Culture, institutions and their interlinkages. How are theories of culture and of institutions applied together?

Table of contents

1. Introduction
1.1 Overview
1.2 Research objectives and questions

2. Concepts and measurements
2.1 Culture
2.2 Institutions

3. Research design and methodology

4. Literature review
4.1 Economics
4.2 Historical events and development
4.3 Public sector
4.4 Sociology and gender studies
4.5 International business
4.6 Corporate governance
4.7 Ethics, CSR and reputation
4.8 Finance
4.9 Summary of the results

5. Discussion
5.1 Findings of the literature review
5.2 Future research
5.3 Limitations

6. Conclusions

References

Appendix

Abstract

This thesis aims to review and synthesise existing journal articles investigating the intersection of culture and institutions to answer the question of how and in which areas theory of culture and institutions have together been applied. In the form of a systematic literature review, we include articles that were divided into one of eight categories, namely economics; historical events and development; public sector; sociology and gender studies; international business; corporate governance; ethics, CSR and reputation; and finance. Results show that cultural and institutional theory have the potential to explain many social phenomena. Culture, in this regard, commonly plays a twofold role by influencing social interaction directly as well as indirectly through formal institutions, whereas the same holds true vice versa for institutions. Articles further suggest, for example, that culture and institutions can create multiple stable equilibria, have a mutual feedback effect, and that some cultural traits are under certain circumstances interchangeable with institutions and lead to the same outcomes. The underlying dynamics are in many cases highly complex as cultural and institutional dimensions interact in a variety of ways. Future empirical and conceptual work can build its foundation upon this research synthesis.

Keywords: national culture, institutions, social interaction

1. Introduction

1.1 Overview

Over the last few decades, an increasing scholarly interest in the topics of culture and institutions has developed and led to a stark rise in published related work. Especially since the early 2000s, the two topics have gained significant ground in a variety of academic disciplines and caused an increase of available literature, which is also confirmed by the results of our search process. Cultural or institutional theory fits into several scientific fields ranging from psychology and sociology to law and economics.

Views on culture and institutions vastly differ. While some academics see them primarily as endogenous (e.g., North, 1990), others perceive them as exogenously given and representing the equilibrium of a game (e.g., Greif, 2006). Bowles (1998), on the contrary, views them as exogenous too, although they lead to the creation of endogenous preferences in society. A change of cultural and institutional settings, however, can be both exogenous (e.g., shocks, like revolutions, political coups and annexations of new territory) and endogenous (e.g., slow change over time; Ogilvie, 2007). In addition, research is also still not completely conclusive about which and how variables affect the formation of culture and institutions. Likely effects stem from geographic factors (climate, resources, food, etc.), language, education, war, pandemics and available technology, amongst others (Alesina & Giuliano, 2015). On the other hand, theory of new institutional economics highlights the implication of path dependence for both concepts (North, 1990). Path dependence suggests that history matters for current and future decisions and potential outcomes. Similarly, random events, in contrast to systematic forces, can have significant impacts (David, 1985). It so implies a certain lock-in effect created by prior events which can figuratively speaking, make it challenging to leave an already adopted path even if there exist superior ones. Also, because culture and institutions interact, they create space for multiple stable outcomes (Bowles, 1998). In sum, as these statements suggest, culture and institutions have the potential to significantly influence not only each other but also many other outcomes. Roland (2016) explains this phenomenon in a way that culture and institutions affect values of a collective which further also changes their preferences. Preferences subsequently become apparent during social interactions when they determine behaviour. This then opens the door for various complex interactions, and we¹ will consider many examples for this in the literature review.

Nevertheless, clear and coherent definitions of culture and institutions should not be taken for granted and they might significantly influence implications of research. Definitions play a crucial role, since the terms refer to rather abstract and broad concepts, which are not necessarily mutually exclusive and can thus carry lots of potential ambiguity and cause confusion. One major difficulty lies in the fact that the concepts of culture overlaps with informal institutions (Alesina & Giuliano, 2015). For now, we simply follow our own general definition of culture where we define it as the sum of values, beliefs, thoughts, social norms and traditions shared between members of the same group . Institutions can be understood as “the rules of the game” (North, 1990, p. 1) and are further divided into formal and informal ones. Formal institutions are made up of binding laws, rules, regulations and constitutions, whereas informal institutions refer to unwritten and shared social rules and therefore overlap with the definition of culture. Due to this conceptual overlap, some papers use the term institutions synonymously with formal institutions (e.g. Guiso, Sapienza & Zingales, 2015). They, moreover, intentionally neglect informal institutions, as such are already included in their definition of culture. Authors of other papers (e.g., Chappell & Waylen, 2013) describe their work as exploring the interplay between informal and formal institutions instead of describing it as exploring the interplay between culture and (formal) institutions. The importance of a clear-cut distinction between the concepts was also pointed out by other scholars (e.g., Alesina & Giuliano, 2015) and is the reason for an extra clarification at the end of Section 2 for this purpose.

