Episodic level: key actors, relations and their behavioural orientation

The episodic level involves focusing on a single temporally confined event. This could be, for instance, the aforementioned closure of a production site in an MNC. Apart from focusing on a temporally confined event, this first level of analysis comprises the identification of the key and relevant actors, their relations towards each other and their behavioural orientations in the event under consideration.

Relevant actors are for us all actors that are in one way or another affected by an event or episode but do not necessarily have a strong influence on its development and outcome (Becker-Ritterspach and Dörrenbächer Reference Becker-Ritterspach and Dörrenbächer2011). These would be regular employees or workers but also their families or private social networks affected by a site closure. Key actors, in contrast, are the ‘political brokers’ (March Reference March1962: 672) in the event. These actors most obviously have the power to influence the development and outcome of the event (Becker-Ritterspach and Dörrenbächer Reference Becker-Ritterspach and Dörrenbächer2011). They are typically executives from the headquarters and the subsidiaries on the managerial side. On the employees’ side, these are typically labour representatives within the company. It is important to note, however, that there may also be key actors who are not members of the MNC. We can think of a wide range of extra-organizational stakeholders or societal actors such as union leaders, politicians and media representatives who substantially influence the development and outcome of a political episode or events (Becker-Ritterspach and Dörrenbächer Reference Becker-Ritterspach and Dörrenbächer2009; Bélanger and Edwards Reference Bélanger, Edwards, Ferner, Quintanilla and Sànchez-Runde2006; Blazejewski Reference Blazejewski2009).

Identifying relevant and key actors is the first step in understanding the development and outcome of a political episode. The second step is to consider the relations among the actors identified. The questions to be posed here are: which key actors oppose or support each other by forming, for instance, a coalition? Who is the constituency of key actors, that is, which relevant actors do the respective key actors represent? Where do relationships to third parties provide access to power bases the focus actor her/himself does not hold?

In political episodes such as a site closure we will often find that subsidiary executives and employee representatives are in an antagonistic relation to each other. In this scenario, regular employees (relevant actors) are in a relationship of support with their labour representative (key actor) who acts on their behalf. Similarly, subsidiary executives may be key political brokers acting on behalf of corporate headquarters. Yet, as we shall see in the specific example below, we should be careful about making a priori assumptions about key actors and actor relations in political events as we might find highly idiosyncratic actor sets and relations of support and opposition that are more complex and cut across a simple management–labour opposition (Blazejewski Reference Blazejewski2009).

Relations among key and relevant actors in political events typically form around the (common) interests and related power sources, that is, behavioural orientations of different actors and actor groups in a given political episode. By behavioural orientation, we mean the interplay of actor interest (motivation) and actor sources of power (ability) that in conjunction influence the development and outcome of an event in a certain direction.

Actor interests are a key concept of the different politics and power perspectives (Bacharach and Lawler Reference Bacharach and Lawler1980; Dlugos et al. Reference Dlugos, Dorow and Farrell1993; Fischer Reference Fischer2005). In the literature interests are alternatively defined as preferences (Fischer Reference Fischer2005), needs, desires and concerns (Fisher et al. Reference Fisher, Ury and Patton2011) of individual or collective actors. They may be confined to material and economic interests as in the work by Allen (Reference Allen2004) and Rothman and Friedman (Reference Rothman, Friedman, Dierkes, Berthoin Antal, Child and Nonaka2001) but we may also extend them to other interests such as political, personal and career interests. We understand actor interests here as the desired course and outcomes of an event as perceived and articulated by relevant and key actors (see Chapter 2). Simply speaking, in a shutdown of a production plant we will find key actors who strive to avert the site closure. These key actors will act in the interest of a larger group of relevant actors. We will find these actors in opposition to other actors whose behavioural orientation is to close the site. These are likely senior subsidiary managers, who act in the interest of headquarter managers and ultimately shareholders.

Identifying relations and the interests embedded in them is only one side of the coin of understanding the behavioural orientation of actors. The other half involves asking about the power of actors. We follow Pfeffer (Reference Pfeffer1981) who suggests that power is a highly ‘context specific’ concept, requiring a context of application. In keeping with this suggestion, we define the power of actors surrounding a temporally confined event as their ability to influence the course and the outcome of the event in their interests (Giddens Reference Giddens1984). Qualifying this power further calls for an understanding of the sources of power available through and employable in the different social relations among actors. Regarding the sources of power, the organizational power literature offers different typologies. Among the first and probably best known is the typology suggested by French and Raven (Reference French, Raven, Cartwright and Zander1960). They distinguish positional power, referent power, expert power, information power, reward power and coercive power. Etzioni (Reference Etzioni1964), who distinguishes three organizational sources of power, presents a similar typology: coercive, utilitarian and normative power. Crozier and Friedberg (Reference Crozier and Friedberg1980) for their part identify four principle zones of uncertainty that define sources of power. These are sources which derive from special skills and from functional specialization, from relations between an organization and its environments, from the control of communication and information, and from the existence of general organizational rules. We shun away from defining a specific set or kinds of power sources. While keeping in mind the different sources discussed, we would like to suggest an empirical openness for the different kinds of power that actors may mobilize and employ. For instance, in the case of the site shutdown subsidiary managers can refer to formal authority and ownership-based decision-making rights to shut down the site. Labour representatives for their part may resort to other rules such as their right to strike or make normative-ethical appeals to social responsibility.

