Institutions and Firms’ Performance: A Bibliometric Analysis and Future Research Avenues

International business scholars have recognized the importance of the contextual embeddedness of firms. However, how they matter remains a contested question. Although recent efforts have been made to review the field, it remains unclear how institutions affect firms’ performance. We aim at answering the following research question: How is the intellectual and the conceptual structure of the institutions and firms’ performance field defined? We searched in the WoS and Scopus databases with pre-determined keywords, and we obtained a sample of 1063 articles that we analyzed by conducting the citation and co-citation analyses, keyword co-occurrence analysis, and thematic map analysis. Our bibliometric results portrayed how the intellectual and conceptual structure of the field has evolved. We contribute to the international business literature by providing a one-stop overview of the field, thus identifying current accomplishments and future research avenues on the relationship between institutions and firms’ performance. By analyzing the articles included on the Emerging and Niche clusters, we discuss future research avenues on the topics of sustainability, entrepreneurship, political ties, and institutional quality.


Introduction
The role of institutions guiding firms' behavior and success is one of the core issues in the international business literature [1]. International business scholars have recognized the importance of the contextual embeddedness of firms [2]. However, how they matter remains a contested question [3]. Unfortunately, differences in the conceptualization of institutions generated some confusion on how institutions affect global strategy and firms [4]. Although recent efforts have been made to clarify the institutional strands [2][3][4] and review the institutional field, it remains unclear how institutions affect firms' performance [5].
Several prior reviews and meta-analyses have examined and reviewed the institutional field, studying, for example, institutional voids and resource commitment [1], institutions and international strategy [6,7], institutions and location choice [8], and institutions and ownership strategy [9]. Only a few consider the performance aspect of firms, such as internationalization performance [10], MNEs' multinationality performance [11], and family firms' performance [12]. However, these studies either consider institutions as a moderator or ignore the effect of institutions on a firm's performance, thereby leaving a gap in our understanding of how institutions affect firms' performance. Therefore, we seek to answer this research gap.
Mainly meta-analyses were conducted to review the field. Thus, our study has taken a step further by offering a different but complementary approach, namely bibliometric research. We argue that bibliometric analyses are particularly relevant for several reasons: (1) Bibliometric research provides a transparent process of disentangling existent knowledge by minimizing bias through an audit trail of authors' procedures and decisions [13]. (2) Bibliometric research summarizes large quantities of bibliometric data, including theoretical and quantitative research [14]. (3) Bibliometric research provides a one-stop overview of the field to endeavor current accomplishments and future research avenues [14].
We follow a recent call that international business scholars need to assess how institutions influence firms and to rethink the methodological approaches [3]. By considering both the home-and host-country effect, we seek to answer the following research question: How is the intellectual and the conceptual structure of the institutions and firms' performance field defined? We followed [13,14] guidelines for conducting a bibliometric analysis. Our results, using a sample of 1063 articles from both WoS and Scopus databases, portrayed how both the intellectual and conceptual structure of the field has evolved, current accomplishments, and future research avenues. By analyzing the articles included on the Emerging and Niche clusters, we discussed future research avenues on the topics of sustainability, entrepreneurship, political ties, and institutional quality.
We contribute in several ways to the international business and bibliometric research literature. First, we contribute to the international business literature by providing a onestop overview of the intellectual and conceptual structure of the field, thus identifying current accomplishments on the relationship between institutions and firms' performance. Second, we contribute to the international business literature by identifying emerging and niche avenues for future inquiry on the relationship between institutions and firms' performance. Third, we contribute to the bibliometric research literature by combining two important databases, namely WoS and Scopus, following the recent call for joint use of the two databases [15]. Therefore, we contribute to the international business literature by offering an up-to-date depiction of the knowledge base and assessing topic tendencies across the field.
The typical structure of a stand-alone systematic literature review includes an introduction/background, a method section, a results section, and a discussion and concluding remarks [13]. Therefore, we provide a background of what we already know in terms of institutional conceptualization. Then, we present the method section, which includes a brief explanation of the bibliometric procedures. The results follow in the second part. Finally, in the third part, we conclude by discussing the results, by providing an overview of the knowledge held and pointing out future research avenues.

