Date of Award

Summer 8-18-2016

Level of Access

Open-Access Thesis

Degree Name

Master of Arts (MA)




Xuan Chen

Second Committee Member

Caroline Noblet

Third Committee Member

Gary Anderson


This paper is separated into three distinct chapters investigating different aspects of the Maine dairy industry. First, the paper provides a broad overview of the United States dairy industry with an emphasis on New England dairy farms. Structural industry changes, dairy legislation, and milk prices are topics of focus. The second chapter considers farm profitability and economies of scale among Maine dairy farms and compares the findings with similar studies from other states and previous studies of Maine dairy farms. The third chapter seeks to estimate the benefits of Maine’s Dairy Relief Program in terms of industry sustainability.

The first chapter provides elaborate background about the United States dairy industry. First, technological advancements in dairy farming in the past 100 years are discussed. The changing distribution is explained, as the number of farms in the past several decades has declined substantially. The apparent trend is that farms remaining in business are getting larger, while small farms are exiting the industry more frequently. Further, milk prices are explored in terms of consumer demand, and price asymmetry

between producers, processors, and retailers. Lastly, a brief overview is provided of risk management programs for dairy farms over the past few decades.

In the second chapter, 79 total farms from the 2010 and 2013 dairy cost-of- production studies were used to assess farm profitability and economies of scale. Estimation of a per-unit cost function indicates that costs fall at a decreasing rate with output and eventually increase. Further, estimation of a Cobb-Douglas production function suggests increasing returns to scale exist in the Maine dairy industry. These two findings reinforce the hypothesis that small farms have higher per-unit costs, as was indicated in the raw data and in similar cost-of-production studies.

The third chapter uses simulations to assess the benefits of Maine’s Dairy Relief program. A sample of 204 total farms from four cost-of-production studies was used to estimate Average Variable Cost (AVC). The AVC function was then used in combination with milk price and output data for all Maine dairy farms to estimate variable profit per cwt for all farms in the state from June 2004 to May 2015, which was necessary to run the simulation program. In each iteration, these estimated variable profits were used to establish a rule-of-thumb filter for removing observations for a probit regression. After removal of filtered observations, a binary probit was estimated with exit decision modeled as a function of seasonality, price lags in the previous six months, and AVC lags in the previous six months. Based on the corresponding probit probabilities of exit, a stochastic exit decision was forecasted without price supports for each farm-month combination. The results suggest that approximately 30% more farms would have exited and exits would have occurred sooner if the price floors had not been enacted in 2004.

Though tier 1 farms observe the greatest difference in number of exits, the three tiers representing larger farms are substantially impacted as well.

Included in

Economics Commons