1 Introduction

The Latin American region has attracted a great amount of interest among management and organizational scholars in recent times (Aguinis et al. 2020). The region comprises 20 sovereign nations plus seven (7) dependent territories distributed over the Americas that share: (a) the societal experience of colonization by the Spaniards and Portuguese in the seventeenth–eighteenth century (Lange et al. 2006) and the subsequent independence in the nineteenth century (Colburn 2002), (b) the primary use of romance languages (i.e., Spanish and Portuguese) (Vassolo et al. 2011), and (c) an influence on their legal systems based on the Code Napoleón (Mirow 2005). The region accounts for 8.37 per cent of the global population (656 million inhabitants) and an aggregated gross domestic product (GDP) of US$ 4.3 trillion in 2020 (O’Neill 2021).

The Latin American context has been defined as a region full of paradoxes (Aguinis et al. 2020; Vassolo et al. 2011). Although in the first two decades of the twenty-first century the region has made improvements in terms of economic development and democratic stability (Bustamante et al. 2021); it continues to have: (a) weak market infrastructure (Estache et al. 2002), (b) vulnerable institutions (Brinks et al. 2019), (c) macroeconomic volatility (Breen and García-Peñalosa 2005), (d) a significant informal economic sector (Smolka and Larangeira 2008), (e) a prominent level of corruption (Lecuna and Chávez 2018), (d) low competition within firms and a polarized labour market (Hermans and Reyes 2020). Some authors have advocated that these distinctive characteristics of the region could serve as a “natural laboratory” (Aguinis et al. 2020; de la Torre et al. 2011; De La Torre and Chacar 2012) to challenge existing managerial and organizational theories eliminating the implicit assumptions often presents in those from developed economies (i.e., solid formal institutions, and efficiency of the market mechanisms). Furthermore, economic, and institutional homogeneity across Latin America offers an alternative to testing existing managerial and organizational theories while reducing the effect of macro-level confounding effects (Cuervo-Cazurra 2016). These unique regional characteristics allow researchers to challenge theoretical assumptions and explore singular phenomena (Aguinis et al. 2020; Vassolo et al. 2011).

Although the Latin American region has been recognized as an opportunity for theory building and testing in management and business research (Vassolo et al. 2011), the contexts remain largely unexplored (Aguinis et al. 2020). While some articles have surveyed the research productivity of authors affiliated with Latin American universities (see Table 1 for a review), our knowledge about how the Latin American context influences organizations and its practices (as a unit of analysis) remains scarce (Amorós et al. 2021). Therefore, we aim to close this research gap by mapping the knowledge about research that has theorized employing the Latin American context. In doing so, we follow the recent calls (Aguinis et al. 2020; Amorós et al. 2021) to provide a research path for theoretical building and testing using this contextual setting.

Table 1 Exemplar bibliometric studies
Full size table

To close this gap and following extant research calls, our article answers the following overarching question: How the research about organizations in the Latin American context has evolved and how could it move forward? This research synthesizes the literature based on a three-stage systematic review process using three complementary bibliometric methods (i.e., co-occurrence, co-citation, and co-authorship network analysis). The systematic review included 1.940 peer-reviewed articles that have theorized using as context the Latin American region. In doing so, we delved into the evolution of the subject by reviewing the leading research organizations, prominent journals, and the main authors that have contributed to this subject from the period 2004–2021. We chose bibliometric techniques because it is one of the most important measures for the appraisal of scientific productions of a field (Block and Fisch 2020; Lim et al. 2022). Further, we developed a longitudinal network analysis (Cortés-Sánchez 2022) to identify the changes along the intellectual, conceptual, and social structures within the field of inquiry.

We found that research about organizations in the Latin American context has gained increasing momentum since 2011, with no signs of slowing down at the time of submitting this article. Furthermore, The Journal of Business Research was identified as the one with a higher quantity of articles in our sample. Besides, the citation analysis reveals that Building Resilience or Providing Sustenance: Different Paths of Emergent Ventures in The Aftermath of The Haiti Earthquake (Williams and Shepherd 2016), Trade Liberalization, Exports, And Technology Upgrading: Evidence on The Impact of Mercosur on Argentinian Firms (Bustos 2011), and The Micro-Determinants of Meso-Level Learning and Innovation: Evidence from A Chilean Wine Cluster (Giuliani and Bell 2005) had the highest number of citations in the periods 2000–2009, 2010–2015, and 2016–2021 respectively. Lastly, Christian Felzensztein was the more productive author based on the number of publications in peer-reviewed journals.

The bibliometric network analysis allowed us to analyze the structure of the conceptual, intellectual, and social structure of the field. First, we found that the conceptual structure of the field presented dissension given the proliferation of new research topics that used the Latin America region as a context for theorizing. Moreover, we observed that the intellectual structure of the field presented a less sparse connection between the nodes and a gradual trend in novel intellectual conversations. Lastly, the social structure of the field showed a process of structural fragmentation between 2004 and 2021.

The following is the structure of this article. Section 2 describes the method and data. Section 3 presents the results of the performance analysis. Section 4 presents the results of three bibliometric analysis techniques (i.e., co-occurrence, co-citation, and co-authorship networks analysis). Section 5 discusses the avenues for further research based on our findings and sets out the limitations of this work.

2 Methods

This study aims to map the research about the organizations in the Latin American context and how could it move forward. Following the best practices in mapping scientific domains, we performed a systematic review and a bibliometric analysis (Cortes-Sanchez 2020; Cortés-Sánchez 2022; Lim et al. 2022; Pizzi et al. 2020). We provide a complete overview of the evolution and trends of the literature by employing a bibliometric analysis and a systematic review of the main publications (Gil-Barragan et al. 2020; Goyal et al. 2021).

2.1 Systematic review protocol

At the time of planning this research, we defined our objective and identified the main keywords for the systematic review. Following Tranfield et al. (2003), our systematic review is a three‐stage process that includes: (i) identification of literature sources, (ii) selection of articles and screening criteria, and (iii) data analysis and codification. The last stage involves a process of coding, analysis, and presentation of the results (Tranfield et al. 2003). Figure 1 synthetizes the systematic review protocol.