Background
Recent research [2][3][4] conceptualized institutions under the following institutional strands: Under the sociology-based approach, institutions are defined as social structures including regulative, cultural-cognitive, and normative elements [16] that provide stability and meaning to social life [2,3]. Institutions determine what is legitimate, meaning what is an acceptable and approved way of conducting certain actions [2]. Thus, firms strive to be legitimate and to conform with granted norms [17]. The increased pressure to achieve legitimacy [18] results in isomorphism, that is, imitation of behaviors that are perceived as widely accepted [4,19]. Therefore, institutions are viewed as inducing mechanisms towards isomorphism [4].
Under the politics-based approach, institutions are defined as the formal and informal procedures and norms embedded in the organizational structure of the political economy [3]. It reflects the processes by which the interest groups express and negotiate their desires with decision makers, to implement norms and regulations they consider beneficial [4]. In this case, the state plays a role as a biased broker among interest groups [3]. The quality of institutions is dependent on representing the interests of the majority and facilitating the achievement of their goals [4].
Under the economics-based approach, institutions are defined as "humanly devised constraints that structure human interaction" [20] (p. 3) divided into formal and informal "rules of the game" [20] (p. 1). In this case, we may discuss the role of the government in the economy, considering the laissez-faire and interventionist approaches [4]. Laissez-faire sees a limited role of the governments, while interventionist argues that governments must take an active role [4]. Considering a more liberal approach, the quality of institutions depends on facilitating market transactions, which can be implemented with relative speed through programs of pro-market reforms [4]. The emphasis in institutional economics relies on the quality of institutions and the degree of institutions supporting effective economic activities [2].
Under the management-based approach, institutions are defined as a framework in which firms may react, adapt, and overcome through learning mechanisms and strategic decisions [4]. The management-based approach focuses on the actions that managers use to steer away from their firms within prevailing institutions [4]. It highlights the importance of context beyond the idea of institutions as setting the scene [3]. In this case, firms may develop an institutional advantage when they are more adept than others at understanding institutions [4]. Firms' success depends on how firms build institutional capabilities as they operate within a particular institutional framework [4].
Under the comparative-based approach, institutions are defined as a set of interdependent institutional arrangements in different areas of socio-economic life that define a given country [2]. The different dimensions of a country's institutional environment are seen as complementary and in combination with each other [2]. The emphasis relies on the differences between configurations of institutional frameworks [2]. Hence, the main advantage is the possibility to capture cross-country differences, not of degree but kind [2].
Independently of the institutional strand considered, studies need to establish the standards used to assess the institutions [4]. Institutions can be assessed through their quality, strength, voids, or distance. Quality refers to institutions that have better or worse characteristics with an explicit standard [3]. Strength refers to an implicit standard of the level of control on the behavior of individuals and companies [3]. Voids refer to countries either having institutions or not [3]. Distance refers to the differences between institutional profiles of two countries [2]. Therefore, the proliferation of definitions and operationalizations has raised a concern about the tightness and rigor of the institutions' construct [2].

Data Collection and Sample
Our sample was built by searching WoS and Scopus with keywords suggested by previous research. We chose WoS and Scopus databases for their comprehensive coverage and indexing of important journals [1], thus a search in just one database would have excluded relevant journals and articles [21]. Keyword families provide an effective way of analyzing keywords that may have been grouped by being close concepts [22]. We analyzed [22], a review of corporate diversification. Their list of keywords, which contains a total of 64 keywords related to "Institutional-issues" and 132 keywords related to "Performance", was the base for the search terms we used. Furthermore, we browsed articles' keywords on institutional topics and iteratively combined various keywords to search the databases [1].
After collecting the articles from WoS and Scopus that met the criteria we removed the duplicates [13], following [15]'s procedure for merging WoS and Scopus databases into a single file, using the R software. A total of 415 articles were removed due to duplication. Therefore, the merged database resulted in 1298 total articles. The studies that do not yield any substance to the research may also be removed [13]. We manually scoped each article, by verifying each title and abstract and, when needed, the article itself [13,23,24]. Manual scoping of each article may safeguard against misclassified articles included in the search results [23]. The studies eligible for inclusion in the sample had to meet the following criteria: directly related to the topic of institutions and including an indication of firms' performance. Therefore, a total of 235 articles were considered misclassified (e.g., [25], which analyzes the environmental impact of palm oil plantation and includes institutional theory in the abstract). The final sample had a total of 1063 articles for further analysis.
The final sample was also normalized for clarity purposes through a manual cleaning of the records [13]. We manually normalized the articles' references list, i.e., cited references used by the authors in our sample. Citation and co-citation analysis require cited references to be in the same format, as they may be considered by the software as two different articles. Moreover, we manually normalized authors' keywords (e.g., RBV and resource-based view). For purposes of visual clarity, authors' keywords that have the same meaning should be considered as a single keyword.
From the 1063 articles in the final sample published between 1983 and 2020, we extracted relevant bibliometric data. The time span displays no restrictions, meaning that the first article on the topic was published in 1983. We extracted information related to the title, authors, publication date, journal, keywords, cited references, and others. This procedure [23] (Table 1).

Procedures
The data was analyzed using the Bibliometrix package from R software. R provides several packages for different types of research. However, Bibliometrix is particularly suitable for science mapping [26], and it has been used in a growing number of publications [13]. Therefore, we performed citation and co-citation analyses (intellectual structure), the keyword co-occurrence network, and the thematic map (conceptual structure) using the Bibliometrix package from R software.

