THE EFFECTS OF DECENTRALIZATION IN META-ORGANIZATIONS:
A CASE STUDY AT THE INTERNATIONAL FEDERATION OF RED CROSS AND RED CRESCENT SOCIETIES
Author: Bas van Rossum UvA Student ID: 10642897
Contact details: bas.rossum@gmail.com
Program: MSc Business Administration – Strategy Track
Institution: Amsterdam Business School, University of Amsterdam Thesis supervisor: Anro Kourula
2
Statement of originality
This document is written by Bas van Rossum who declares to take full responsibility for the contents of this document. I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it. The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
3 Table of contents Abstract 4 1. Introduction 5 2. Literature review 6 2.1 Meta-organization 7 2.1.1 Impact of meta-organizations 7 2.1.2 Membership 8 2.1.3 Decision-making structures 10 2.1.4 Inertia 11 2.2 Decentralization 12 2.2.1 Effective decision-making 14 2.2.2 Local responsiveness 15 2.2.3 Innovation 16 3. Theoretical framework 17 3.1 Local responsiveness 18 3.2 Effective decision-making 19 3.3 Innovation 21 4. Methodology 22 4.1 Archival research 22 4.2 Interviews 23 4.3 Data analysis 24
5. Results archival research 26
5.1 Moving towards a New Operating Model 26 5.2 Accenture decentralization review 29 5.3 Implementation of the decentralization review in the Europe Zone 31
6. Results field research 33
6.1 Local responsiveness 36 6.2 Effective decision-making 38 6.3 Innovation 40 6.4 Performance evaluation 41 7. Discussion 41 7.1 Local responsiveness 42 7.2 Effective decision-making 44 7.3 Innovation 45 8. Conclusion 46 9. Managerial implications 46
10. Limitations and future research 47
4
Abstract
Traditionally organizational research has focused itself on individual-based organizations which are composed of individual members mainly driven by monetary incentives. Meta-organizations, often referred to as umbrella Meta-organizations, however, are composed of other organizations that have joined on a voluntary basis based on its collective goals or to strengthen its own organization. This difference gives a different dynamic in an organization and therefore makes the existing research on individual-based organizations not per se applicable in the context of a meta-organization. Which is an organization form that has received little attention from scholars therefore this paper has been exploratory of nature. It has questioned the effects of decentralization in meta-organizations on its operations in terms of local responsiveness, effective decision-making and innovation. The theoretical framework has been studied in close cooperation with the International Federation of Red Cross and Red Crescent Societies – the world’s largest humanitarian network which has recently decentralized its operations. Rare access was provided to hundreds of documents internal documents and interviews with high placed representatives in the global and European management teams in the execution of the case study. Based on the provided data it can be concluded that in this particular case decentralization increased its local responsiveness and effective decision-making as its center of support and analysis are closer to its members operations. However, decentralizing a meta-organization’s operations is not sufficient to capture these benefits – enabling mechanisms need to be put in place in order to let its members benefit and promote participation of its members in the new, more intense, structures. Decentralization has also an effect on innovation – whether it is beneficial for a meta-organization depends on whether it aims for globally standardized or locally standardized innovation. Meta-organizations are not innovators, but knowledge brokers that provide a platform for exchange of innovations created within its membership.
5
1. Introduction
The term meta-organization has been relatively new in the field of organizational research. It was first introduced by Ahrne and Brunsson (2005) who defined it as “an organization that is
composed of other organizations”, later on refined by Gulati et al. (2012) as “an organization whose agents are themselves legally autonomous and not linked through employment relationships” whereas traditional organizational research is often based on the assumption
that members of organizations are individuals. Meta-organizations, sometimes also named an “umbrella organization”, use various types of names to describe them, such as union, federation, confederation, commission, network and club but all can be considered to be associations of which members join on a voluntary basis and may withdraw membership at any time. Its existence and often prominent place in the industry is becoming more common in today’s world and the number seems to be growing with globalization (Scholte, 2000). In 2003 there were about 10.200 international and 200.000 national meta-organizations in Europe, of which two third was founded in the last 30 years (Ahrne & Brunsson, 2005). Organizations are increasingly acknowledging the importance of joining meta-organizations as “organizations are inescapably bound up to the conditions of their environment” (Pfeffer & Salanick, 1978).
Despite the increasing importance of meta-organizations in industries, and the exceptional growth figures in the last decades (Ahrne & Brunsson, 2005), the interest of scholars in this specific type of organization has been remarkably little. The general organizational theory tends to focus on the assumption that the members of organizations are individuals – whereas those of meta-organizations are other organizations. Therefore the assumptions and conclusions under which traditional organization theory has been founded may not apply in the context of meta-organizations. The fact that the concept of meta-organizations was only first introduced in organizational literature by Ahrne and Brunsson in 2005, indicate that much research still needs to be done to better understand the mechanisms of this type of organization and its effect in today’s industries. Ahrne and Brunsson (2005) and (2008) have mainly explored the core concept of meta-organizations, whereas other researchers have taken on their foundational research to have a closer look at meta-organizations in combination with variables such as decision-making processes (Casula Vifell & Thedvall, 2012), structural inertia (Köning, Schulte, & Enders, 2012) and its organizational design (Gulati, Puranam, & Tushman, 2012).
This research takes a closer look into meta-organizations that have decentralized their operations to better assist the needs of its members or the environment they operate in. The increasing importance of meta-organizations in industries stimulates their growth as more organizations are joining them. Causing some meta-organizations to becoming extensive bureaucratic organizations making their decision-making complex, its local responsiveness difficult and its innovation relatively far from the reality their members face. Decentralizing, defined as “the extend of power and authority extended down through the hierarchy” (Hage &
6 Aiken, 1970), their operations would be one way to address this problem. For instance the International Federation of Association Football (FIFA) has decentralized its operations through the creation of administrative bodies for each of the six continental confederations (e.g. Union of European Football Associations (UEFA) for Europe and parts of Asia). Similarly, BirdLife International, the world’s largest bird conservation organization, has decentralized its operations through regional coordination offices. Where the global secretariat coordinates and facilitates BirdLife international partnership strategies, programs and policies and the regional coordination offices are responsible for regional implementation and assisting member organizations in conservation, planning, capacity building and fundraising. This research specifically looks at “how does (de)centralization affect the local responsiveness,
decision-making structures, and innovation in the context of meta-organizations”. A theoretical
framework will be tested these variables through a single case study at the International Federation of Red Cross Red Crescent Societies, a meta-organization that has decentralized its operations in 2007 through the establishment of regional offices.