Considering the above-mentioned implications and uncertainties in science, the aim of this paper is to conduct a systematic literature review on the topic of culture and institutions with a special focus on the interactive forces driving the mechanisms between them. By laying out and adhering to a specified search strategy, we made use of scientific search engines to obtain high-quality search hits. Thereafter, we continued to review the chosen literature, and synthesise and discuss their findings. Focus is laid upon work that explores the nexus of culture and institutions and explicitly disregards literature dedicated to only one of the concepts. Most of the articles investigate cultural and institutional effects together with a third factor. This means we not only cover how the two main pillars of this thesis have mutual effects but also which impacts culture and institutions have on third factors. These potential third factors, which do not belong within the concepts of culture or institutions, might be, for example, economic performance and development (e.g., James, 2015; Keating, 2001; Roland, 2016; Tabellini, 2010; Yerznkyan, Gassner & Kara, 2017), opportunity entrepreneurship (e.g., Petrakis & Kostis, 2014), corporate social responsibility (CSR; e.g., Graafland & Noorderhaven, 2020) or human health (e.g., Lamont & Hall, 2009). Work, which is reviewed, does not depend on the certain academic subject but depends on its scientific quality based on a journal ranking.

The thesis is further divided into the following different sections. The remaining part of Section 1 dwells on the research objectives aimed to be achieved and questions to be answered through the literature review. The underlying theoretical concepts of culture and institutions are outlined in Section 2. It discusses the definitions of these two central terms as well as different possible measurements or dimensions for them. The next section, Section 3, explains the research design and the methodology used to find, categorise, review and synthesise existing journal articles. As part of a systematic literature review, this part comprehensively outlines the process and search steps taken, including search terms, tools and additional criteria. At the core of the paper lies the following literature review in Section 4. It entails eight groups of literature based on their primary theme, which can be either about economics; historical events and development; public sector; sociology and gender studies; international business; corporate governance; ethics, CSR and reputation; or finance. This is followed by a synthesis of the results. Section 5 covers a discussion of the incorporated literature in regard to the research questions and comments on limitations and potential future research. Section 6 concludes.

1.2 Research objectives and questions

The central research objective of this paper is to elaborate on existing scientific journal articles, which investigate the nexus of culture and institutions and can be of either empirical or conceptual nature and includes other literature reviews as well. This is done through applying a systematic literature review to first find and identify relevant pieces of work, then review them, and finally synthesise their findings. Papers included in the literature review all share the commonality of investigating the intersection of culture and institutions. Focus of this thesis lies on the rather broadly defined research question of how and in which areas cultural and institutional theory have been jointly applied in prior literature. Deconstructing this research question, we look at articles to find out how and which theories are applied, how linkages between the two concepts are conceptualised, and which causalities and effects are suggested. In this regard, this paper aims to go beyond the simple statement that culture and institutions matter. This has been shown and mentioned before (e.g., Guiso et al., 2015; Tabellini, 2010), so we will rather concentrate on synthesising the findings of journal articles looking at how the underlying mechanisms work. To answer the overarching research question, we examine literature which we categorised, based on their primary focus, in one of eight groups, namely economics; history; public sector; sociology and gender studies; international business; corporate governance; ethics, CSR and reputation; and finance. Elaborating on the findings and implications of former literature should help build a foundational knowledge. Besides presenting the current status-quo of research concerned with the interplay of culture and institutions, we try to reveal existing research trends and directions. Future research can then take into account this synthesised knowledge and expand on it, be it in the form of empirical or theoretical work.