Citation and Co-Citation Analysis
A citation refers to the author's recognition of prior published works [27]. Authors explicitly recognize an intellectual debt through citations [27]. We performed a citation analysis to identify works that had a major impact, that were inspired on or were drawn upon when conducting focal research [23,24,27]. Using the reference list of the 1063 articles of our sample, we computed the 25 most cited works. The list was computed using Local Citation Score (LCS), i.e., the number of times a reference was cited in the articles from our sample [13]. Therefore, the citation analysis may help portray the knowledge base of the institutions and firm's performance topic, by identifying leading works, approaches, and theories.
Co-citation analysis provides insights into the intellectual structure of a field by exploring how works and theories interconnect [23,27]. Co-citation analysis inspects the reference list of the articles in the sample to identify and count the frequency of using simultaneously two given works [23,24,27], thus identifying their interrelationships [27]. The more often two articles are cited together, the stronger their connection [23]. The co-citation network represents each work in a node and the lines represent the strength of the connection between the nodes [23]. The diameter of the circles represents relevancy: a larger circle indicates a reference more often cited [27]. Therefore, the co-citation analysis is based on the proposition that co-cited articles share a bond or conceptual similarity [24].

Keyword Co-Occurrence Network
Keyword co-occurrence network provides insights into the conceptual structure of a field by exploring how keywords interconnect. Two keywords are said to co-occur if they both occur in the same author's keywords list [28]. A stronger relationship is expected between keywords that represent the core issues [29]. Therefore, the keyword co-occurrence network is based on the rationale that keywords share a bond or relationship when they co-occur.

Thematic Map
The thematic map provides insights on the patterns, trends, seasonality, and outliers of research topics, by demonstrating the evolution of topics in different periods [26]. Thematic maps are very intuitive as topics are grouped in four different quadrants, based on their centrality (plotted on the X-axis) and density (plotted on the Y-axis) [30]. The centrality measures the level of connectedness between topics, and thus is significative in a specific domain [30]. On the other hand, density measures the level of development in terms of intra-cluster cohesion [30]. Depending on the quadrant, topics will be classified as Motor Topics meaning topics that are both able to influence the research field and are well developed [30]. Basic Topics mean topics that are transversal for a discipline [31]. Niche Topics mean specialized topics amongst scholars [30] and thus are relatively hermetic and do not influence nor are they influenced by other topics-contrary to Basic Topics that are intertwined in the extant research. Emerging Topics mean topics that are weakly developed or emerging to be further developed [31].

Results
Authorship analysis portrays an illustration that authors have been conducting more research on the topic, thus driving the field ( Figure 2) [24]. We conducted author's name manual disambiguation to overcome possible name ambiguities in the dataset [32]. We identified the USA, China, and United Kingdom as the most prolific countries (Table 1). Our results have shown Mike W Peng, Yuan Li, En Xie, and Garry Bruton as the most prolific authors in the topic of institutions and firms' performance. However, production over time differs across authors, as some authors were more prolific at the early stages than others ( Figure 3). Therefore, Mike W. Peng and Garry Bruton contributed the most at the early stages, while Mark van Essen, Ekrem Tatogulo, and Pursey Heugens produced the most during the most recent stage. Nevertheless, Mike W. Peng is the most prolific author, as his production has been key throughout the years.

Citation and Co-Citation Analysis
The works that are more often cited are likely to be perceived as holding the greatest impact on a field of study [27]. Jointly, the 1063 articles in the sample used a total of 63,747 references. Examining such a large list would be unfeasible [27], thereby the results of the 25 most cited references identified from the reference list of the 1063 papers of our sample are presented in Table 2.
These were the references that had the greatest impact on the field ( Table 2). The most relevant cited references (e.g., top 10) may be grouped in two different categories: institutional conceptualization and strategic choices. On the one hand, we identified the most influential works on institutional conceptualization, mainly sociology-based approach and economics-based approach, being [16][17][18][19][20]33]. The resource-based view [34] was also commonly used jointly with institutional theory. On the other hand, we identified the