2. Literature review
The literature review is divided into two parts: first, a closer look is given to the concept of meta-organizations and its impact in today’s society and second the review elaborates on the literature on decentralization, particularly linked to the variables of local responsiveness, effective decision-making and innovation.
2.1 Meta-organizations
So far organizational research has mainly been built on the assumption that its members are individuals: the individual-based organization as identified by March and Simon (1958). However, as mentioned, an increasing number of organizations are composed of other organizations making them meta-organizations. A concept that has received little attention in organizational research. It can however be argued that in some respect the knowledge gained from traditional organizational research can be applied to organizations regardless of the nature of their membership – this does however not apply to all of the research as meta-organization have common characteristics that are considerably different from those of individual-based organizations. Whereas, for instance, an individual-based organization is composed of a clear hierarchal structure with a clear agency relationship where the principal can delegate tasks to the agents based upon an incentive system where one aligns each other interests. (Barney & Hesterly, 1996). A meta-organization is composed of autonomous members that have joined the organization on a voluntary basis and often pay membership fees and therefore cannot be forced to perform tasks. Nor are they sensitive to monetary incentives, as a member organization often has access to more resources than a meta-organization of which they are a member. Member meta-organizations therefore, different from individuals, do not join a meta-organization because of monetary incentives, but rather because of their shared purpose and interest, to bundle strengths and its effects on their social status (Ahrne & Brunsson, 2005). Formal authority through employment contracts have
7 often been central in the organizational literature (Coase, 1937), while in fact the absence of such formal authority is a defining feature in the structures of a meta-organization.
One does not only need to treat organizations different from individuals because of the nature of their membership, but also because they both resemble different unique aspects. It can be stated that members in individual-based organizations are rather homogenous, they have the same basic needs, duties and rights, whereas members of a meta-organization are unique and therefore often incomparable in many ways. They represent different organizational forms, objectives, sizes, structures and grassroots members – resulting in a tangle of different organizations which need to work voluntary and coordinated towards joint objectives without imposing actions on them that affect their autonomy and therefore their identity as an organization (Ahrne & Brunsson, 2005). In some ways it can be argued that its members are a constraint to the work of a meta-organization, as their participation is often not their core-interest – while at the same time participation of its members are the only asset of meta-organization that gives them legitimacy to take a firm stand in the industry environment.
2.1.1 Impact of meta-organizations
In the little research that has been done on meta-organizations, it might appear as it is a relatively weak organizational form because of, among others, its member’s autonomy and consensus decision-making. While this is undoubtedly a constraint in the effectiveness of meta-organizations, they also have been able to put their stamp on many industries as they exist today.
Meta-organization’s main purpose is work in the interest of its members whom share some kind of common characteristic or objective (Burnsson & Ahrne, 2008). This purpose is often put in practice by first, creating an optimal environment for its members and second, support its members in their operations by functioning as a platform for engagement (Burnsson & Ahrne, 2008; Petrova, 2007). They often play a significant role in their industries as they influence its evolution and with that reduce uncertainty for its members, creating a stable environment to work in. Meta-organizations are often established to form a higher, overarching body in which members join efforts to speak with a unified voice in their interactions with the field and environment they operate in (Köning, Schulte, & Enders, 2012). They often issue guidelines, organize conferences and carry out research that contributes to the development and promotion of standards, definitions and the broader identity of a certain industry. In the recent years meta-organizations have played an increasingly important role in policy-discussions as governments and other institutions often prefer to negotiate with meta-organizations over individual-based organizations as they are considered a knowledgeable partner that is able to represent a large proportion, if not all, of the organizations in the industry making the policy-making process considerably more efficient (Storgaard Bonfils, 2010; Petrova, 2007). Whether the government or other relevant
8 authorities find the meta-organization a good partner largely depend on its status within the industry. The status of a meta-organization is often depended on its members, and vice versa. A high standard of the membership reflects on the reputation of meta-organizations as a knowledgeable partner with a large and respected membership, whereas a respectable meta-organization has a positive reflection on the member organization. The latter, for instance, is the case at the Global Alliance in Management Education (CEMS), former Community of European Management Schools and International Companies, which is a highly respected meta-organization which only accepts the best management schools in the world in their membership (Ahrne & Brunsson, 2005).
Other impacts of meta-organizations can be related to unifying regulations, for instance in the aviation industry through the International Air Transport Association (IATA), or collaborative military action through the North Atlantic Treaty Organization (NATO).
2.1.2 Membership
Members of meta-organizations can have different organizational forms. In some cases they are states, such as in the United Nations, World Trade Organization or the International Monetary Fund, but in other cases it are firms, as in industry associations such as the European Automobile Manufacturer’s Association. The process of welcoming new members into the organization differs from organization to organization. Many meta-organizations use a closed membership in which prospective member hands in an application to join. This application can be approved by either the founder or director of the meta-organization or by its members collectively (Gulati, Puranam, & Tushman, 2012). For members of meta-organizations that operate in a highly competitive market, the guarantee of a closed membership is a vital reason to join as it gives them a strategic advantage over its competitors. If, for instance, every airline could join a certain alliance it would reduce, if not destroy, its competitive advantage and with that its entire purpose. Contradictory, a meta-organization as the United Nations benefits from an increased number of members as it increases the states it represents and with that its legitimacy in negotiations. There is however also a small group of meta-organizations that uses an open membership in which organizations self-select themselves (Gulati, Puranam, & Tushman, 2012). An example of such are IT platform provides that only require ICT firms to accept their terms of reference. While open membership meta-organizations often grow rapidly, it is difficult to find skilled people, ensure quality and it is hard to coordinate them as joining the organization is as easy as leaving (Gulati, Puranam, & Tushman, 2012).