Fig. 1
figure 1

Systematic review: strategy and steps

Full size image

2.2 Identification of literature

We chose nine keywords from organizations based on previous literature (Alayo et al. 2021; Bužavaitė et al. 2019; Kücher and Feldbauer-Durstmüller 2019; López-Fernández et al. 2016; Parastuty 2018) and 26 keywords to capture each of the Latin American’ countries (e.g., countries in South America, Central America, the Caribbean, and México that speak a Romance language) (Gil-Barragan et al. 2020; Jaen et al. 2017). Table 2 shows the keywords used with its support in the literature.

We selected the Web of Science (WoS) Core Collection (Li et al. 2021) for our literature search because prior research highlights the robustness of the repository (Du et al. 2021; Goyal et al. 2021; Li et al. 2021; Mallawaarachchi et al. 2020; Wu et al. 2018; Ye et al. 2020; Zhang and Liang 2020; Zhang et al. 2017), as well as their reliability and transparency in contrast of other sources (e.g. Google Scholars) (Mingers and Lipitakis 2010). WoS Core Collection possess over 21,100 peer-reviewed high-quality journals with 74.8 million records in 254 disciplines (Clarivate 2022). The repository includes 3400 high-impact journals in social science with 9 million records in 58 disciplines (Clarivate 2022). WoS has been historically the most widely used database for bibliometric and scientometric research because it has the oldest and most complete records of citation indexes (Ellegaard and Wallin 2015; Pranckutė 2021). Some authors have recognized the WoS repository as the leading high-quality source in the management and economic field (Bar-Ilan 2018; Bonilla et al. 2015; Ellegaard and Wallin 2015; Herrera and de las Heras-Rosas 2020; Tanco et al. 2018).

Table 2 Keywords used in the query
Full size table

Although other repositories have a higher coverage of journals and articles [e.g., Scopus repository has a coverage of 25,100 peer-reviewed journals with 77.8 million records; including 12,464 journals in social sciences (Elsevier 2020)] some authors have pointed out a quality issue given questionable peer-review practices of some journals that could bias the bibliometric analysis (Cortés-Sánchez 2020a, b, c, d). That later point -quality articles with a reliable peer-review process—is relevant in the selection of the WoS repository. Moreover, universities in the Latin American context are incentivizing their faculty to publish in high-quality journals given a set of external pressures (e.g., international accreditation, global ranking position, and allocation of research sources from governments and other funding agencies) (Aguinis et al. 2020; Ronda-Pupo 2016). Therefore, publishing in high-quality journals -mainly indexed in the WoS Core Collection—has been acknowledged as one of the main criteria in the community (Seipel 2003).

After selecting the repository, we executed a Boolean search in the articles’ titles and abstracts. Besides, we follow a retrospective look at prior literature (Aliyev et al. 2019; Cancino et al. 2020) to include valuable articles outside the WoS repository. Overall, we identified 35.730 articles.

2.3 Selection of articles and screening criteria

We selected and applied the inclusion and exclusion criteria (Table 3). We have just included publications published in peer-reviewed journals (Block et al. 2020). Moreover, we only included articles in the WoS categories of Business, Business Finance, Economics, and Management. We also generated an extra filter for articles in the WoS research area of Business and Economics. Finally, we excluded the articles after December 2021 to have complete years for analysis. In total, 33.307 articles were excluded from the sample, and 2.423 articles remained for the next stage.

Table 3 Inclusion and exclusion criteria
Full size table

In the screening stage, we reviewed each of the remaining abstracts manually to corroborate the focus of the article. We considered that the publication did not fit the criteria if the article focused on comparative analysis using Latin America as a reference, or if the Latin American context was part of a comparative analysis with other regions. In total, 1.940 articles met the criteria and were subject to quantitative and qualitative analysis. The final dataset used in this research is publicly available at https://doi.org/10.17632/pmwfhndv72.1 (Diaz Tautiva et al. 2022).

2.4 Bibliometric analysis

Bibliometric analysis is a rigorous method for exploring and evaluating quantitatively large volume of scientific material (Donthu et al. 2021; Lim et al. 2022). Bibliometric analyses are useful for untangling and mapping the cumulative knowledge of emerging or established research areas (Aliyev et al. 2019; Block and Fisch 2020; Booth et al. 2020; Donthu et al. 2021; Tanco et al. 2018; Wang et al. 2018). Employing this method scholars have proved that it is possible to identify knowledge gaps in the literature, to recognize novel ideas for further research, and to provide a full overview of a field (Caiazza and Simoni 2019; Linnenluecke and Griffiths 2013; Vallaster et al. 2019). Two complementary techniques must be applied to obtain robust results (Donthu et al. 2021; Gutiérrez-Salcedo et al. 2018; Noyons et al. 1999): (a) performance analysis, and (b) science mapping.

First, we implemented a performance analysis. It is a technique that aims to evaluate scientific actors (e.g., authors, research organizations, countries, and journals) and their research impact (i.e., citations and quantity of publications) based on the bibliographic information (Donthu et al. 2021; Noyons et al. 1999; Waltman et al. 2010). Following prior performance analysis (Block et al. 2020), we provide an overview of the yearly production based on the number of publications, the foremost journals, the major research organizations, the leading authors in the field of inquiry, and the central articles on the sample.

Second, we implemented a science mapping analysis. It is a technique that examines the relationships (i.e., citation, coauthorship, co-citation, bibliographic coupling, and co-occurrence) between research components (i.e., research organizations, articles, journals, words, or authors) (Donthu et al. 2021; Noyons et al. 1999; Yan and Ding 2012). Through this analysis, it is possible to inform the structures among the research components by considering their networks of interactions (Donthu et al. 2021; Yan and Ding 2012). Under this perspective, the research components are the nodes of a network, while the relationships represent the links between them. Network-based bibliometrics must consider three elements: (a) the network types, (b) the aggregation level, and (c) the approach analysis (Yan and Ding 2012).

Prior research points out about the complementarity effect of using several bibliometric networks to provide a consistent analysis (Yan and Ding 2012). Yan and Ding (2012, p. 1231) suggested that using three complementary types of bibliometric networks (i.e., co-word, co-citation, co-authorship) could provide a broad view of a field because those have a low level of network similarities -based on cosine distance—while providing a different perspective of a field based on social and information networks.