It should also be noted at this point that the above power perspectives have been criticized for seeing power as a property of actors, disregarding that power could only be mobilized through and materialized in social relationships (Clegg et al. Reference Clegg, Courpasson and Phillips2006). While we concur that social relations or exchange relations are a sine qua non for mobilizing and enacting power (the concept of resources as a source of power is useless when these resources are socially irrelevant, i.e. they cannot at least potentially be exchanged with other actors), we also recognize that the sources of power that can be mobilized and enacted in social relations may be of different kinds and are unequally or asymmetrically distributed among actors (some actors may have more contextually relevant resources that can be traded). Actors are positioned in different contextual fields and situations and are embedded in diverse social relations, which structure the episodic availability and employability of different kinds of resources.

Application

Returning to our example, we may identify that among the relevant actors in the Bochum plant shutdown are all employees and managers who will be affected by the closure of the production site. Affectedness could mean, in this specific case, job losses for workers or for senior managers having to move to another place and organizational unit. Clearly not all actors are affected by the event equally, which has a bearing on their interest in the event. What is more, as not all actors affected by a development can also influence the event, it is crucial to identify those actors that have the power to influence the development in their interest. In the case of the closure of the Opel site, we may therefore investigate the different actors that were directly involved in the negotiations that resulted in the site closure. For instance, the actors that were involved in the last episode of site shut down were the members of the German Opel works council, local shop stewards at the Bochum site and top management at the Bochum site, Opel Germany, GM Europe and GM Detroit respectively. Once key actors are identified, we inquire about their different behavioural orientations in the development of the event. In this case, the works council as well as management in Germany pursued established paths or rituals to voice their objections against the impending job cuts (Blazejewski Reference Blazejewski2009). Their shared perspective was based on formal authorities and competences and a long-standing cooperative stance between workers’ representatives and local/national managers at Opel. They therefore mainly pursued a strategy of negotiations with the MNC headquarters in Detroit. In turn, the shop stewards in Bochum who did not have access to the negotiations or other formally legitimated ways of protest but had strong personal ties to the workers on the shop floor, used this social power base to persuade the shop floor to stage an illegal wild-cat strike at Opel Bochum.

Limitations

The first level of basic micropolitical analysis involves understanding how the developments and outcomes of a specific temporally confined organizational event are connected to specific sets of key and relevant actors, their behavioural orientations and relations. It involves, in particular, asking about the interests as articulated by actors and identifying the different sources of power employed in and available through their relations. Such an analysis serves to explain why certain actors are able or unable to realize their interests. While this perspective allows us to develop a fine-grained understanding of a specific temporally confined event, it falls short of explaining how the development and the outcome of a given event is enabled and constrained by its wider temporal and contextual conditions. These contextual conditions become salient when we want to understand why similar organizational events across time and space result in very different trajectories and outcomes. For instance, our above analysis does not explain why shops stewards in Bochum pursued a strongly divergent path of protest vis-à-vis the members of the Opel works council – albeit they are ‘partners in arms’ in that both groups represent the workers’ interests, are members of the same union (in this case IG Metall) and are strongly embedded in the German industrial relations system. Also, limiting our analysis to the episodic level, we are unable to explain why the wild-cat strike took hold in some of the three Opel plants in Bochum but not in others. Answering these questions requires a deeper understanding of how socio-spatial conditions or their change over time constitute key and relevant actors, their behavioural orientations and their relations.

The socio-spatial and temporal constitution of behavioural orientations

The second level of analysis involves a deeper understanding of how the socio-spatial and temporal embeddedness constitutes actors, their behavioural orientations and relations and, thereby, the development and outcome of an event. It involves essentially shifting from a mere episodic perspective of an event to its wider systemic embeddedness and conditioning. By socio-spatial embeddedness or systemic embeddedness we mean the embeddedness of actors in a particular social position. We draw here on the work of Schütz and Luckmann (Reference Schütz and Luckmann1973) and Giddens (Reference Giddens1984). Schütz and Luckmann (Reference Schütz and Luckmann1973) suggest that the situation defines an actor's position with respect to their contextual conditions. Schütz argues that a situation is defined relative to and dependant on an actor's standpoint. This standpoint or position corresponds with certain interests, stocks of knowledge and experiences. In this view the standpoint also corresponds with the access to specific sets of resources (Blazejewski Reference Blazejewski2009). Giddens (Reference Giddens1984) theorizes actor ‘situatedness’ or ‘positioning’ in three respects, that is, chronological (position in daily life, within a lifespan, within a historical period), spatial (physical place or location) and social positioning (with respect to a network of social relations). Similar to Schütz (Schütz and Luckmann Reference Schütz and Luckmann1973), Giddens (Reference Giddens1984) sees positioning closely related to the behavioural orientation of actors (Becker-Ritterspach Reference Becker-Ritterspach2006). Drawing on Schütz and Giddens, we suggest that analysing the situation or position of actors in context helps us to understand both actor interests as well as the actor's access and employability of resources in and through social relations or social capital (Becker-Ritterspach and Dörrenbächer Reference Becker-Ritterspach and Dörrenbächer2011).

