RT Dissertation/Thesis
T1 Theoretical analysis and preference modelling for the valuation of ecosystem services from native pollinators in selected Thai rural communities
A1 Narjes,Manuel
WP 2019/01/28
AB Until now, the existing microeconomic models concerned with pollination markets have not accommodated the global diversity of beekeeper-farmer interactions. The most prominent of such theoretical models is dedicated to describing the determinants of colony stocking densities and of equilibrium wages that farmers have paid to commercial beekeepers for decades in the highly bee-pollination reliant almond monocultures of California. This cumulative dissertation generalizes this basic model by taking into account the marginal productivity of a given agro-ecosystem’s wild bees and the opportunity costs that farmers incur when assigning labor time to beekeeping. In that regard, we assessed the economic potential of on-farm beekeeping, which can involve several bee species, by juxtaposing this activity’s net benefits against those from hiring commercial pollination services. In addition to serving as a classification tool for a plurality of farmer-beekeeper-nature interactions and related optimization problems, the resulting analytical framework helps identifying the institutional settings that are most likely to lead to a specific bioeconomic equilibrium supply of pollination. What is more, it illustrates the interplay of the pertinent economic and agro-ecological factors, thus assisting the postulation of empirically testable hypotheses.
We also conducted two separate discrete choice experiments (DCEs) with orchardists from the Thai provinces of Chiang Mai (N = 198 respondents) and Chanthaburi (N = 127), in order to elicit their preferences for changes in the population of local wild bees that would hypothetically result from a conservation policy consisting (along with a per-household implementation fee) of at least one of the following three measures: (i) offering farmers bee-friendly alternatives to conventional agro-chemicals, (ii) enabling the protection and/or rehabilitation of natural bee habitats near cropland, and (iii) fostering the husbandry of native bee species by transferring technical knowledge on the practice of on-farm beekeeping. 
In this context, we fitted random parameter logit models on the Chiang Mai dataset. They yielded a significant willingness to pay (WTP) for the presented conservation measures and suggested that the disutility the respondents perceived for a 50% decline in the local population of native bees was greater than the utility they would derive from experiencing a bee population increase of the same magnitude. Moreover, comparing our aggregated WTP estimates to the expected production losses, showed that orchard farmers underestimated the true use value of pollination. On the other hand, the average WTP for all conservation measures combined by far exceeded the costs that, according to our calculations, each household would incur for such a project to be implemented. Our models also indicated a significant preference heterogeneity in the sampled population, which we could partly explain with idiosyncratic variables such as the respondents’ attitudes towards native bees and beekeeping. 
Finally, we examined further sources of randomness in the observed choice behavior, by modelling the unknown choice decision-relevant influences that could not be captured during the DCEs. To that end, we fitted generalized mixed logit (GMXL) models on the pooled datasets, which allowed comparing, on a common utility scale, the part-worth (value) estimates from Chiang Mai and Chanthaburi, where different experimental designs were applied. Our results reveal that farmers in Chanthaburi, who reported having experienced crop declines that they attributed to insufficient pollination, introduced less subjective factors into their choices than their Chiang Mai counterparts, who may have been less familiar with the importance of conserving bees. Moreover, the GMXL results also suggest that Chanthaburi farmers placed a significantly higher value on the above-mentioned measures (i) and (ii), while caring comparatively less about a 50% decline in local wild bee colonies. One can thus hypothesize that an actual local pollinator decline may have made Chanthaburi farmers more aware of the importance of conserving native bees, while paradoxically making them more independent from the provision of wild pollination services, as they started managing crop pollination with stingless bees.
K1 Biene
K1 Agrarökosystem
K1 Bestäubung
K1 Ökosystemdienstleistung
K1 Erhaltung
K1 Bewertung
PP Hohenheim
PB Kommunikations-, Informations- und Medienzentrum der Universität Hohenheim
UL http://opus.uni-hohenheim.de/volltexte/2019/1563