2. Concepts and measurements

This section starts with an overview of how the concepts of culture and institutions are most often incorporated in empirical studies. The next two sub-sections are then devoted to defining the concepts of culture and institutions and contain a brief overview of possible definitions used in existing papers. Different academic disciplines might also use slightly diverging definitions, concepts and measurements. Economics, for example, commonly adopts a game-theoretic point of view (e.g. Bellow & Bowles, 2013). As such, appropriately defining culture and institutions is no straightforward task, as we will see. Nevertheless, in both cases we also discuss possible dimensions or measurements of them, which also indicates how the two concepts can be more concretely implemented in empirical work. This foundation also helps with gaining a deeper understanding of the subsequently reviewed journal articles. In the end of the section, we ultimately isolate the further use of the two terms in the thesis for more clarity. Empirical papers tend to incorporate cultural and institutional features by transforming them into measurable dimensions to make them mathematically or statistically useable. Fernández (2008) elaborated on this and found three primary ways that have for this purpose been taken by scholars. These include conducting social experiments, incorporating survey data through questionnaires or relying on data from immigration. Each type comes with its own advantages and disadvantages. Through experiments researchers compare the way specific games (e.g., ultimatum game, cooperation games, etc.) are played differently by subjects from various cultural backgrounds, such as from a different nationality or ethnic group. Besides the way the games are played, diverging outcomes of the games are of interest as well. Using survey data, on the other hand, is the most widespread approach to measure effects of culture and institutions. Here, participants' answers are aggregated based on cultural affiliation and then regressed against the specific dependent variable of interest. Scholars can also resort to already existing survey data coming from the World Values Survey (WVS), General Social Survey (GSS), European Social Survey (ESS) and other sources (Alesina & Giuliano, 2015). The third and last approach Fernández (2008) mentions is by observing the behaviour of immigrants from a range of origins but immigrating into the same country. Another way to empirically incorporate culture and institutions is by looking at historical data from cultural or institutional changes, which regularly created natural experiments (Acemoglu & Robinson, 2012). Examples for this are the fall of the Berlin Wall, where Eastern and Western Germany were—after almost 30 years of separation—again unified, or cases of colonisation, where an overseas power began (or stopped) to exert power over less developed nations. In case of colonisation, this mostly led to a sudden and significant change in formal institutions, while cultural aspects of indigenous people remained relatively stable, at least in the short to medium-term. Many scholars, however, are not solely interested in correlation but rather causation (e.g., Stephan, Uhlaner & Stride, 2015). One common challenge, therefore, remains the possibility of omitted variables, which has been addressed by introducing suitable instruments from a variety of sources to control for additional effects and estimate a causal relationship. Furthermore, empirical models usually have to find a trade-off between external and internal validity (Alesina & Giuliano, 2015). While experimental games usually score high on internal validity because of specifically designed treatments and control variables, it is disputable to what degree their findings can be generally applied to the outside world. The opposite is true when investigating the effects of historical shocks on cultural and institutional arrangements. All those measurements enable cross-country research of cultural and institutional distances, where a high distance implies a higher difference between countries (Beugelsdijk, Ambos & Nell, 2018).

2.1 Culture

Culture can be specified in many ways, also depending on the academic discipline. We, therefore, give attention to some recurring definitions used in scholarly literature. To begin with, Hofstede (1991), for example, used the following definition:

Culture is a collective phenomenon that is shared with people who live or lived within the same social environment, which is where it was learned. It is the collective programming of the mind which distinguishes the members of one group or category of people from another. (p. 5)

Another possible conceptualisation of culture, which is often adopted in economic literature, was created by Guiso, Sapienza and Zingales (2006, p. 2), who defined culture as “those customary beliefs and values that ethnic, religious, and social groups transmit fairly unchanged from generation to generation.” Here, culture is understood as part of an intangible heritage, passed along from parents to their children. This is also why another core concept in cultural theory is represented by family values and ties since families in general play a crucial role in an individual's development and value formation. Besides this vertical transmission of culture, there also exists a horizontal transmission process between peers (Roland, 2016). Regardless, the former is generally recognised as the more influential way of how culture is transmitted and also explains why culture changes more slowly over time, compared to formal institutions, which have the possibility to radically change over a short period of time (Roland, 2004). Adopting a game-theoretic point of view, culture can be seen as a focal point leading to a Nash equilibrium (Greif, 1994). Culture, from this perspective, thus forms the common beliefs and actions for several people and influences their interactions. Culture, as a consequence, steers cooperation.

Those presented different definitions are only some examples and are not necessarily mutually exclusive but rather highlight different points of views on culture and can together create a more complete picture of the concept. More generally, we can follow our own broad definition of culture where it can be simply seen as a far-reaching term, standing for the inherent values, beliefs and thoughts, social norms and traditions shared between members of the same group. Distinct cultural groups can further be encountered on multiple levels in society, since culture can be affiliated, for example, with a family, group of friends, corporation, ethnic group or nation. Either way, culture determines social choices and governs behaviour and by doing so, directly shapes preferences. Some scholars explicitly split the term culture and prefer the terms values, beliefs and preferences (Alesina & Giuliano, 2015). We, however, do not distinguish between them for the sake of this paper and rather use the umbrella term culture for a more coherent narrative and simplicity. Preferences and their implication are also of direct interest for all scholars studying social interactions and is, for example, the reason for the emergence of the relatively new field of cultural economics, which tries to incorporate formerly neglected cultural variables into the study of economic activities and outcomes (Ogilvie, 2007).