Giddens (Reference Giddens1984) maintains that actors’ positioning entails many very different contextual dimensions. This is also a key reason as to why structurally determistic models of actor behaviour are misplaced. For actors have through their multiple social positioning (e.g. in a family, a community, a firm or other organization, a region or a country), a wide range of different contextual reference points (e.g. local/firm, regional or national institutions) the relevance of which is subject to the specific situation of actors in time and social space. What is more, actors may not only draw on different contextual reference points based on their specific social position, but they may also come to vastly different usages and interpretations of the very same reference points. In this view, actor behaviour in micropolitics reflects a creative search of useful contextual reference points as well as a creative interpretation and employment of those contextual reference points selected (Blazejewski Reference Blazejewski2009; Karnøe and Nygaard Reference Karnøe and Nygaard1999; Whittington Reference Whittington1992).

As a starting point for a socio-spatial analysis, we consider the positioning of actors in context as multidimensional and comprising different levels of analysis. This leads us to ask the question about the key dimensions of social embeddedness in a micropolitical analysis. We suggest that the firm (or focal organizational) embeddedness of actors is the most crucial dimension to understand the behavioural orientation of actors in MNC-focused micropolitics. However, as the behavioural orientation is probably never solely structured by an actors’ position within the organization we have to inquire about the actors’ embeddedness outside the focal organization, that is, his or her broader societal embeddedness. Such extra-organizational embeddedness includes, for instance, an individual's embeddedness in a family, a community, a firm or other organization, a region or a country, etc. As we will discuss in more detail below, these organizational and extra-organizational dimensions can be looked at different levels of analysis, that is, the micro-, meso- and macro-levels. In a last step, we will then introduce the relevance of temporality to a micropolitical analysis.

Organizational position

The most obvious context for understanding the micropolitics in an organization is probably the divergent organizational positioning of actors within the very same organization. After all, it is the very nature and intention of formal organizational design to align organizational positions with specific behavioural orientations, in other words, to allocate organizational goals and sources of power to organizational positions. And even if we disregard formal structures and position and turn to informal structures and positions of actors, it is clear that an individual actors’ position in the organization informs the interests and sources of power, hence, the behavioural orientation in a micropolitical event.

Now, considering the position or situation of actors in the organization involves different analytical levels. Moving from macro to micro, we can see actors located at a certain kind of macro-organizational unit. What we have in mind here is organization-level embeddedness such as subsidiary or headquarters-level embeddedness. Inquiring about how this level of embeddedness structures the behavioural orientation of actors we can ask to what extent actors align with the interest of a certain site and can draw on sources of power that are related to this level of embeddedness. Actors may, for instance, fight for the survival of their production site and may draw on a specific source of power that is related to their centrality in the corporate network. Identifying sources of power at this level encompasses the question about the site's formal authority within the MNC, its position, centrality and importance in the global value chain, corporate division of labour or strategic importance with regard to crucial input and output markets (see related dimensions as developed in strategic contingency frameworks as presented by Hickson et al. (Reference Hickson, Hinings, Lee, Schneck and Pennings1971)). For instance, key actors who are located downstream in the value chain in a strategically important subsidiary with high network centrality may be able to draw on a substantial set of relations and resources to defend or expand their position.

Moving on to the next level of embeddedness, we have to consider that actors are positioned in organizational units, departments or work teams. Here we have to ask about the position of actors within a specific organizational entity in terms of his or her functional, professional and hierarchical work role. For example, knowing whether an individual works in a research and development (R&D) or production department, as an unskilled worker or a master craftsman with a rare production skill, as a worker with a labour representative function or senior manager, gives us an indication why some actors are key actors and why they follow one behavioural orientation rather than another. The organizational position of an individual actor or actor groups impacts actor interests as well as access to power bases such as social relations, important networks or organizational resources.

Application

So applying the relevance of the organizational context to our example of the Bochum plant, we may inquire how the different actors’ situation or position structures their interest and available sources of power vis-à-vis the threatened closure of the production site. Specific questions may be: do the actors differ in their interests depending on site location, hierarchical or functional position? Answering these questions gives us a strong indication about the interests of the actors involved as well as the sources of power available to them. In the Opel case, the three Opel sites were assigned diverging positions in the Opel production system in Europe (Blazejewski Reference Blazejewski2009). Whereas site I was at that time an assembly site working at full capacity and site III was largely a support unit shipping out parts to repair shops in Germany, site II produced gearboxes and axles for all other GM Europe production plants. Due to its central and essential position in the production system, workers at site II had not only a strong interest in the strike in order to fight for their jobs, they also had the means at hand to deal a strong blow at GM management by blocking the entire European car production for six days. So, depending on the site, workers were well placed to use their central position in the production system as an effective power base. Or they perceived their position to be rather weak and peripheral and also as largely unthreatened by the impending job cuts, so they withdrew from the increasingly heated struggle.