Initially, we generate a co-word network based on the number of times a keyword co-occurred at the article-level. Co-word networks embody the conceptual structure of a field as it identifies the main research topics on the community (Sedighi 2016; Su and Lee 2010; Yan and Ding 2012). To perform the co-word network, we select the author keywords as the unit of analysis as those are a more specific descriptor of the article’s content (Li et al. 2009; Liu 2013; Zhang et al. 2015). Particularly, the author keyword analysis could offer information regarding the research trends which are concerned by the researcher community (Block et al. 2020; Li et al. 2009).

Then, we generate a co-citation analysis to reveal the intellectual structure of the field and its underlying theoretical perspectives (Donthu et al. 2021; Rossetto et al. 2018; White and Griffith 1981). The analysis considers the frequency that two units (i.e., authors, references, or journals) are co-cited thus indicating their affinity (Rossetto et al. 2018; Small 1973). To perform the co-citation network, we select the single author as the unit of analysis. Employing this macroscopic view, we could reveal the theoretical frameworks and positions that the authors have employed (Rossetto et al. 2018). We used the number of publications and the frequency of citations as a measure of relevance within the network (Block et al. 2020).

Lastly, we generate a co-authorship analysis to examine the academic interaction among academic units (e.g., authors, organizations, or countries) (Donthu et al. 2021; Van Eck and Waltman 2013). To perform the co-citation network, we select the single author as the unit of analysis. Scholars that collaborate form a network named “invisible college” that defines the social structure of the field (Crane 1977; Zuccala 2006). Donthu et al. (2021, p. 290) advocated that the co-authorship analysis at the author-level could shed lights on geographical clusters of scholars that ignite new research in underdeveloped areas. Besides, it can guide future scholars to reach out and cooperate with established scholars in the field (Donthu et al. 2021).

Given our research objective, we employ an enriched longitudinal three-level (i.e., macro-, meso-, and micro-level) approach analysis (Yan and Ding 2012). We calculated the macro-level statistics (i.e., density, average path length, and transitivity) to identify the global structure of the network, the meso-level clustering indicators (i.e., clustering coefficient, communities, and modularity) to explore the groups’ behavior within the research components, and the micro-level individual characteristics (i.e., betweenness centrality and cluster classification) to identify the research component’s power and position (Borgatti et al. 2009; Cortés-Sánchez 2022; Iacobucci et al. 2019). Table 4 provides a formal definition of each of the measures and their references in the literature.

Table 4 Approach of analysis
Full size table

Given the size of the sample and the complexity of the networks, we used two complementary software to perform the bibliometric analysis. Initially, the networks (i.e., co.-occurrence, co-citation, co-authorship) were obtained using the VosViewer software (Van Eck and Waltman 2013; Waltman et al. 2010). Then, we imported the matrixes in the R software to calculate the macro-, meso-, and micro-network measures. The three-level network analysis was performed using the igraph R package (Csardi and Nepusz 2006) in each network using the commands edge_density (network density), mean_distance (average path length), modularity (modularity), cluster_louvain (community structures based on the Blondel et al. (2008) algorithm) and betweenness (betweenness centrality). Lastly, the cluster classification was obtained using the Waltman et al. (2010) modularity-based technique incorporated in the VosViewer software (Waltman et al. 2010).

3 Performance analysis

Section 3 presents a descriptive overview of our final sample. We show the yearly productivity and the most prolific journals and authors. Besides, we uncover the leading publications and the research organizations with the most articles published in the field.

3.1 Yearly productivity based on the number of publications

Figure 2 illustrates the yearly productivity of the field based on the number of publications. Research about organizations in the Latin American context was a minor research field between 2004 and 2009, with an average of only 27 articles per year. From 2010 to 2015, the publication average was 71 articles per year.

Fig. 2
figure 2

Articles per year. Notes: Yearly productivity based on the number of publications from 2004 to 2021 (N = 1940)

Full size image

The research context gained momentum and achieved recognition from 2011 onwards, yielding 61 articles. Whether we consider the timeframe from 2010 to 2015, Fig. 2 illustrates an increase of 162 per cent compared to the period from 2004 to 2009 with 427 articles published. Moreover, the level of activity from the years 2016 to 2021 represents a 215% increase over 2010–2015, with approximately 225 articles per year. Furthermore, the year 2019 had a productivity pick with 275 articles published. Notably, the years 2020 and 2021 maintained a level of publications above the average for the period considered, demonstrating scholars' continued interest in the study of organizations in Latin America.

3.2 Main journals

The final sample comprises 1940 publications in 460 journals. The top 15 journals represented 34.46 per cent of total publications. Table 5 exhibits the 2020 impact factor and additional metrics of the journals. The Journal of Business Research is the top one with 114 publications. Academia Revista Latinoamericana de Administración (top two) and Revista Brasileira de Gestão de Negócios (top three) have 66 and 46 publications correspondingly. The Journal of Business Research has the highest impact factor among the top 3 journals.

The impact factor (IF) of a journal indicates the number of citations that the journal's articles receive and is an indicator of the quality of the journal (Garfield 1999). According to Table 5, Supply Chain Management-An International Journal (IF: 9.012), Journal of World Business (IF: 8.513), and Small Business Economics (IF: 8.164) had the highest impact factors with a total of forty-seven (47) publications.

Table 5 Top 15 journals based on the number of publications
Full size table

Table 5 also shows that The Journal of Business Research received 1777 citations with an average of 16 per publication. Academia Revista Latinoamericana de Administración received 298 citations with an average of 5 per publication. Revista Brasileira de Gestão de Negócios received 189 citations with 4 cites per publication in average. Furthermore, Research Policy had the highest average citations per publication (97), followed by those published in the Journal of Developed Economies (42), Journal of International Economics (33) and Supply Chain Management-An International Journal (33).

3.3 Main research organizations

Our final sample contains 1940 publications published by 1807 research organizations. The top 15 research organizations published 36.94 per cent of the total. Table 6 illustrates the total publications and citation metrics for each. The University of Sao Paulo, Brazil, is the most productive university with 110 publications. The University of Chile, Chile, and Tecnologico Monterrey, Mexico, have 42 and 41 publications accordingly.