In the following, concrete traits of culture adopted in theoretical as well as empirical work are illustrated. The measures discussed include Hofstede's (2011) cultural dimensions, trust, morality, family ties and social capital. This is only a non-exhaustive list of cultural aspects as measured by scholars. Still, these examples were chosen since they are of high importance for this thesis because they either account for a wide scholarly adoption overall or were specifically adopted by the reviewed literature in Section 4 and therefore represent the necessary conceptual foundation for the later parts of this thesis.

One pioneering work in the field of culture comes from Hofstede (1991, 2011). He developed a framework with altogether six distinct dimensions to describe and compare cultures. These dimensions are power distance (PDI), collectivism vs. individualism (INV), uncertainty avoidance (UAI), masculinity vs. femininity (MAS), short-term vs. long-term orientation (LTO) and restraint vs. indulgence (IND). Each dimension is measured on a national level on a scale from 0 to 100. First, the index for power distance determines to what extent members of a group accept an unequal distribution of power. Nations scoring high on this trait show a stricter hierarchical order and tend to accept this. Concerning individualism (high score on the scale) vs. collectivism (low score), individualistic countries value self-determination and a looser integration of people into groups, whereas collectivist societies emphasise the “we” instead of the “I” and can be described as having strong ties within groups. Uncertainty avoidance, on the other hand, measures how tolerant a society is for ambiguity. Nations with a relatively high level of uncertainty avoidance tend to create stricter rules and codes, and adhere to them, while the opposite is true for nations on the lower end of the scale. Another cultural trait is expressed by the way a society views themself with respect to the past, present and future, which is accounted for by the dimension long-term orientation vs. short-term orientation. Nations of a low score (short-term oriented) try to maintain old, accustomed traditions and norms, which they highly value. Long-term oriented societies, in contrast, have a tendency for embracing development and change when necessary. Next, on the masculinity vs. femininity scale, masculinity is defined as “a preference in society for achievement, heroism, assertiveness and material rewards for success”, while femininity is defined as “a preference for cooperation, modesty, caring for the weak and quality of life” (Hofstede, 1991). The index ultimately stands for the preferred values of a society in relation to gender. The sixth and last cultural dimension is indulgence vs. restraint. A low level (restraint) describes “a society that controls gratification of needs and regulates it by means of strict social norms”, while in contrast, indulgence illustrates “a society that allows relatively free gratification of basic and natural human desires related to enjoying life and having fun” (Hofstede, 2011, p. 15).

When using Hofstede's (2011) framework, one must note that there are no clear-cut categories for any of the dimensions, but they are rather blurred as the continuous scale suggests. Taking the masculinity vs. femininity dimension as an example, there is no pre-defined score, or range of scores, representing a masculine country per se. If a country is deemed to be rather masculine is only seen when comparing it with another country. The author himself points out how the most important feature of the framework is the possibility to compare countries with each other. Each of Hofstede's dimensions can also be used in empirical work. Articles in the literature review, for example, build upon each dimension with the exception of indulgence vs. restraint.

The next cultural characteristic, trust, is also commonly used in studies (Alesina & Giuliano, 2015). A certain level of trust is needed for ultimately every social interaction or transaction. It can be measured through surveys or experiments as either trust in other members of the same or another social group, or as trust in existing formal institutions. Many differences in this trait have been found between nationalities, ethnic communities and different socioeconomic groups. Guiso, Sapienza and Zingales (2009), for example, found that trust levels between individuals are higher within a nationality than between nationalities. There is also general evidence for members of the same or a similar group trusting each other more compared to members of a distant group. Trust can further be split into generalised and particularised trust, which is also called limited trust by some authors (James, 2015). Trusting only one's close acquaintances and one's family is understood as particularised trust, while its generalised counterpart means trusting the majority of people in the community. Related to trust is also the concept of morality, which can again be divided into generalised and limited morality. James (2015) explains the relationship between trust and morality as trust being the belief about the degree other members of a group can be trusted, while morality determines how trustworthy people in reality are. Groups exhibiting high morality are less likely to behave opportunistically (Rose, 2011). Like trust, morality is most commonly measured with the help of surveys or laboratory games (Alesina & Giuliano, 2015).

Trust and morality are part of the bigger concept of social capital. According to Putnam (1993), this concept refers to the relationships between single members of a group, which are in turn based on and create reciprocity and trustworthiness. The author argues that a high social capital leads to the effective functioning of society and the establishment and maintenance of well-functioning institutions, such as a democracy. Furthermore, he counts factors such as trust, social interactions, cooperation and participation as some of the key components of social capital.

