INSTITUTIONS AND INSTITUTIONAL CONTEXT.doc

THE INSTITUTIONAL CONTEXT OF MULTINATIONAL ACTIVITYMauro F. GuillénThe Wharton School and Department of SociologyUniversity of PennsylvaniaandSandra L. SuárezDepartment of Political Science, Temple UniversityPrepared for the Second Edition ofOrganization Theory and the Multinational CorporationEdited by Sumantra Ghoshal and Eleanor WestneyNew York: St. Martins Press.The comparative theory of institutions is based on the premise that the underlying assumptions guiding economic behavior vary across time and space. As an economic actor, the multinational enterprise (MNE) is exposed to a variety of institutional contexts (Westney 1993), which creates the need for it to diagnose their peculiarities, and decide whether to adapt its strategy and organizational structure accordingly. This chapter reviews the main ways in which scholars have looked at cross-national variations that are relevant to the study of foreign direct investment (FDI) and the MNE. Five main cross-national approaches are identified: cross-cultural, comparative authority and business systems, political economy of foreign investment, comparative corporate legal traditions, and political hazards. The theoretical, methodological, and empirical accomplishments of each approach are systematically compared. INSTITUTIONS AND INSTITUTIONAL CONTEXTUnlike ten or twenty years ago, most social scientists agree today that institutions and institutional context matter when it comes to understanding economic, political, or social behavior by individual actors and organizations. Organizationsincluding MNEsare constrained and supported by institutional forces (Scott 1995:55). MNEs are born into a specific institutional context in their home country whose features MNEs carry with them when expanding abroad (Doremus et al. 1998). And the institutional context in the host country is relevant to understand the level of FDI, the choice of entry mode, and the way in which the MNE is organized and managed. Although institutions are relevant to the study of multinational activity, however, no attempt has been made to review the various ways in which scholars have defined the institutional context.The study of institutions has a long history in the social sciences, going back to at least the second half of the 19th century (Scott 1995). Institutions are sense-making frames that (1) constitute actors as such, and (2) guide their action in appropriate and effective ways towards legitimate and meaningful ends. Institutions refract motives and behavior, goals and means, into meaningful action. Institutions come under many different guises. Laws and other “rules of the game” are highly formalized institutions, imposed by legitimate actors (the state) or powerful organizations. Institutions, however, also include taken-for-granted assumptions, developed through habit and rooted in collective understandings. Formal and informal institutions have the same effect: providing for stability and meaning to economic, political, and social life (Scott 1995:33). Whether formally sanctioned or not, institutions are historically developed and relatively resilient to change. There are three key aspects of institutions that deserve careful attention. First, it is important to underline that institutions are simultaneously constraining and enabling. They constitute actors, guide action, and pose limits on behavior. Choice is informed as well as constrained by the way knowledge is institutionally constructed (Scott 1995:51). The sociological concept of institutions as constituting actors as well as shaping action stands in contrast with the economic view of institutions. Economists tend to see institutions as constraints on behavior, as “the rules of the game in a society the constraints that shape human interaction” (North 1990:3; emphasis added). Given human tendencies to cheat one another, institutions become mechanisms to overcome anomalies, e.g., market failure due to the costliness of measurement and enforcement (North 1990, 1997; Williamson 1985). Institutions, however, are more than just constraints on behavior. They constitute and enable actors to engage in socially meaningful action by making them legitimate and knowledgeable in a given situation. Thus, local law or culture may view foreign firms as intruders whose activities need to be curtailed, or they may see them as contributors to local prosperity. As a result, MNE behavior will differ depending on the nature of the institutions in the host country.Second, institutions shape not only the means of social action but also the ends, that is, preferences are endogenous to the institutional understanding of the world (Scott 1995:51). Institutions affect to what extent organizationsincluding MNEspursue profitability, growth, technological advancement or employee well being as their primary goals. It is important to note that an MNE may be born into a country in which profitability is the overriding goal of firms, and expand into countries in which technological development and employee welfare are also relevant considerations. Third, institutions are embedded in carriers or repositories, which facilitate their continuity over time. Scott (1995:52-55) distinguishes among three main institutional carriers, namely, cultures, social structures, and routines. Cultures are symbolic representations of the world, of problematic social reality (Geertz 1973:220), and they provide not only values and beliefs, but also strategies of action (Swidler 1986). Social structures consist of constellations of roles and positions in which actors operate. They create channels for the articulation and deployment of institutions because actors tend to behave according to the institutions recognized as such in the social structure in which they are embedded (Granovetter 2001). Routines are learned, nonreflective patterns of behavior that provide for continuity, stability, and predictability. Unlike other organizations, MNEs are by definition exposed to two or more cultures, social structures, or sets of routines, one in the home country, and one or more in the host countries in which they operate. Thus, MNEs are exposed to possibly different and potentially conflicting institutional demands (Westney 1993). In organizational terms, the demands of the home-country environment stem from the “imprinting effect” (Stinchcombe 1965), while the institutional conditions in the host country may be conceptualized according to the resource-dependence, neoinstitutional, or transaction-cost perspectives (see Scott 1998; Perrow 1986).When it comes to the study of FDI and the MNE, institutional contexts tend to be defined at the country level of analysis. Rare is the country, however, that contains only one relevant institution. Thus, institutional contexts are more likely systems of somewhat mutually consistent and coherent institutions that have co-evolved over a relatively long period of time. The different approaches that scholars have used to characterize cross-national institutional contexts adopt one of two strategies to accommodate this complexity. Some of them make simplifying assumptions and focus on a relatively narrow type of institutions. Others deal with the complexity of institutional systems head on. The former strategy generally is in a better position to provide testable propositions and quantitative indicators, while the latter