian Franchising Association (ABF) indicate a decline in the number of sectors, such as the number of franchise networks, falling by almost 7.5% between 2015 and 2017 (Associação Brasileira de Franchising – ABF, 2017). This decline is even greater in 2020, with a reduction of 8.6% compared to the previous year (Associação Brasileira de Franchising – ABF, 2022). However, these same ABF reports indicate a rapid recovery (between 2017 and 2019, and from 2021 onwards), indicating that some sectors seem to have suffered less over the period. In an exploratory inquiry, we analyzed the locational concentration of franchise networks at the city-level in Brazil (Fischer, Meissner, Vonortas, & Guerrero, 2022).

Our results reveal a certain heterogeneity from both the locational perspective (cities where the franchise networks operate) and from the perspective of the networks themselves. Additionally, we found a significant relationship between the analyzed dimensions of EE and the concentration of local entrepreneurial activity. These results help us to understand the vectors of EE as elements related to the performance of franchise networks in Brazil. However, when considering the specific EE drivers of franchise concentration, we perceive a certain level of detachment between the concentration of entrepreneurial businesses and typical EE components, a situation that largely validates concerns about developing countries’ fragilities in terms of resource endowments and the existence of severe institutional voids (Cao & Shi, 2021).

The contributions of this study are threefold. First, there is no record of previous work going into this level of detail regarding the relationship between EE contextual factors and the locational concentration of franchise network units. Second, we contribute to the debate about governance arrangements in the face of crises from the perspective of entrepreneurship and the business environment surrounding franchised network units. Third, our main contribution extends to the discussion of EE by incorporating franchising arrangements into the analysis, thus broadening the understanding of the dynamics and challenges faced by these relevant actors in the Brazilian economic context spread across the national territory.

The proposition of franchising as an entrepreneurial practice has been a recent development. Initially, the high levels of standardization and replicability typical of franchise networks made the franchisor-franchisee relationship resemble an employment relationship more than a composition between two legally independent firms. However, the increasing presence of multi-unit or multi-brand franchisees (Grünhagen, González-Díaz, Hussain, & Monteiro da Silva Filho, 2022) and the need for more capable local managers to handle greater operational complexity have made franchisees with entrepreneurial profiles increasingly desirable (Watson et al., 2020).

Entrepreneurship and franchising models can enjoy interesting synergies. A franchising arrangement allows local entrepreneurs to benefit from both a well-known brand and a tested business model, which could reduce the risks associated with starting a new business (Gillis & Castrogiovanni, 2012). Beyond the effort towards an operation in which they are residual claimants (Sharma, Patel, & Pandey, 2021), franchisees contribute local knowledge and a more immediate response to environmental changes.

The logic discussed above aligns with works such as Dada (2016), who proposed characterizing franchise networks as a coalition of entrepreneurs. However, this proposal has so far focused on the internal aspects of the franchisor-franchisee relationship. Additional contributions have been obtained by analyzing the environment around franchise network units, not only from a market perspective but also in terms of how the observed dynamics and characteristics in these locations affect the development of entrepreneurial activity. Melo, Borini, Isaac, and Correa (2023) contribute to the debate by highlighting the role of the institutional context in the presence or absence of franchise networks in small- and medium-sized Brazilian cities. A potential contribution within the field developed in this work refers to the interpretation of the presence of franchise networks in municipalities through EE logic.

The concept of EE is based on one possible interpretation of the broader theme of ecosystems in entrepreneurship literature (Flechas, Takahashi, & de Figueiredo, 2022). The core of EE discussions lies in understanding how the internal attributes of these ecosystems are configured to enable the development of entrepreneurial activity in a specific location as well as how local entrepreneurs access and exploit available resources (Spigel & Harrison, 2018).

When analyzed in the context of EE, entrepreneurial activity is related to orchestration among actors, thus promoting the sharing of knowledge between organizations (Bonomi Santos, Fischer, Cherubini Alves, & Roundy, 2025). According to the literature, this process involves cooperation, complementarity, and interdependence, ultimately fostering entrepreneurial activity and allowing value co-creation. This perspective aligns with the view of O'Connor and Audretsch (2023), who define entrepreneurial activity as an endogenous aspect within EE, although susceptible to changes triggered by both external and internal shocks.