It was also indicated that many cultural characteristics are interconnected. Gorodnichenko and Roland (2015), for instance, argue that the distinction between individualism vs. collectivism influences many other forms of individual and group behaviour and is arguably the most important cultural dimension. There also appears to be a significant correlation among cultural characteristics, such as trust, individualism and family ties, amongst others (Alesina & Giuliano, 2015).

Further approaches to formalise culture come, for example, from Schwartz (1992) or Project GLOBE (House et al., 2004).² Articles in the literature review might also utilise additional cultural traits, which will be further explained accordingly, but the above-mentioned ones account for the majority of studies adopting cultural dimensions in Section 4 and can be seen as a brief, yet important overview of main cultural characteristics.

2.2 Institutions

Institutions, on the other hand, are prominently defined as “the rules of the game in a society or, more formally, [institutions] are the humanly devised constraints that shape human interaction. Thus, they structure incentives in exchange, whether political, social, or economic” (North, 1990, p. 1). These constraints can further be classified into formal and informal constraints, where formal ones are explicitly stated and consist of laws in the legal sense, rules, regulations and constitutions. Contrariwise, implicit social rules and norms together form the informal institutions of a society. The informal constraints consist of socially shared norms, rules and conventions and are unwritten compared to their formal counterpart. Informal institutions can modify formal constraints. North (1990, p. 37) further specifies that formal rules are formed by the polity, while informal ones are “part of the heritage that we call culture”. Here, a definitional blur between culture and institutions becomes further apparent. Compared to formal institutions, which can be fundamentally transformed in a relatively short period of time, informal ones change and develop only very slowly (North, 1997). When using the word institutions, we do not refer to organisations per se, such as businesses, NGOs or international organisations (North, 1990). However, such organisations are themselves subject to and create or influence informal and formal institutions. Both types of institutions have their own enforcement mechanism in place. An example for enforcing formal constraints is through legal courts when they decide on a lawsuit. An example for the enforcement mechanism of informal institutions is committing a social taboo and then being confronted with social consequences or social punishment. A social taboo can be raising a wrong topic to speak about, inadequate table manners or being overly late for an appointment. Both formal and informal institutions interact, although informal institutions are “arguably more primary and deep-seated than formal institutions'' (Crossland & Hambrick, 2011, p. 4).

Institutions can, on the other hand, be split into political or economic and inclusive or exclusive institutions (Acemoglu & Robinson, 2012). Political institutions determine the amount of political power individual groups have and the incentives politicians face. Such institutions are responsible for legislation and the subsequent application and enforcement of the law and have great influence on the economic and social system of a society. With legal titles like property rights, they also influence economic institutions, which are structures governing economic incentives and activities. Economic institutions can take the form of different types of regulated markets, property rights or family allowances, just to name a few instances. Political institutions directly shape economic ones, so the two types are deeply intertwined and together have a great impact on society. Flachaire, García-Peñalosa and Konte (2014) point out that political institutions influence economic development of nations only indirectly through economic ones. Economic institutions, in turn, influence economic growth, development and wealth. Furthermore, there exists a differentiation between inclusive and extractive institutions. Inclusive institutions are categorised by giving individuals the allowance or right to participate in political or economic activities, whereas extractive institutions serve only selected people or groups in power by granting them exclusive rights (Acemoglu & Robinson, 2012). A well- established constitutional democracy, as present in many modern states, is an example for an inclusive political institution. A dictatorship, on the other hand, falls within the realm of extractive political institutions, since political power is concentrated in the hands of a single person or small group.

In the subsequent paragraphs, because informal institutions are conceptionally embedded in culture—as one can see by comparing the section about informal institutions with the section about culture—we only reserved space to address measurements of formal institutions. Exemplary measures of informal institutions are therefore included in the discussion of culture above.