Joining a meta-organization often gives a paradox to member organizations – especially for firms which compete with other members within the meta-organization. It is stated by Ahrne and Brunsson (2008) that “the more similar the members are the stronger the
meta-organization” and “the higher the dependence of members on the meta-organization for their operations the stronger will be the meta-organization” as it enhances their efficiency,
9 strengthens its identity and increases the internal consistency which enables them to influence the environment more efficiently. Members will benefit from having a strong meta-organization, therefore increasing similarity with its co-members and dependence towards the meta-organization would be assumed a positive development. However, such development could possibly threaten the existence of the members as it might damage their uniqueness and identity from which they derived their competitive advantage – a development that especially concerns firms which compete with other members within the organization. However, it is argued that the longer an organization has joined a organization, the more similarity will occur between them and its co-members as the meta-organization will impose directives and standards on its members (Burnsson & Ahrne, 2008). So arguably the process to unify the membership is only a matter of time. While it is often argued that member organizations are much stronger in terms of available resources, capacity and status – still their membership towards meta-organizations is rarely without consequences (Burnsson & Ahrne, 2008). It creates fundamental challenges for the identity and autonomy of members as only through unifying the membership of the meta-organization is able to optimally impact the environment.
Despite the pressure on the member’s identity and autonomy the number of organizations joining meta-organizations is rapidly increasing (Scholte, 2000). To explain this phenomenon it is important to acknowledge that membership commitments work both ways. There are mutual written agreements, often in the meta-organization’s statutes, what one can expect from the other party to reach a goal that benefits both parties. Members however should carefully consider whether it is worth to partly sacrifice their identify and autonomy to fully benefit of de added value the organization can bring. In the reasoning to join a meta-organization, four main categories can be identified.
1. Ideological reasoning 2. Active reasoning 3. Passive reasoning 4. Destructive reasoning
Many members join a meta-organization out of an ideological reasoning. They operate from a similar mindset or ideology from which they proclaim their identity. Joining a meta-organization that operates from a similar mindset strengthens their identity as well as facilitates cooperation with other members that work from a similar reasoning. In the extend of this, some members join a meta-organization because of active reasoning to gain concrete benefits from their participation. These firms often have a shared purpose and interest in the activities and/or objective of the meta-organization. Participation is deemed as a good and low cost opportunity to influence the environment in which they operate to maximize their businesses. Since members can often concentrate their resources that will enhance the impact of joining efforts such as lobbying and public relations campaigns. Combining these resources goes further than financial commitments – it is about working
10 together on a horizontal hierarchy level to pursue joint interests. Members are expected to contribute by giving free access to their knowledge, personnel and network. There is also a more passive reasoning for joining a meta-organization. Sometimes it is expected and therefore simply not a question in the first place. If one does not, it might raise suspicion – for instance in the case of the Dutch Association of Travel Companies (ANVR). If a travel agency is not a member, the consumer might prefer booking their holidays at an alternative travel agent. In other cases membership is not considered as a necessity but as recognition of a certain quality, like in the case of the Global Alliance in Management Education (CEMS) which was mentioned before. Membership to such a meta-organization affects the status of the organization. Members that join a meta-organization out of passive reasoning do not necessarily have clear objectives of what they like to gain from their membership other than affiliating their name with an often respected meta-organization. They do not value participation much and are likely to become free-riders (Ahrne & Brunsson, 2005). At last there are organizations that have a destructive reasoning for joining a meta-organization. They do not aim to gain positive effects of their participation, but want to minimize any negative affect by internal lobbying to stop certain activities the meta-organization otherwise might undertake (Ahrne & Brunsson, 2005).
2.1.3 Decision-making structures
While meta-organizations actively attempt to unify their membership, there remains a high degree of differentiation between its members that hinders effective decision making (Burnsson & Ahrne, 2008). Many meta-organizations have put a system in place in which each member gets one vote regardless of their size and/or the resources they put in the meta-organization. This, however, is often criticized by members at the end of the spectrum that are much larger and pay significantly more than the membership average. For instance, the United States’ annual contribution covers 22% of the annual budget of UNESCO, but its vote only counts for 0,5% as UNESCO got 195 members of different sizes. Voting in general rapidly politicizes member’s involvement in a meta-organization which creates tension and conflict which will in turn have a negative impact on the overall aim of the meta-organization. Therefore many meta-organizations try to limit voting to the utmost necessity, instead attempts to find compromises among each other through consensus decision making as there is no hierarchy between members (Gulati, Puranam, & Tushman, 2012; Solansky, Beck, & Travis, 2014). While this harmonizes the internal relations in a meta-organization, the reached consensus might lack a clear direction as members will always have to find compromises to reach an agreement. Whether this is an acceptable trade-off depends on the importance of a meta-organization to retain its members and whether the members find it of crucial importance to preserve their own autonomy and identity. Consensus decision making is often referred to as soft rules that are not legally binding and are followed by members on a voluntary basis (Brunsson & Jacobsson, 2000). Meta-organizations therefore often use non-binding agreements that are presented as recommendations, guidelines, policies and protocols. Casula, Vifell and Thedvall (2012) describe this as “governance through
11
bureaucratization” that allows “action plans become more precise, formal and visible”. While
the set rules might be relatively soft, meta-organizations are able to put heavy authority over their members to force them to comply with the set agreements – which in turn allows them to, based on a unified membership, strongly affect policies that shape the environment. Casula, Vifell and Thedvall (2012) have identified three dimensions that decide whether the meta-organization is successful in this. The first phase, formalization of the process, requires to adopt organizational procedures that are embedded in formal structures. It defines the hierarchal structure of decision making and often enables participation of members through e.g. committees. The second phase, precision of scope, uses these set formal structures to clarify and redefine concepts that shape the environment. The meta-organization, as a powerful actor in the environment, can set these or adjust the status quo based on the input of its members. The third and last phase, visibility of policy output, encourages the meta-organization and its members to take collective action in disseminating and applying the (re)defines concepts to pioneer in and reshape the environment.
