Joining Policy Forums Together to Develop Ki-no-Eki, a Community Currency System for Forest Management in Japan: Dynamics of Policy Communication Networks

Abstract

In some mountainous areas of Japan, the Ki-no-Eki system, in which wood is collected to thin the forest and is exchanged for community currency, has been specifically designed and implemented as a solution to current and emerging forest governance issues. This study aimed to capture the evolutionary processes of a complete communication network consisting of organizations that joined policy forums to help develop the Ki-no-Eki system. A total of 26 policy forums were held from 2011 to 2019 to discuss the adoption and implementation of the Ki-no-Eki system across Japan, and coattendance and the resultant policy discourses among 62 participating Ki-no-Eki organizations in these forums were regarded as dynamic communication network processes. We analyzed how policy communication networks formed and evolved to understand the underlying network dynamics driven by not only endogenous network processes—bonding and bridging social capital—but also exogenous effects defined by actors’ attributes. We employed the stochastic actor-oriented model for network dynamics to manage the collected longitudinal undirected network data. We found (i) the emergence of bonding social capital and (ii) homophilic and heterophilic connections in communication networks, which provided insightful explanations of the driving forces of social cohesion among Ki-no-Eki organizations engaged in forest management in Japan.

1. Introduction

As a policy tool for fostering social capital and revitalizing the local economy in Japanese local communities, a variety of community currency schemes have been adopted and implemented over several decades [1,2,3]. Community currency, a complementary currency system, is a medium for the exchange of goods and services within a limited regional boundary [4] and has been advocated as a socially and environmentally sustainable and innovative form of money in local communities [5,6]. In some mountainous areas of Japan, the Ki-no-Eki method, in which wood is collected to thin the forest and then exchanged for community currency, has been adopted and implemented to make effective use of forest resources and promote regional development [7]. Addressing current and emerging forest governance issues in Japan—such as the decline of self-harvesting forestry and rural community development—and designing innovative and sustainable solutions using this community currency system for forest management require collective action by and mutual understanding among diverse stakeholders [8,9,10,11]. Policy forums could be an impetus for fostering collaboration and deliberation among these actors [12].

Policy forums are communication venues for formal and informal interactions and discussions among stakeholders involved in a specific policy issue to exchange critical information among participants and reinforce their policy ideas and beliefs [13]. By voluntarily joining policy forums, participants can learn from other participants’ experiences, facilitate collaborative behaviors, and enjoy club goods or exclusive benefits from their attendance and the resultant collaboration [14,15]. Since coparticipants in a policy forum have many opportunities for policy discourse and share common interests, it is important to investigate forum participation and communication dynamics among Ki-no-Eki organizations in discussing and developing an innovative forest governance tool, i.e., the Ki-no-Eki community currency system for forest management in Japan. Although there have been many empirical investigations of the kinds of structural patterns of networking that emerge among stakeholders and how they affect stakeholders’ actual decisions and behaviors in diverse examples of environmental governance and dealing with environmental challenges [16,17,18,19], the use of these forums in forest governance is still not fully understood. We aim to fill this gap by focusing on Ki-no-Eki organizations’ spontaneous interactions for policy discourse in the Japanese context of forest governance and ask the following research question: What are the underlying mechanisms of the development of policy communication networks among Ki-no-Eki organizations through participation in policy forums to address forest governance issues using a community currency over time?