Forms of formal institutions are usually, just as cultural dimensions, transformed into scales and so made measurable for empirical purposes. Concrete observations of formal institutions, which have also been partly adopted by articles in the literature review, encompass indexes related to legal systems (e.g., rule of law, civil law, corporate law, private law, etc.), property rights, corruption levels, bureaucracy and regulation (e.g., market regulation, labour regulation, etc.). The World Bank, for instance, compiled comprehensive data about a range of institutional aspects under the name Worldwide Government Indicators (WGI), which are available for 214 countries over 20 years.³ Included measures describe corruption levels, government effectiveness, political stability and the rule of law, amongst others (The World Bank, 2019). The Ibrahim Index of African Governance (IIAG), as another example, specialises in providing data about overall governance performance for African countries by looking at the quality of different institutions.⁴ Key components include measures about safety and the rule of law, participation and human rights, sustainable economic opportunity and human development (Mo Ibrahim Foundation, n.d.). Beugelsdijk et al. (2018) built their work on other indexes of institutional quality. They use indicators from the Quality of Government Institute (QOG), Index of Economic Freedom and the International Country Risk Guide (ICRG).⁵ All three data aggregators offer a wide range of tools and variables about different formal institutions across countries and years. Additionally, Fainshmidt et al. (2018) developed their own classification of countries based on institutional and also cultural arrangements. The country-specific qualitative data they used for this include measures about state intervention and regulation, labour unions, generalised trust and family ties, amongst others and were primarily obtained from experts in these respective fields. Further existing data about formal institutions have been gathered and prepared by scholars, the OECD and other organisations.⁶

In the case of both culture and institutions, more instances and examples exist and are going to be mentioned in the literature review in Section 4. In other words, some of the articles also include different cultural dimensions and institutional measurements than the ones already discussed. Henceforth, to avoid any ambiguity in the further sections, we ultimately refer to formal institutions when using the term institutions. When referring to informal institutions, we explicitly use this exact wording, but otherwise only refer to their formal counterpart. We, hence, as far as possible put informal institutions under the umbrella term of culture, unless otherwise specified. It seems necessary to make a clear distinction between culture and informal institutions because informal institutions conceptionally reside in the sphere of culture, or vice versa as some scholars suggest. Despite the individual authors of the journal articles regularly using different definitions, we will try to follow our own definitions for a more coherent narrative. The same approach of laying out these concepts was also adopted by other scholars (e.g., Alesina & Giuliano, 2015; Guiso et al., 2015). An uncareful use of the words would be a possible cause for otherwise avoidable confusion.

3. Research design and methodology

This section of the thesis is devoted to the specific research design and the methodology that was used over the course of the literature review. In order to answer the defined research question of how and in which areas theory of culture and institutions have been jointly incorporated in existing research, a systematic literature review seemed to be the most suitable approach. The reason for this is that a systematic literature review is designed to identify all scholarly papers dedicated to a specific research question or topic which fulfil some special criteria. Criteria might be based on different dimensions, such as special focus of the paper, year of publication, quality of journal or its used methodology. A systematic literature review, therefore, represents a special type of literature review where a certain search strategy is defined and then strictly adhered to. Compared to a more general literature review conducted on an ad-hoc basis, the more rigorous methodology minimises the risks of failing to include important studies and the risk of selection bias (Snyder, 2019). In the following, the methodological concept to identify relevant work shall be outlined.

In short, a multi-step approach to find relevant academic literature has been applied. After defining rather generalised key terms and the overall search criteria, Google Scholar was used to obtain a broad range of search hits to begin with. The findings of this first step helped with developing a more systematic and more precise approach to find and identify journal articles to include in the thesis. Based on the search results, a more comprehensive list of terminology to search for has been compiled. Subsequently, we made additional use of the databases EBSCOhost and ProQuest. By considering the advanced search option these platforms offer and combining them with the updated search criteria, we generally obtained good results of scholarly work in terms of quality and quantity. Regardless, in a last step, these obtained results had to be further limited on an individual basis to only consider literature of the right research focus (e.g., only societal culture in contrast to organisational culture), and to not go beyond the scope of this thesis. The remaining literature meeting those criteria was, as a result, classified by their main research focus and incorporated into the literature review. In the following part of this section, we will outline the search process that has led us to the ultimate search strategy in detail.

To find relevant research papers, we set a clear inclusion as well as exclusion criterion in respect to the specific theme or topic that must be covered by the articles. Relevant literature is defined by examining the intersection of culture and institutions. Pieces of work that focus solely on one of the two concepts is the explicit exclusion criterion. Vice versa, this implicates the sole inclusion of literature explicitly assessing culture and institutions, which is also of importance for the further search steps and, more concretely, for defining appropriate search terms. Regarding the time span of relevant literature, we initially did not set any constraints, since it was unclear at that time how many search results we are going to obtain. However, after applying the search strategy by searching for the predefined search terms with the help of the predetermined search tools and considering the finding of a manageable number of relevant literature, it became evident that a subsequent introduction of at time constraint remained unnecessary. We, therefore, argue to have incorporated high-quality research into this thesis without limiting the sources to certain years of publication only. Moreover, we did not actively discriminate against any research from certain academic disciplines as long as they examine effects of culture and institutions. Found literature, thus, stems from a variety of areas, such as economics, business and management, sociology, anthropology and politics. Academic areas are only insofar biased, as the journal ranking—which we used as a guideline to receive high-quality articles—focuses on business, management and economics journals. Regardless, individual articles published by those journals can still be of a wider range of scientific disciplines and are hence sufficiently broad in topics. In this regard, many instances of reviewed literature can also be classified as being multidisciplinary and combining aspects of more than just one subject area. The criterion causing to exclude some otherwise seemingly relevant literature based on the research area is, consequently, the quality of the journal itself. Furthermore, we draw a clear line between societal and organisational culture, and do not review articles with an exclusive bent for organisational or corporate culture. Cultural aspects of organisations differ from both the individual and the culture at the national level and is consequently affiliated with alternative ways of thinking in respect to empirical models and theory (Hofstede, 1991).