Citation and Co-Citation Analysis
The works that are more often cited are likely to be perceived as holding the greatest impact on a field of study [27]. Jointly, the 1063 articles in the sample used a total of 63,747 references. Examining such a large list would be unfeasible [27], thereby the results of the 25 most cited references identified from the reference list of the 1063 papers of our sample are presented in Table 2.
These were the references that had the greatest impact on the field ( Table 2). The most relevant cited references (e.g., top 10) may be grouped in two different categories: institutional conceptualization and strategic choices. On the one hand, we identified the most influential works on institutional conceptualization, mainly sociology-based approach and economics-based approach, being [16][17][18][19][20]33]. The resource-based view [34] was also commonly used jointly with institutional theory. On the other hand, we identified the most influential works on strategic decisions, being [35][36][37]. Hence, based on the citation score, results showed that articles related to institutional conceptualization and strategic choices hold the greatest impact on the field. Note: LCS (Local Citation Score) is the number of times a reference was cited in the articles from our sample; GCS (Global Citation Score) is the number of times a reference was cited in Google Scholar at the end of 2020. Source: Computed with RStudio-Bibliometrix with data retrieved from WoS and Scopus.
The networks (Figures 4 and 5) were drawn using the Kamada-Kawai layout due to symmetric drawings, small number of edge crossings, and congruent drawings of isomorphic graphs [53]. Kamada-Kawai layout is based on the theoretical distance between vertices being related to the geometric distance between them in the drawing [53]. Relevant papers are displayed as nodes [13]. The co-citation connections, shown as arrows, correspond to the intellectual ties between the works [13,27]. In addition, the thickness of the line connecting a pair of works represents the strength of the tie [27]. The diameter of the circles represents relevancy: a larger circle indicates a reference more often cited [27]. Therefore, the results' analysis should focus on identifying nodes drawn near each other and measuring the thickness of the connection between the nodes.
To identify the intellectual ties between works, we decided to focus on the 25 most relevant articles for purposes of visual clarity and brevity (Figure 4 and detailed in Table 2 [13]. We observed two main clusters in the co-citation network, namely the sociology-based approach cluster (blue) and the economics-based approach cluster (red). On the blue cluster, we observed the thickest tie between [17] and [19], as both are seminal articles on the conceptualization of institutions under the sociology-based approach. In addition, [17] and [19] showed a strong connection with [18] as firms may obtain legitimacy by behaving in a socially accepted way. Furthermore, [16] is linked to the sociology-based approach by defining institutions as regulative, normative, and cultural-cognitive pillars. Corporate social responsibility is commonly analyzed under the sociology-based approach [42,45,47,48], such as social environment and performance. As the most relevant methodological procedures, we identified structural equation models [41] and common method bias [43].
On the red cluster, [20] conceptualized institutions under the economics-based approach, i.e., formal and informal constraints, and showed a strong connection with most works. In addition, we observed a strong tie between [20] and the strategic tripod [40], the resource-based view [34], and the agency theory [38]. The economics-based approach has been applied to emerging economies [34,36,39,46,50]. It is worth noting issues such as guanxi [46], institutional voids [34], entry-modes [50], and strategic decisions [36,39,50] applied to the emerging economies. On the other hand, the economics-based approach has also been used to explain the transition economies [34,44]. As the most relevant methodological procedure, we identified multiple regression analysis [49].
Publications 2021, 9, x FOR PEER REVIEW 9 of 21 Note: LCS (Local Citation Score) is the number of times a reference was cited in the articles from our sample; GCS (Global Citation Score) is the number of times a reference was cited in Google Scholar at the end of 2020. Source: Computed with RStudio-Bibliometrix with data retrieved from WoS and Scopus.
The networks (Figures 4 and 5) were drawn using the Kamada-Kawai layout due to symmetric drawings, small number of edge crossings, and congruent drawings of isomorphic graphs [53]. Kamada-Kawai layout is based on the theoretical distance between vertices being related to the geometric distance between them in the drawing [53]. Relevant papers are displayed as nodes [13]. The co-citation connections, shown as arrows, correspond to the intellectual ties between the works [13,27]. In addition, the thickness of the line connecting a pair of works represents the strength of the tie [27]. The diameter of the circles represents relevancy: a larger circle indicates a reference more often cited [27]. Therefore, the results' analysis should focus on identifying nodes drawn near each other and measuring the thickness of the connection between the nodes.

Keyword Co-Occurrence Network
The keyword co-occurrence network involved observing how keywords are used together (i.e., co-occur) in a research field. The interpretation of the results is somewhat like the co-citation analysis. We applied the Kamada-Kawai layout to Figures 6 and 7 [53]. The relationship between two different keywords is identified with an arrow. The thickness of the arrow reflects the strength of the tie. When keywords are placed close to each other and connected with an arrow, they share a relationship and a conceptual meaning. The diameter of the circles represents frequency: a larger circle indicates a theme more often researched [27]. Figure 6 highlighted the centrality of three research topics: institutional theory, firm performance, and institutional environment. These are the core keywords because we are studying the relationship between the institutions and firms' performance. It is worth noting that China and emerging economies also showed a strong tie with the core concepts. Nonetheless, results showed three main clusters in the network, namely the corporate We further analyzed the latest 5 years (2016-2020) to capture novel trends ( Figure 5). On the one hand, we noticed that articles related to multinational corporations' subsidiaries [52], guanxi [46], and resources [51] have lost relevance. On the other hand, we identified the increased importance of articles related to political ties [54,55] and corporate social responsibility [56]. Therefore, up-to-date state of the art is probing into ways of moving beyond the main institutional strands of both sociology-and economics-based approaches, by considering the politics-based approach and including the effect of political ties on the relationship between institutions and firms' performance.