Apart from the site structure, the international dimension is also crucial for understanding the political episodes at Opel Bochum in 2004. On the one hand, as part of the European production system Opel Bochum was at that time competing against other sites in Belgium and Poland for the assignment of the new Delta production platform – it would ensure important investments into the successful site and provide a perspective for employment there. This situation made it more difficult for management and labour representatives at Opel Bochum to openly counteract decisions from the Detroit headquarters so as not to endanger the assignment of the platform. On the other hand, the Opel sites in Antwerp and Gliwice hosted an active labour force that on the international day of labour solidarity sent workers from Belgium and Poland to Bochum in order to support the local opposition against the impending job cuts. Consequently, an analysis of the contextual embeddedness of the wildcat strike at Opel Bochum discloses that there were opposing forces (competition and solidarity) at play at the same time – thus undermining a unanimous strategy in response to the headquarters’ restructuring plans.

Limitation

The key limitation of the organizational-level analysis is to reduce actor behaviour in micropolitics to their organizational embeddedness. Actors are not solely embedded in organizational contexts. Organizations are embedded in extra-organizational and societal contexts and so are the actors in them. Both the societal embeddedness of organizations and actors inform the behavioural orientation of actors within organizations. Such perspectives become relevant in two ways. They explain, for instance, how macro-level institutions at the national level structure the behavioural orientations of actors in rather similar organizational positions differently across countries. Moving to the micro-level of societal embeddedness, such as analysing for instance the family situation of actors or their position within a local community, may also explain why actors in a similar position even within the same unit may come to show quite different behavioural orientations. Hence, moving beyond a mere organizational positioning perspective calls for a societal embeddedness analysis at different analytical levels.

Extra-organizational context/situation

Although important, the actors’ position within the organizational context is not the only relevant context. The wider societal context or extra-organizational context, be it at the macro-, meso- or micro-level, shade into and inform the actors’ behavioural orientation in micropolitical organizational events. Extra-organizational contexts provide additional power bases such as private or public social networks, access to discourses that can serve to sustain arguments and positions inside the organization as well as resources that can be brought into the political arena. Importantly, not only do extra-organizational contexts inform the behavioural orientation within organizations but they constitute a wider set of key actors outside the organization with a stake in and a potential influence on the micropolitical developments within the MNC.

The relevance of societal context in terms of organizational and actor embeddedness can be theorized in a number of ways. While many more perspectives are thinkable, we restrict ourselves to two major theoretical strands, that is, new and comparative institutionalism, as both of them have been connected to micropolitical processes in MNCs in the literature.

Scholars from comparative capitalism emphasize the structuring effect of macroinstitutions that differs across countries (Almond and Ferner Reference Almond and Ferner2006; Ferner et al. Reference Ferner, Almond and Colling2005; Kristensen and Zeitlin Reference Kristensen and Zeitlin2005; Morgan and Kristensen Reference Morgan and Kristensen2006; see also Chapter 3). Focusing on the macro-systemic level, these approaches have often discussed the structuring role of national institutions in micropolitical processes. A typical example is how the actors’ embeddedness in a specific industrial relations system or the vocational training system informs their actors’ behavioural orientations (Edwards et al. Reference Edwards, Rees and Coller1999). For instance, unlike American top managers, German top managers would probably see a site closure and related lay-offs only as a last resort due to stringent labour legislation and the related cost of plant closure in Germany. Workers for their part may see a chance of saving their site based on an industrial relations framework which provides them with specific sources of power. In this view, national institutions provide actors with resources that can be effectively employed in micropolitical battles (Edwards et al. Reference Edwards, Coller, Ortiz, Rees and Wortmann2006). While comparative institutionalists have mainly considered the influence of national institutional systems such an analysis can easily be extended to the regional or local level (Morgan and Kristensen Reference Morgan and Kristensen2006). For instance, comparing different states or regions in the US, we would probably find that legal frameworks and institutions differ from state to state.

New institutional analysis opens up further ways of seeing the structuring effects of society on micropolitics (Blazejewski Reference Blazejewski2009). New institutionalists emphasize institutional or isomorphic pressures in organizational fields, which contribute to organizational homogenization of organizational structures, forms or practices (DiMaggio and Powell Reference DiMaggio and Powell1983). The probably most widely adopted concept of institutions within new institutionalism stems from Scott (Reference Scott1995), who sees institutions as consisting of normative, cognitive and regulatory pillars. As we suggest below, seeing organizations as embedded in and influenced by other organizations in the organizational field, as well as seeing normative, cognitive and regulatory elements of institutions at different levels of societal analysis, can inform a micropolitical analysis that takes into consideration the impact of the wider societal environment.

Clearly, other organizations, be they located at the national, regional local level, such as media, unions or governments, often introduce further (relevant and key) actors into the micropolitical arena and, thereby, influence the development and outcomes of micropolitical events. In a similar vein, other business organizations or business networks of suppliers and clients have a stake in strategic decisions and may try to influence the process and have an influence on the behavioural orientations of key and relevant actors within the MNC.

