Chapter 2: Literature Review
5. Discussion
5.4 Layers of Agency
國
立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
distance, though both indicated a willingness to continue using the property for
community based projects in the future. While it is likely that time and money constraints do indeed play a part, there are two other possible interpretations that arise from two separate branches of literature. While much agency literature focuses on the need to monitor agents to protect against “shirking”, others indicate that shirking can also occur on the part of the principal (Wiseman et al., 2012). This may be exacerbated in situations with multiple principals and lightly enforced informal contracts. Monitoring on the part of agents in this case may also be problematic if hierarchical relations or ingroup-outgroup biases are especially salient.
A second interpretation of this situation can be found in principal-principal issues regarding weak or unclear property rights. This mirrors a similar situation in many remote indigenous areas in Canada where infrastructure (housing and schools often built at some expense by the national government) tends to deteriorate more quickly than expected. Canadian parliamentary hearings in 2015 suggest this results from unclear sense of ownership obligations (Vining 2016). It may be the case that unclear ownership (of the property or of the daycare program) could be addressed in the future, perhaps though a continuation of formal contract or a mobilization of human resources on a volunteer basis. Although the ownership of the property is clear by title, the question of exactly who should be responsible for property maintenance could be addressed through renewed formal contract or other informal means. A visible presence of continued
“outsider” action could go a long way to maintain the sense of long-term researcher involvement in the community, though this depends on both funding and manpower.
5.4 Layers of Agency
While at their most basic level, principal-agent relationships consist of a single agent representing the interests of one principal, in the building of a development initiative through access to state funds and university support in a mixed indigenous/non-indigenous community, this was far from the case. Multiple principals were served by
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
multiple agents at multiple levels, and each relationship at each level can experience agency difficulties. Managing these relationships and reducing agency costs in this complex web was challenging, which is something that can be seen in the difficulty of project implementation.
The head researcher and the research team in the Lokah initiative served concomitantly as an agent for two main principals. First, the research team acted as an agent serving the indigenous community of Wulai, providing expert knowledge in science, economics, organization, and media relations. Secondly, the research team also acted as an agent on behalf of the government of Taiwan. In a position as gatekeeper with access to funding the team operates as the agent charged with fulfilling governmental policy. According to agency theory, problems of informational asymmetries goal conflicts can be expected to exist within both of these agency relationships. In addition, the agent working with two principals also finds him or herself with the further and complicating possibility that the two principals, in this case the government funders and the community leaders, may have divergent or incompatible goals.
Traditionally, principal and agents are seen as having divergent interests. The divergence may in reality be small, but the perceived divergence of interests between principals and agents creates an uncertain environment for a development program. Aligning goals was a difficult task, especially at the beginning of the project. This seemed to stem from information asymmetry relating to perceived motivations. As the team took first steps to demonstrate their intentions through communication, time spent in the community, and action, their good motivations began to become apparent, and community members reciprocated with their own participation. The community members’ participation benefitted the team by giving access to information and especially through the
involvement of respected elders, access to bonding social capital of the community. The actions that were identified in interviews as helpful were respectful two-way
communication, reciprocal sharing of information, and the inclusion of respected elders.
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
“lots of talk, lots of time, and humility,” backed by the confidence to follow through.
Yueh-po Huang emphasized including community leaders in a consultative and
collaborative role: “...we also invite some local people to be a consultant to this project, senior local people, headmen, a lot of people who have senior position. A lot agreed to be informal members of our research team.”
The community is often seen as a homogenous collective principal; however, in practice communities are made up of individuals who have their own particular needs, wants, and attitudes, and ideas. Often they can work together for their common interests, and when their own interests are aligned, viewing aggregate levels of social capital in terms of a collective principal at a meso or macro level is suitable. Yet in reality, even in small communities individuals play multiple roles and are involved with many overlapping subgroups; for example those based on gender, interests, jobs, family and kin, age, and so on. These various group memberships, while indicative of bonding social capital within groups, can still lead to conflict when members attempt to act collectively with
community outsiders. There may be difficulties that arise in dealing with agents if actors within a collective principal do not agree on how to align their own internal interests, for example with regard to policy changes or institutional reforms (Nielson & Tierney, 2003) figure 3. Intra-group variations in power, status, expertise, or other factor then create their own agency relationships and costs.