Keyword Co-Occurrence Network
The keyword co-occurrence network involved observing how keywords are used together (i.e., co-occur) in a research field. The interpretation of the results is somewhat like the co-citation analysis. We applied the Kamada-Kawai layout to Figures 6 and 7 [53]. The relationship between two different keywords is identified with an arrow. The thickness of the arrow reflects the strength of the tie. When keywords are placed close to each other and connected with an arrow, they share a relationship and a conceptual meaning. The diameter of the circles represents frequency: a larger circle indicates a theme more often researched [27]. Figure 6 highlighted the centrality of three research topics: institutional theory, firm performance, and institutional environment. These are the core keywords because we are studying the relationship between the institutions and firms' performance. It is worth noting that China and emerging economies also showed a strong tie with the core concepts. Nonetheless, results showed three main clusters in the network, namely the corporate governance' cluster (green), the multinational corporations' cluster (red), and the entrepreneurship' cluster (blue). First, on the corporate governance' cluster, we identified issues such as agency theory, ownership structure, board of directors, and family-owned firms. Second, on the multinational corporations' cluster, we identified issues such as internationalization, corporate social responsibility, and business groups being applied to multinational corporations in developing and emerging economies. Third, on the entrepreneurship' cluster, we identified issues such as strategic choices being applied to the small and medium enterprises, and state-owned firms, in transition economies, which are usually defined as having low institutional quality and high levels of corruption.
Publications 2021, 9, x FOR PEER REVIEW 12 of 21 governance' cluster (green), the multinational corporations' cluster (red), and the entrepreneurship' cluster (blue). First, on the corporate governance' cluster, we identified issues such as agency theory, ownership structure, board of directors, and family-owned firms. Second, on the multinational corporations' cluster, we identified issues such as internationalization, corporate social responsibility, and business groups being applied to multinational corporations in developing and emerging economies. Third, on the entrepreneurship' cluster, we identified issues such as strategic choices being applied to the small and medium enterprises, and state-owned firms, in transition economies, which are usually defined as having low institutional quality and high levels of corruption. In recent years, the state of the art has changed into several different research topics (Figure 7). On the one hand, we identified the decrease in the relevance of topics such as business group, strategic choice, and ownership structure. Thus, the keywords business group, strategic choice, and ownership structure were not represented in the network. On the other hand, we identified the uptrend of sustainability, Africa, and political ties. Therefore, the state of the art is probing into ways of being more sustainable and environmentally friendly, understanding business ties such as political ties, and considering different research settings such as Africa.

Thematic Map
Although the keyword co-occurrence network is prominent at identifying relevancy and relationships, it requires additional evidence to capture emerging trends and future research tendencies. Thematic map may provide an unbiased method of classifying keywords into clusters accordingly to the strength of their internal (density) and external associations (centrality) [31]. In this sense, the upper-right quadrant (Motor Topics) contains mainstream topics [31], topics that are both able to influence the research field and are well developed [30]. The lower-right quadrant (Basic Topics) shows topics that are transversal for a discipline, being able to influence other topics but being weakly internally established [30,31]. The lower-left quadrant (Emerging Topics) highlights topics that are weakly developed or emerging to be further developed [31]. Finally, the upper-left quadrant (Niche Topics) includes specialized topics amongst scholars [30]. Figure 8 highlights how the field (1983-2020) is grouped in terms of research topics. As Motor Topics, we identified the cluster of corporate social responsibility and resourcebased view. In between Motor and Niche topics, we identified the cluster of corporate governance, agency theory, ownership structure, and board of directors. In between Motor and Basic topics, we identified the cluster of China, state-owned firms, and institutional transition. As Basic Topics, we identified the cluster of firm performance, institu- In recent years, the state of the art has changed into several different research topics (Figure 7). On the one hand, we identified the decrease in the relevance of topics such as business group, strategic choice, and ownership structure. Thus, the keywords business group, strategic choice, and ownership structure were not represented in the network. On the other hand, we identified the uptrend of sustainability, Africa, and political ties. Therefore, the state of the art is probing into ways of being more sustainable and environmentally friendly, understanding business ties such as political ties, and considering different research settings such as Africa.