From a slightly different angle, but also from a new institutionalist perspective, we can explore how regulative, normative and cultural-cognitive elements of institutions at different levels inform actors’ behavioural orientation. Again, we could identify such institutional elements at the national, regional or local level. For instance, actors situated in a certain country, region, local community or organization may differ in their behavioural orientation from other actors. For instance, actors from a certain region may be renowned for their norms of resistance or revolutionary stances in industrial conflicts surrounding plant closures.

Lastly, the societal analysis at the micro-level calls for the specific understanding of the micro-level situation or social situation of individual actors. For instance, a key actor's possible allegiance to and position in a specific educational, professional, ethnic or religious community, as well as an individual's position in his or her family, can be quite crucial to understanding his or her behavioural orientations. Individual actors, even when holding the same organizational role or position and even when embedded in the same macro- and meso-societal context, may differ substantially in their behavioural orientation in organizational events. For instance, the question of how much an actor is affected by the job loss of a site closure also depends on the question of whether he or she has to sustain a family or has strong/weak local or regional roots. For example, top managers or other key actors who started their career as workers at the local production site and gradually moved into top management positions at the local site may have a different behavioural orientation vis-à-vis the site compared to managers who were sent from a foreign corporate headquarters. Hence, differences in the societal embeddedness of key actors can have a strong bearing on their behavioural orientations in micropolitical events.

Application

In the Opel case, the legal framework governing industrial relations in Germany was a key contextual element on the macro-/societal level. On the one hand, it limited the possibilities for strategic action of the works council, e.g. by expounding the conditions under which strikes are legally possible in Germany. On the other hand, it provided the basis for a dual system of worker representation inside the firm: the works council (codified in the Betriebsverfassungsgesetz) as the employer representation at the organizational level and the union-affiliated shop stewards (Vertrauensleute) at the shop floor level who were protected by the German Constitution (Grundgesetz) and industry-wide collective labour agreements (Tarifverträge). In the Opel case, both groups drew on the respective legal frameworks and associated resources as bases of power and in legitimating their respective paths of action – albeit according to highly differing interpretations of the formal regulations.

Another important contextual element in the Opel case was the local identity and the culture of solidarity in Bochum and the Ruhr region at large. Rooted in a history of harsh labour conditions in the mining and steel industry and the joint fight for worker rights in the nineteenth century, as well as the shared experience of the industrial decline in the second half of the twentieth century, particularly in Bochum, workers and their families saw themselves as ‘fighters’, even ‘radicals’, who were strongly committed to the region and local solidarity. This extra-organizational identity construction provided a considerable resource during the fight for jobs at Opel in 2004.

Limitation

Integrating the societal and communal level in a micropolitical analysis only makes sense when connected to an organizational level analysis. In other words, a societal embeddedness analysis as such will probably never suffice to understand the behavioural orientation of actors in a specific focal organization. Only when combined with the organizational embeddedness of actors do we understand the relevance of the wider societal context. The societal analysis also shares a limitation with a mere organizational-embeddedness-based analysis. Both types of analysis make it rather difficult to understand how temporal effects play into micropolitical events. For instance, we learn in these cross-sectional perspectives little about how past events inform current events. By the same token, we may find it difficult to understand how a current political episode structures future political scenarios.

Socio-temporal extensions

In this last step, we introduce the relevance of temporality to micro-political analysis. The effect of temporality on micropolitics has to be looked at in conjunction with the social spatial embeddedness. In other words, a concern with temporality entails asking how the contextual embeddedness of actors has changed over time and how this affects their behavioural orientation and thereby the development and outcome of a micropolitical event. We hold that temporality is relevant for the analysis in two ways that are intertwined but shall be kept separate for analytical reasons. These are organizational and societal changes on the one hand and positional changes of actors over time on the other hand.

The first aspect entails a closer look at how organizations and societies change over time. This implies that actors who have not changed their social position face different contextual conditions over time. It also implies that contextual and organizational changes go along with different key and relevant actor-sets. An example for an organizational change would be mergers and acquisition by an MNC (e.g. possible devaluation of exclusive sources of power held before). Another example would be the effect of a plant closure (withdrawal of certain actors) on the chance of survival for other corporate sites. Here we look at how contextual changes over time and past micropolitical events in the organization inform current micropolitical events.

An example for a societal change would be changes in national legislation such as changes in labour law or the introduction of a minimum wage. We can easily imagine that the acquisition of another production site in a low-wage country with weak labour protection could potentially undermine the bargaining position of actors embedded in a production site of a high-wage and high-labour-protection country. A temporal analysis may also involve considering potential future events as they may inform current events. A planned change in corporate strategy may inform the behavioural orientations of key actors. Hence, embedding a micropolitical analysis in temporal terms involves asking how past and potential future events inform the behavioural orientations in current events as well as how contextual changes explain different dynamics and outcomes of events at different points in time.

Organizational and societal changes are only one effect of temporality on micropolitics, however. The other relevance of temporality lies in the changes of actors’ social positions across time and space. Such trajectories are typically captured in the actors’ biographies but should also include a closer look at their socio-positional aspirations for the future, that is, in organizational (career) and societal terms (intent to emigrate etc.) (Becker-Ritterspach and Dörrenbächer Reference Becker-Ritterspach and Dörrenbächer2011).