A deeper understanding of these elements allows the view of EE as an open system influenced by external conditions (Wurth et al., 2022). Since EE is a unique structure, each with its own peculiarities, characteristics, and idiosyncrasies (Brown & Mason, 2017), we understand that part of the discussion on EE reinforces the idea that there is no single model of entrepreneurship to follow, which would allow the understanding of entrepreneurial activity as diverse and multifaceted (Alves et al., 2021; Audretsch, 2021).

Accordingly, literature has increasingly recognized that the relationship between geographical aspects and entrepreneurial activity is complex, involving various possible configurations of actors and factors (Bennett et al., 2025; Schrijvers et al., 2023; Alves et al., 2021). This means that empirical evolution in the EE field has increasingly embraced the notion of causal complexity, a feature that diminishes the sense of validity of predetermined, isomorphic views on how EE is configured across space and time (Roundy, Bradshaw, & Brockman, 2018; Brown & Mason, 2017). As outlined in the introduction, these features seem to be particularly relevant in the case of developing countries since these institutional and macroeconomic environments can nurture entrepreneurial activity even in the face of myriad socioeconomic maladies and voids (Cao & Shi, 2021).

This heterogeneous view of EE has fueled discussions on the elements that constitute such arrangements. Stam and Van de Ven (2021) proposed a framework that includes two sets of elements: (1) resource endowment and (2) institutional arrangements. The first set comprises physical infrastructure, finance, leadership, talent, knowledge, intermediaries, and demand, whereas the second set includes formal institutions, culture, and networks. In the authors’ model, the output is defined as productive entrepreneurship, which they describe as: “Any entrepreneurial activity that contributes (in)directly to the net output of the economy or to the capacity to produce additional output” (Stam & Van de Ven, 2021, p. 814). This view aligns with the less restrictive understanding of entrepreneurial activity advocated by Audretsch (2021). In this sense, understanding the configurational heterogeneities of EE paves the way for interpreting the influences that such dynamics exert on the development patterns of the actors involved, including the performance of organizations operating within these ecosystems.

Based on the models adopted in studies addressing the composition and dynamics of EE (Stam & Van de Ven, 2021), we propose our Analytical Model (Figure 1), which encompasses, in addition to the Dynamics of EE and the locational concentration of franchised network units, a theoretical proposition to be validated through empirical observation. In this model, we adopt a simplified view of the connection between core EE elements and their respective direct effects on entrepreneurial activity, following the canonical literature. Our theoretical proposition is as follows:

Theoretical Proposition: The dynamics of Entrepreneurial Ecosystems affect the locational concentration of franchised networks.

With the goal of assessing the connection between EE dimensions and the locational concentration of franchised networks, our simplified model combined elements attached to Demand, Knowledge, Networks, Leadership, and Talent. Although these elements do not cover all typical EE dimensions (e.g. Stam, 2015; Stam & Van de Ven, 2021), they offer a workable structure with good coverage of EE processes associated with resource endowments and institutional arrangements.

Here, we follow the logic originally outlined in Stam (2015) and refined by Stam and Van de Ven (2021). In this case, demand conditions create an environment conducive to absorbing value propositions from new businesses, thus generating a market for innovation (Fischer, Queiroz, & Vonortas, 2018). The concentration of knowledge sources in the region also stands as a critical feature in defining the aggregate capabilities of EE to nurture the emergence of new businesses, a situation that has been particularly associated with the dynamics of knowledge spillovers (Acs, Audretsch, & Lehmann, 2013). In turn, networks represent a key condition for agents and organizations to establish productive interactions, an element that lies at the core of the very notion of EE (Spigel & Harrison, 2018). Leadership conveys the idea that places require some degree of guidance or orchestration to build coherence among agents, thus fostering the densification of connections and the emergence of shared views about EE-level development goals (Bonomi Santos et al., 2025). Finally, talent comprises the existence of human capital, typically understood as a quintessential pillar of any kind of productive entrepreneurship (Fritsch, Obschonka, & Wyrwich, 2019).