Thematic Map
Although the keyword co-occurrence network is prominent at identifying relevancy and relationships, it requires additional evidence to capture emerging trends and future research tendencies. Thematic map may provide an unbiased method of classifying keywords into clusters accordingly to the strength of their internal (density) and external associations (centrality) [31]. In this sense, the upper-right quadrant (Motor Topics) contains mainstream topics [31], topics that are both able to influence the research field and are well developed [30]. The lower-right quadrant (Basic Topics) shows topics that are transversal for a discipline, being able to influence other topics but being weakly internally established [30,31]. The lower-left quadrant (Emerging Topics) highlights topics that are weakly developed or emerging to be further developed [31]. Finally, the upper-left quadrant (Niche Topics) includes specialized topics amongst scholars [30]. Figure 8 highlights how the field (1983-2020) is grouped in terms of research topics. As Motor Topics, we identified the cluster of corporate social responsibility and resource-based view. In between Motor and Niche topics, we identified the cluster of corporate governance, agency theory, ownership structure, and board of directors. In between Motor and Basic topics, we identified the cluster of China, state-owned firms, and institutional transition. As Basic Topics, we identified the cluster of firm performance, institutional environment, and innovation. As Niche topics, we identified the cluster of corporate social performance, sustainable development, and corporate environmental responsibility. As Emerging Topics, we identified the cluster of emerging economies, business groups, institutional voids, and India and the cluster of financial performance, institutional pressure, and sustainability.
Publications 2021, 9, x FOR PEER REVIEW 14 of 21 tional environment, and innovation. As Niche topics, we identified the cluster of corporate social performance, sustainable development, and corporate environmental responsibility. As Emerging Topics, we identified the cluster of emerging economies, business groups, institutional voids, and India and the cluster of financial performance, institutional pressure, and sustainability. To capture up-to-date results, we provide a thematic map for papers published between 2016 and 2020 ( Figure 9). We identified major topic changes being the topics of emerging economies and corporate governance. On the one hand, the development of the emerging economies topic has increased, thereby being classified as an influential and well-developed topic. On the other hand, corporate governance moved from a specialized topic to be a transversal topic to the field. Therefore, as Motor topics, we identified strategic choice, human resource management, and corporate social responsibility. As Basic topics, we identified corporate governance and China. As Niche topics, we identified institutional quality, profitability, foreign direct investment, and institutional factors. As upto-date Emerging topics, we identified the cluster of firms' ties (e.g., political ties, business ties, managerial ties, and government support) and the cluster of sustainability (e.g., environmental performance, green supply chain management, and sustainability). In addition, we identified India and Africa as emerging research settings. To capture up-to-date results, we provide a thematic map for papers published between 2016 and 2020 ( Figure 9). We identified major topic changes being the topics of emerging economies and corporate governance. On the one hand, the development of the emerging economies topic has increased, thereby being classified as an influential and well-developed topic. On the other hand, corporate governance moved from a specialized topic to be a transversal topic to the field. Therefore, as Motor topics, we identified strategic choice, human resource management, and corporate social responsibility. As Basic topics, we identified corporate governance and China. As Niche topics, we identified institutional quality, profitability, foreign direct investment, and institutional factors. As up-to-date Emerging topics, we identified the cluster of firms' ties (e.g., political ties, business ties, managerial ties, and government support) and the cluster of sustainability (e.g., environmental performance, green supply chain management, and sustainability). In addition, we identified India and Africa as emerging research settings.

Discussion and Concluding Remarks
In this paper, we analyze the extant research on the relationship between institutions and firms' performance, by conducting bibliometric analyses of the articles published between 1983 and 2020 to unfold both the intellectual and conceptual structure of the field. To answer the research question, we conduct the citation and co-citation analyses, the keyword co-occurrence network, and the thematic map analysis. Our results summarize the data to present both the intellectual and conceptual structure of the field, providing scholars the opportunity to encompass the broad picture of the literature. We contribute to the international business literature and bibliometric research literature in several ways. First, we contribute to the international business literature by providing a one-stop overview of the intellectual and conceptual structure of the field, thus identifying current accomplishments. Second, we contribute to the international business literature by identifying emerging and niche avenues for future inquiry. Third, we contribute to the bibliometric research literature by combining two important databases, namely WoS and Scopus [15].
We contribute to the international business literature by taking a step further on understanding the knowledge yield and the current accomplishments on the relationship between institutions and firms' performance. Prior reviews and meta-analyses mainly considered institutions as a moderating effect on phenomena related to the firms' performance [10,11]. Given that understanding the contextual embeddedness of firms remains a contested question [2,3], we argue that there is a gap in our knowledge on the effect of institutions on firms' performance [5]. Thus, our bibliometric analysis complements prior research by further enhancing our understanding of the relationship between institutions and firms' performance.
Our results provide several current accomplishments related to the most used institutional strands and the most centralized and developed topics of the field. On the one hand, based on the top 10 most cited references, we identified sociology-based approach [16][17][18][19] and economics-based approach [20] as the most used institutional strands to explain the differences across firms' performance. In addition, we identified the most influential works on strategic decisions as [35][36][37]. On the other hand, our results provide several Motor and Basic topics. The discussion on corporate governance suggests including the institutional factors to further understand the contextual nature of corporate govern-

