« Back to Results

Automation and Digitalization in Agriculture

Paper Session

Friday, Jan. 3, 2020 2:30 PM - 4:30 PM (PST)

Manchester Grand Hyatt, Mission Beach A
Hosted By: Agricultural and Applied Economics Association
  • Chair: Xiaoxue Du, University of Idaho

Toward Economic Foundations of Supply and Demand for Automation in Production Agriculture

David Hennessy
Michigan State University


Two general purpose technologies have dominated innovation since World War II. Information
Technology emerged as an applied discipline and economic sector during the
1950s. At about that time too discoveries in genetics laid the foundations for precise biological
engineering as a tool-set available to the health, agriculture and industrial sectors.
Our interest is in how these innovations have a↵ected labor and capital use in production
agriculture. Labor and capital inputs have di↵ered in their comparative provision
of two critical services, physical work and information processing. Labor’s comparative
advantage has traditionally been in Bayes-type information processing rather than
Newton-type work. Thus the laborer can adjust the point at which a stem is cut and
will discard potentially problematic produce. Capital, being unable to adjust so readily,
was limited in where it could replace labor. Over time, however, raw ingredients were
adapted to become more uniform and capital smartened by way of sensors and related
technologies. These innovations have decreased demand for labor-sourced information
processing capacity. The proposed analysis will construct a Bayesian model of energy
and decision-making needs in food production so as to understand how agriculture’s capital
and labor requirements have evolved over time. Raw product heterogeneity will be
parameterized as a determinant of demand for information processing while both factors
will be endowed with decision-making types I and II error probabilities to parameterize
information processing capabilities. Factors will also di↵er in the unit cost of work done.
An equilibrium displacement model may be used to calibrate welfare e↵ects.

Adoption of Mechanization Solutions for Harvesting Fresh Market Blueberries

Karina Gallardo
Washington State University
Liang Lu
University of Idaho
Jill J. McCluskey
Washington State University


Blueberries to be sold in the fresh market represent a great example of a specialty crop
industry highly dependent on seasonal labor, where harvest mechanization appears as a
promising alternative to enhance the long-term economic profitability. In this study, we
analyze the expected profits of three harvesters aiming to replace the manual harvesting
of fresh market blueberries: the hand-held shaker, the over-the-row, and the modified
over-the-row harvester. We first compute the net present value of the three technologies,
to assess outcomes under risk neutrality. Then, we apply stochastic dominance to account
for risk. Across the 7 states included in the analyses, the most profitable alternative is
the modified over-the-row, followed by the over-the-row harvester, and the hand-held
shaker, compared to manual harvesting. Di↵erences in revenues across states, are due
to the variability of fresh market blueberry prices and yields per acre. The stochastic
dominance curves show a pattern consistent with the expected profits results, with some
di↵erences across states. Findings in this study should be useful for growers in their decision
making to invest in mechanization alternatives to manual harvesting. This is also
helpful to scientists and entrepreneurs to improve the reliability and efficiency of mechanical
harvesters. Finally, this could help extension professionals thriving to facilitate the
adoption and di↵usion of mechanical harvesters, by showing potential adoption patterns
across di↵erent blueberry production regions in the United States.

A General Equilibrium Model of Technology Adoption-Theory and Evidences from Robotic Milking Systems in Idaho

Xiaoxue Du
University of Idaho
Thomas Reardon
Michigan State University
Hernan Tejeda
University of Idaho
Philip Watson
University of Idaho


Vast price fluctuations, labor availability issues, and trade agreement uncertainties threaten
the livelihood of dairy farms, especially small- and mid-sized dairies. Automatic milking
systems (AMS) are quietly yet rapidly gaining popularity in response to this risky and uncertain business environment. A stylized fact regarding AMS adoption pattern in Idaho (3rd largest milk production state) suggests that the early adopters of AMS are the small and mid sized farms instead of large ones. This paper first provides a conceptual framework of AMS adoption. In particular, we develop a general equilibrium model of technology adoption. The dairy owner has to make not only production decisions, but also leisure (or o↵-farm activity) time allocation. The main observation from this model is that decline in labor availability has direct e↵ect of reducing dairy productivity, as well as indirect e↵ect of forcing the dairy owner to allocate time di↵erently. As a consequence, adopting robotic milking systems not only increases milking productivity, but also allows the dairy owner for more o↵-farm activities (or leisure). We test this hypothesis using
first-hand collected survey data from dairy farms in Idaho.

Strategic Outsourcing and Precision Agriculture: Towards a Silent Re-Organization of Agricultural Production in France?

Genevieve Nguyen
Julien Brailly
François Purseigle


In France as in other European countries, farm outsourcing has been developing for the past twenty years. Today, this phenomenon concerns both small and large farms. What is surprising is the growing number of farmers who outsource precision farming operations that involve sophisticated technologies and specialized expertise. This stylized fact is rather counter-intuitive to the known result of transaction cost theory, according to which in the presence of specific assets, ownership prevails over outsourcing. The objective of our study is to analyze the determinants of these new agricultural outsourcing practices associated with precision agriculture. We start with the transaction costs and property rights frameworks, then discuss recent theoretical contributions of relational contracts to explain the possibility of outsourcing in the presence of high asset specificity. Empirical evidences are provided for France using a mixed research methodology. Based on original data from surveys of 1200 farmers and of 20 of medium and large custom operators, our methodology combines an estimation of discrete choice models of outsourcing for different levels of asset specificity and case studies of major farm outsourcing organizational schemes. Our results show that in the presence of high specific assets, outsourcing can be preferred to ownership for strategic reasons. This is possible when one considers possible ex-ante incentive mechanisms (expectation of specialization gains, inclusion of a bonus based on the value of the output in the formal contract, participation of a third party), and informal incentive mechanisms built through repeated interactions.
JEL Classifications
  • O3 - Innovation; Research and Development; Technological Change; Intellectual Property Rights