Notwithstanding, we abide by previous literature and encompass the notions of causal complexity (e.g. Alves et al., 2021) and the view of EE as Complex Adaptive Systems (Roundy et al., 2018), meaning that these direct associations may fail to underscore the extent to which EE logics apply when looking at a phenomenon that looks into (1) a different strand of entrepreneurship than what has been observed previously in the literature (a condition explored by Bennett et al., 2025 in the United States), and (2) a macro environment that is unlikely to resemble the socioeconomic dynamics observed in developed countries (Cao & Shi, 2021), that is, the case of franchise spatial distribution across Brazilian cities. Hence, we build a novel picture to unveil how and to what extent EE components affect the emergence of a singular type of productive entrepreneurship within the reality of a developing economy.

The first step involved establishing a sample of franchise chains operating in Brazil. We selected a set of firms associated with the Brazilian Franchising Association (ABF) for the analysis. We included franchised chains that in 2016 had clear information on the location of their units nationwide on their institutional websites. The sample included 239 franchized networks. Owing to the absence of information, it was not possible to classify the percentage of company-owned and franchised units in the sample of the chains analyzed. At the time of sample selection, this number represents approximately 7.9% of the networks operating in Brazil. For each chain, the municipalities where the network operated and the number of units per location were tabulated. For this selected group, information was collected for the years 2019 and 2022.

Information from the SEBRAE Local Development Index (ISDEL) was used to assess EE dynamics at the municipal level. The dimensions presented in the index (Competitive Insertion, Productive Organization, Business Fabric, Governance for Development, and Entrepreneurial Capital) can be interpreted as proxies for the elements proposed in the model by Stam and Van de Ven (2021), Demand, Knowledge, Networks, Leadership, and Talent. The indicators presented in the ISDEL are formulated based on the evaluation of official sources and comprise an analysis of 106 variables grouped into five attributes. The index values were normalized to values between 0 and 1 using min-max procedures.

The variables related to Population and Municipal GDP include control elements aimed at estimating econometric models because there is an expectation that the size of the socioeconomic system at the municipal level will directly impact the number of franchise units in the territory. Additionally, based on the premise that Brazil is a country of continental dimensions with significant regional asymmetries in economic behavior dynamics, binary markers (dummies) are included for each macro-region. The descriptions and sources of each variable are shown in Table 1. Aiming to deepen the understanding of the relationships outlined in the theoretical model, the empirical stage incorporates the level of the subdimensions of the SEBRAE Local Development Index (Column Subdimension of Table 1). This analytical detailing allows us to extract more concrete information regarding the elements that comprise the ecosystem dimensions.

The analytical strategy was developed following the typical estimation structure to identify the association between components of EE and elements related to outcome conditions in terms of the concentration of entrepreneurial activity (Siqueira, Fischer, Bin, & Kickul, 2023; Fischer et al., 2018). Thus, we can describe the basic equation to be estimated as follows:

where $E$ refers to entrepreneurial activity, represented by the variable Total Units. EC, BF, GD, PO, and CI describe the effects of Entrepreneurial Capital, Business Fabric, Governance for Development, Productive Organization, and Competitive Insertion, respectively. The combined term for controls refers to the Population, GDP, and regional dummies to control for region-specific effects. The subscript “i” denotes each territorial unit, while “t” refers to each period. Thus, it is worth noting that the dependent variable is lagged by one year for each ecosystem dimension, thereby allowing for a more accurate approximation of the causal flow between contextual conditions and entrepreneurial activity. Specifically, the franchise unit data cover the years 2016, 2019, and 2022, while the ISDEL dimension data refer to the years 2015, 2018, and 2021. The term “ε” encompasses the residuals of the model. This basic structure is maintained for estimations, including the detailed ISDEL subdimensions.

For the estimation stage, we estimated the models using Random Effects for panel data. We also applied a series of consistency tests to deal with the count data. In this respect, all models were initially estimated as Poisson distributions. We then applied likelihood ratio tests of the alpha parameters to identify the potential effects caused by overdispersion in the dependent variable. We then compared the quality of these estimates based on the AIC and BIC information. These procedures are warranted by visualizing the total unit histogram, which indicates high levels of skewness in the distribution. Given the low number of observations with a zero value in the sample (15.41%), an appropriate technique for negative binomial distributions was used. Robustness tests were conducted for the two levels of analysis (dimensions and subdimensions) by excluding data for municipalities located in the State of São Paulo. These alternative estimations were considered based on the identification of this federative unit as a hub of franchise unit concentration, representing 36% of the total for this variable over the analyzed periods. Please refer to Appendix I for a complete description of the dataset. All procedures dealing with data curation and analysis were performed using Stata/SE version 15.1.