Discussion and Concluding Remarks
In this paper, we analyze the extant research on the relationship between institutions and firms' performance, by conducting bibliometric analyses of the articles published between 1983 and 2020 to unfold both the intellectual and conceptual structure of the field. To answer the research question, we conduct the citation and co-citation analyses, the keyword co-occurrence network, and the thematic map analysis. Our results summarize the data to present both the intellectual and conceptual structure of the field, providing scholars the opportunity to encompass the broad picture of the literature. We contribute to the international business literature and bibliometric research literature in several ways. First, we contribute to the international business literature by providing a one-stop overview of the intellectual and conceptual structure of the field, thus identifying current accomplishments. Second, we contribute to the international business literature by identifying emerging and niche avenues for future inquiry. Third, we contribute to the bibliometric research literature by combining two important databases, namely WoS and Scopus [15].
We contribute to the international business literature by taking a step further on understanding the knowledge yield and the current accomplishments on the relationship between institutions and firms' performance. Prior reviews and meta-analyses mainly considered institutions as a moderating effect on phenomena related to the firms' performance [10,11]. Given that understanding the contextual embeddedness of firms remains a contested question [2,3], we argue that there is a gap in our knowledge on the effect of institutions on firms' performance [5]. Thus, our bibliometric analysis complements prior research by further enhancing our understanding of the relationship between institutions and firms' performance.
Our results provide several current accomplishments related to the most used institutional strands and the most centralized and developed topics of the field. On the one hand, based on the top 10 most cited references, we identified sociology-based approach [16][17][18][19] and economics-based approach [20] as the most used institutional strands to explain the differences across firms' performance. In addition, we identified the most influential works on strategic decisions as [35][36][37]. On the other hand, our results provide several Motor and Basic topics. The discussion on corporate governance suggests including the institutional factors to further understand the contextual nature of corporate governance problems [57][58][59][60][61]. The research on human resource management (HRM) is reviewing the institutional forces on HRM practices (see [62] for a detailed systematic literature review and future research avenues on the topic). The research on internationalization is discussing the contingencies guiding firms' internationalization [63], such as technology-based barriers [64].
We contribute to the international business literature by identifying avenues for future inquiry on the relationship between institutions and firms' performance. We argue that thematic maps provide a transparent process of disentangling future research avenues, such as Niche topics that may hold theoretical contributions to the field due to low centrality and Emerging Topics that may provide scholars with underdeveloped topics. Our results provide three Emerging clusters: (1) sustainability, (2) entrepreneurship, (3) political ties and one Niche cluster: (4) institutional quality.
The literature on sustainability defines corporate sustainability as a combination of environmental, social, governance, and economic dimensions often called the quadruple bottom line of sustainability [65]. Although largely inconclusive, there is an ongoing discussion on whether it pays to be green [66]. This means that whether corporate environmental responsibility results in superior financial performance remains a contested question [67]. However, the most pressing question is not whether it pays but instead when and under what circumstances [66]. It is equally significant to know how corporate sustainability is influenced by the institutional context embeddedness [67], as the sustainable business practices between economies are far from a mature research topic [65].
The literature on entrepreneurship is discussing how complex institutional environments affect entrepreneurship around the world. Social entrepreneurship, female entrepreneurship, and informal entrepreneurship are a few phenomena that may require further attention. First, at the bottom-of-pyramid economies, institutional quality positively moderates the relationship between the social impact and financial performance [68]. Second, in the more hostile environments, female entrepreneurship faces unique challenges that can affect negatively the business [69]. Third, in India and Turkey, firms that delay their business registrations are seeing an increase in their financial performances [70,71].
The literature on political ties proposes that government affiliation enhances firms' profitability and innovation [72]. Adequate support, favorable policies, privileges, and protection against unlawful behavior are a few benefits of firms' political ties [73]. However, as the institutional environments alter, the benefits of political ties may change accordingly [74]. For example, political ties are more useful under low legal enforceability [73] and inefficient property rights [72]. Nonetheless, recent research is arguably over-considering the effect of political ties under inefficient institutional environments, especially emerging economies [75][76][77], thereby leaving the question of whether political ties lose strategical value as institutional environments improve [78]. Contradictory perspectives propose that, despite the institutional transition in China, the state is still considered a powerful player [74]. As the institutional environment evolves, what matters is the social exchange between the firms and governments, as firm government ties move towards prosocial activities and contributions rather than proself [78].
The literature on the niche topic of institutional quality states that our understanding of how institutions influence firms' performance is rather limited [5], especially as the effect of pro-market reforms on firms' performance has been mixed thus far [79]. One theoretical explanation is that institutional change leads to rent distribution, meaning winners and losers [5]. Based on recent research, the mixed findings may be due to: (1) the progression of the institutional change and (2) contingent on the firms' ownership. On the one hand, in the initial stage, the institutional change imposes greater uncertainties and increased transaction costs, while as the reform process progresses, formal rules and regulations start providing greater support in the market economy [80]. On the other hand, institutional development tends to benefit the performance of domestic firms and familyowned firms as opposed to foreign firms and nonfamily-owned firms respectively [5,81]. In addition, new entrants and incumbents respond differently to the institutional environment development [82].
Third, we contribute to the bibliometric research literature by combining two important databases, namely WoS and Scopus. We followed the recent call for joint use of the two databases [15]. Scholars mainly assess one or the other [23,24,27], and only a few use both [1,83]. WoS provides great coverage across years and a varied range of information [15]. The major advantage of the Scopus database is its greater publication range [15], in terms of journals, languages, and sources (e.g., conference proceedings and books). Nonetheless, there are relevant journals included in just one of the two databases or only partially included in one of them [21]. Thus, a search in just one database would have excluded relevant journals and consequently many important articles and authors [21].