One element that leads to agency difficulties is the presence of multiple principals. For the action researcher, he or she falls in the position as an agent for the local community
Figure 3. Common Agency Types (adapted in part from Neilson and Tierney, 2003)
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
on one hand, and of the greater society (represented by government) on the other. The agent must provide expert information to both sides, though this information must be presented in much different ways. The ability to navigate both the community and governance environments requires different skill sets and communication styles that are not necessarily complimentary. This situation is made even more difficult when the interests of these two different principals are not in alignment.
Such was the situation faced by the researchers in the land regulation situation.
Representing the interests of the community meant supporting protests, but doing this too strongly or too publicly could lead to alienation of government, which could in turn cause difficulties in accessing funding. The team accompanied community members and
provided political, geographical, and media relations information that helped the
community come to a satisfactory solution with the government. This challenge in acting in the best interests of the community while still acting as mangers for national
government funding. This assistance was appreciated by the community. According to Yueh-po huang,
“So we had a debate at the monthly meeting. Are we taking sides to show our position on this matter against the regulation or support this regulation? So the final decision, we just provide opinion from all the teachers and the, provide as much as possible for the local Atayals. We decided not to just take a laissez-faire approach on this matter, kind of work with them but not overtly. We had to keep some distance. The program was officially funded. If you take a radical position or too overt to support local Atayals, you can be drawn into a catch 22 situation”.
Another aspect of agency theory that is clear from the property development regulations situation is multiple interests of the community. Often principals are treated as a single entity with shared vision, and sometimes indigenous communities are romantically viewed this way. Yet in reality this is rarely the case, as indigenous communities have
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
as either a collective principal in the process of reaching consensus, as multiple principals within the community vying for the agents’ services, or as a fragmented principal with dispersed stakeholders sharing broadly similar goals. A closer look at motivations for protesting land regulations provides some insight.
Community members were not all in agreement about the proposed changes to zoning legislation. While most wanted a relaxing of regulations to allow small developments on ancestral lands, others were pushing for complete abolition of existing restrictions. The divergence of interests in the context of previous property sales led to the question of background actors.
“Some local Atayal are saying that the regulation on land use is too complex. we should Abolish this regulation so we can really develop land in our traditional way. but in reality you know maybe someone behind the local Atayal [is] manipulating the scene and they maybe that person is entrepreneurial urban [resident].” Yueh-po Huang.
The motivations of these hidden principals required careful consideration by the team, as moral hazard on the part of principals they served could lead to substantial agency costs11.
This situation is less commonly addressed in agency theory literature, but interviewees did bring up the skill which some community members had developed over a long period of dealing with outside agents for bringing financial resources to the community although there can be tradeoffs in terms of autonomy. According to a program staff member, “It’s easy to get resources from outside through the government, the government projects, and
11 For an interesting discussion of moral hazard on the part of the principal, see Dohler’s “Discovering the Dark Side of Power: The Principal’s Moral Hazard in Political-Bureaucratic Relationships”
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
also the research projects, but more resources means less ability to decide for ourselves because of the funding structure.”
In general, however, it may be more accurate to describe the situation as one of
fragmented principals, where the interests of principals are broadly similar rather than a multiple principal situation where principals have contradicting interests that place them in conflict. For example, there was no indication that the interests of any two community groups were in direct conflicting, yet at the same time bringing people from the various interest and age groups to take part in collective action did not occur immediately. While it is not yet clear if the Lokah Initiative’s work seeking to increase perception of common interest and action among the community has led to continuing and increased
commonality of interest between community members, the opportunities provided by the Initiative, for example through the “Old House”, did provide a space for bringing
together principals for information sharing. Beyond this kind of interaction being indicative of increasing bonding social capital within the community, it could also be an avenue that helps increase interest alignment between the principals, while the instigation of this subproject through cooperation shows the importance of social capital exchange between respected community members and action researchers.