Analyzing the evolution of the total number of units per city over the selected period, we observed a 6.6% reduction in the number of units between 2016 and 2022. Because of the proposed methodological design, we cannot conclude an absolute decline in the number of units in cities, considering that the results presented here do not capture the emergence of new franchise networks over this period. Nevertheless, some movements are noteworthy. Within the analyzed time interval, the decline in the number of units in cities that are capitals (states or federal) is considerably greater than that in non-capital cities (9.3% versus 4.4%). Similarly, taking as references the cities that had more than 100 units in 2016, a decline of 10.5% was observed, while in the others, this reduction was only 1.3%.

With regard to the size of networks, from the perspective of franchise networks, some interesting movements can be discussed. In the sample, 60.9% of the chains experienced a reduction in the number of outlets between 2016 and 2022, representing a net negative balance of 4,651 units. Within this group, 17 networks ceased to operate through franchising. Among the growing chains (27.6% of the sample), the net balance was 3,016 units. These six chains remained stable. These results are somewhat predictable, considering the context of the period analyzed. The fact that some networks have grown reinforces the idea presented by Bretas and Alon (2020) that the effects of crises are heterogeneous for franchised companies. In general, the literature points to attributes derived from this business format that help face a crisis (Dermonde & Fischer, 2021). The analyzed data did not allow for definitive conclusions on this topic. However, the rate of operation closures through this business model (only 7.1%), despite the crisis of the period, suggests a convergence between our results and the aforementioned literature.

The results for the net effects at the first level of analysis (ecosystem dimensions), controlling for the size of the economy (Population and City GDP) and the macro-region of the country, are presented in Table 2. We provide a series of statistical diagnostics to guarantee the robustness of our estimation. First, we outline a direct comparison between Poisson and Negative Binomial estimations. The likelihood-to-ratio test of the alpha parameter identifies the significant presence of overdispersion in the dependent variable, a feature that indicates that Negative Binomial models are more appropriate for the interpretation of coefficients and significance levels. The AIC and BIC parameters offer further support to this aspect, indicating a better fit of the Negative Binomial Models. In any case, as can be inferred from a direct comparison between models, the main results associated with the independent variables of interest remain unaltered. In this respect, while all analyzed dimensions have significant coefficients, the negative effect associated with the Entrepreneurial Capital dimension contradicts expectations based on the literature regarding the dynamics of EE. These results remained consistent once the data for cities located in the State of São Paulo were excluded, suggesting a good level of robustness associated with these conditions. This finding is interpreted as a consistent indicator of the specificities related to franchise-based entrepreneurship compared to other forms of entrepreneurship investigated in the literature.

On the other hand, the coefficients associated with competitive insertion, business fabric, and productive organizations emerged as critical elements for the establishment of franchised network outlets. Governance for Development also presents positive and significant coefficients, although at a lower level than the other dimensions.

Table 3 shows the second level of analysis, which considers the sub-dimensions of EE as predictors of the number of franchised network outlets. Again, we have proceeded with auxiliary tests to generate a more robust exploration of findings. In this case, we notice some differences between Poisson and Negative Binomial estimates, but we favor the interpretation of the latter due to the presence of overdispersion, as outlined in the likelihood ratio test of alpha. Furthermore, this approach allows for the observation of a high degree of specificity regarding the elements that influence the dynamics of entrepreneurship at a local level. In this case, the negative effects associated with the components of Entrepreneurial Capital are exclusively linked to the variables “Entrepreneurial Education”, although this relationship manifests specifically when São Paulo State ecosystems data are excluded. In terms of “Business Fabric”, the occurrence of Solidarity Values presents a positive relationship with the dependent variable in both stages of the estimations, while Business Networks have a positive and significant coefficient only for the restricted sample. Regarding the dimension of Governance for Development, there are differences when comparing models III and IV, where the Planning sub-dimension is a predictor for the complete sample, while Participation and Social Control are present only in the robustness test. In the case of Productive Organization, the models are aligned, demonstrating negative effects for Productive Structure and positive for Consumption and Credit Potential, Innovation, and Environmental Impact. Regarding the dimension of Competitive Insertion, the results are significant and consistent for the Tourism and Creative Economy vector.