The analytical relevance of such an actor-focused change perspective over time has again two aspects to it. First, we can ask how an actor's biography (past societal and organizational positions held) and future aspiration feed into behavioural orientations in current events. We can ask, for instance, whether a subsidiary manager who is on a foreign assignment has had a past career in corporate headquarters and wishes to return to corporate headquarters in his or her home country in the future. Answering this question is likely to give a strong indication as to why he or she supported the plant closure or resisted it, and what kind of sources of power were mobilized in the event. In a second and related perspective, this type of temporal analysis may help us to understand why the same key actors display contrasting behavioural orientations in similar events and under similar contextual conditions across time.

Application

Going back to our example, we find that Opel Bochum was faced with repeated rounds of job cuts and threats of closure in the past that had so far been put off successfully through e.g. workers’ concessions regarding employee benefits and the cooperative relationship between the works council and the German management, which buffered German interests in the negotiations with the US headquarters of GM. This shared experience of numerous rounds of impending closure and negotiations influenced the political behaviour of both works council and management in the 2004 episode which – as in the past – relied on negotiations and concession making as the only option to cope with the current threat.

Regarding the temporal effects of the current episode on the future, the Opel case provides another interesting example: as a result of the 2004 wildcat strike, GM rearranged its European production system and established a multiple-site strategy regarding its parts production. In effect, since 2005 the Bochum plant site II was no longer the only site producing gearboxes and axles for the European assembly plants. It thus lost its central and unique position in the European production system, thereby forfeiting important bases of power and strategic leverage for its political agenda.

Introducing a temporal dimension we are thus able to uncover how the organizational and societal context changes over time and how these changes constitute shifts in the behavioural orientations of actors and their chances to realize their desired ends in current and future episodes.

Limitation

Limitations regarding the temporal extension of the political perspective mainly result from methodological challenges. When we start our research process on politics in the MNC with an episodic perspective as recommended here, the primary focus tends to be on the current political situation in the MNC. Often, interviews with actors involved in the current episode are the main source of data at this stage. If we use these same data to sustain the temporal extension of the political perspective, the information available regarding past influences on current political events is necessarily retrospective in kind. In retrospect, past events tend to be (unconsciously) rearranged, re-evaluated and modified as subjective constructions rather than unbiased reports of the past. It is therefore imminent to take account of and present data on temporal developments of political situations as retrospectively constructed and also seek to understand why actors reconstruct their past in this particular way. Constructing one's past (as well as one's future) as such is therefore a political act and a worthy subject of political analysis in itself. Apart from interview data, historical accounts of the past events under consideration – when they are available in public reports (journal articles, newspapers) or company archives – can be a productive source of information and help us to understand how and why actors in current political episodes actively construct and make use of temporal developments as they might serve their present and/or future interests.

Starting the research

In the following, we will examine aspects of power and politics in defining the research question, selecting the appropriate research site and negotiating and renegotiating access to the MNC.

Data collection

The majority of qualitative research in management and international business published in prestigious journals is based on case studies (Welch et al. Reference Welch, Piekkari, Plakoyiannaki and Paavilainen-Mäntymäki2011). In the field of international business, the typical case study uses interviewing as the data collection technique (Piekkari et al. Reference Piekkari, Welch and Paavilainen2009). We will therefore focus on the challenges of interviewing which can be defined as ‘talking – usually face to face – with those knowledgeable about what is being studied’ (Macdonald and Hellgren Reference Macdonald, Hellgren, Marschan-Piekkari and Welch2004: 264). We embrace this definition as it implies questions about ‘who is knowledgeable’ and ‘what is deemed knowledge and why’. In other words, it provides a useful platform to scrutinize a data collection technique from a politically aware perspective.

Data analysis

The fact that data are often captured in several languages is characteristic of research on the MNC (e.g. through interviews, company material and public documents). While knowledge about MNCs is predominantly disseminated and published in English (in itself this is indicative of hegemonic practices; Tietze Reference Tietze2004; Steyaert and Janssens Reference Steyaert and Janssens2013; Tietze and Dick Reference Tietze and Dick2013), many researchers engage in a hidden process of sensemaking and translation when analysing multilingual data sets. Researchers may translate themselves if they have the requisite language skills or rely on professional translators – at least when researchers who are non-native speakers of English reach the stage of publishing in English. The challenge is whether the richness of quotations can survive multiple translations (Macdonald and Hellgren Reference Macdonald, Hellgren, Marschan-Piekkari and Welch2004). In a recent article, Chidlow et al. (Reference Chidlow, Plakoyiannaki and Welch2014) treat translation as a social practice and suggest that it be reframed as a process of intercultural interaction rather than a technical exercise of following the protocols of back translation. So far, the way researchers cross the language boundary in data collection and analysis is rarely documented in published work or discussed in methodological texts (see Chidlow et al. Reference Chidlow, Plakoyiannaki and Welch2014; Marschan-Piekkari and Reis Reference Marschan-Piekkari, Reis, Marschan-Piekkari and Welch2004; Welch and Piekkari Reference Welch and Piekkari2006, for exceptions).