Table 6 Top 15 research organizations based on the number of publications
Full size table

Table 6 also shows that The University of Sao Paulo, Brazil, received 997 citations with an average of 9 per article. The University of Chile, Chile, received 342 citations with an average of 8 per article. Tecnológico of Monterrey received 432 citations with an average of 11 per article. The University of San Andres had the highest average number of citations per publication (44), followed by those published by the World Bank (28), Florida International University (26), Nber (26), and Pontificia Universidad Catolica of Chile (23).

3.4 Important authors

Our final sample contains 1940 publications published by 4452 authors. The top 15 authors published 9.3 per cent of the total. Table 7 shows the total publications and citation metrics of each author. Felzensztein, C is the most productive researcher with 19 publications, followed by Ciravegna, L and Bianchi, C with 13 and 12 publications accordingly.

Table 7 also shows that Felzensztein, C received 544 citations with an average of 29 per article. Ciravegna, L received 535 citations with an average of 41 citations per article. Bianchi, C received 386 citations with an average of 32 citations per article. Husted, B has the highest average number of citations per article (130), followed by those published by Galiani, S (82), Ciravegna, L (41), and Martincus, Cv (35).

Table 7 Top 15 authors based on the number of publications
Full size table

3.5 Important articles

Table 8 exhibits the most relevant articles in terms of WoS citations. As older publications are more likely to be cited (Block et al. 2020), we split the sample into three periods (i.e., 2004–2009, 2010–2015, and 2016–2021) based on the publication’s trends observed in Sect. 3.1. Giuliani and Bell (2005) published in Research Policy the most cited article between the years 2004–2009 with 640 citations. Giuliani and Bell (2005) examined the influence of the absorptive capabilities of individual firms on the functioning of the intra-cluster knowledge system and its interconnection with extra-cluster knowledge, concluding that knowledge flows within a core group of Chilean firms characterized by advanced absorptive capabilities.

Table 8 Top five publications in the periods 2004–2009, 2010–2015, and 2016–2021
Full size table

The article written by Husted and Allen in 2006 and published in the Journal of International Business Studies is the second most cited. Husted and Allen (2006) studied the relationship between corporate social responsibility and international organizational strategies. They showed that institutional pressures guide corporate social responsibility decision-making in Mexican firms. The article by Arellano (2018) published in The American Economic Review is the third most cited with 339 citations. Arellano (2008) developed a small open economy model based on Argentina to study default risk and its interaction with output and external debt.

Bustos (2011) published in The American Economic Review the most cited article between the years 2010–2015 with 515 citations. Bustos (2011) studied the impact of the MERCOSUR regional free trade agreement on the technological improvement of Argentine firms. He revealed that an increase in income produced by the trade agreement can induce exporters to improve their technological base. Isenberg (2010) published in Harvard Business Review the second most cited article during this period with 404 citations. Isenberg (2010) described nine key principles for generating a strong entrepreneurial ecosystem using Chile and Colombia as examples.

Fernández-Villaverde, Guerrón-Quintana, Rubio-Ramírez and Uribe published the third most cited article during this period with 228 citations. This article was published in the National Bureau of Economic Research in 2011. The authors, using the cases of Argentina, Brazil, Ecuador, and Venezuela, showed how changes in the volatility of the real interest rate have important effects on real variables (e.g., consumption, investment, and work). Schnabl published in The Journal of Finance the fourth most influential articles with 180 citations since 2012. Schnabl (2012), studying Peruvian banks, explained the international banks liquidity shock process. The fifth most influential publication is the Muller and Kolk article (2010), with 174 citations. Muller and Kolk (2010) found evidence analyzing Mexican firms that management commitment to ethics is necessary for corporate social performance.

The most cited article during the 2016–2021 period is the one published by Williams and Shepherd in the Academy of Management Journal with 131 citations. Williams and Shepherd (2016) developed a theoretical model of resilience in the face of adversity. They used as a theoretical sample the creation of enterprises initiated by Haitian locals in response to the Haiti earthquake in 2010. Helpman et al. (2017) published the second most cited paper in the Review of Economic Studies during the 2016–2021 period. Using Brazilian data, Helpman et al. (2017) confirmed that a significant part of the wage inequality appears within sectoral occupations and for comparable individuals.

Frank, Cortimiglia, Ribeiro and de Oliveira (2016) published in Research Policy the third most cited paper during this period with 88 citations. Frank, Cortimiglia, Ribeiro and de Oliveira (2016) studied innovation in the Brazilian industry and explained how firms have a propensity to implement two opposing innovation strategies (e.g., market orientation and technology acquisition). Marodin et al. (2016) published in Supply Chain Management was the fourth most influential article with 76 citations. Marodin et al. (2016) explained how lean practices were executed and implemented in diverse positions within the supply chain in Brazilian companies. Van Marrewijk et al. (2016) published in Organization Studies the fifth most cited paper during the 2016–2021 period with 71 citations. Van Marrewijk et al. (2016) analyzed the political struggles over role patterns in the temporal organization of the Panama Canal Expansion Program. They argued that the diffuse hierarchy led to engaging in social and discursive practices of harmony-seeking and conflictual negotiations with authority.

4 Science mapping

Section 4 describes the results of the co-occurrence, co-citation, and co-authorship network analysis. Based on it, we describe the conceptual (Sect. 4.1), the intellectual (Sect. 4.2), and the social (Sect. 4.3) structure of this area of interest. Each section provides a summary of the networks' evolution based on the macro-, meso-, and micro-level characteristics; as well as an in-depth cluster analysis grounded in the sample.

4.1 Conceptual structure

4.1.1 Co-occurrence network: evolution in the period 2004–2021

Table 9 displays the number of keywords per analysed period. The total number of keywords observed in 2004–2009 was 233, whereas in 2016–2021 was 2644: a more than 11-fold increase. The average author keywords per article increased by 9.55% between the 2004–2009 and the 2016–2021 period.