In addition to fragmented principals, approaching the Lokah Initiative’s work by incorporating the concept of informal principals may provide some insight. Whereas formal principals have in place principal-agent relationships based on formal contracts or agreements, accountability to informal principals is based on informal, implicit, or hypothetical delegation of authority. Relative power differences between agents and informal principals, be they individuals or groups who do not have a formal contract with the agent in place, can affect the relationships and accountability between principals and agents, bring with them issues of how to clarify accountability(Steets, 2010). The task of clarifying expectations and accountability can be difficult when the principal is a
community rather than an individual, and where group heterogeneity and changeable or
‧
A further dimension to this situation is the researchers acting as agents with government funding agencies as principals. While academia enjoys a good deal of freedom and autonomy, when funds are needed to support research or developmental initiatives, the balance of power can shift. The principal, in this case government funding agencies, may seek to gain higher levels of agent performance through monitoring and providing or withdrawing incentives, the decision on whether or not to support a program may be based on criteria that may not give the most accurate impression of the project’s progress as determined by either the researchers or the community, as was brought up by Ching-ping Tang. These agencies, or actors within agencies may also attempt to control agent behavior through access to financial support that can be suddenly withdrawn with little prior notice or reason. The position of the research team relative to their positions as academics seeking funding was acknowledged by Lokah volunteer Yuakan: “Things in the tribe might move slowly, but because it’s an initiative, they have to do something within six months. They need to show achievement. They have their own pressures.”12 However, a sudden or unexpected withdrawal of resources can present a serious challenge to building strong ongoing academia-community relationships.
At a different level of principal-agent relationship, the project’s volunteers can also be seen as agents acting on behalf of multiple principals, and thus faced their own problems with agency costs. They were motivated and worked enthusiastically in support of the project and the local community.13 They wanted to do a good job, though it is possible they had some interests that did not fully align with the researchers. They didn’t always perceive high levels of fairness in the project, though this is possibly due to a lack of understanding certain project challenges. Although it may be the case that there was some lack of interest alignment or background information asymmetries, the volunteers seemed to gain an awareness of project goals. Some of this knowledge and activation led to
12 Interview 9.
13 They likely had their own self-interests as well, though bounded self-interest of volunteers is beyond the scope of this work.
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
increased activity outside of the project’s research area that included cultural and
environmentally friendly entrepreneurial activities, a network that could be utilized in the future. With at least some volunteers, experience in the project led to improvements in human capital, as well as planting a seed of cross-regional social capital growth.
As was discussed earlier, there were some indications that different researchers engendered different levels of trust. Although it could be expected that if a collective agent presented a disunited front a lack of trust could occur, interviews gave no indication this was the case. Perhaps it would be more accurate to say that different behaviors of researchers engendered different levels of trust. Thus trust levels differing between agents are more likely to be the result of personal traits within the observer, the actor, or both. As mentioned earlier, trust (as an indicator of social capital) has been related to the agreeableness factor in the Big Five personality model. While
agreeableness has been related to prosocial behaviors that help facilitate group interactions, people with high agreeableness may find competitive fields more challenging (Judge & Zapata, 2015). While increasing the level of agreeableness in principal or an agent is likely an inefficient, or even impossible endeavor, training the observable behaviors that imply agreeableness could be used to increase impressions of trust.
There is some literature on leadership and action research that suggests trust can be engendered in such a way. It is likely worth considering that training trust engendering behaviors is something that could be done to help increase social capital growth in future projects. This could be incorporated in a kind of leadership training for team members.
Indeed, this idea of skills training could also be applied within the community as part of future programs.
As stated by Cargo and Mercer,
The integrity of partnerships rests on the presence of mutual
‧
國立 政 治 大 學
‧
N a
tio na
l C h engchi U ni ve rs it y
environments that support diversity and allow partners to express and accommodate their different points of view through ongoing, open, and honest dialogue. Effective leadership, a strong predictor of partner- ship synergy, facilitates this type of communication.
(Cargo & Mercer, 2008)
On the part of the researchers, there also exists different levels of trust in community members. Although this remains to be studied in depth, this could be due to personality factors or due to weaknesses in developing clear and flexible lines of communication, or due to the influence of outside actors.