2.1.4 Inertia
Organizations that have a generally slow response rate to adjust to the changing environments are considered to have symptoms of inertia. For meta-organizations this is not different from traditional individual-based organizations, but they do have different drivers of inertia based on the nature of the organizational form (Köning, Schulte, & Enders, 2012). Köning, Schulte and Enders (2012) have identified the five core problem areas that stimulate inertia in the conditions under which meta-organizations work. First, it is argued the traditional culture of consensus decision-making comes from a tendency to avoid conflict and a lack of hierarchy due to its horizontal internal relations makes it unable for organization to adequately respond to changes in the environment. Second, many meta-organizations are argued to have an elitist identity which is defined as “a collective sense of
superiority within a meta-organization that is primary stems from the meta-organization’s tradition, heritage and culture as well as from its socio-economic importance” (Köning, Schulte,
& Enders, 2012). While a certain amount of elitism is naturally embedded in the organization to establish the legitimacy of its members. Too much reinforces the creation of a fixed collective identity that refuses adaption to changes in the environment and ignores external advice. Third, the disability of meta-organizations to attract champions. Champions are typically young, passionate and charismatic managers who drive organization change (Beath, 1991; Zaltman & Duncan, 1997). They however tend to prefer a fact-based and problem-oriented style to apply their knowledge, skills and flexibility to solve complex organizational issues. As meta-organizations often tend to be governed through bureaucratization (Casula Vifell & Thedvall, 2012) and have relatively little resources available (Burnsson & Ahrne, 2008), it is not surprising that these champions seek for a career elsewhere. Fourth, limited environmental intimacy which argues that meta-organizations tend to be unsuccessful in “the
amount and quality of information gathered by an organization about external developments within their respective fields” (Köning, Schulte, & Enders, 2012). Possibly because of the
12 internal relations that ask for collaboration within the meta-organization while the members often compete outside of the meta-organization. Thus, members are expected to only share their information and experiences when they foresee some personal benefit – or at least not a negative impact compared to their direct competitors (Cabrera & Cabrera, 2002). Lastly, protracted decision-making, which argues the internal decision-making procedures are too lengthy and complex before it can adequately respond to changes in the environment leading to organizations coping with high structural inertia.
In order to relax the identified inertia mechanisms Köning, Schulte and Enders (2012) argue meta-organizations should attempt to pursue five organizational changes. First, there should be constitutional reforms limiting the election period of representatives in decision-making bodies. A high fluctuation of representatives discourages the establishment of routines and interpersonal relations which are a source of inertia as sensitive issues are preferred to be avoided instead of addressed. Second the decision-making processes should be simplified to make them faster and efficient – this includes, among others, limiting the need for approval of regional decision-making structures. Third, meta-organizations should prioritize their efforts regarding environmental observations to increase their environmental intimacy. In order to respond adequately to changes in the environment one should constantly monitor its developments to stop opportunities and threats as early as possible. (Henderson, 2006). Fourth, the working structures should be designed to efficiently use the time of representatives. As argues by Ahrne and Brunsson (2005) participation in the meta-organization is often not the first priority of a member and therefore is reluctant to invest their time if it is not efficiently used. Furthermore Köning, Schulte and Enders (2012) argue that “meta-organizations that are able to establish time-efficient structures and procedures
have more champions in their ranks and generally make decisions faster than MOs without these structures” making it easier for them to respond to changes in the environment. At last,
meta-organizations are advised to moderate their meeting frequency considering that too little meetings will slow down organizational processes that are in need of governance approval and/or direction while a too high frequency will remove the sense of urgency to make such decisions.
2.2 Decentralization
Carter and Cullen (1984) formulated three perspectives on decentralization: hierarchical, concentration and participation. The hierarchical perspective makes the distinction between centralization and decentralization based upon a one dimensional argument; whether the power in an organization is located at senior management. Centralization is therefore described as “the extend of power and authority held at upper levels of the organization’s
hierarchy” and decentralization as “the extend of power and authority extended down through the hierarchy” (Hage & Aiken, 1970). Following this argumentation, more research has been
conducted that have resulted in different perspectives. The concentration perspective argues that it is not about whether the organization’s power is located high or low in the
13 organizational structure. Instead, it is about the concentration of power that defines the intensity of (de)centralization in organizations, in which a high concentration would mean that the decision making authority is allocated to a small group of people regardless of their level in the organization. The participation perspective takes a step further in building on this reasoning. It argues that it is not solely about the organizational level or the size of the group that makes decisions; it is about who participates in the decision making process and whether this involves people from across the organization (Carter & Cullen, 1984).
The opinion that decentralization will lead to an increased efficiency in organizations dates back decades (Oechssler, 1997). It is generally believed that decentralization has a positive effect as, among other arguments, management should not handle too many routine matters and it empowers lower-level employees to become problem-solvers (Perrow, 1977; Richardson, Vandenberg, Blum, & Roman, 2002). Furthermore, decentralization enables effective decision-making (Wooldridge & Floyd, 1990), increases local responsiveness (Enright, 2005) and stimulates innovation (Leiponen & Helfat, 2011). On the other hand, it is important to take into account that when decentralizing the organization’s operations you add another bureaucracy layer and have information spread across the firm, making it more difficult to make decisions based on all available information (Perrow, 1977). Decentralization is especially claimed to be of importance to organizations that percentagewise employ a high number of professionals dealing with complex regional or local environments (Richardson, Vandenberg, Blum, & Roman, 2002). As the number of knowledge based firms rapidly increases (Alvesson, 1993), the organizational landscape are slowly moving to a new way of governing organizations. This new governance structure is characterized by the emerge of decentralized authorities at the cost of central authority (Kersbergen, 2004). This shift towards horizontal and multilevel decision making allows organizations to better adapt to changes in both the internal and external environment (Richardson, Vandenberg, Blum, & Roman, 2002) and is often done by organizations that are currently performing under their aspiration level (Ashmos, Duchon, & McDaniel, 1998; Richardson, Vandenberg, Blum, & Roman, 2002). Though it is important to note that decentralization in itself is not the cure to make organizations able to perform better relative to its competitors in the environment. What organizations need to seek for is a balance between centralization and decentralization that are in implemented parallel to each other depending on the needs and characteristics of different levels of the organization (Brown & Eisenhardt, 1998). Regardless of the organization’s decision, it will always end up in a paradox in which the organization has to make certain trade-offs. When, for instance, an organization decides to decentralize its responsibilities, it may aim to increase utilization of local knowledge, but at the same time this might lead to an over consideration of local characteristics at the cost of the overall picture, also known as the self-regulation paradox (Carter & Cullen, 1984).
