Social capital, agricultural innovation and the evaluation of agricultural development initiatives

In this thesis, I show that social capital has an important role in the evaluation of development initiatives targeting agricultural innovation. Social capital and agricultural innovation are naturally linked from an innovation system perspective in which innovations result from the integration of knowledge from various actors and stakeholders. In chapter 1, I identify the three research questions upon which this thesis is based. First, how are social capital and agricultural innovation related? Second, can development initiatives increase agricultural innovation by building social capital? Third, does the initial level of social capital increase the success of these development initiatives in enhancing agricultural innovation? These question mainly relate to the increasing number of policies, programs and project that include beneficiaries in the design, management and decision making process. In chapter 2, I elaborate on the main concepts underlying this thesis including social capital, how it relates to development initiatives in the agricultural sector, and how it can be measured. I broadly define social capital as the participation of individuals in formal and informal networks, the norms that define these networks and the trust these individuals have within and outside these networks. Participation in networks is structural social capital, whereas norms and trust within and between these networks is cognitive social capital. I distinguish four dimensions of social capital: structural bonding, structural bridging, cognitive bonding and cognitive bridging. In this thesis bonding and bridging social capital is akin to social capital inside and outside the village. Agricultural innovation is defined in terms of improved land and crop management practices, an important area of agricultural innovation for small scale producers. In chapter 3 till 7, I empirically investigate the relationships between social capital, agricultural innovation and two types of development initiatives. The first initiative is the implementation of agricultural research through the Integrated Agricultural Research for Development (IAR4D) approach. IAR4D was adopted by the Sub Saharan African Challenge Program (SSA CP) and implemented in eight different countries.  The core of this approach is the development of Innovation Platforms (IPs), which can be described as an informal coalition and alliance of conventional agricultural research and development actors. Using the semi experimental data collected in this context, I could investigate the important role of social capital in different contexts. The second type of initiative is implementation of sustainable certification schemes through group-based experimental learning approaches. I investigate four sustainable coffee projects in Vietnam, of which two adopted the interactive Farmer Field School training approach. The data of these four projects allow me to verify some of the conclusions in a different context and for a different development initiative. In chapter 3, I use baseline data from the IAR4D initiative to explore the association between different forms of social capital and uptake of various agricultural innovations, for a sample of 2500 households in seven countries in SSA. I find that structural bridging social capital is associated with more extensive adoption of agricultural innovations. This result is true for the pooled model as well as for four of the seven country models. This form of social capital captures agriculture-related links creating access to knowledge and resources and is considered an important dimension of economic development. I find a negative association between cognitive bonding social capital and the innovation index. This finding could represent a potentially harmful side of social capital in terms of agricultural innovation.  In chapter 4, I investigate the impact of IAR4D on social capital. I narrow my focus on the border region between Rwanda, Uganda and the DRC. Because the SSA CP data set consist of randomized data of participating and non-participating villages, before and two years after implementation, I can investigate the impact of the program. Many participatory projects in rural Africa are efforts to enhance development indirectly by promoting cooperation in formal or informal networks, and by encouraging trust and norms of behaviour towards mutually beneficial action. But it remains unclear whether external interventions can actually influence social capital, especially in the short term. I show that IAR4D has had a positive impact on structural bridging social capital in DRC and Uganda. There was no impact on structural bridging social capital in Rwanda, or on the other dimensions of social capital. Finally, I showed that traditional agricultural extension has been less successful in increasing structural social capital than IAR4D. In chapter 5, I use data from a survey I conducted among IP coordinators to measure the extent to which IPs were implemented according to the principles of IAR4D across the three sub regions. Linking these data to the main survey data, I find that the extent to which IPs were implemented according to IAR4D principles  is associated with the success of IAR4D in increasing the level of household food security, although not through increased adoption of agricultural innovation or increased levels of social capital at household level. Looking at the sub-components of these principles, especially involvement of IP stakeholder is crucial. Tentative results suggest that this involvement is higher in communities with a higher level of education, a higher percentage of female headed households, and a higher level of village social capital. In chapter 6, I analyse how different indicators used to represent social capital are related in the border region between Rwanda, Uganda and the DRC. I focus on the relationship between various indicators of trust, an important component of cognitive social capital, and group membership, an important component of structural social capital. The indicators used are based on questions I added to the follow up survey of the SSA CP in 2010. I find that different indicators of trust and group membership cannot be empirically captured by an overarching social capital factor, and are not even necessarily associated to each other. In chapter 7, I present evidence that the relationship between social capital and agricultural innovation is not only evident for the IAR4D approach, but also for a different development initiative in a different context: sustainable coffee certification in Vietnam. I use data collected among 240 randomly selected project participants and 150 comparable farmers that did not participate in the projects. I focus on the role of bonding and bridging cognitive social capital, defined as trust. I find a significant positive relation between trust and the uptake of sustainable agricultural training practices. This relationship mostly results from high levels of bridging trust, and is even higher in combination with high levels of bonding trust. I also find tentative evidence that participation in the sustainable coffee projects positively influenced bonding trust in one project whereas it negatively influenced bridging trust in another project. In chapter 8, I give an overview of the three main findings. First, social capital is associated with agricultural innovation. Second, development initiatives can influence social capital. Third, the existing level of social capital is associated with the success of development interventions. However, the effect was not necessarily positive and depends greatly on the dimension of social capital. I also present several implications for policy. First, stimulating social capital, especially bridging social capital, may be a natural leverage point for policy makers to promote agricultural development. Second, increased levels of social capital can indeed be an outcome of development initiatives, either intentionally or not. At the same time, I show that this impact requires specific efforts and is not necessarily positive. The third implication is that social capital matters for the success of certain development initiatives, either as a source of heterogeneous implementation or impact.  Fourth, it is vital to take into account the multi-dimensional nature of social capital and the fact that these dimensions might have different relations to agricultural innovation and development initiatives. Combined these implications mean that indicators of social capital should be included in the design and evaluation of agriculture-related development initiatives. Finally, I give suggestion for future research. First, to further unravel the chains of causation between different dimensions of social capital, agricultural innovation, and development initiatives. Second, to validate the indicators and indices of social capital using experimental games, more advanced survey questions, or better embedding them in existing theories. A third area of future is to advance in the measurement of innovation as a truly interactive and participatory process. Fourth, to address whether the importance of social capital, as a catalyst for success or as an outcome variable, depends on the nature of the development initiatives or the context in which it is implemented. Finally, research yet has to address the long-run effect of development initiatives on social capital