In their study of how language diversity affects the work processes of corporate boards, Piekkari et al. (2015) explain how translation was part of the data analysis stage. They conducted a total of thirty-two personal interviews with former and current board members, managing directors and executive managers in nine Nordic firms. Out of these interviews, twelve were carried out in English and twenty in various Nordic languages, i.e. Finnish, Swedish, Danish and Norwegian. Thus, the large majority of the interview quotes had to be translated from a local Nordic language into English which was indicated in the article.

The translation process followed several steps. First, in order to share interesting insights and powerful quotations with other members of the research team, Rebecca summarized the interviews in English. The interviews that had been conducted in Finnish, in particular, could not be understood in the original language by the other members of the research teams (see also Meriläinen et al. Reference Meriläinen, Tienari, Thomas and Davies2008). Second, once the authors had selected the interview quotes for publication they themselves iterated several times between the original and the target language to maintain the intended meaning and the richness of the interviewee's verbatim expressions during the course of the translation. Third, two professional proofreaders, who were native speakers of English, carefully read all the quotes translated by the researchers into English and the specific meanings and nuances of different words and expressions were discussed. Taken together, the translation process itself deeply immersed the researchers in the qualitative data and triggered novel insights and interpretations of the phenomenon at hand (Piekkari et al. 2015).

Writing-up for publication

The requirements of multiple audiences often pull researchers in different directions. While academic peers look for the theoretical contribution, managers expect practical recommendations and research participants are primarily concerned with maintaining anonymity. These expectations can be difficult to align, thereby further challenging the ‘text-book connection’ between research questions and methods (Buchanan and Bryman Reference Buchanan and Bryman2007: 492).

‘Being political’ in writing and published accounts can have very serious consequences. O'Connor (Reference O'Connor1995) studied written accounts of change in a high-technology company. These accounts praised the efforts of the organization development function whose role was presented as pivotal in change initiation and implementation. In her conclusions, however, O'Connor observed how involvement in key decisions was limited to a small group of key managers, how disagreement was treated as resistance and lack of understanding rather than as involvement and how change narratives revolved around a heroic figure and his adversaries. The high-technology company did not welcome O'Connor's interpretation. Her gatekeeper denied her account, described it as shocking, outrageous and unacceptable, and never met with her again. While such candid accounts are unusual, they almost certainly reflect a relatively common experience in fieldwork. A critical research agenda may also prevent a research project from taking place altogether. A colleague of ours confessed the following in an informal conversation: ‘How can you ever say anything critical about Goldman Sachs when 50 per cent of your MBA students are sent by them!’ Corporate sponsorship in its various forms raises the delicate question of whether the alliance between business schools and companies is unholy, even corrupt, and ‘contaminates’ research findings.

Evaluating the quality of research

Given that many researchers undertake context-sensitive ‘inquiries from the inside’ (Evered and Louis Reference Evered and Louis1981) when studying power and politics in the MNC, it is not meaningful to evaluate them as to whether they are generalizable, replicable or objective. In fact, such criteria would be inappropriate for a research project that does not subscribe to positivist ideals. Nevertheless, the use of positivist quality criteria dominates management research (Gibbert et al. Reference Gibbert, Ruigrok and Wicki2008) and these criteria are often employed as ‘commonsense benchmarks’ to evaluate qualitative research (Johnson et al. Reference Johnson, Buehring, Cassell and Symon2006: 136), as if one size would fit all.

How to evaluate the quality of qualitative research is a contentious issue, representing a longstanding area of discussion and debate among methodological authorities (Creswell and Miller Reference Creswell and Miller2000; Kirk and Miller Reference Kirk and Miller1986; Lincoln and Guba Reference Lincoln and Guba1985; Symon and Cassell Reference Symon, Cassell, Symon and Cassell2012). Johnson et al. (Reference Johnson, Buehring, Cassell and Symon2006: 133) argue that evaluation criteria are not neutral or ‘value-free’ but ‘constituted by particular philosophical conventions’. They propose a contingent criteriology that ‘enables different sets of evaluation criteria to be contingently deployed so that they fit the researcher's mode of engagement’ (Johnson et al. Reference Johnson, Buehring, Cassell and Symon2006: 134).

Given the diversity of views of what constitutes appropriate quality criteria, calls have been made for increasing the transparency of qualitative research (Bansal and Corley Reference Bansal and Corley2011; Bluhm et al. Reference Bluhm, Harman, Lee and Mitchell2011). This requirement does not only cover describing the range of data sources used or the steps taken in data analysis, but also an explanation of how the findings were generated. As Bansal and Corley (Reference Bansal and Corley2011: 236) state, ‘[b]ecause discovery can be serendipitous, methodological rigor is conveyed through the authenticity and candor of the text. It is important that researchers be able to describe how they discovered their insight’. Authors may also illuminate why they selected particular theoretical and ideological frames for their projects. In addition, researchers of MNCs may include details about the languages used during the research process (to gain access, in the work of the research team, during data generation with different gatekeepers, translators or interpreters).