14
2.2.1 Effective decision-making
Fredrickson (1986) describes decentralization as an important influencing factor that shapes the decision-making process. It is stated that efficient decision-making through minimizing knowledge transfer costs and utilizing local knowledge (Christie, Joye, & Watts, 2003) is the main value-adding feature and often the initial reason, of starting a decentralization process that gives authority to lower-level employees who normally are not consulted, but are nevertheless able to contribute a great deal to the making of certain decisions (Richardson, Vandenberg, Blum, & Roman, 2002). It is however important to note that making the final decision is only the end of a process that went through several stages often already involving lower-level employees. Mintzberg (1976) has identified three main phases in the decision making process. First, the decision recognition phase in which one is confronted with an issue that needs a decision. Second, the development phase in which the issue is discussed and several solutions are mentioned, often in consultation with lower-level employees who have a better idea of the reality. Third, the selection phase in which based on internal discussions a final decision is taken. Mintzberg’s (1976) process has similarities to the process identified by Gioia and Chittipeddi (1991). It argues that before taking decisions one must first make sense of the environment, by understanding both internal and external conditions that influence the situation. This can best be done by engaging in a dialogue with lower-level employees whom often have a better micro view on the issue – whereas the higher-level employee can, based on the input of several people/departments, shape the macro view and take an informed decision. Whether such indirect or input generating involvement is sufficient depends on the organizational context. Organizations operating in predictable environments prefer central authority making as their performance is based upon stable decision-making to best fit their environment (Wooldridge & Floyd, 1990; Mintzberg & Waters, 1985). In such case solely input from lower-level employees might be sufficient to adequately respond to small changes in the environment. Rapidly changing environments, however, are argued to prefer decentralized decision-making to enable them to proactively respond to changes in the environment and to empower employees to go beyond what is expected from them (Koch & McGrath, 1996). In such cases shifting decision-making authority downwards has proven to have an attitudinal effect on employees (Koch & McGrath, 1996). However scientific prove that decentralization is directly related to the enhancement of the actual performance outcomes, for instance financial performance and growth rates, have been relatively weak (Porter & Lawler, 1965). Nonetheless Wooldridge and Floyd (1990) argue that decentralizing decision-making activities to middle managers impact the quality of decisions and thus, very likely also the overall organizational outcomes. It is however important to note that organizations are not to be considered either centralized or decentralized. As mentioned earlier, most firms are encouraged to find the right balance between centralization and decentralization that is most appropriate in both their internal and external environment (Brown & Eisenhardt, 1998). There is a wide range of options to choose from, from the completely autocratic decision-making process which is used in fully centralized organizations to the completely decentralized decision-making structure. Many firms have
15 characteristics of both and therefore categorized somewhere in the middle of these two extremes depending on the concentration of power and the participation of lower-level employees (Carter & Cullen, 1984). Whereas high participation in decision-making would indicate joint decision-making between managers and their lower-level employees, a high power concentration would indicate a dynamic participative philosophy when a select group from across firm levels is included in decision making (Leana, 1986).
2.2.2 Local responsiveness
The hype around globalization has led researchers to overlook trends in the world economy that indicate the emerging importance of regional economies, strategies and organizations. Organizations have increasingly invested in regional strategies that have led to multiple advantages such as more regional information on the differences of markets and government regulations (Enright 2005; Rugman & Hodgetts, 2001). Furthermore research has indicated that most multinational organizations compete for market share within their own region (Rugman & Verbeke, 2004) supporting the argument that the world economy is semi-globalized (Ghemawat, 2003). This argument is especially viable in the service sector, which currently employs approximately 70% of the work force in the Western world (Rugman & Hodgetts, 2001), and asks for customized products that meet local needs.
It is argued by Rugman and Hodgetts (2001) that successful multinational organizations are those that decentralized its operations with a high local autonomy to be able to adjust to the local environment and provide local responsive products and services. Global headquarters are often argued to lack local knowledge when making strategic decisions that shape the direction of the organization, while they have a tremendous impact on all levels of the organization (Ambos & Mahnke, 2010). While a global operating organization gives firm, location and region-specific advantages. To exploit those it is key to have local knowledge fed into strategic decision making (Goold, Campbell, & Alexander, 1998). Organizations are therefore increasingly applying regional strategies through the set-up of regional management centers that are either designed to utilize a decentralized multidivisional structure (M-form) according to sub-regions, lines of business or activities (Leiponen & Helfat, 2011). The literature divides two types of regional management centers: regional headquarters and regional offices. Regional headquarters are defined by Enright (2005) as
“an office that has control over the operations of one or more other offices and/or subsidiaries in other economies or countries in the region and which does not need to make frequent referrals to overseas parent companies”. Regional offices have a similar definition, apart from
the last sentence “does not need to make frequent referrals to overseas partner companies” which is omitted – distinguishing the two concepts by the level of local authority to make autonomous decisions (Enright, 2005). Regional headquarters have the authority to develop long-term regional strategies, control and coordinate regional subsidiaries and evaluate their operational and managerial performance (Paik & Sohn, 2004) and is deemed to influence the corporate global strategy through the organizational structure, active signaling behavior
16 and/or incentives from the global headquarters to do so (Mahnke, Ambos, Nell, & Hobdari, 2012) whereas regional offices have the sole mandate to support subsidiaries in execution of the global strategies and to feed local information back to the global headquarters.
2.2.3 Innovation
The literature distinguishes two views that innovation seeking firms should take into account when designing an organizational structure. The knowledge-based view argues that decentralization of research and development (hereafter R&D) though multiple locations will allow firms to better adapt to new markets through using existing product and market knowledge to create imitative innovations (Grant, 1996). Organizational economic research, in contrast, argues that firms should pursue a centralized R&D approach as this will lower the cost of communication and coordination and increase the generalizability of innovations which will in turn allow organizations to enter markets with entirely new innovations (Argyres & Silverman, 2004).
