Global finance as an action research domain: Testing the boundaries

Introduction

In January 2008 a team began a project to develop a long-term process to integrate social, environmental and economic concerns into the logic of the global financial system. A first set of interviews asked whether we can talk about ‘global finance’, or whether we have to talk about sub-industries such as insurance, investment firms and banks. Interviewees were almost unanimous in the belief that those industries are so different and that national boundaries are so important that they absolutely cannot be thought of as one integrated system. Within months, such divisions were rendered meaningless as the global ‘financial’ system careened towards collapse.

The research team was assembled through an NGO founded by the author dedicated to supporting development of global multi-stakeholder change strategies (the NGO no longer exists). The author had worked for several years with Verna Allee (Allee, 1997, 1999, 2000, 2008) to develop a new mapping approach called Value Network Analysis; Allee and her colleague Oliver Schwabe joined the team. Jem Bendell and Sanjeev Khagram are other long-term colleagues who helped. Bendell focuses on corporate change and sustainability (Bendell, 2000, 2002, 2006) and Khagram on large change topics (Khagram, 2005, 2006; Waddell & Khagram, 2007). In addition, the authors of Democratic dialogue (Pruitt & Thomas, 2007) supported the meeting development; two junior researchers also assisted. Interviewees included decision-makers and NGO innovators, described in more depth below.

The 15-month initiative was structured as an action research project to produce a strategy and leadership group that could pursue the long-term goal of global financial system transformation. Our initiative had five distinctive features. Each individually made it a quite unusual action research project; collectively they made it a unique one.

The project focused on a global system (broadly defined as ‘finance’; defining it in an actionable way was a project goal), rather than an organization, a geographically constrained grouping, or a network. Organizational action research has tended to happen in the ‘organizational learning’ tradition (Argyris, 1999; Argyris & Schön, 1978) or geographically narrow communities such as with ‘community development’ (Stringer, Guhathakurta, Masaigana, & Waddell, 2008). There has been important action research with networks, such as transnational research networks (Brown & Gaventa, 2010), national development networks (Engelstad & Gustavsen, 1993; Gustavsen, 1998), and multi-site projects driven networks (Fuller-Rowell, 2009). A recent change noted in health care is ‘a shift of research focus away from the organizational structures and processes of specific care functions towards studies of the boundaries and boundary objects between them’, (Huzzard, Ahlberg, & Ekman, 2010). Action researchers increasingly embrace organizational complexity and systems as qualities of their fields of investigation.

Action researchers are urged to think about scale (Bradbury-Huang, 2010). However, no examples have been found by the author of a project that focuses on a truly ‘global system’. Scale and diversity with this focus are unparalleled. This scale raises particular questions and challenges, such as about the resources and methodologies that might make such an approach effective and how to ‘make sense’ of the scale for people in an actionable way when an underlying premise is that diversity and participation are valued.

A second distinction is that the system was very ‘under-defined’ (Brown, 1983; Trist, 1983). As the opening paragraph describes, people did not even recognize the proposed system boundaries as legitimate when the project began. At one extreme nearly everyone in the world could be considered a financial system stakeholder. Any decisions by researchers to identify a tiny group to engage would inherently suggest boundaries that could create trajectories in unpredictable ways. This presented particular challenges in defining the community and participants, and engaging them meaningfully. Indicators of ‘meaningful’ engagement would be a consensual sense of the general structure of the system and production for participants of useful tools to share insights, build relationships and take joint action.

The goal to support changing the ‘logic’ of the system goes beyond single- and double-loop learning (Argyris, 1976) processes, to better attend to triple-loop learning also (Issacs, 1993; Nielsen, 1993; Romme & van Witteloostuijn, 1999; Yuthas, Dillard, & Rogers, 2004). Triple-loop learning has drawn various definitions. Some refer to it simply as ‘learning to learn’. Yuthas et al. ‘use the phrase ‘‘triple-loop’’ learning to capture the notion of continual reflection on the learning process, the contexts within which learning occurs, and the assumptions and values motivating the learning and influencing its outcomes’ (Yuthas et al., 2004, p. 239).