Table 9 Authors keywords statistics
Full size table

Table 10 exhibits the maso and meso indicators of the three networks modelled. We observed a reduction in the network density between 2004–2009 and 2016–2021 of 1.9 percentage points, and a reduction on the modularity of 0.14 caused by the increase on the total of keywords within the sample. Moreover, the number of communities detected increased on 73.09% in the 2004–2021 period. Besides, we identified an increase in the clustering coefficient of 20.14%. In sum, the macro changes in the network suggest conceptual progressive dissension given the proliferation of new research topics that used the Latin America region as a context for theorizing.

Table 10 Co-occurrence network: macro- and meso-level statistics
Full size table

Table 11 presents the top 10 betweenness keywords for each period. Following prior research (Cortés-Sánchez 2022), Table 11 omitted countries (e.g., Brazil, Colombia, Chile) as well as regions (e.g., Latin America) to avoid redundancy in a regional-focused study. Exploring the co-occurrence networks, we can observe the conceptual structure of this field. Concepts such as innovation, entrepreneurship, and strategy were central in the network structure and showed a consistent appearance. Lastly, the emergence of new concepts in the 2016–2021 period (i.e., sustainability, corporate governance, internationalization, and taxes) suggests a change on research themes with a potential impact on the institutional and economical regional’ paradoxes.

Table 11 Co-occurrence network: top-10 keywords based on betweenness centrality
Full size table

4.1.2 Co-occurrence network: cluster analysis

Figure 3 displays the co-occurrence network, its evolution during the 2004–2021 period, and the keywords density within the network. We set a minimum number of occurrences of 10. In total, 160 keywords had at least 10 occurrences in the sample. We omitted the keywords used in the research query (see Table 2) to avoid redundancy in a focused study (Block et al. 2020). A final set of 53 keywords remained after the cleaning. Innovation and emerging markets had 100 and 96 occurrences respectively. Other widely used keywords included entrepreneurship, governance, performance, corporate social responsibility, and institutions. Moreover, we used the clustering modularity-based technique incorporated in the VosViewer software (Waltman et al. 2010) to obtain the clusters within the network. Using this technique, we identified five (5) keywords’ clusters (Table 12). We labelled each cluster using an open coding technique based on the main themes observed (Aliyev et al. 2019). Besides, the appendix section provides a broad overview of the results using the keyword plus.

Fig. 3
figure 3figure 3figure 3

Co-occurrence of keywords. Note: the first panel identified the keywords’ network, the second panel identifies the time evolution in the network, and the third panel identifies the network’ density (N = 1.940 publications). The network visualization was created using VOS Viewer version 1.6.17

Full size image
Table 12 Co-occurrence network: cluster analysis
Full size table

Cluster I (red) is labelled the Development, Entrepreneurship, and Institutions cluster. The cluster focused on macro- and micro-level phenomena as economic growth, institutions, entrepreneurship, and corporate social responsibility. The most frequent keyword in the cluster was entrepreneurship (n = 63). Top publications on the cluster investigated how institutional and national-level governance logics toward corporate social responsibility are inducing organizational strategies (Husted et al. 2016); how organizations select the time and pace of internationalization after introduction on the market (Lopez et al. 2009); how multinational organizations balance their global and local strategies with all stakeholders (Gifford et al. 2010); and how entrepreneurship and innovation theories adapt in emerging economies (Chaston and Scott 2012).

Cluster II (green) is labelled as the Internal Capabilities and Outcomes cluster. The cluster focused on micro elements of the organizational structure such as productivity, competitiveness, efficiency, innovation, and creativity. The most frequent keyword in the cluster was innovation (n = 100). Top publications on the cluster investigated how meta-innovation systems emerged and evolved through multiple initiatives (Etzkowitz et al. 2005); how the interaction between public research organizations and industries could be articulated (De Fuentes and Dutrenit 2012); what attributes and assets increase the likelihood to engage in innovation activities (Chudnovsky et al. 2006); how innovation activities increase the organizational survival (Fernandes and Paunov 2012); and how national labour regulation affects the organizational characteristics and its further growth (Almeida and Carneiro 2009).

Cluster III (blue) is labelled the Strategies and Contexts cluster. The cluster explores the strategic interaction of the organization with its environment. The most frequent keyword in the cluster was emerging markets (n = 96). Top publications on the cluster investigated how organizations achieve a global competitive advantage through their know-how in their establishment country (Del Sol and Kogan 2007); how individual, organizational, and contextual factors contribute to a higher likelihood of early internationalization (Amoros et al. 2016); how organizations disclosure their social impact to gain the support of the stakeholders (Ching and Gerab 2017), and how the value creation and capture perspective regarding international business are shaped by the Latin American context (Hermans and Reyes 2020).

Cluster IV (yellow) is labelled Governance and Business Performance cluster. The cluster includes elements associated to how a firm organizes plus the role of internal mechanisms toward organizational value creation. The most frequent keyword in the cluster was governance (n = 62). Top publications on the cluster investigated how formal institutions are enforced in the field of corporate governance (Siegel 2005); how internal corporate governance mechanisms (e.g., board of directors) are affected by high ownership concentration (Lefort and Urzua 2008); how contextual factors influence the implementation of new organizational practices (i.e., lean production) (Marodin et al. 2016); how organizations could improve their overall performance through the adoption of a strategic view of their processes (McCormack et al. 2008); and how organizational actors establish role structures on the face of a complex project (van Marrewijk et al. 2016).

Finally, Cluster V (purple) is labelled as the Organizational Development cluster. The cluster contains the elements associated with the development and growth of the organization. The most frequent keyword in the cluster was supply chain management (n = 17). Top publications on the cluster investigated how participation in supply chains can improve impoverished communities (Hall and Matos 2010); how CEO's personality influences strategic change (Herrmann and Nadkarni 2014); how the implementation of the strategic process leads to the firm strategic positional advantages (Kotabe et al. 2007); how organizational cultural characteristics affect the use of internal quality techniques (Gambi et al. 2015); and how firms could evaluate their supply chain practices and assess them by maturity level.