2. Ki-no-Eki, a Community Currency System for Forest Management

2.1. Reforestation Initiatives, Self-Harvesting Forestry, and Community Currencies

Approximately 70% of Japan is forested, and people have been using these forests for energy sources, building materials, and other resources since ancient times [20] (pp. 75–76, p. 96). However, the use of forests in Japan changed dramatically during the 1960s. The popularization of chemical fertilizers and importation of crude oil and natural gas sharply reduced the use of grass, firewood, and charcoal. With the liberalization of timber imports, the demand for domestic timber declined, and Japan’s timber self-sufficiency rate fell below 50% in 1970 and reached its lowest point (18.8%) in 2002. Not using domestic timber can cause problems in planted forests, which account for approximately 40% (approximately 10 million ha) of Japan’s forested area. For example, because Japan is located in the Asian monsoon region, where summers are hot and rainy, activities such as undergrowth cutting and thinning are essential for the cultivation of planted forests [21]. If such activities are not performed, forests can become dark and experience poor understory growth, which can lead to the deterioration of the forest environment, as evidenced by soil outflow and a decline in biodiversity. In addition, the repetition of thinning at a low rate is considered a reasonable practice for increasing total carbon storage, as thinned trees are supposed to be effectively exploited, including treetops and stumps, branches and leaves, and debris, during the sawing and product processing stages [22].

In recent years, however, this situation has started to change. Since the self-sufficiency rate hit its low point in 2002, it has recovered, exceeding 30% in 2014. Two specific trends are driving this increase. First, there has been a shift away from the use of foreign timber for the raw materials used in large-scale wood processing plants. This shift, coupled with an increase in the number of biomass power plants operating in Japan, has increased domestic timber production. Second, a self-harvesting forestry movement has spread as the antithesis of the large-scale production course taken by the government. In self-harvesting forestry, forest owners use primarily family labor to grow their own trees and harvest and transport them. Since 2005, this form of forestry has been practiced by self-employed small-scale migrant forestry workers, who do not own forests, and volunteers from cities. Self-harvesting forestry accounts for an estimated 10% or more of Japan’s material production (approximately 1.8 million m³ per year in 2015). Compared with large-scale forestry, where clear felling occurs every few decades, self-harvesting forestry is characterized by small-scale thinning that does not exceed the rate of forest growth. Compact machinery, which has a limited effect on the environment and is therefore preferred for purposes of environmental conservation, is used to remove logs [20] (p. 27).

Tosa no Mori Kyuen-tai (henceforth Tosa) in Kochi Prefecture created this new self-harvesting forestry system around 2005 [23] (p. 42). Tosa is a nonprofit organization founded in 2003 by a group of local residents. The forest maintenance project that they organized started by picking up the timber discarded as forest cooperatives or production forest associations thinned the forests. As their skills improved, they expanded the scope of their work to include thinning as well as collecting and transporting logs. Participants in the project received a ticket certifying their participation each time they participated in these activities and could then exchange these tickets for local products. This system was funded by local businesses, and the tickets functioned as a community currency. Tosa’s participation in the Biomass Plant Experiment Project of the New Energy and Industrial Technology Development Organization (NEDO) in 2005 truly established self-harvesting forestry as a system.

One of the seven projects in NEDO’s Biomass Energy Regional Systematization Experiment Project (2005–2009) was the Systematization of Energy Self-Sufficiency in the Niyodo River Basin, Kochi Prefecture. Discarded timber was collected and processed into chips to be used in a gasification plant to produce electricity (with a rated output of 150 kW). The main goal was to collect discarded timber from local forests. Instead of entrusting the collection to forest cooperatives and other businesses, Tosa advocated self-harvesting forestry, with the inhabitants of the basin cutting down the trees on their own mountains and transporting them to intermediary timber depots. As a result, much discarded timber was collected. During the first year, participants were paid JPY 3000 per ton; however, some believed that this remuneration was too low. Thus, Tosa members received an additional JPY 3000 in community currency, bringing the total to the equivalent of JPY 6000 [23] (pp. 41–42). The system created by Tosa has since evolved from a single-use gift certificate to a coin-type local currency that can be used multiple times.

Efforts to establish this biomass power plant business were novel; however, penetration into other regions was limited because of the requirement for plant installation. Kenji Niwa promoted forest health checks, with local residents and researchers working together to investigate the state of planted forests in the basin. He was impressed during his observations of this initiative in May 2009 and expanded it nationwide. He gave it an attractive name—Ki-no-Eki (“Tree Station”)—and revised the system so that anyone could implement it by reading the manual. For example, the weighing of thinned timber was simplified and no longer required expensive equipment. In December 2009, the first Ki-no-Eki was opened, and the system has been spreading throughout Japan since 2012.