Scharmer does not use the term ‘triple-loop learning’, but nevertheless distinguishes his ‘Theory U’ in juxtaposition to single- and double-loop learning that are ‘learning from the past’ with what is sometimes referred to as ‘the Kolb learning cycle’ (Kolb, 1984). Theory U involves shifting from the inner place of one’s functioning, to a future space of possibility that one feels is wanting to emerge (Scharmer, 2009). This requires different tools such as scenario planning where people can free themselves of unrecognized assumptions or emphasize them to varying degrees to imagine how things can be different. Waddell (2011) presents these three learning processes from various dimensions as incremental change (single-loop), reform (double-loop) and transformation (triple-loop). How to create processes to support the transformation raises deep questions about methodologies and the relationship between the three types of changes. For example: can sufficient incremental change result in transformational change? How do we shift ‘natural science’ to a new paradigm (Kuhn, 1962) or operating logic that supports transformation?

The fourth distinctive feature of the work described here is that it was the researchers who initiated the project, rather than having others come to the researchers for help (Schein, 2001) or being contingent upon system participants first agreeing to work with the researchers (Senge & Scharmer, 2001). The researchers did not profess to have substantive expertise (in finance), but rather expertise in system organizing and change. This raises particular challenges around expert and discursive legitimacy (Huzzard et al., 2010). How can people without these types of legitimacy effectively act in an issue domain (Trist, 1983)?

Perhaps no issue domain is so wrapped up with extreme issues of power, with such entrenched interests that are so central to power structures; this is the project’s fifth distinctive quality. The long-term aspirations would seem to require a ‘. . . positive kind of power that can be called ‘‘the power of balance’’ . . . (where leaders):

But how can this be developed?

These distinctive qualities connect the project to emerging strategies to address pressing issues, actions and ideas. Although they do not usually have formal ‘action research’ frameworks, this includes actions to address climate change, popular practitioner notions such as the ‘tipping point’ (Gladwell, 2002) and strategies to address health issues represented by the Global Fund to Fight AIDS, Tuberculosis and Malaria. Believing that action research has an important contribution to make to these sorts of efforts because it reflects a new type of participatory action at a global scale (Frieze & Wheatley, 2011; Hawken, 2007), the work is understood as global action research. We are, in fact, developing new global ‘institutions’ (Khanna, 2011; Waddell, 2011) and action research makes a contribution to ensuring the values that it reflects are core to what is emerging.

The project goals and action research

Box 1 outlines the goals associated with the long-term activities of the Global Finance Initiative (GFI) as the project was called. The first 18-month phase had three goals articulated in a proposal to the Ford Foundation, which provided seed funding of US$180,000. The first goal was to create a definition of ‘the global financial system’ that would be useful to its participants and provide an actionable framework for those interested in its transformation. The second goal concerned defining a strategy for the system’s transformation with the leadership of the system’s change advocates. And the third goal was about developing methodologies that would be useful to support the first and second goals.

Box 1: GFI Project Goals

The research team recognized the audaciousness of its goals. It approached the project as an initial step in what would quite possibly be a multi-decade initiative that would give rise to a new governing logic. The GFI was a natural development of the author’s long-term interest of in the finance system and its role in defining social opportunities. During the 1980s Waddell led in reframing the world’s largest community-based credit cooperative’s (credit union in Canada) goals to provide both leading financial services and community well-being; in the 1990s some work focused on bank–community relationships. Work on multi-stakeholder relationships and questions about change processes to support realization of people’s highest aspiration led to developing the concept and practice of ‘societal learning and change’ (Waddell, 2005). After contributing to a report to UN Secretary-General Kofi Anan on the future of global governance, Waddell then globalized the work and coined the term ‘Global Action Networks (GANs)’ (Glasbergen, 2010; Waddell, 2003, 2011; Waddell & Khagram, 2007) to describe an emerging global, multi-stakeholder change and governance structure.

GFI can be described as global action research project in terms of the characteristics and quality choicepoints associated with action research (Reason & Bradbury, 2001):

However, by emphasizing the importance of developing new tools and methodologies, the project aimed to treat action research itself as an emergent developmental form. The scope and complexity of the global finance issue domain was beyond anything value network analysis had addressed.

The project activities

There are many important action research questions associated with this project. Given the brevity of this article, the author has chosen to focus on the challenge of creating a common sense of what ‘the global financial system’ is. The importance of this is associated with the belief that an important quality for effective action is a shared sense of community and the roles of people and institutions in it. For global finance, the complexity of the community can easily lead to disempowerment of those not at its power center, as people can not understand the current power structures and therefore are at their mercy.