Future Research Avenues
We provide future research avenues from our critical analysis of the obtained results. First, the conceptualization of institutions under both politics-and comparative-based approaches remain relatively unexplored. On the one hand, the politics-based approach highlights how interest groups express and negotiate their desires to get these implemented into the laws and regulations [4]. Thus, the ability to achieve each firms' interest may depend on their relationships with policymakers and decision makers. Our results corroborate to describe that both the intellectual and conceptual structure of the field is probing into ways of understanding firms' connections, i.e., political ties, government support, and business ties. Therefore, future research may find it relevant to further understand how firms' insidership [84] may help firms overcome institutional disadvantages.
On the other hand, the comparative-based approach emphasizes the system of interdependent institutional arrangements [2]. Thus, the main advantage of this approach is allowing capturing cross-country differences, not of degree but kind [2]. This goes in line with a recent line of research calling for thick approaches, i.e., methods that examine the interrelationships of institutions in terms of their complementary substitutive effects [3]. In addition, the key comparative theoretical ideas behind [85]'s work have not been sufficiently developed and employed in subsequent international business research [2]. Therefore, future research may find it interesting to further understand the interrelationship between institutions, by applying thick approaches under the comparative-based institutional strand and those effects on firms' performance.
Second, there has been a considerable focus on China, India, Latin America, and the US and Canada as the research setting [79]. However, based on the thematic map, our results contradict prior results by identifying India as an emerging research setting. We corroborate to describe that China, Latin America, and emerging economies are the most used research settings. While defining Africa and India as emerging research settings. Even though pro-market reforms started in the 1970s and 1980s for regions such as Europe [79], little attention may have been given to Europe as research a setting. Therefore, future research may find it useful to understand how institutions affect firms' performance in other regions such as Europe and Africa.
Third, following the analyses of the thematic maps, we propose several research avenues for future inquiry. Although inconclusive, there is an ongoing discussion on whether it pays to be green [66]. However, the most pressing question is not whether it pays but instead when/under what circumstances [66]. In addition, social entrepreneurship, female entrepreneurship, and informal entrepreneurship are few phenomena that are facing unique challenges regarding the complex institutional environments. The ongoing discussion may be inspiring scholars and policymakers to identify mechanisms to address these challenges [68], as business success is imperative to countries' recovery [69]. Future research may probe into examining the nation-level background governance conditions that may influence corporate sustainability and international entrepreneurship.
As the institutional environments alter, the benefits of political ties may change accordingly [74], leaving the question of whether political ties lose strategical value as institutional environments improve [78]. However, contradictory perspectives propose that the state is still considered a powerful player [74,78]. Future research may further understand how the firm-government relationship is shaped by the institutional environment, rather than if political ties lose value over time. On the other hand, the effect of pro-market reforms on firms' performance has been mixed thus far [79]. The mixed findings may be contingent on (1) the progression of the institutional change and (2) the firms' ownership. Future research may examine how the timing and speed of institutional development affect the complex relationship [81] and how institutional quality affects different types of firms, as firms' profitability may not be uniform across firms.
Fourth, there are relevant journals included in just one of the two databases (WoS and Scopus) or only partially included in one of them [21]. Conducting bibliometric analyses in just one database would have excluded relevant journals and consequently many important articles and authors [21]. Thus, merging WoS and Scopus provides greater coverage. As this information is valid for any field, future research may consider using both databases as they complement each other [15,21].

Limitations
There are some limitations to this study. First, we collected exclusively peer-reviewed articles published in journals classified as business/management outlets [23,24]. However, other journals also publish institutions and firms' performance research, such as disciplinary journals in economics and finance. Yet, research on economics and finance tends to be quite hermetic and arguably seeks little contribution outside the core discipline [27]. On the other hand, we do not expand to other sources of knowledge, such as theses and conference proceedings [23]. In addition, we do not account for the differences between British and American English in the list of pre-determined keywords. Nevertheless, we are confident that the articles collected are a representative sample of the research on institutions and firms' performance. Future studies may extend the number of journals sampled and prospect into different disciplines [27] while extending to other sources of knowledge [23].
Second, citation and co-citation network mapping requires the researcher to select a minimum citation threshold for including publications into a network [13]. Thus, this threshold is most likely to exclude more recent publications that have not met the minimum number of citations, and thus biases the display of citations to older papers [13]. However, by conducting analyses for different periods, we believe that we may have identified more up-to-date topics. Nonetheless, future research may examine recently published citations for both empirical and conceptual development of the field.
Third, while our bibliometric analysis methodology seems reasonable, it is just a proxy that does not preclude the complete reading of the papers to unveil both the intellectual and conceptual structure of the field [27]. Thus, other forms of analysis such as bibliometric content analysis [1] may be able to explore in a greater depth the field. Therefore, future research may use different research techniques to further understand how research on the topic has evolved and detect additional gaps and future research directions.