2.2. Ki-no-Eki Based on a Community Currency System

In the Ki-no-Eki system, irregular discarded timber and thinned wood are purchased at a price slightly higher than the market rate. Payment is made in the community currency, which can be used only in local shops and not at large supermarkets. An overview of the activities involved in a typical Ki-no-Eki system is shown in Figure 1 and includes the following: registered shippers thin trees on the local mountains and collect thinned timber that has been cut and left behind by forest cooperatives (A). The collected timber is transported to a predetermined intermediary timber depot, where it is exchanged for community currency according to weight or volume. Although the sales route varies depending on the individual Ki-no-Eki, the thinned timber is typically sold to chip suppliers outside the region or purchased as fuel for firewood boilers at local bathing facilities (B and C). The shippers then use the community currency at local grocery stores, eateries, gas stations, etc. (D, E, and F). The community currency received by business operators, such as local shops, is cashed in (H and I) or reused at other stores (G). This flow creates additional economic benefits and develops social capital between volunteers and local residents who know each other only by sight, as well as volunteers and local residents interested in Satoyama (traditional Japanese rural landscapes), to foster new relationships. To facilitate the smooth operation of this system, an executive committee, consisting of registered shippers, registered shops, and local governments, is established for each local Ki-no-Eki. Since 2011, Ki-no-Eki systems have been launched in various places, and in 2012 alone, 17 new stations were launched. By the end of 2020, Ki-no-Eki systems had been established in 50 locations throughout Japan.

Although one of the major features of Ki-no-Eki is the use of a community currency as a means of payment, community currency initiatives in Japan originally began in the late 1990s. Similar initiatives were introduced overseas, such as local exchange trading systems (LETS) in the UK and Time Dollars in the United States. Through the mid-2000s, interest in community currencies increased in Japan, and the concept continued to expand [1,2]. When it peaked around 2006, there were community currencies in more than 300 locations. However, that number gradually decreased, partly because the effects of introducing community currencies into local communities were unclear; by the end of 2020, the total number had dropped to 184 [7]. However, while the number of community currencies as a whole has been decreasing, the number of Ki-no-Eki using community currencies continues to increase, accounting for approximately 27% of community currencies in Japan (see Figure 2).

A questionnaire survey conducted by Izumi and Nakazato [7] (pp. 159–161) in 46 Ki-no-Eki in 2019 revealed that of the 45 organizations that responded, 40 were issuing community currencies. In 2018, these 40 groups had a combined timber shipment volume of approximately 11,700 tons (15,600 m³). The largest shipment volume for a single organization was 1099 tons, and the lowest was 18 tons. The community currency was paid according to the weight or volume of the shipped materials. The total amount issued in 2018 was equivalent to approximately JPY 65 million, although the average amount issued decreased from 2016 to 2018 (see Figure 3). In 2018, the highest amount issued per organization was JPY 6.493 million, the lowest was JPY 30,000, and the median was JPY 1.11 million. Although a community currency is eventually redeemed in JPY, most community currency issued (more than 95%) is used at least once. Although the Ki-no-Eki system recommends using the same community currency token multiple times, the average turnover rate is actually 1.24 (n = 24), revealing that secondary circulation is not common and many shops cash the currency quickly.

Although the material production of the 40 Ki-no-Eki issuing community currencies accounted for only approximately 1% of the total of self-harvesting forestry, the community currency system has some unique features. First, since the community currency can be used only in the local community where the specific Ki-no-Eki is located, it can contribute to the increased exchange of local products and revenue for local shops. Second, registered shippers include not only local residents but also forest volunteers. By using the community currency instead of cash, forest volunteers can be connected with the local residents and shops that have agreed to collaborate with the Ki-no-Eki [24,25].