To meet this challenge, several methodologies were used. One type involved ‘visual diagnostics’, a term being used with increasing frequency in the social sciences to describe methodologies that present data with an emphasis upon diagrams and drawings rather than words. One form of this better-known to system change experts is mind-mapping (Buzan & Buzan, 1996). Another kind of visual diagnostics is social network analysis (SNA) where diagrams of arrows and nodes describe relationships between people, roles, or organizations. Mapping is a tool of growing importance in network development (Krebs & Holley, 2004; Waddell, 2011).

The project team decided to combine three mapping methodologies because of their promise to complement one another and to contribute to insights about particularly large issue system. Mapping was chosen as a key methodology for two reasons. The more prosaic one is its ability to identify stakeholders and their relationships in order to identify who to strategically engage in the project and future stages of it. The other was based in the belief that if people could see the financial system as maps, it would lead to empowering insights about relationships that would produce both a stronger sense of community and more strategic action.

With web-crawl mapping the key input is a number (usually 10 to 30) of ‘seed’ website addresses of organizations known to be active in an issue arena. Software developed by Richard Rogers at the University of Amsterdam (issuecrawler.net) generates maps describing the relationships between URLs. These maps are themselves useful as a way to easily portray relationships, but they also generate useful lists of URLs to suggest other organizations for interviews. They help identify key ‘bridging’ organizations (Brown, 1991; Westley & Vredenburg, 1991), hubs and social structures that helped determine who to invite to stakeholder meetings. Of course, the tool is predicated upon relatively well-developed web-sites; anyone that is a serious player in global finance could be expected to have one.

Value Network Analysis (VNA: see valuenetworks.com/) (Allee, 2011) is a mapping methodology developed by Allee that represents systems in terms of roles and exchanges between roles (‘deliverables’). This is very useful when you want to understand what roles and exchanges are necessary for a healthy issue system, which roles need more attention, and which might be so well resourced that they are creating problems such as needless redundancy or bottlenecks (Allee, 2008). There are two forms of VNA: participant-based VNA is similar to SNA where entities (individuals, organizations or stakeholder groups) are the key unit of analysis that is connected by tangible and intangible exchanges.

With role-based VNA nodes are specific roles of individuals, stakeholders or organizations rather than those entities. For example, there may be a role of ‘policy advocate’ or ‘evaluator’; although the role may be played by more than one entity, it will only appear once, with linkages to it raising important insights about what is needed for that role to be played well. This moves fully into the 21st-century network-centric rather than organization-based world. It also allows moving beyond the confusion and emotional loyalties that organization-based analysis in a network can produce.

Of course, these mapping methodologies are only as good as the conversation and dialogue that they produce. The maps themselves are like books: full of information, although the information presented in one map can fill a written chapter. They make ‘seeing’ a system and relationships much easier, to be able to understand its dynamics and how to take strategic action. The question is how to use that information to produce insights and action. A key activity for this is ‘validation’ of the maps – creating community discussions about them and refining them to reflect these discussions.

The team undertook the following steps:

Step 1) Initial interviews: These 31 interviews identified and explored different frameworks, such as ‘global finance’ and ‘ESG’ (environmental-social-governance) as ways to define the project boundaries and stakeholder groups. The interviews were with people in leadership in various stakeholder groups (see Box 3 for the eventual list of groups). They were identified through document review and the project leader’s knowledge of key organizations in the domain, along with references in interviews.

Step 2) Initial web maps: The maps were developed based upon the organization websites of people interviewed.

Step 3) Report 1: Steps 1 and 2 resulted in an initial report in May 2008 and shared with those interviewed. One major finding was that the number of multi-stakeholder networks in the issue arena had grown substantially in the past five years, suggesting that the domain was maturing with a new capacity for interaction (Waddell, 2011). A second finding was lack of confidence across the board in ‘the system’s’ ability to respond to the emerging crisis. The report noted that it would be impossible to engage senior people in the system in the GFI, given the crisis. On the other hand, people were much more prepared to talk about major change. In part, this is the reason that the project began to emphasize connections with the critical voices outside the traditional power structures, in particular with the multi-stakeholder networks that already had capacity to work across divides.

Step 4) Stakeholder meeting 1: In August at the London School of Economics a stakeholder meeting was held with 17 people, including five from the project team with facilitators. A decision was made to focus on people outside of the traditional powers of the finance industry and public regulators for two reasons: 1) the belief that the ‘outsiders’ needed organizing, and 2) fear that with insider-experts the conversation would devolve into intimidating technicalities. The participants included people from multi-stakeholder organizations dealing with finance since the team felt they had more of a ‘system’ perspective, two leading academics, and some NGO representatives. The meeting was presented as follows:

The meeting ran into challenges, despite preparation that included interviews in advance of all participants and professional facilitation. ‘Participants: (1) challenged initial investigation findings, (2) identified some bridges that could help to change the current global financial landscape, and (3) suggested next steps’ (GFI, 2008). ‘One of the great learning was that this global financial infrastructure is not on people’s agenda’ (Pruitt, 2008). People did not think about it, and the concept was vague.