Thus, in a global evaluation of these results, and based on the observed robustness in the behavior of the variables, it is concluded that ecosystems that are more economically thriving (combining low levels of vulnerable and low-income populations with high levels of consumption potential), with high innovation potential and low levels of environmental impact are more likely to produce high numbers of franchised network units. This is associated with the negative effects of high levels of sectoral specialization and the presence of productive clusters and the positive effects of the density of Tourism and Creative Economy activities, a situation that indicates the centrality of diversified and service-oriented urban centers, even when controlling for the population size and GDP of the territorial units. In this sense, although our research indicates degrees of overlap between the concept of EE and the concentration of franchised network units, the association dynamics between these concepts are essentially different from those observed in the literature dedicated to examining other types of productive entrepreneurship.

This study aimed to understand the relationship between the vectors of Brazilian EE and the locational concentration of franchised networks in Brazilian cities. The results mostly converge with the understanding that local contextual elements are related to the development of entrepreneurial activity. In Models 1 and 2, presented in Table 2, a significant relationship for the five tested dimensions was noted. For four of them – Networks, Knowledge, Demand, and Leadership–in the model of Stam and Van de Ven (2021), this relationship is positive. This is in line with the majority of the literature discussing the relational dynamics between EE and entrepreneurial activity (Spigel & Harrison, 2018; Stam, 2015). These findings reinforce the importance of local contextual factors in fostering franchised network activity, as Melo et al. (2023) suggest.

On the other hand, the negative association observed from the Entrepreneurial Capital Dimension – which is equivalent to “Talent” in Stam and Van de Ven’s (2021) model – suggests an apparent contradiction with the current literature on EE. From a simplistic view, such a result could be interpreted as a refutation of the understanding of franchising models as an entrepreneurial activity (a view defended by authors in the field, such as Watson et al. (2020) and Dermonde, Fisher, and Moraes (2024)). However, when delving into this dimension, subdimensions such as education, entrepreneurial education, and business conditions were considered. The negative relationship indicates that the worse these indices, the lower the capacities for classic entrepreneurship are assumed to exist, as discussed by Spigel, Kitagawa, and Mason (2020). In the context of the analyzed period, notably marked by crises of various natures in Brazil and worldwide (manifesting as a hostile environment for entrepreneurial activity), the findings may suggest that the option of undertaking through franchises reflects a lower-risk choice, considering the replication of a tested and validated model (Gillis & Castrogiovanni, 2012). In addition, we might consider that such conditions may point towards a situation in which lower-risk entrepreneurial strategies may be more attractive in areas with limited autonomous entrepreneurial activity. These features are intrinsically connected with the nature of different strands of entrepreneurial activity, a topic that deserves further attention from the dedicated literature (O’Connor & Audretsch, 2023). For instance, recent inspections comparing EE components on a diverse set of entrepreneurial measures have underscored the co-existence of different configurational trajectories in the United States (Bennett et al., 2025). These interpretations align with discussions on EE that consider entrepreneurial activity to be multifaceted and diverse (Alves et al., 2021; Audretsch, 2021). Thus, from the analyses conducted, we advocate understanding the franchise system as an entrepreneurial activity but with its own specificities and characteristics. Such conditions warrant the relevance of moving forward with EE-based approaches that deal with heterogeneous forms of entrepreneurial activity, thus shedding light on the sectoral dimension of the entrepreneurial phenomenon (Fischer, Alves, Vonortas, & Brown, 2024).

A more in-depth look at the EE subdimensions with the highest relationship coefficients with entrepreneurial activity can be carried out. These are cases of Solidarity Values and Consumption and Credit Potential in the complete model. In the model excluding the State of São Paulo, these same sub-dimensions are accompanied by Business Networks and Innovation. These results reinforce the understanding that consumer markets with available resources and access to credit are attractive for choosing municipalities as destinations for franchised networks. Excluding São Paulo, a locality with the greatest weight in Brazilian economic activity, the density of businesses and business services is important, indicating the relevance of this contextual attribute.

Overall, the results of this study confirm our previously presented theoretical proposition. Through empirical analysis, one can verify the significant relationship between EE dynamics and the locational concentration of franchised networks. Such considerations advance the debate about the relationship between contextual elements and the dynamics of franchise network activity (Melo et al., 2023; Bui et al., 2022).