4.2 Intellectual structure

4.2.1 Co-citation network: evolution in the period 2004–2021

Table 13 exhibits the maso and meso indicators of the three networks modelled. We observed an accelerated reduction of 6.1 percentage points on the network density between 2004–2009 and 2016–2021. Additionally, we detected a decrease in the clustering coefficient of 20.14%, a reduction in the modularity of 29.97%, an improvement of 11.87% in the nodes (co-cited authors) average path length, and an increase on 6.88% in the number of communities through the 2004–2021 period. In sum, the macro changes in the network suggest a less sparse connection between the nodes and a gradual trend to novel intellectual conversations.

Table 13 Co-citation network: macro- and meso-level statistics
Full size table

Table 14 presents the top-10 betweenness co-cited authors for each period. Exploring the co-citation networks, we can observe the intellectual structure of this field. Multilateral economic research organizations such as the World Bank and the Organization for Economic Co-operation and Development (OECD) lead the intellectual conversation. Furthermore, the community is employing novel methodological approaches (i.e., structural equation modelling, robustness analysis and methods in business research, and case study approach) as a source to explore the complex organizational phenomena in the Latin American context. Lastly, we observe a risen interest in the community for joining the conversation about how organizations generate and implement strategies (e.g., innovation development, international management, and capacities building) to survive and grow in paradoxical contexts.

Table 14 Co-citation network: top-10 cited authors based on betweenness centrality
Full size table

4.2.2 Co-citation at the author level cluster analysis

Figure 4 displays the co-citation network, its evolution during the 2004–2021 period, and the co-cited author's density within the network. A total of 50,037 co-cited authors were identified in the sample. We set a minimum number of occurrences of 50 to obtain a parsimonious view of the network based on the leading co-cited authors. A final set of 68 authors meet the threshold. World Bank, the Organization for Economic Co-operation and Development (OECD), and Hair, J were the most co-citations with 204, 200, and 190 respectively. Moreover, we used the clustering modularity-based technique incorporated in the VosViewer software (Waltman et al. 2010) to obtain the clusters within the network. Using this technique, we identified four (4) co-cited clusters (Table 15). We labelled each cluster using an open coding technique based on the co-cited authors’ research areas (Aliyev et al. 2019).

Fig. 4
figure 4

Co-citation analysis at the author level. Note: Created using VOS Viewer version 1.6.17 (N = 1940)

Full size image
Table 15 Co-citation network: cluster analysis
Full size table

Cluster I (red) labelled Business and organizations methods and foundational theories contains 28 authors. The most cited authors in the cluster are Hair, J (190); Porter, M (176); Eisenhardt, K (149); Teece, D (141); and Zahra, S (128). Authors in the cluster have developed foundational theories in fields such as strategic management, entrepreneurship, international businesses, innovation and technology, digital transformation, complexity, and economic development. Top authors on Cluster I have theorized using the following frameworks: (a) the dynamic capabilities framework (Eisenhardt and Martin 2000; Teece 2007; Teece et al. 1997; Zahra and George 2002), (b) the competitive advantage framework (Porter 1990), and (c) the economics of competition (Porter 1998). Besides, we observed that advances in Partial Least-Squares (PLS) and multivariate data analysis (Hair Jr et al. 2017), the case study as a research design (Yin 2009), and the developments in reducing common method bias (Podsakoff et al. 2005) have grounded the empirical research in the field.

Cluster II (green) labelled International Business contains 18 authors. The most cited authors in this cluster are Cuervo-Cazurra, A (180); Khanna, T (165); Hofstede, G (131); Johanson, J (120); and Peng, M (85). Authors in the cluster have explored phenomena as international strategies, export behaviour, international market discovery, cultural differences and its impact on business performance, and organizational capabilities toward internationalization. The authors in Cluster II have theorized under the lenses of: (a) network and knowledge diffusion theory (Coviello et al. 2017; Johanson and Wiedersheim-Paul 1975); (b) agency costs theory (Cuervo-Cazurra and Dau 2009); (c) institutional settings, contexts, and changes theory (Cuervo-Cazurra and Genc 2008; Khanna and Palepu 2000; Peng 2003); (d) culture theory (Hofstede 2001); (e) resources-based theory (Peng and York 2001); and (f) technological capabilities theory (Dunning 1994).

Cluster III (blue) labelled Corporate Governance and Financial Econometrics contains 15 authors. The most prominent authors in the clusters are La Porta, R (182); Jensen, M (121); Fama, E (101); Claessens, S (87); and Arellano, M (80). Authors in this cluster have studied phenomena such as equity financial development, corporate governance structures and mechanisms (e.g., ownership structures and managerial incentives), and market regulations. Top authors in Cluster III have theorized under the lenses of: (a) theory of corporate finance (Claessens et al. 1999; La Porta et al. 1999); (b) agency theory (Fama and Jensen 1983), and (c) behavioural theory (Gómez-Mejía et al. 2007). Besides, we observed that advances in cross-sectional and panel data estimation (Arellano and Bond 1991; Blundell and Bond 1998; Wooldridge 2010) have grounded the empirical econometric strategies to explore these phenomena.

Cluster IV (yellow) labelled International Trade, Economic Dynamics, and Policy Evaluation contains seven authors. The most prominent research organizations in the cluster are the World Bank (204) and the OECD (200); while the top authors were Bernard, A (124), Acemoglu, D (77), and Melitz, M (68). Top organizations in the cluster have analysed phenomena such as macroeconomic dynamics, international economic activity, international trade trends, and macro-economic productivity. Besides, top authors have theorized about the international organizational productivity of the firms (Bernard and Jensen 1999), and the intra‐industry effects of international trade (Melitz 2003).

4.3 Social structure

4.3.1 Co-authorship network: evolution in the period 2004–2021

Table 16 exhibits the maso and meso indicators of the three networks modelled. We observed a low dense network (less than 1%) along the three periods. The network reduced their density in 0.4 percentage points between 2004 and 2021. The process is observed as the total number of communities detected in 2004–2009 was 150, whereas in 2016–2021 was 937: a more than sixfold increase. Besides, the average path length increased from 1071 in 2004–2009 to 3376 in 2016–2021, implying a longer communication between the communities. Lastly, we observe slight changes in the clustering coefficient and modularity in the 2004–2021 period with a variation of 1.02% and 0.5% respectively. In sum, we observed a structural fragmentation process of the network.