3. Research Methods

3.1. Data on Two-Mode Coparticipation Networks and One-Mode Communication Networks among Ki-no-Eki Organizations

We aimed to capture the evolutionary processes of a complete communication network of Ki-no-Eki organizations that joined policy forums for the development of the Ki-no-Eki system. Using the relevant documents on these forums, we identified a total of 26 summits and/or block councils, which we call policy forums and which were held to discuss the adoption and implementation of the Ki-no-Eki system across Japan from 2011 to 2019, and 62 participating Ki-no-Eki organizations.

In this study, a policy forum is any kind of venue where Ki-no-Eki organizations can gather, share their experiences and practices, and exchange relevant information about how to design and maintain Ki-no-Eki systems. Attending the policy forums provides opportunities for participants to communicate and interact and facilitates collaboration among them [14,15]. According to the Ki-no-Eki Project Report 2010 [26], a briefing session on the experimental results of the Nakanoho-cho Project in Ena City, Gifu Prefecture, was held in March 2010, and the first meeting had 175 participants. Then, the first to third Ki-no-Eki meetings were subsidized by the Toyota Foundation in July, September, and December 2013, respectively. Since then, regional meetings have been held several times in Tohoku, Hida Mino, Hokuriku, Chugoku, Shikoku, Kinki and other locations. The most recent conference, the 8th Ki-no-Eki Summit, was held in Takayama City, Gifu Prefecture, in September 2019 and was attended by 22 Ki-no-Eki organizations.

The coparticipation of a group of Ki-no-Eki organizations in a series of policy forums can be represented as a two-mode network (or affiliation network) [27], where an organization is linked to any forum it participates in. The affiliations of Ki-no-Eki organizations with the forums can be transformed into a one-mode network, where coparticipating organizations in the same forums are linked to each other. That is, ties among Ki-no-Eki organizations in the undirected one-mode networks created in this study represent the linkages among them for communicating about the development and improvement of Ki-no-Eki systems in their regions by attending the forums. The database, comprising 26 forums held between 2011 and 2019 and involving 62 Ki-no-Eki organizations, was transformed into three undirected one-mode communication networks over three consecutive time periods for the development of policy discourse on Ki-no-Eki systems (i.e., early policy discourse (t₁, 2011–2013), middle policy discourse (t₂, 2014–2016), and recent policy discourse (t₃, 2017–2019)). The networks represent the coparticipants in the forums in each three-year period and the resultant mutual communications among them.

In addition to the abovementioned longitudinal relational dataset, the 2019 Ki-no-Eki Census was used to identify the following attributes of individual Ki-no-Eki organizations: (i) the year of establishment, (ii) the location (i.e., 8 regional divisions of prefectures: Hokkaido, Tohoku, Kanto, Chubu, Kinki, Chugoku, Shikoku, and Kyushu), and (iii) the annual amount of community currency issued by each organization [7].

3.2. Stochastic Actor-Oriented Modeling

We regarded the abovementioned longitudinal communications among 62 Ki-no-Eki organizations as dynamic social network processes and analyzed how communication networks formed and evolved using an advanced method of inferential social network analysis, the stochastic actor-oriented model (SAOM) for network dynamics [28,29], which was implemented in the RSiena package [30]. This statistical model simulates evolutionary network changes between discrete observation points and estimates parameters for underlying network dynamics driven by not only endogenous network processes—for example, triad closure and popularity—but also exogenous effects defined by actors’ attributes—such as homophily [31]. Some critical assumptions underlie the SAOM [28]: First, the time horizon of network data among the same actors is treated as continuous and subsequent between the observation points; second, the network evolution is driven by a Markov process, which means that current network changes at each observation point depend only on the previous, not the future, state of the network; third, each actor can decide its own formations and dissolutions of ties; and fourth, network changes occur in increments over observation time points.