By decentralizing through using multiple R&D locations, organizations will be able to better adapt its innovation to the local market, also known as home-based-exploiting R&D (Kuemmerle, 1999). Furthermore, it might create spillover effects of new location-specific knowledge to other locations, called home-base-augmenting R&D (Kuemmerle, 1999). Though arguable a large part of knowledge can be identified as explicit knowledge which is hard to transfer, and even tacit knowledge will be hard to transfer as it requires understanding of the local context in which the knowledge has been created (Kogut & Zander, 2003). This will make sharing of knowledge between different locations challenging for the organization. When also taking behavioral research into account, this might even seem more challenging. Often organizations reward their employees based upon individual or team performance – however, if one is dependent on the sharing of information of another individual, that particular person will have very little incentive sharing that knowledge, which will create social dilemmas (Cabrera & Cabrera, 2002). Also the argumentation regarding the cost of communication and coordination between multiple locations can be put in question. In some cases this may be arguable, however the need for such coordination might vary depending on the purpose of R&D. When the knowledge is market and customer oriented the possibility of building economies of scope will be less. This indicates little need for coordination. In contrast, knowledge that is more generally applicable can be a catalyst for a range of innovations which will need coordination but also has a much larger potential to create ground breaking innovations (Nelson, 1990).
In short it can be concluded that organizations can best use a centralized R&D department if they are seeking generalizable innovations with a wide range of applications that will lead to new-to-market innovations. Organizations that aim to be market and product oriented are best advised to decentralize their R&D operations in order to be close to the customer’s
17
Figure 3.1: Three types of (de)centralization Source: Enright (2005)
needs. This will likely lead to imitative innovations based upon competitors whom have already introduced the new-to-market innovation earlier (Leiponen & Helfat, 2011).
3. Theoretical framework
As mentioned, most research in organizational theory has focused on individual-based organizations – also when it comes to the effects of (de)centralization. Meta-organizations, however, have a different composition and different purpose towards its members which imply that the traditional (de)centralization literature does not fully apply (Ahrne & Brunsson, 2005). To explore the effects of (de)centralization on meta-organization’s operations three variables: effective decision-making, local responsiveness and innovation have been chosen to build a theoretical framework. Though, argued by Brown and Eisenhardt (1998), an organization is not either centralized or decentralized as every organization has characteristics of both depending on several variables. However, to conduct this research the level of decentralization of an organization is analyzed against the three types of (de)centralization distinguished by Enright (2005) as displayed in Figure 3.1.
Global headquarters is a centralized
governance mechanism that functions as an overarching office of an organization and develops long-term strategies, controls and coordinate regional offices and evaluate their operational and managerial performance.
Regional office is a mechanism to
decentralize operations and has the role to support subsidiaries in execution of the global strategies and feed local information back to the global headquarters.
Regional headquarters is a decentralized
governance mechanism and acts as an
autonomous office that has the authority to develop long-term regional strategies, control and coordinate regional subsidiaries and evaluate their operational and managerial performance.
When using these three management structures of decentralization for the analysis, it is important to take into account that the precise function differs in the case of meta-organizations due to a lack of hierarchal power – as the power lies with its members (Burnsson & Ahrne, 2008). Therefore when talking about a global or regional headquarters it is often an elected board with member representatives that give guidance to the (regional) organization.
18
3.1 Local responsiveness
The world is argued to be semi-globalized (Ghemawat, 2003) and that successful multinational organizations are increasingly investing in regional strategies to better exploit the local knowledge that gives a competitive advantage and helps providing local responsive products and services (Enright, 2005; Rugman & Hodgetts, 2001; Goold, Campbell and Alexander, 1998). Arguably meta-organizations should follow this trend, as they aim to provide a favorable environment and support its members – whom apparently are working mostly on a regional level base. Being local responsive goes a step further when comparing to regional decision-making as it includes Mintzberg’s (1976) first decision-making phase: the decision-recognition phase, where one is confronted with an issue that needs a decision. It acts upon local issues instead of responding to global issues in a regional matter. It is argued (P1) that meta-organizations can be more local responsive through decentralizing to regional structures. Meta-organizations will be better able to respond to local trends by exploiting the region-specific advantages that they gain from their members (Goold, Campbell, & Alexander, 1998). The theoretical framework makes a distinction between global headquarters and regional offices (P1a) and regional offices and regional headquarters (P1b) recognizing there is not one type of decentralization (Brown & Eisenhardt, 1998). In the first proposition (P1a) it is argued that the regional office structures have the preference over global headquarters as the centre of analysis and support is brought closer to its members. In the second proposition (P1b) the regional headquarters has the preference over regional offices as its analysis directly informs its operations that can be, without any further consultation, be guided to the specific needs of the region. In comparison the regional offices may only feed information to the global headquarters that they may take into consideration in global strategic decisions. Arguably being local responsive may not be in the interest of meta-organizations as Ahrne and Brunsson (2008) state that “the more similar the members
are the stronger the meta-organization” as it will be better able to influence the
macro-environment. Following the regionalization trend will make the meta-organization better able to support its members directly, but may damage its purpose to influence the environment because its operations are fragmented based on regional conditions.
19
Figure 3.1: Level of local responsiveness in different types of (de)centralization
Proposition 1 (P1): Through general decentralized structures meta-organizations can be more local responsive in comparison to centralized structures.
Proposition 1a (P1a): Increasing local responsiveness in meta-organizations is more effective through decentralization to regional offices in comparison to centralized global headquarters.
Proposition 1b (P1b): Increasing local responsiveness in meta-organizations is more effective through decentralization to regional headquarters in comparison to decentralization to regional offices.