Table 16 Co-authorship network: top-10 authors based on betweenness centrality
Full size table

Table 17 presents the top 10 betweenness authors for each period. By exploring the co-author's networks, we can observe the social structure of this field. Although the 2004–2009 period presented the densest network structure; just six authors were central in the community. The structure in the first period suggested that those authors were pioneers in the use of the Latin American context to theorize regarding organizational phenomena. From 2010 onwards, a new generation of scholars positioned themselves in the community using the regional context as part of their research design. We observe that scholars in the fields of entrepreneurship (e.g., Christian Felzensztein, Luciano Ciravegna, Marcelo Nagano, José Amoros), strategic management (e.g., Esteban Brenes, Jorge Carneiro, Hsia Hua Sheng) and public policy evaluation (e.g., Alessandro Maffioli, Rodolfo Stucchi, Christian Martinicus) consistently used the Latin American context in their research. Their position as central authors in the network remained unchanged between 2010 and 2021.

Table 17 Co-authorship network: top-10 authors based on betweenness centrality
Full size table

4.3.2 Co-authorship network at the author level-cluster analysis

Figure 5 displays the co-authorship network, its evolution during the 2004–2021 period, and the author's density within the network. A total of 4452 authors were identified in the sample. We set a minimum of four papers published to obtain a parsimonious view of the network based on the recurrent authors. A final set of 70 authors meet the threshold. Christian Felzensztein, Luciano Ciravegna, and Constanza Bianchi were the most frequent authors in the sample. Moreover, we used the clustering modularity-based technique incorporated in the VosViewer software (Waltman et al. 2010) to obtain the clusters within the network. Using this technique, we identified 12 co-authorship clusters (Table 18). We labelled each cluster using an open coding technique based on the topics researched by the authors (Aliyev et al. 2019).

Table 18 Cluster analysis of the co-authorship analysis at the author level
Full size table

Cluster I (red) labelled Innovation and Organizational Performance contains 11 authors. Scholars in the clusters have explored the effectiveness of public policies toward innovation incentives (Crespi et al. 2016), the innovation strategies used by organizations in the region (Zuniga and Crespi 2013), the determinants of performance in micro and small enterprises (Berrone et al. 2014), plus leadership (Castano et al. 2015) and cross-cultural negotiation (Ogliastri and Quintanilla 2016) in Latin American context. Cluster II (green) labelled Corporate Governance and Disclosure contains 10 authors. Scholars in the clusters have explored the relationship between corporate governance (Garay and Gonzalez 2008) and disclosure (Garay et al. 2013) on firm value, the role of the board of directors on international behaviour (Herrera-Echeverri et al. 2016), and how the business environment and corporate governance mechanisms shape the organizational behaviour (Gaitan et al. 2018).

Cluster III (blue) labelled Global Strategies and Local Environment contains seven authors. Scholars in the clusters have investigated the role of the local economic environment on the internationalization performance of the firms (Cuervo-Cazurra et al. 2018); the nature and behaviour of Multilatinas organizations (Aguilera et al. 2017); and the evolution of the internationalization and capacities of Latin American organizations to compete abroad (Carneiro and Brenes 2014). Cluster IV (yellow) labelled Internationalization Capabilities and Export Performance contains seven authors. Scholars in the clusters have explored how international networks could facilitate the reach on number of markets (Felzensztein et al. 2015); how firms implement generic strategies in the market (Brenes et al. 2014); and how newly born international firms use internal capabilities, innovation strategies, and corporate advantage to increase their export performance (Martin et al. 2017).

Cluster V (purple) labelled Internet Marketing and Technological Capabilities contains six authors. Scholars in the clusters have explored how consumer’s online purchase is affected by the perceived risk and trust of the firm (Bianchi and Andrews 2012), how Internet marketing capabilities influence export market growth (Bianchi and Mathews 2016); and how managerial and technology-related capabilities affect the international performance of SMEs (Bianchi et al. 2017). Cluster VI (light blue) labelled Networks, Collaboration, and Regional Clusters contains six authors. Scholars in the clusters have explored how geographical proximity influences the marketing cooperation between firms (Geldes et al. 2015); how clustered-based firms generate cooperative marketing strategies (Felzensztein et al. 2014); and how geographical proximity contributes to the building of social capital on industrial clusters (Geldes et al. 2015).

Cluster VII (orange) labelled Organizational Enabling Factors and Decisions contains four authors. Scholars in the clusters have explored how organizations decide about new markets expansion (Magnani et al. 2018); how sources and resources for innovation are related to innovation outcomes (Seclen-Luna et al. 2020), and how research organizations responded to an event disruption on their business model (Cordova et al. 2021); Cluster VIII (brown) labelled Organizational Reputation and Attractiveness contains seven authors. Scholars in the clusters have explored the factors that impact the foreign operation (i.e., talent acquisition) of firms in the Latin American region (Newburry et al. 2014); how home-country economic openness impacts the organizational reputation (Borda et al. 2017); and whether organizations should adapt their human resources management practices to cross-cultural differences (Figs. 5 and 6).

Fig. 5
figure 5figure 5figure 5

Main component of the Co-authorship network analysis at the author level. Authors with at least four papers. Note: Created using VOS Viewer version 1.6.17 (N = 1940)

Full size image

Cluster IX (pink) labelled Business Groups and Economic Elite contains four authors. Scholars in the clusters have explored how multinationals in key economic sectors employ connections to legitimate their activities (Bucheli and Salvaj 2018); how government relationships with foreign multinationals will depend on the host country's political strategy (Bucheli and Aguilera 2010); and why business groups in emerging countries prevail after pro-market economic reforms (Bucheli et al. 2019). Cluster X (light pink) labelled Entrepreneurship and Regional Development contains three authors. Scholars in the cluster have explored the relevance of the national level of entrepreneurship on the development of the nations (Amorós et al. 2012); what set of economic conditions are needed to foster entrepreneurship in peripheral and central cities (Amoros et al. 2013); and what factors determine the international behaviour of new ventures (Amoros et al. 2016).