The desirability of a new state of the network is captured by the objective function indicating how likely each actor is to change the network in a specific way and is defined as a linear combination of a set of the abovementioned endogenous network effects and exogenous effects as follows, where $f_i\left(\beta ,x\right)$ is the degree of desirability that each focal actor (i) assigns to a certain state (x) of the network and $s_{ki}\left(x\right)$ are effects specified as independent variables in the model [32] (p. 33):

$$f_i\left(\beta ,x\right)={\displaystyle \sum }_k{\beta }_k{s}_{ki}\left(x\right)$$

(1)

When each actor (i) has the chance to alter the network to one of the possible new states (C), the probability of changing to this new state is defined as follows [32] (p. 34):

$$\frac{\mathit{exp}\left(f_i\left(\beta ,x\right)\right)}{{\Sigma }_{x^{\prime }\in C}\mathit{exp}\left(f_i\left(\beta ,x^{\prime }\right)\right)}$$

(2)

3.3. Model Specification

As communication between two actors is not a one-sided interaction, ties in this study are viewed as not directed but undirected connections. This setting represents a process in which an actor decides to join a specific policy forum with prior expectations that more or less voluntary ties for communication with other participating actors will be established in a forum and that these ties will be beneficial for setting agendas for Ki-no-Eki systems and learning from each other’s experiences and know-how [13]. For the model specification in this study, we considered two types of effects on communication network formations: (i) network structural effects endogenous in evolutionary processes of communication ties and (ii) homophilic effects driven by the shared attributive properties held by actors [29], as shown in Figure 4.

With regard to network structural effects, first, as the most basic effect in the SAOM, the degree effect estimates an actor’s tendency to have communication ties (see Figure 4a) and indicates whether forging and maintaining communications with other actors in the network is beneficial or costly [28]. Second, to assess transitivity effects, i.e., whether the tendency of actors in the networks to have a single or multiple communication partners in common is likely to increase tie formations, we used transitive triads and geometrically weighted edgewise shared partners (GWESP) effects, respectively, in the model (see Figure 4b,c). Communicating through shared or indirectly connected partners functions as bonding social capital—forming socially cohesive and dense connections [33,34,35]—within the network and is critical for sharing/confirming further information with, reinforcing trust in, and minimizing uncertainty about not only these newly added third-party partners but also the former directly connected partners [36,37,38]. The final effect tested was popularity, which was modeled using the square root of the degree of alters (see Figure 4d). Popularity represents actors’ propensity to communicate with a few central actors possessing critical resources in the network, as these popular actors can perform a coordinating role and function as bridging social capital—forming socially loose or weak connections [33,34]—among the actors based on their resources [17,38]. In this study, as the number of links to many actors formed by coattending forums increases, actors can accumulate more resources, i.e., information about formulating and carrying out Ki-no-Eki systems, and thus become more attractive members of their networks.

Regarding homophily effects indicating mutual attachment between actors based on shared properties [17,36,39], the organizational characteristics—such as the regions and ages of Ki-no-Eki organizations—and the operational characteristics of Ki-no-Eki systems performed by these organizations—such as the sizes of shippers and amounts of community currency issued—were considered drivers for tie formations. That is, we aimed to examine whether two Ki-no-Eki organizations having the same regional basis or similar organizational ages, sizes of shippers, and amounts of community currency issued would increase or decrease their propensity to join the same forums and engage in communication with each other within the networks (see homophily-based effects in Figure 4).

4. Results

Figure 5 graphically depicts the development of policy communication networks among Ki-no-Eki organizations in every three-year period from 2011 to 2019, illustrating which Ki-no-Eki organizations attended policy forums together to discuss the design and improvement of their Ki-no-Eki systems during each period.

Table 1. Structural characteristics of policy communication networks among Ki-no-Eki organizations over time.

Observed Period	Number of Ties	Average Degree	Density
Early policy discourse (t1)	234	7.54	0.12
Middle policy discourse (t2)	424	13.67	0.22
Recent policy discourse (t3)	405	13.06	0.21

presents the descriptive statistics of the changes in degrees and densities of communication networks over the three time periods, indicating large increases in dyadic relationships among Ki-no-Eki organizations (from an average degree of 7.54 at t₁ to 13.67 at t₂ and 13.06 at t₃) and a shift toward denser network structures (from a density of 0.12 at t₁ to 0.22 at t₂ and 0.21 at t₃) in later years.

Table 2. Changes in policy communication networks among Ki-no-Eki organizations between subsequent observations.