We developed a multi-step search process to determine potentially relevant studies. In a first search step, we used Google Scholar to gain a first insight into existing literature on culture and institutions. For this step, no systematic approach was yet taken, as search results from Google Scholar should rather give a broad overview of related work about the topic and should additionally help with creating a structured search strategy for further search tools. Key terms used in this early stage of the search process are quite broad and contained the terms culture, cultural, institution and institutional, which were combined with Boolean operators. Applied queries were “culture AND institutions” and “culture OR cultural AND institutions OR institutional”. No special journals were taken into consideration at this point. Both searches returned nearly the same number of results with about 4.770.000 and 4.880.000 accordingly and revealed the vastness of related literature. Yet, it must be noted that both culture and institutions are very broad terms, used in many contexts and subject areas and are, as a consequence, commonly included in articles. To significantly reduce the number of results, we for this reason had to introduce additional criteria to make the literature review manageable. However, within the realm of Google Scholar remain some challenges which make the online search engine less suitable for a systematic search. For example, one major drawback are the limited options for an advanced search, which are usually implemented through the use of truncations, or also called wildcards (Halevi, Moed & Bar-Ilanc, 2017). On the other hand, Halevi et al. (2017) state that Google Scholar uses a process called automatic stemming, which is similar to using truncation symbols but leads to less accurate search results because truncations can neither be inserted manually nor intentionally. For this and other reasons, Google Scholar only seems to be a good choice for a first quick search, before other academic databases are utilised for a more targeted approach.

After concluding the first search stage with Google Scholar and screening some of the first articles, we expanded the search terms, since we got an impression of further terms related to the initially defined words. Table 1 below contains the extended relevant terms, where included asterisks act as placeholders for different possible word endings. The part preceding the asterisks is the main word stem. Some word stems can, for this reason, be transformed to a noun, adverb or adjective correspondingly (e.g., cultur* can become culture, cultural or culturally, among others). The placeholders can also account for a term in plural (e.g., policy and policies). Furthermore, we took into account how the same term can be written differently depending on the use of British or American English (e.g., generalised morality or generalized morality). Section 2, including the definitions and measures of culture and institutions as implemented in existing studies, helped with formulating the search terms applied in the second stage of the search process. Regarding culture-related terminology, different dimensions, possible measurements and related prominent scholars or studies have been added to the list of search terms. In case of institutions, more specific types of informal and especially formal institutions have been joined. In both cases, adding related words should help in finding all potentially relevant scholarly work and leads us to ten search terms (or groups of terms) for culture and 15 for institutions. Due to the conceptual overlap of culture and informal institutions, some of the search terms could have been allocated to the respective other column too.

Table 1

Abbildung in dieser Leseprobe nicht enthalten

For a more targeted search, we combined the newly created list of search terms with Boolean operators (mainly AND and OR) and other advanced search options outside the realm of Google Scholar. We used the EBSCOhost search engine as well as ProQuest. Although, EBSCOhost provides access to more content, the search engines offer access to different academic library databases. The exact same query in ProQuest also resulted in fewer results compared to EBSCOhost. In terms of advanced search options, they offer, at least for our purposes, the same features. Through the EBSCOhost interface, we selected the two databases EBSCOhost Academic Search Premier and EBSCOhost Business Source Premier.⁷ ProQuest, in contrast, offers access to nine different academic databases. Due to the wide use of the words culture and institutions, the first searches returned results numbering in the many thousands. To sharply limit the number of search hits and to obtain more relevant work overall, we introduced the ranking of the Academic Journal Guide (AJG) 2018,⁸ compiled by the Chartered Association of Business Schools (CABS), for the search strategy. The AJG rates business and management journals based on peer-review, editorial and expert judgements, and information about citations (CABS, 2018). Although the ranking concentrates on journals from the spheres of business, management and economics, individual articles published in these journals can be classified as coming from other disciplines as well or can take the form of multidisciplinary articles. Because we want to focus solely on high-quality literature, this ranking gives us an adequate foundation to constrain the number of found articles. Hence, only journals with a grade of 4* or 4 have been chosen for the review.⁹