Cluster XI (light green) labelled Indigenous Entrepreneurship contains three authors. Scholars in the cluster have analysed the cultural context of Indigenous entrepreneurship (Tretiakov et al. 2020); and how the political and economic environment shapes Indigenous entrepreneurial behaviour (Macpherson et al. 2021). Lastly, Cluster XII (grey) labelled Travel and Tourism contain two authors. Scholars in the cluster have analysed the factors that increased a country's total quantity of visitors (Raventos 2006); and how the airline industry evolved in Central America (Raventos and Melgarejo 2016).

5 Concluding remarks and limitations

5.1 Implications, empirical gaps and further research opportunities

Employing a performance analysis of the literature plus three complementary bibliometric methods (i.e., co-occurrence, co-citation, and co-authorship network analysis), we have provided an overview of this field of inquiry. The results pointed out the productivity trends (Sect. 3.1), the main journals (Sect. 3.2), the leading research organizations (Sect. 3.3), the prominent authors (Sect. 3.4), and the most-cited articles per period (Sect. 3.5). Besides, we have delved into the longitudinal evolution of the conceptual (Sect. 4.1), intellectual (Sect. 4.2), and social (Sect. 4.3) structures of the field. In sum, the analysis allows us to understand the field and to recognize potential avenues of research employing the Latin American context for theory building and testing.

Based on the cluster analysis of each bibliometric network, we have identified several research opportunities for management and organizational scholars. Table 19 provides an overview of possible research topics that can expand each cluster observed on the co-occurrence (see Sect. 4.1.2) and co-authorship networks (see Sect. 4.3.2). To motivate the theory building and testing, we have pointed out an exemplar paper that could be used as a guideline for further theorizing using the Latin American context.

Table 19 Potential research avenues using the Latin American context
Full size table

Lastly, further research could also develop our bibliometric analysis including additional databases (e.g., Scopus and Google Scholar), types of academic documents (e.g., book chapters and conference papers), as well as non-English articles. Moreover, further projects could use a framework analysis (e.g., Antecedents, Decisions, and Outcomes framework), and a thematic in-depth literature review could provide additional insights into this research area. Besides, for a more detailed analysis, a domain review (Lim et al. 2022) on a specific topic with a narrower scope would be desirable (e.g., focusing on organizational networking, resource mobilization, or institutional effects for organizations in Latin America).

5.2 Conclusions and limitations

Although the Latin American region has attracted a great amount of interest among management and organizational scholars, there is still a potential set of fields that could use the context as a “natural laboratory” for theory building and testing. We contribute to the literature by conducting a state-of-the-art subject review. Our research answer how the research about organizations in the Latin American context has evolved and how could it move forward. In doing so, we delved into the evolution of the subject by reviewing the leading research organizations, the most prominent journals, and the main authors that have contributed to this subject during period 2004–2021.

Several conclusions can be drawn from our performance analysis. First research using the Latin American context gained momentum and achieved recognition from 2011 onwards. Besides, the Journal of Business Research is the most prominent journal with 114 publications. Also, the University of Sao Paulo, Brazil, is the most productive university with 110 publications. Moreover, Christian Felzensztein is the most productive researcher with 19 publications. Lastly, Giuliani and Bell (2005) published in Research Policy the most cited article between the years 2004–2009 with 640 citations, Bustos (2011) published in The American Economic Review the most cited article between the years 2010–2015 with 515 citations, and Williams and Shepherd (2016) published in Academy of Management Journal the most cited article between the years 2016–2021 with 131 citations.

In addition, several conclusions can be drawn from our bibliometric analysis. First, the conceptual structure of the field presented dissension given the proliferation of new research topics that used the Latin America region as a context for theorizing. Besides, concepts such as innovation, entrepreneurship, and strategy were central in the network structure. Also, the clustering analysis suggests the existence of five clusters: (I) Development, Entrepreneurship, and Institutions, (II) Internal Capabilities and Outcomes, (III) Strategies and Contexts, (IV) Governance and Business Performance, and (V) Organizational Development.

Moreover, the intellectual structure of the field presented a less sparse connection between the nodes and a gradual trend to novel intellectual conversations. Multilateral economic research organizations such as the World Bank and the Organization for Economic Co-operation and Development (OECD) lead the intellectual conversation. Furthermore, the community is employing novel methodological approaches as a source to explore the complex organizational phenomena in the Latin American context. Besides, we observed an increasing interest of the community regarding how organizations generate and implement strategies. Lastly, the clustering analysis suggests the existence of four clusters: (I) Business and organizations methods and foundational theories; (II) International Business; (III) Corporate Governance and Financial Econometrics; (IV) International Trade, Economic Dynamics, and Policy Evaluation.

Lastly, the social structure of the field presented a structural fragmentation process between the 2004 and 2021 period. Besides, we observe a change in the central authors from 2010 onwards. A new generation of scholars positioned themselves in the community using the regional context as part of their research design in fields such as entrepreneurship, strategic management, and public policy evaluation. Lastly, the clustering analysis suggests the existence of twelve clusters: (I) Innovation and Organizational Performance; (II) Corporate Governance and Disclosure; (III) Global Strategies and local environment; (IV) Internationalization Capabilities and Export Performance; (V) Internet Marketing and Technological Capabilities; (VI) Network, Collaboration, and Regional Clusters; (VII) Organizational Enabling Factors and Decisions; (VIII) Organizational Reputation and Attractiveness; (IX) Business Groups and Economic Elite; (X) Entrepreneurship and Regional Development; (XI) Indigenous Entrepreneurship; (XII) Travel and Tourism.

This research is not without limitations. First, we have just considered peer-reviewed publications and excluded books and other types of documents. Besides, our sample is limited to English-language journals. Moreover, we followed prior bibliometric analysis and omitted data from other repositories to enhance the quality of the review. Also, we focused the in-depth analysis in the top publication on each period and cluster, excluding documents with limited impact. Second, as the co-citation analysis assigns weights to each article, our rankings are likely to be biased towards older articles, since they are more likely to be cited than recent articles. Therefore, our analyses could be biased in favour of older articles and potentially lead to an underestimation of emerging and recent trends in the field. Likewise, the keyword co-occurrence analysis does not capture the meaning of the keywords, it is possible that some articles were associated to the same cluster even if those had a different purpose and orientation. Finally, our review is deliberately very descriptive, a more detailed analysis is beyond the reach of this sole project (see Fig. 6).