3.2 Effective decision-making
When talking about effective decision-making the word effective is defined as “adequate to
accomplish a purpose”. Meta-organization’s purpose is often considered to be twofold: 1)
creating an optimal environment for its members and 2) support its members in their operations by functioning as a platform for engagement (Burnsson & Ahrne, 2008). It is argued (P2) that meta-organizations are better able to achieve these goals when they have decentralized their operations. Decentralization of an organizational decision-making structure goes from a complete autocratic decision-making process to a certain level of decentralized decision-making structure (Carter & Cullen, 1984) giving authority to lower-level employees whom were formerly not consulted but, nevertheless, have valuable information to add that will help making effective decisions (Richardson, Vandenberg, Blum, & Roman, 2002). Arguably (P2a) to contribute to the achievement of the meta-organization’s purpose it is important to build in some level of decentralization in its decision-making, such as the establishment of regional offices to allow lower-level employees to participate in the decision-making and with that being able to make better informed decisions at a global level leading to the accomplishment of its purpose (Carter & Cullen, 1984). Depending on the level of decentralization it requires active involvement of lower-level, regional based employees in
20 both the development and selection phase identified by Mintzberg (1976). When using Enright’s (2005) regional management centers distinction the regional office feeds information into the decision-making process at the global headquarters, whereas, if established, a regional headquarters would be authorized to make a final decision in the selection phase without consultation with the global headquarters. When making a distinction between which is most beneficial for stimulating effective decision-making in the organizational context of a meta-organization it is argued (P2b) that the regional offices have the preference over regional headquarters. When explaining this proposition it is important to take into account the purpose of the meta-organization. Whereas both the regional office and regional headquarters can provide a supportive platform for membership engagement, creating an optimal external environment can only optimally be done if an organization can speak with one unified voice that is free of any regional differences. Furthermore, as argued by Ahre and Brunsson (2005) the participation of members in meta-organizations is often not the first priority, therefore adding another formal governance layer on regional level will require further engagement from its members which increase the lengthiness of the decision-making progress. Therefore Köning, Schulte and Enders (2012) advocated for more simplistic decision-making to make better use of the membership’s limited time for engagement in the meta-organization.
Figure 3.2: Level of effective decision-making in different types of (de)centralization
Proposition 2 (P2): Through general decentralized structures decision-making in meta-organizations can be more effective in comparison to centralized structures.
Proposition 2a (P2a): Decision-making in meta-organizations is more effective through decentralization to regional offices in comparison to centralized global headquarters.
Proposition 2b (P2b): Decision-making in meta-organizations is more effective through decentralization to regional headquarters in comparison to decentralization to regional offices.
21
3.3 Innovation
The literature distinguishes two views on (de)centralized innovation: the knowledge-based view (Grant, 1996) and the organizational economics view (Argyres & Silverman, 2004). In which the knowledge-based view argues innovation through decentralizing R&D departments has more potential as it will allow organization to better adapt to markets, the organizational economic research, in contrast, argues a centralized R&D approach will lower the coordination cost and increase generalizability of innovations. In the case of meta-organizations its purpose is to support its members (Burnsson & Ahrne, 2008), these members are argued to work mainly on a regional basis (Rugman & Hodgetts, 2001). Therefore it is argued (P3a) that meta-organizations should decentralize their R&D structures to better fit the regional context of its members. Despite adjustment to a specific regional context it is argued (P3b) that through working with regional offices over regional headquarters its potential to scale will be bigger. Regional offices have a stronger linkage to the global headquarters. Collaboration between the two will take into account both the holistic macro-view from the global headquarters as well as the regional focus of the regional offices. This will allow both views to be embedded in the innovation which will increase the ease of transferability both because its knowledge is present at both the global headquarters and the regional offices as well as the innovation may be region specific but has flexibility to adjust to similar challenges in a different context. R&D that is solely executed by regional headquarters is more effective to address a certain regional challenge, however, its potential to scale is limited as it is relatively fixed. Increasing the effectiveness of R&D through establishing regional headquarters is more effective on a regional basis, but also more costly as its application is limited. R&D through regional offices, on the other hand, has the potential to scale, decreasing its overall costs and its application wider but faces challenges in transferring knowledge to different contexts. Innovation through the meta-organization’s global headquarters are argued to be most suitable if its purpose is to address global challenges and/or come up with break-through inventions (Leiponen & Helfat, 2011).
22
Proposition 3a (P3a): Through decentralized structures meta-organization’s innovations will better fit the regional context of its members in comparison to centralized structures.
Proposition 3b (P3b): Innovations through decentralized regional office structures are more likely to scale in comparison to regional headquarters.
4. Methodology
Organizational theory has traditionally focused itself on individual-based organizations – while the majority of the organizations belong to this type of organizational form it has a somewhat narrow-minded view. This research, instead, specifically focuses on meta-organizations. The research is exploratory of nature as it aims to seek new insights regarding three variables in the concept of meta-organizations, which are generally spoken about in individual-based organizational settings (Saunders, 2009). It questions whether decentralization of a meta-organization’s operations has a positive effect on the local responsiveness, efficient decision-making and innovation and, if so, explores which regional management form is most suiting in execution. The research is deductive of nature as its theoretical framework derives from existing organizational literature. This framework tests the effects of decentralization on the three variables in the context of a meta-organization. For this purpose a qualitative research approach was considered most suitable due to the richness of its data in understanding the dynamics and theories in one single setting (Eisenhardt, 2009). The research has been conducted through a single case study, which is described by Saunders (2009) as “a research strategy that involved the empirical investigation
of a particular contemporary phenomenon within its real life context using multiple sources of evidence based on qualitative research methods which are based on numerical or non-qualified data”. Yin (2003) argues that a case study is a suiting research design if the case
meets three set criteria. First, the research should aim to answer a how or why question. Second the phenomena should occur in a real life situation and last the control of the research on this situation is limited. The International Federation of Red Cross Red Crescent Societies (hereafter the Federation) has been selected as it meets these criteria and has decentralized its operations in 2007 with the establishment of regional offices to enhance their support towards its members (hereafter National Societies) in their specific needs in the region. Whether the decentralization process has contributed to this will be tested through the theoretical framework. The single case study will be explorative of nature and be a first step to better understand the mechanisms of meta-organizations (Rubin & Rubin, 1995). The case study tests the theoretical framework in the organizational context of the Federation. It will be guided by two main sources of data.