The GFI Team aimed to get the participants’ support for filling in the VNA survey, but this was not in the cards. People left very skeptical about the value of the methodology itself. Rather, the meeting produced a plan to further consultations and nurture development of a Stewardship Team.

Step 5) Community consultations: Following the workshop both individual and collective virtual conversations were held with those who attended the meeting. This produced an initial group of three significant stakeholders who offered to help ‘steward’ the emergence of a more formal group and advise the team. They also identified four important meetings of finance system stakeholders and ensured invitations to the Project Team. Luckily these coincided with other team work, so travel expenses to the Geneva, Singapore, Washington, DC and Washington State meetings were minimal. This allowed further conversations and even presentations with stakeholders about the GFI, and further development of the project team of the dynamics and development of relationships.

All of this work produced what proved to be very useful descriptions of the ‘global finance system’. The description took two forms. One was based on the concept of global public policy for finance as described in Box 2, which identified nine key institutions.

Box 2: A Definition of the Global Finance System

This project aims to map how global public policy (GPP) for finance is made. By GPP for finance we mean institutions that:

We want to understand the roles and exchanges in that system (Role-Based VNA) and the relationship between key organizational actors (Participant-Based VNA).

The second definition was by stakeholder group, defined by their roles and goals. These were divided first into two groups: the insiders or current power-holders who are the ones effectively making the decisions. The second group is the outsiders who are pushing for innovation and integrating some aspect of the social-environmental-economic impact concerns. This division reflects a typology proposed by Mitchell, Agle, and Wood (1997).

Box 3: The Global Finance System by Stakeholder Group

Insiders/Decision-Makers

Outsiders/Innovators

Step 6) Mapping part 2: The project team remained convinced that a collective understanding of ‘the system’ and current actors’ roles in it, is essential to develop an effective change strategy. The two definitions provided a framework for much more effective mapping. Therefore, both web crawls and VNA were taken up again. Individual web crawls were undertaken for each stakeholder group, to both better identify participants in the stakeholder group and their relationships to the global public policy financial institutions.

Figure 1 is an example from a crawl seeded with URLs of what we identified as the critical and influential actors in the global finance public policy arena. For ease of viewing, we have limited the number of nodes appearing on the map. This map is based on co-links between the URLs, with each arrow representing outlinks from one URL to another. The seed URLs are triangles; these site nodes are well connected, centrally located and large in size. In addition to their presence in the map, the relative location of the nodes is also telling. The web crawl shows the bis.org node (Bank for International Settlements) to be in the center of the map. Their central location and size of the node is an indicator that BIS has many inlinks from a wide range of other web sites (even more obvious in a many where the restriction on the number of nodes appearing is removed). OPEN IN VIEWER

The lack of support for the VNA mapping by the stakeholders meant the planned data-gathering through online surveys was impossible – people simply would not respond without some high-legitimate figures asking them to do so. Consequently, the team opted for a much less robust, but surprisingly useful process, of generating VNA maps through website analysis and some follow-up interviews.

The VNA mapping had never been applied to a domain of the scale and complexity of the global finance system, and considerable progress was made with application issues. Of course, you do not want all the roles and participants in a system, but only the ones who are key to addressing your problem statement. In the GFI case, these were ones associated with policy-making. To arrive at this type of simple representation requires addressing two particular challenges. One is to define the roles – both conceptually and in terms of the words used to describe the roles and exchanges, so that they are meaningful to diverse stakeholders. This should be guided by the ‘cut of analysis’ that will produce a number of roles that is insightful: National? International? By industry? Sub-issue? A second challenge is data aggregation/reduction. A system like global finance must be made both simple and complex enough to be actionable. For GFI, global and national were important distinctions.

