Date of Award


Document Type


Degree Name

Master of Science (MS)


Food Systems

First Advisor

Christopher Koliba

Second Advisor

Sidney Bosworth


A comprehensive literature review was undertaken in order to define and assess the sustainability and resiliency characteristics associated with grass-based and confinement dairy farming. Primarily as a result of reduced input costs, grass-based dairy farming often enhances profitability over confinement systems, especially on small farms. Further, conversion of tilled soil to permanent pasture has been shown to significantly reduce harmful sediment and nutrient transport into waterways. Perennial forage also acts as a carbon sink, curtailing or even negating a grass-based farm's carbon footprint. Finally, social benefits derived from enhanced nutrition and higher quality of life are also associated with grass-based dairy farming. Given that policy goals of the State of Vermont include both bolstering farm viability and reducing farm-related runoff, two questions are then raised. What is the most effective way to incentivize the adoption of rotational grazing in Vermont? And what types of farms are best suited to its use?

A series of interviews with dairy experts and farmers was conducted as a preliminary investigation into these questions. This qualitative evidence suggested that farmers generally adopted grass-based dairying after observing a peer's success with the method, suggesting that a key leverage point may be peer-based learning.

A behavioral economics game was developed to evaluate the role of peer networks in facilitating decision-making under conditions of uncertainty. A computerized game platform simulated networks of small dairy farm enterprises, with participants acting as farm managers. Treatments varied the size of peer networks, as well as the inclusion of a perfectly-performing automated 'seed player.' Participants could base their decisions upon the successes of their peers. They received a cash incentive based on their farms' performance. Results indicated that players with higher numbers of peers made better economic decisions on average. The inclusion of a 'seed player' within a network, which modeled the ideal behavior, also facilitated better decision-making. Both of these correlations were statistically significant. Furthermore, the shape of the 'diffusion curve' of new adoptees confirmed literature on the dynamics of innovation diffusion. Public policy implications from this work include an increased focus on facilitating peer-to-peer learning among farmers where Best Management Practice adoption is a policy goal.

To further evaluate the potential for peer learning to facilitate positive change, the Dairy Farm Transitions Agent Based Model (DFTABM) was developed. The model was calibrated using existing datasets along with the qualitative and quantitative results described above. It forecasts effects on farm profitability, attrition, and soil loss arising from varying assumptions about peer network connectivity, peer emulation, macroeconomic trends, and agri-environmental policy. Nine experimental treatments were assessed. Overall, it was found that high rates of emulation coupled with high rates of connectivity'especially targeted connectivity among smaller farms'yielded the best balance of farm viability and reduction in soil loss. The establishment of a performance-based tax credit had no clear correlation with the resulting soil loss figures predicted by the model. Policy implications from this study include the finding that direct payment schemes for reduction in environmental harm may not always have their intended effects, whereas policies that enhance peer-to-peer learning opportunities, especially among the proprietors of smaller farms, may present an effective and relatively affordable means by which to bolster farm profitability while also reducing environmental degradation.



Number of Pages

160 p.