4.1 Archival research
The archival research will help to better understand the organizational context and the processes the organization went through before, during and after the decentralization process of 2007. This archival research was done through the analysis of secondary data
23 related to the decentralization process. In the first stage 243 documents from governing board sessions, statutory meetings and other pre-post documentation that are related to the decentralization were identified. These documents have been scanned on their relevance and have informed the author on the broader organizational context. In the second stage the documents, directly related to the decentralization, were re-examined and its relevant content coded systematically to prepare them for further in-depth analysis. An overview of the documents that have been used can be found in Table 4.1. As can be seen in this table, a first decentralization review has been conducted by Accenture in 2013. This document mainly focused on the implications towards the Global Senior Management Team. Differently, this research has a more regional and concrete focus to test the theoretical framework and by doing so it looks specifically into the situation in the Europe Zone and whether the organization, with the establishment of regional offices has successfully enhanced the Federation’s support function towards National Societies in terms of local responsiveness, decision-making structures and innovation. The archival research has helped to gain a better understanding of the organizational context and a deeper understanding on the implementation process. It has provided an overview of the theoretical plans, whereas the interviews gave an insight into the practical outcomes of the implementation. Furthermore, it has enabled the author to better guide the interviews as well as the interpretation of its findings.
Documents Release data Stage 1 Stage 2
IFRC Governing board 11th session (2005) un till the 31th session (2015)
incl. extraordinary sessions
Letter of the President
Decision sheets Summary records 26 26 26 5 5 5
IFRC General Assembly 14th session (2003) un till the 19th session (2013)
Decision sheets Summary records Incoming documents 6 6 147 4 4 18 Strategy 2020 1 1
Accenture Decentralization review 2013 2 2 IFRC Europe Zone Office Operational Plan 2014, 2015 2 2 IFRC Europe Zone Office Paper on Implementation of
decentralization review recommendations
2014 1 1
Total 243 47
Table 4.1: Used internal documentation 4.2 Interviews
To get a better insight in the effects of the decentralization process in the Federation, interviews have been conducted to gather qualitative data. As shown in Table 4.2 the data for the research has been gathered by interviewing 10 representatives from various components
24 of the Federation. The respondents were carefully selected through critical sampling ensuring senior level representation of the IFRC Global Secretariat in Geneva, the IFRC Europe Zone Office in Budapest and Europe Zone National Societies whom have organizational experience in both the centralized and decentralized organizational structure and participate in or are directly affected by management and governance decision-making. The data has been collected through semi-structured interviews which have allowed for the questions to stream rather fluid while having a consistent line of thought in the background (Rubin & Rubin, 1995). The interviews have been conducted via Skype as all respondents worked in different places across Europe. The purpose of these interviews have been twofold. First to test the theoretical framework in the Federation’s organizational context and second to understand the process of decentralization and whether it has been beneficial for the Federation and National Societies in terms of enhancing the support function of the Federation. The guiding, questions were open-ended and focused on testing the theoretical framework and getting a better understanding of the context in which the decentralization took place. They have been drafted in close cooperation with the Federation and the University of Amsterdam to ensure practical and academic relevance. The interviews have been conducted anonymously to prevent social desirable answers and to create a safe environment to speak.
Position Abbreviation of respondent Length Page Transcript Words of transcript IFRC Global Secretariat (Geneva)
Global management team
Head of department R1 R2 R3 R4 54m10s 48m55s 46m37s 38m59s 10 8 9 7 5.634 4.286 4.745 3.631
IFRC Europe Zone Office (Budapest)
Regional management team R5 R6 R7 41m57s 1h7m56s 1h2m43s 9 12 10 5.194 7.159 6.331 National Societies Head of department R8 R9 R10 46m15s 38m45s 1h2m33s 7 8 9 3.763 3.792 5.536 Total 8h38m50s 89 50.071
Table 4.2: Profile of the respondents 4.3 Data analysis
The analysis of the Federation’s archives had the objective to better understand the process of decentralization and how this effected the organization internally – allowing for more specifically directed interviews and putting the data gathered by the interview into an organizational context. Furthermore, it provides a brief introduction to the Federation’s
25 structures and objectives for readers whom may not be familiar with the Federation as a meta-organization.
The interviews, with permission of the respondents, have been recorded and, as shown in Table 4.2, transcribed. Before preparing them for further analysis the content has been checked with the respondents for possible adjustments or clarifications to their statements. The author coded the different sections and quotes in the interviews. As argued by Pratt (2009) the author himself can best transcribe and code the interviews to ensure that the context is taken into consideration during the data analysis. Usage of software for qualitative analysis such as Nvivo, was considered, however with only ten interviews the amount was considered to be manageable to do its analysis manually. Five broad categories have been identified: 1) process of implementation, 2) local responsiveness, 3) decision-making structures, 4) innovation and 5) performance and evaluation. These have been defined before the categorization process as they have been based on the interview questions. Based on the answers of the respondents subcategories were identified (see Tables 6.1 until 6.6 in Annex III) giving a better insight into the broad categories. After categorization the codes were analyzed one by one aiming to find patterns in the effects of decentralization in the set variables seen from different perspectives in the organization. These patterns have been tested to the propositions in the theoretical framework to analyze whether they are supported in the context of the Federation. The findings have been illustrated with several quotes from the respondents and an abbreviation so it can be linked to a respondent in Table 4.2. The validity of these results have been secured through triangulation as multiple types of sources of evidence have been used to look at the theoretical framework from different perspectives. The critical sample has been characterized by its diversity in terms of organization component and geographical coverage to show different perspectives and to minimize any biases (Yin, 2003). By using this research set-up the external validity is limited as the results are specific to the Europe Zone in the Federation and therefore cannot be generalized to meta-organizations in general. To overcome this limitation more research with a broader range of meta-organizations needs to be done to present generalizable findings. Furthermore, the Federation is a meta-organization that has decentralized to regional offices and has only been discussing about taking the decentralization to a next level through the establishment of regional headquarters. The propositions that include a distinction between regional offices and regional headquarters (P1b, P2b and P3b) in the theoretical framework can therefore not be tested, but can only be discussed based on expectations. The interviews, however, have provided valuable information on these propositions and are therefore discussed in this paper without making any conclusions. It is important to note that these propositions are not tested and further research need to be done in meta-organizations that had have such a development to present valid findings.