Observed Period	0 → 0 (Continued Nonrelationship)	0 → 1 (Developed Relationship)	1 → 0 (Terminated Relationship)	1 → 1 (Maintained Relationship)
From t1 to t2	1345	312	122	112
From t2 to t3	1220	247	266	158

summarizes the changes in ties for communications between the two consecutive time periods, showing that Ki-no-Eki organizations forged new relationships (0 → 1; 312 ties) rather than terminating previous relationships (1 → 0; 122 ties) from the earlier period (t₁) to the middle period (t₂). In contrast, Ki-no-Eki organizations broke prior links (1 → 0; 266 ties) rather than initiating new links (0 → 1; 247 ties) between the middle period (t₂) and the recent period (t₃).

The results of the analysis of network dynamics for policy communication among Ki-no-Eki organizations over the three observed time periods are shown in

Table 3. Results of the stochastic actor-oriented model testing the formation of policy communication networks among Ki-no-Eki organizations.

Parameter	Estimate	Standard Error
Rate
1. From t1 to t2	22.50 **	11.32
2. From t2 to t3	5.06 ***	1.50
Network structural effects
3. Degree (density)	−2.03 **	0.86
4. Triad closure (transitive triads)	0.11 ***	0.01
5. Network closure (GWESP)	1.89 ***	0.48
6. Popularity (square root of degree of alter effects)	−0.63 ***	0.08
Homophily-based effects
7. Region	0.22 **	0.09
8. Organization age	0.25 ***	0.08
9. Number of shippers	0.14	0.22
10. Value of community currency issued	−0.29 **	0.12

Note: ** p < 0.05. *** p < 0.01 (all two-tailed).

. When the estimate divided by its standard error is in absolute terms similar to or larger than 2.58, 1.96, or 1.64, each result is statistically significant at the 1%, 5%, or 10% levels, respectively [30]. Rate parameters (Parameters 1 and 2) indicate that the propensity to change the communication relationship between two successive networks—i.e., new communication ties between organizations that did not previously interact appear, and extant communication ties disappear—varied among the different time periods, with many more stable periods occurring later (from 22.50 (p < 0.05) between the early policy discourse stage (t₁) and the middle policy discourse stage (t₂) to 5.06 (p < 0.01) between the middle policy discourse stage (t₂) and the recent policy discourse stage (t₃)). That is, as Ki-no-Eki based on a community currency system became more institutionalized and formalized over the decade, the Ki-no-Eki organizations’ relations for policy discourse also became more stabilized.

Regarding network structural effects, first, the degree effect demonstrates Ki-no-Eki organizations’ cost of establishing or maintaining communication ties with other organizations; thus, simply maximizing the numbers of connections was not beneficial for them (Parameter 3; est. = −2.03, p < 0.05). Second, Ki-no-Eki organizations were more likely to communicate with partners of their partners, as explained by both the triad closure effect (Parameter 4; est. = 0.11, p < 0.01) and the network closure effect (Parameter 5; est. = 1.89, p < 0.01), representing bonding social capital. Thus, Ki-no-Eki organizations’ preference for initiating communication with indirectly connected organizations and developing one or more partners in common indicates the importance of fact-checking or confirming shared information and trust building, which leads to strong bonds and densely connected subgroups among the Ki-no-Eki organizations in the communication networks. Last, the negative popularity effect (Parameter 6; est. = −0.63, p < 0.01) shows that communications in the networks were not hierarchical or centralized around a few very pivotal Ki-no-Eki organizations. That is, Ki-no-Eki organizations tended to join forums and communicate with diverse organizations, regardless of their positions or centralities within the networks, rather than with a few central leading organizations. This finding implies that information, knowledge, and policy ideas about the development of Ki-no-Eki systems are distributed and circulated among organizations rather than being dominated by or centralized around a few popular organizations bridging or coordinating other organizations connected to them.

