Abstract
We develop a test for the presence of the monopsony power of the livestock integrator (principal) on the market for contract growers (agents) and estimate the model with the data on swine industry contract settlements. A natural test for the monopsony power of the principal would compare the estimated values of the marginal revenue products with the actual payments that agents receive for their services. The problem with implementing this approach comes from the fact that agents’ abilities and actions are unobservable. Our approach is based on estimating the slope of the inverse supply function for grower input using generalized method of moments (GMM) estimators. The model specifies the relationships between the observable consequences and unobservable grower characteristics imposing the first order conditions for principal’s profit maximization. The results show that the null hypothesis of no market power cannot be rejected.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The measurement of the cost of moral hazard in swine production contracts has been attempted by Dubois and Vukina (2004).
Notice, however, that in reality the price received by a contract grower and the internal price of captive supplies are never going to be the same even in the perfectly competitive setting. This is because the payment per unit of output that growers receive (settlement price) does not reflect the entire marginal cost of producing that output but rather represents the compensation for only those inputs and services that growers provide.
However, these models are quite different in many other respects. For a good review the monopsony in the labor market literature see Boal and Ransom (1997).
Over the years the industry has experienced many mergers and acquisitions such that many firms which existed during the period covered by this data set are no longer here today. The largest swine producer in the U.S. today is Smithfield Foods. Other big players are, for example, Tyson Foods and Premium Standard Farms.
The reasons for unequal number of observations across growers in not known. We confidently rule out the possibility that the facilities of those growers with only few observations in the data set were predominantly idle during that period.
As a consequence, we occasionally encounter more than one observation per grower i in period t.
Commingled pigs are feeder pigs bought at an auction or from an outside source. Other feeder pigs come from the breeding stock controlled by the integrator hence are deemed to be of superior quality.
The procedure to convert the quarterly prices into monthly figures and the exact matching of the monthly prices to contract settlement dates is explained in detail in Martin (1997).
For simplicity, we ignore the question whether the swine producer is an integrator or a fully vertically integrated company engaged in all stages of production and processing. Here we define an integrator as a firm that is engaged only in the production of finished hogs. The company buys feeder pigs and feed on an open market and sells live finished animals to the processing plant.
This result holds for a wide class of payment mechanisms including simple piece-rates and piece-rate tournaments (for risk-neutral and -averse agents) as long as the stochastic production technology is additive (see Levy and Vukina (2004)). Of course, if ability affects the marginal cost of effort or the marginal return of effort, then agents of different ability would choose different levels of effort.
Similar approach has been used by Scully (1974) to test for the monopsony power of major league baseball teams in the market for professional athletes.
We ignore the fact that the feed efficiency bonus is truncated at zero, which is rather harmless since all values of \({\left( {\phi - \frac{{F_{i} }}{{q_{i} }}} \right)}\) in the data set are strictly positive.
We also allow the weight of the finished hogs (K it ), and the number (H it ) and the weight of feeder pigs (κ 0it ) to vary across growers and time, whereas in the theoretical model all those variables were constant across growers and time.
Notice that even if we can identify the parameters exactly, estimating the supply elasticity may require obtaining the forecasts of the latent ability variable a i . Because the number of ability variables increases as the number of growers increases, any estimator of the ability variable will be inconsistent in short panel data like ours. This is a common problem in econometric models with fixed effects (see Wooldridge (2001) for example).
Matching of variable quality inputs with growers of different abilities have been studied by Leegomonchai and Vukina (2005) in the context of dynamic (renewable) broiler production tournaments characterized by both moral hazard and adverse selection problems. They rule out the presence of the ratchet effect type of dynamic incentives that would be caused by matching of high ability types with low quality inputs due to the emergence of pooling equilibria where the growers mask their types to prevent being stuck with the low quality input in the next tournament.
Our dataset is unbalanced in that the number of time series observations per grower varies across growers. As long as the selection variable indicating whether or not data are available is strictly exogenous, the standard theory carries through; see Wooldridge (2001), pp. 552–556. For further discussion on the time aggregation of moment conditions see Hayashi et al. (1996), pp. 270–271.
All other specifications produced worse results either in terms of higher J statistics, or wrong signs of the estimated coefficients, or both.