Further limiters for an advanced search were the availability of the full text of the article and articles in the English language only. No limits of the publication date have been set. Arising from definitional difficulties as mentioned above, we also took into account how some scholars strictly prefer the term informal institutions over the term culture and came up with another set of queries. Here, we intentionally omitted the word culture and instead used the term informal institutions. Taking into account these limitations and thoughts, the final two queries used in the EBSCOhost and ProQuest search interface can be found in Appendix A. Besides EBSCOhost and ProQuest, another search was additionally conducted in Google Scholar, this time considering the journals too.

Nevertheless, an even more fine-tuned query based on the advanced terms presented in Table 1 was entered in the search engines as well but produced even less results. For this reason, the additionally defined key words were eventually not considered in the last queries. We regarded the number of found relevant papers to separately examine from the preceding queries as perfectly manageable, and this was the reason for further omitting an even more targeted query at this stage. At the same time, we did not have the intention to neglect otherwise relevant literature due to an overly specific search. Less specificity resulted in a possibly higher quality of search results in this case, meaning eventually all articles are based on only the two queries in Appendix A.

In a next step, the sum of resulting articles from all three search engines was then inspected individually to ensure the definite relevance for our purposes. Most of the rejected articles were omitted because of a focus on organisational culture, a general focus on higher educational institutions or the sole concentration on either culture or (formal) institutions. Screening potentially eligible studies was a challenge by itself, mainly due to an abundance of subject­specific technical terms and definitional causes. For example, some authors refrained from using the term culture and instead favoured the term informal institutions, or vice versa. Others widely disregarded the umbrella terms of culture and institutions and instead only concentrated on and named specific examples of them. Through the search process, some articles were also found in more than one database, but finally, this left us with 35 individual articles to have a closer look at in the literature review in the next section of this thesis, thereof seven articles stemming from Google Scholar, 21 articles from EBSCOhost and seven articles from ProQuest. Years of publication range from 1972 until 2020, with the vast majority being published after the year 2000.

While reviewing the relevant papers, we have also classified them into one of the following themes: economics; historical events and development; public sector; sociology and gender studies; international business; corporate governance; ethics, CSR and reputation; and finance. In spite of the available classification of journals from the AJG, individual articles were commonly from a multidisciplinary field or focused on a special topic not necessarily conforming to the AJG's classification. For this reason, the AJG's classification appeared to be somewhat unfitting for our purposes yet acted as a foundation to create or own categories to provide a more coherent narrative based on the shared or similar themes of articles and not based on the overarching research trend of the journals. We regarded creating a new category of articles as appropriate, if at least three articles shared the very same or similar research focus. In the following, we further explain the process of allocating articles to the defined thematic areas.

[...]

¹ In the sections devoted to culture, institutions and their dynamics, we ultimately refer to the author of this paper together with the reader, if we use the word we. In the section where we explain what we did over the course of the thesis (e.g., methodology), we refer only to the author of the thesis when we use the word we.

² See the reference list or https://globeproject.com/

³ See https://databank.worldbank.org/source/worldwide-governance-indicators

⁴ See https://mo.ibrahim.foundation/iiag

⁵ See https://www.gu.se/en/quality-government, https://www.heritage.org/index/ and https://www.prsgroup.com/

⁶ In case of the OECD, see https://www.oecd.org/statistics/listofoecddatabases.htm, where data about formal institutions can be found scattered in multiple databases.

⁷ See http://support.ebsco.com/help/?int=ehost&lang=en&feature id=Databases&TOC ID=Always&SI=0&BU=0&GU =1&PS=0&ver=live&dbs=aphjnh,aph for information about EBSCOhost Academic Search Premier and http://support.ebsco.com/help/?int=ehost&lang=en&feature id=Databases&TOC ID=Always&SI=0&BU=0&GU =1&PS=0&ver=live&dbs=buhjnh,buh for EBSCOhost Academic Search Premier.

⁸ See https://facultystaff.richmond.edu/~tmattson/AJG%202018%20Joumal%20Guide.pdf, for the most up-to-date ranking from 2018.

⁹ CABS (2018) defined journals marked with 4* as “journals of distinction”. These represent the top-tier journals world-wide with the highest scientific standards and subsequent impact. Journals awarded 4 are in the second­best tier and are still regarded as top publishers in their specific field. They are categorised by a high submission rate, a low acceptance rate and their articles being cited frequently.