Depending on the researcher's philosophical assumptions, reflexivity can be a relevant quality criterion. Reflexivity refers to research that takes into account ‘the social production of scientific “facts”’ (Hardy et al. Reference Hardy, Phillips and Clegg2001: 532). ‘Whereas objective knowledge claims to be unsituated – true any time and any place – reflexive knowledge is situated and includes a recognition of the multiple translation strategies that bring it into being’ (Hardy et al. Reference Hardy, Phillips and Clegg2001: 554). Johnson and Duberley (Reference Johnson and Duberley2003: 1279) add that reflexivity ‘entails noticing, evaluating and being suspicious of the relationship between the researcher and the “objects” of research’. In this regard, researchers could probe the outcomes of their own inquiry in terms of which data ‘get translated’ in a way that audiences can access them and which data disappear and remain invisible. Moreover, who are the agents of translation/interpretation and how is their work acknowledged and even incorporated into the research outcomes?

Overall, while researchers engage in political actions during various steps of the research process, this is rarely acknowledged in the methodological writings or in journal publications. Easterby-Smith et al. (Reference Easterby-Smith, Thorpe and Jackson2012) encourage researchers to critically reflect on all the political influences on their research, and to make these thoughts available to their peers. They argue that such reflexivity ‘should increase, rather than decrease, the credibility of the results’ (Easterby-Smith et al. Reference Easterby-Smith, Thorpe and Jackson2012: 86).

Implications for researchers

In this chapter, we have deliberately not offered a blueprint for conducting research on MNCs from a political perspective or advocated particular research methods as the best or even appropriate tools to collect empirical data, because we believe it is not possible to do so. Instead, we have focused on the research process and rehearsed examples from the MNC literature to demonstrate how researchers have dealt with political challenges. As we have shown, the entire research process is political because different stakeholders at different stages of the research inform and influence the study. This causes some dilemmas which the researcher has to address.

If one wants to conduct empirical research on micropolitics in MNC settings one needs access to data. The dilemma a researcher or research team faces is whether to declare one's political interests to potential funding bodies and research participants up front. We have seen that it is possible to gain support from funding bodies while pursuing a critical agenda (e.g. the work by Geppert and his colleagues). Here, being aware of which funding body to target is a necessary pre-requisite for receiving funding for emancipatory, critical research. Nevertheless, in MNC settings gaining access to local organizations can be difficult and can include activation of non-organizational networks. Alternatively, the researcher may decide to use secondary data such as media texts to address problems of access (e.g. the work by Vaara and his colleagues), or, like Dick and Collings (2014), bring a critical mindset to the data only when they began to conceptualize their findings. These examples show that it is possible to conduct empirical, critically oriented research in MNC settings, but the process is precarious. Researchers need to reflect on the sources of data, the research questions they wish to explore, the funding and sponsoring bodies that support their projects, and to what extent they will make their findings available to which group of stakeholders and at what stage in the research process.

Our opening vignette has shown that collaborators and gatekeepers do not always take critical research findings favourably. Controversy, conflict and unequal power relations may be accepted as normal, interesting and unavoidable phenomena by academic audiences and many practitioners will agree with this view, too – particularly if it concerns organizations other than their own. However, having organizational politics publicly discussed and analysed is a different matter and may lead to a hostile reaction by influential members of the researched collective.

In our opening vignette, the researchers’ critical engagement with the research scene unfolded in the data analysis stage. The senior gatekeeper rejected the findings as ‘untruthful’ and damaging prior to publication. There are no easy answers in situations like this. Researchers may use their institutional ethical protocols and processes to safeguard themselves and their projects from interventions by MNC representatives, e.g. by adhering to standards of anonymity and confidentiality. However, research that detects malpractices may need to name the MNC participant in order to challenge these behaviours. By drawing on secondary material and using personal or social media networks, access to information may be gained while over-dependence on organizational sponsors is avoided. While these are not infallible prescriptions for managing the political aspects of the research process in MNCs, they do provide some guidance as to how to prepare for their occurrence.

Finally, in some cases it may be appropriate to adjust one's writing strategy beyond the procedures of rendering participants and organizations anonymous and treating sources as confidential. The approach we employed in the opening vignette is advocated by Watson (Reference Watson2000). It relies on a combination of social science writing with fictional elements which enables researchers to disguise people, situations and contexts while drawing on analytical concepts and rigour. Nevertheless, writing and publishing semifictional accounts can also detract from conveying powerful messages about micropolitical behaviour, its ties with wider historical and societal processes and the darker sides of how influential corporate élites exercise power. Also, some academic journals may not accept this approach as suitable for the treatment of empirical data and the furtherance of knowledge.

In sum, we invite researchers interested in studying the MNC from a micropolitical perspective to address a host of questions which are not only of an academic nature but also fundamental to the implementation of their research projects. In particular, management researchers whose research is located in MNCs or other international settings meet with additional complexities that form part of the political backdrop to all research. Understanding these additional complexities also from a political perspective is an essential part of the research process from instigation to execution and publication.