Regarding homophily effects, having the same geographical basis (Parameter 7; est. = 0.22, p < 0.05) and similar organizational ages (Parameter 8; est. = 0.25, p < 0.01) increased the tendency of two Ki-no-Eki organizations to attend the same policy forums, which allowed them to communicate with and learn from similarly situated organizations. On the other hand, Ki-no-Eki organizations preferred to communicate with other organizations within the networks whose community currency circulation size differed from their own (Parameter 10; est. = −0.29, p < 0.05), as such organizations might be spread across different developmental stages of issuing community currency and therefore might have more information to share.

5. Discussion

The first main finding, that is, observing bonding social capital (both the triad closure effect and the network closure effect supported in this study) but not bridging social capital (the negative popularity effect), indicates that Ki-no-Eki organizations may experience low information resource scarcity due to the series of policy forums held over a long period of time [34]. That is, these policy forums provide a useful way to share information and hold discussions and have reduced the organizations’ reliance on centralized information resources. Instead, Ki-no-Eki organizations facing few difficulties in accessing information resources due to these interactions and their networking in many policy forums were inclined to build trust, form commonly agreed-upon norms and rules, and help self-monitor the information exchanged through their dense and close connections [41], as explained by both the triad closure effect and network closure effect noted in this study.

The second main finding supports not only some Ki-no-Eki organizations’ homophilic tendency to share information with organizations within their own regions and of similar ages but also the heterophilic tendency to communicate across organizations with different community currency circulation sizes. That is, social cohesion in the communication networks among Ki-no-Eki organizations is attributable not only to similarities but also to differences in the organizational attributes [36], meaning that they sought and shared information with other organizations with different know-how and information resource needs as well as those with similar needs and know-how.

6. Conclusions

The Ki-no-Eki system is spreading across Japanese communities as an innovative and sustainable policy instrument for promoting self-harvesting forestry, revitalizing regional forestry, and restoring social capital in rural communities. This study empirically investigated how communication ties among Ki-no-Eki organizations were forged by coattending policy forums where they shared information and discussed issues related to the development and improvement of the Ki-no-Eki system for forest management in Japan. Using a longitudinal inferential network analysis of the evolution of connections for communication among 62 Ki-no-Eki organizations, we found (i) the emergence of bonding social capital rather than bridging social capital and (ii) homophilic and heterophilic connections in these organizations’ communication networks.

This study allowed for the in-depth investigation of key policy communication processes and helped to provide insightful explanations of the driving forces of social cohesion for communication and deliberation among Ki-no-Eki organizations engaged in forest governance in Japan. More specifically, we noted that communication networks among Ki-no-Eki organizations may be sustained by the development of common norms and values fostered by the network structural effects identified in this study, such as the voluntary formation of bonding social capital and homophilic relations. On the other hand, the unsupported role of bridging social capital should be used to guide any stakeholders or activists engaged in forest governance through the Ki-no-Eki system. Considering that the number of Ki-no-Eki organizations is increasing and that late adopters of Ki-no-Eki based on a community currency system may need to rely on some leading boundary-spanning organizations that could efficiently transfer information or know-how to them, the concern about this lack of coordination could be mitigated by the heterophilic relations observed in this study.

This study also has some obvious limitations that should be addressed in future research on Ki-no-Eki organizations. First, regarding social capital, this study considered both bonding and bridging social capital metaphors in analyzing the endogenous network process but omitted the linking social capital process that mainly engages with the exogenous network process [16,42]. The next research agenda must include the role of Ki-no-Eki organizations’ connections with external authorities outside the networks examined in this study. Second, in this study, we mainly investigated the structural patterns of coengagement among stakeholders in a Japanese forest governance case—i.e., which stakeholders and with whom they were involved in policy discourses over time—but did not examine how their engagement affected each stakeholder’s ability to manage the forest governance issues they faced. Beyond the question of how distinct network structures emerge, which is addressed in the current study, future research should focus on answering the question of how the emerging structures of networks coevolve with each engaging stakeholder’s behavioral changes or contribution to solving forest governance issues in the community in order to gain a better understanding of the nexus between the dynamics of communications networks and actual outcomes of participatory forest governance [3,43].