is best equipped to offer a more complete, nuanced, and historically meaningful understanding of the phenomenon under examination. The next section compares examples of one and the other strategies for characterizing institutional contexts. FIVE APPROACHESWhile there are many approaches to the comparative study of institutional contexts, only a few have been pioneered by, or influenced, students of FDI and the MNE. Table 1 summarizes the main features of the five approaches considered here, namely, cross-cultural, comparative authority and business systems, political economy of foreign investment, comparative corporate legal traditions, and political hazards. Each approach is systematically compared to the others in terms of its main line of argument, basic concepts, methodology, resulting classification or measurement of different institutional contexts, main research questions related to FDI or the MNE, and leading proponents. Illustrative empirical studies are also discussed.The Cross-Cultural ApproachPerhaps the most influential way of capturing cross-national institutional differences relevant to the study of FDI and the MNE is the one pioneered by Geert Hofstede, a Dutch social psychologist and organizational consultant. His 1980 book, Cultures Consequences, is one of the most widely cited monographs in the management literature. Its empirical analysis is based on 117,000 individual responses from 88,000 IBM employees in 66 countries obtained through surveys at two points in time, 1968 and 1972. He focused his data analysis on 40 countries. In a later book, Hofstede (1991) expanded the analysis of the data to cover 50 countries.Hofstede sees culture as the “collective programming of the mind” (1980:13). His main argument is that culture shapes thinking and action. “People carry mental programs which are developed in the family in early childhood and reinforced in schools and organizations, and these mental programs contain a component of national culture” (1980:11). Since people from different nationalities have different values and perceive things in different ways, designing and managing an organization that spans different countrieslike IBMrequires gaining an understanding of cross-national cultural differences. Hofstedes approach to characterizing institutional context is indebted to several key sociological conceptsincluding power and authority (Weber 1922; Pareto 1976), and community and society (Tönnies 1887)as well as psychological ones like the distinction between intrinsic and extrinsic work goals (Herzberg et al. 1959). Methodologically, Hofstede (1980:14, 17) argues that mental programs cannot be directly observed, so one needs to rely on observable and measurable verbal and nonverbal behavior, or on answers elicited by well-crafted questions. He proposes to use a questionnaire whose items capture different values and attitudes towards work and management. He finds that differences in responses across national groups of interviewees are statistically significant. It is important to keep in mind that the responses are also statistically different across occupation, gender, and age groups (1980:71-72). Hofstede then uses these responses to calculate aggregate scores denoting certain key cultural characteristics or dimensions. He focuses on four: power distance, uncertainty avoidance, individualism, and masculinity. Power distance is defined as the extent to which a boss can determine the behavior of a subordinate. Hofstede argues that power distance is accepted by both boss and subordinate, and shaped to a large degree by national culture (1980:99). The power distance index is a composite measure combining responses to questions about subordinates fear of disagreeing with their manager, and subordinates preferred and perceived types of decision-making behavior by their manager. Similarly, he measures attitudes towards organizational rules, employment stability, and stress to calculate an index of uncertainty avoidance. The index of individualism focuses on the relationship between the employee and the organization, i.e. the individual and the collectivity, and is measured in terms of perceptions as to whether the job provides the respondent with personal time, freedom, challenge, use of skills, adequate physical conditions, and training. Finally, the masculinity sic index attempts to capture the extent to which the respondent emphasizes job content and rewards at the expense of interpersonal relations, security, and the company (Hofstede 1980: 99-103, 153-164, 214-222). He also classified countries into groups based on a cluster analysis of the data: Anglo, Germanic, Nordic, Near Eastern, More Developed Asia, Less Developed Asian, More Developed Latin, and Less Developed Latin (Hofstede 1980:332-339).Students of FDI and the MNE have adopted Hofstedes cross-cultural concepts and measures to provide answers to several important questions. They have used his cross-national cultural indexes and clusters to answer several questions about the foreign expansion of the firm. In a frequently cited paper, Kogut and Singh (1988) found that joint ventures would be preferable to wholly owned subsidiaries the greater the cultural distance between the host and the home country. They calculated distances between pairs of countries based on Hofstedes four cultural dimensions, an indicator that has been used subsequently by other scholars. In addition, they found that companies prefer greenfield entry over acquisitions when uncertainty avoidance in their home country is high. Barkema, Bell, and Pennings (1996) predict and show empirically that the longevity of foreign ventures decreases with cultural distance between the home and the host country, especially in the cases of joint ventures and acquisitions. Finally, a myriad researchers has used Hofstedes approach and measures to explore how the MNE is structured (e.g. Lubatkin et al. 1998), and uses its human resources across different institutional contexts (Liberman and Torbiörn 2000.Scholars have been using Hofstedes measures of individualism, power distance, and uncertainty avoidance for nearly two decades, although the data were originally collected between 1968 and 1972. While cultural attitudes and values are supposed to be relatively stable, it is hard to deny that there is cultural change in the world, especially as countries develop economically. Drawing on comprehensive surveys of the general population in 38 countries between 1981 and 1997, Inglehart and Baker (2000) document that national cultures and values change over time. Somewhat consistent with Hofstedes general approach, however, they find that most cultural change is “path-dependent” rather than convergent. Thus, cultural change does not seem to erase cultural differences across countries. Hofstede (1980:342-371) himself looked into this issue and found that his indexes evolved somewhat between 1968 and 1972, but, like Inglehart and Baker, found no convergence across countries. Hofstedes cross-cultural approach suffers from a number of shortcomings. First, his approach is overly deterministic, and he “oversocializes” actors (Granovetter 1985), as is evident in his definition of culture as a mental program. Second, many social scientists are uncomfortable with the idea of aggregating individual responses to characterize a culture or a national institutional setting, and would rather take history and i