Our research contributes to the literature by exploring the relationship between contextual factors and the locational concentration of franchise network units, an area largely uncharted by academics. In doing so, we engage in a broader perspective on the interconnection between EE elements and the emergence of productive entrepreneurship. A key aspect that we identified is the mismatch between the talent dimension and franchise-related entrepreneurial activity, that is, the consistent and negative association between these two variables. This is in clear contrast to prior literature, but it generates a potential avenue to understand how entrepreneurial activity can thrive even in places with a lack of human capital. Perhaps this is not the case when considering high-tech firms, but franchise units appear as an alternative. This has important implications as it underscores that dense EE may emerge through different forms of output. As Complex Adaptive Systems, the resulting trajectories are unpredictable, but franchises can stand for an interesting way to foster the emergence of an entrepreneurial culture in the region (which might mitigate EE-level voids).

Additionally, the context of the analyzed period contributes to the debate on the governance arrangements of EE in a developing-country context. Not only practice but also research done in these environments is heavily inspired by what goes on in developed nations. While we do not disagree that this is a fair and legitimate strategy, it often neglects some specific features of less-developed markets, particularly in the institutional context. By tackling this phenomenon, we provide new evidence on EE dynamics within an emerging economic environment, bringing lessons that can and should be taken seriously by practitioners and policymakers in Brazil and in developing countries facing similar constraints. This means embracing the institutional complexity upon which EE emerges and develops, thus requiring bespoke strategies that may be aligned with overall EE structures but which require adaptations. This is valid not only for place-level approaches, but also for specific kinds of entrepreneurial activity (considering their respective requirements). For practitioners, another element of interest that arises from our research concerns the evaluation of key EE drivers in establishing expansion strategies.

As usual, this study is not free of limitations and opens the way for new empirical research. The first limitation refers to the number of franchise networks analyzed and the lack of information regarding the ownership structure of the units. Future studies could focus on a larger number of networks, considering the differences between the operating segments of this type of organization. Analyses that consider the ownership structure of the operating units (i.e. company-owned versus franchised) are also encouraged, and will be considered valuable contributions. Furthermore, the analysis of networks operating in different national markets could contribute to the debate on EE dynamics at a global level. Perhaps more importantly, this is a limitation associated with our transactional view of EE. In this case, our research is only capable of addressing the spatial co-occurrence of the elements observed in cities. This means that our assessment, although complying with best research practices, is unable to unravel potential linkages across different ecosystems (i.e. spatial spillovers) that can affect how specific local configurations can be complemented or supplemented by elements located elsewhere (Fischer et al., 2025). This can be of particular interest when addressing franchise networks, as we expect that managerial and operational knowledge may flow across territories as a function of the nature of such entrepreneurial businesses.

In this study, we aimed to investigate the relationship between vectors of local EE and the locational concentration of franchise chains in Brazilian cities. We conducted longitudinal studies spanning 2016, 2019, and 2022. This study analyzed the relationships between the characteristics of Brazilian EE, as measured by the five dimensions of ISDEL, and the number of units in 239 selected Brazilian franchise networks at the municipal level.

Our results indicate different movements when analyzed from the perspective of municipalities as well as from the perspective of networks. However, our focus was on understanding the relationship between contextual indicators and the concentration of network operations. We observed a significant relationship between all the analyzed EE dimensions and the volume of units per city. However, it is noteworthy that contrary to the literature on EE, we observed a negative relationship for the Entrepreneurial Capital dimension. At a more in-depth level of analysis, we can further verify that the subdimensions of Solidarity Values, Consumption and Credit Potential, Business Networks, and Innovation have the highest significant positive coefficients for the dependent variable.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

EJSB and MD contributed to the design and implementation of the research, BBF to the analysis of the results, EJSB, MD and BBF to the writing of the manuscript. MD conceived the original and supervised the project.

Eugenio Bitti is grateful for the grant #2020/13301-2 from the São Paulo Research Foundation (FAPESP). Matheus gratefully acknowledges funding from the Office of Research and Innovation of the University of São Paulo through the Support Program for New Faculty Members. Bruno Fischer acknowledges financial support from the Basic Research Program at the National Research University Higher School of Economics (HSE) and from the Russian Academic Excellence Project “5-100”.

The supplementary material for this article can be found online.