References
Altonji JG, Segal L (1996) Small sample bias in GMM estimation of covariance structures. J Bus Econ stat 14:353–366
Appelbaum E (1982) The estimation of the degree of oligopoly power. J Econom 19:287–299
Azzam AM (1998) Captive supplies, market conduct, and the open-market price. Am J Agric Econ 80:76–83
Azzam AM, Anderson DG (1996) Assessing competition in meatpacking: economic history, theory and evidence. Prepared for the USDA grain inspection, packers and stockyards administration, report GIPSA-RR 96-6, Washington, District of Columbia
Black DA, Loewenstein MA (1991) Self-enforcing labor contracts with costly mobility: the subgame perfect solution to the chairman’s problem. Res Labor Econ 12:63–83
Boal WM, Ransom MR (1997) Monopsony in the labor market. J Econ Lit 35:86–112
Bresnahan TF (1987) Competition and colussion in the American automobile industry: the 1955 price war. J Ind Econ 35:457–482
Dubois P, Vukina T (2004) Grower risk aversion and the cost of moral hazard in livestock production contracts. Proceedings from the ASSA 2004 Annual convention, San Diego, California, January 3–5, 2004. Am J Agric Econ 86:835–841
Grimes G, Plain R, Meyer S (2003) U.S. hog marketing contract study. National Pork Board, January
Hansen LP (1982) Large sample properties of generalized method of moment estimators. Econometrica 50:1029–1054
Hayashi F, Altonji J, Kotlikoff L (1996) Risk-sharing between and within families. Econometrica 64:261–294
Kliebenstein JB, Lawrence JD (1995) Contracting and vertical coordination in the United States pork industry. Am J Agric Econ 77:1213–1218
Leegomonchai P, Vukina T (2005) Dynamic incentives and agent discrimination in broiler production tournaments. J Econ Manage Strategy 14:849–877
Levy A, Vukina T (2004) League composition effect in tournaments with heterogeneous players: an empirical analysis of broiler contracts. J Labor Econ 22:353–377
Martin LL (1997) Production contracts, risk shifting, and relative performance payments in the pork industry. J Agric Appl Econ 29:267–278
Murray BC (1995) Measuring oligopsony power with shadow prices: U.S. markets for pulpwood and sawlogs. Rev Econ Stat 77:486–498
National Pork Producer Council (2000) Guide to contracting: production. Des Moines, Iowa
Paarlberg PL, Boehlje M, Foster K, Doering O, Tyner W (1999) Structural change and market performance in agriculture: critical issues and concerns about concentration in the pork industry. Purdue University, Department of Agricultural Economics, Staff Paper, No. 99–14
Rebitzer JB, Taylor LJ (1995) The consequences of minimum wage laws: some new theoretical ideas. J Public Econ 56:245–255
Schroeder TC, Jones R, Mintert J, Barkely A (1993) The impact of forward contracting on fed cattle transaction prices. Rev Agric Econ 15:325–337
Schroeter JR (1988) Estimating the degree of market power in the beef packing industry. Rev Agric Econ 70:158–162
Scully GW (1974) Pay and performance in major league baseball. Am Econ Rev 64:915–930
Tauchen G (1986) Statistical properties of generalized method-of-moments estimators of structural parameters obtained from financial market data. J Bus Econ stat 4:397–416
U.S. Department of Agriculture (USDA), Economic Research Service (ERS) (2001) Commodity costs and returns: U.S. and regional cost and return data. <www:ers:usda:gov=Data=CostsAndReturns=testpick:html>
U.S. Department of Agriculture (USDA), Grain Inspection, Packers, and Stockyards Administration (GIPSA) (2001) Packers and stockyards statitical report, 2001. <www:usda:gov=gipsa=pubs=stat00=Part2:pdf>
Wooldridge JM (2001) Econometric analysis of cross section and panel data. MIT: Cambridge, Massachusetts
Acknowledgements
This research was partially funded by the North Carolina Attorney General-Smithfield Foods Agreement. This paper benefited greatly from insightful comments by Brian Murray, Mary Muth, Wally Thurman, Mike Wohlgenant, and two anonymous referees. We are thankful to Laura Martin for the permission to use her contract settlements data set.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Inoue, A., Vukina, T. Testing for the principal’s monopsony power in agency contracts. Empirical Economics 31, 717–734 (2006). https://doi.org/10.1007/s00181-005-0041-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00181-005-0041-6