Figure 2 is a VNA map of the roles in global public policy (gp) development represented by nodes and the tangible exchanges represented by the arrows between them. VNA also maps intangibles, which are at least as important, not shown here for simplicity concerns except for the illustrative ‘influence’ dashed arrow between ‘gp developer’ and ‘secretariat’ nodes. The greatest contribution of this map is to make discussable a complicated system like global finance. It shows that there are a modest number of critical roles in developing the gp for finance. These roles can be organized in a number of ways, and by distinguishing between them and their interdependence, new ways of organizing the gp system for finance can be proposed. As well, this map about the current system facilitates discussion about the strength and weakness of some roles – such as the stakeholder one (referring to the formal role of stakeholders for influencing gp development). OPEN IN VIEWER

Step 7) Second stakeholder meeting: In January 2009, 10 stakeholders and the project team convened at Harvard Business School with the core question ‘How to realize large system change in global finance?’ The financial crisis had deepened by this time. Despite Obama’s election there was great pessimism about the potential for change of the needed scale – everyone was simply focused on trying to hold together the old, rather than develop something new.

Only two of the second meeting stakeholders had also attended the August meeting. The invitations were more clearly focused and the work in the intervening time produced new contacts – in only one case did the project extend an invitation to a person at the previous meeting who declined. We were clearer about the second meeting conversation, and could more clearly identify appropriate participants who were envisioned as potential ‘stewards’ for subsequent steps. As in the earlier meeting, these were people from leading organizations in the ‘outsider stakeholder’ category.

This meeting began with people sharing what they were doing in response to the financial crisis, and the opportunities they were developing. The project team then briefly introduced its definitions and maps of the global finance system, and then conversation turned to ‘innovations to change the system’ with questions such as ‘what’s the leverage point we should be aiming for?’

The mapping proved useful to this discussion. It visually demonstrated that most NGOs who think they are out to change the financial system are not even focused on the most powerful of global public policy institutions: the Bank for International Settlements and the Financial Stability Forum. Instead, they are focused on the UN as a potential ally (but by purposeful design it has no role in the system) and as an adversary the World Bank (which indeed is an important system actor). Comments illustrated that this was a new and valuable perspective for the participants. Notable in the subsequent GFI action plan developed by the participants was absence of reference to traditional targets like the World Bank, IMF and UN.

The VNA maps also produced useful discussion. For example, one participant asked ‘why aren’t the rating agencies on the maps?’ The answer: the maps are of the global public policy financial system – and rating agencies are private sector actors. But should they then be part of the public system?

The closing session was about ‘What role can GFI or others play to realize your goals?’ The key meeting outcome was a strategy for moving ahead. A key element of this was to create a global discussion about ‘What should be the purpose of the global financial system, and on what principles should it be based?’

Step 8) Project summarizing. Over the next couple of months the project team brought together all the information and with further consultations with the January meeting participants wrote a final report. At 20 pages, the report was designed for easy consumption by stakeholders, rather than as an academic document. It included a strategy to build a coalition of the ‘outsider stakeholder groups’ and a six-step action plan that was quite feasible to develop over the next three years. It was designed to continue with the action research strategy of creating regional discussions hosted by key organizations to inter-actively build understanding of ‘the system’ and the ties between stakeholder groups so they would be in a good position for action when the next crisis hits – as it surely will within the next few years.

The concept of ‘stewardship’ was a guiding one behind the project (Block, 1993). January meeting participants would have stewardship roles in implementing the plan. We never intended to have a ‘representational’ group at the meetings, given global diversity that would suggest a much too large of a group. Rather, we aimed for a group that had sufficient diversity in voice and legitimacy with key stakeholder groups, as well as capacity to work together with commitment. Leadership would continue to grow and shift as the projects development was ‘stewarded’.

Project learnings

In terms of the three project goals (Box 1), this project may be considered successful although the relationships did not extend to ongoing financing. There were important learnings for future similar efforts.

Conclusions

The recent financial crisis followed a series of earlier ones of increasing magnitude. I feel compelled to refer to the inevitable future global financial crisis, almost certainly of even greater scale, that I have become convinced we now face. We have undergone minor reforms. There has been no apparent embrace of the level of systems transformation that is required to prevent crisis.

Thus, to use our time well is to organize in preparation for the next global crisis. With the recent one, two key elements were missing to turn the crisis into opportunity. One was vision – there is no rallying vision about how the system can be different. The one behind this project – a system of multi-stakeholder control grounded in a clear sense of purpose and values – is a start, but requires much more work to inspire with confidence. The second missing element is a network of committed advocates for change who can generate the political will (pressure) to realize the vision when the next crisis strikes. Developing such a network that can face the massed powers of the financial system insiders is a huge, but not impossible task. Action research can make an enormous contribution to bringing the discipline, rigor, insight and human connections to address this challenge.