Market and welfare effects of the U.S. Livestock Mandatory Reporting Act.

• Author: Njoroge, Kenneth

1. Introduction

   The Livestock Mandatory Reporting Act (the Act) was enacted in 2001 to provide market participants with information on all cash and noncash transactions reported by packers to the Agricultural Marketing Service (AMS) on a daily basis. The information is aggregated under specific confidentiality guidelines (1) to preserve the anonymity of the source (2) and published in the Mandatory Livestock and Meat Market News Reports (Reports). (3) AMS is responsible for setting the confidentiality guidelines as well as for enforcing the Act. As stated in the Act (Federal Register 2000), the Reports intend to provide information "that can be readily understood by producers, packers, and other market participants, ... [improve] the price and supply reporting services of the Department of Agriculture and [encourage] competition in the marketplace for livestock and livestock products." The rationale is that "as more animals have been purchased under marketing arrangements, neither the arrangements nor the final purchase prices are publicly disclosed." (4)

   About two years after the Act went into effect, Schroeder, Grunwald, and Ward (2002) reviewed the early literature on livestock mandatory reporting (LMR). A consensus appeared to be forming that, while LMR could be beneficial by reducing uncertainty and increasing competition in cattle markets, revealing too much information about rival pricing could facilitate collusion (and reduce competition) among packers (Schroeder, Grunwald, and Ward 2002, p. 11).

   The uncertainty-reducing effect of LMR was deduced from two experimental studies by Anderson et al. (1998) and Bastian, Koontz, and Menkhaus (2001). In the first experiment, price and quantity information in the fed cattle market, simulated through the Fed Cattle Market Simulator (FCMS), (5) was varied, and market outcomes were observed. Uncertainty, measured by price dispersion, was found to increase with the amount and type of information withheld. In the second experiment, also using the FCMS, market participants were provided with price and quantity information on cattle sold through the spot markets as well as through forward contracts. The authors observed a substantial reduction in the variance of forward contract prices. The effect of the forward-market information on the variance of spot prices was negative but not statistically significant.

   Maintaining the hypothesis that mandatory reporting reduces uncertainty through the provision of more information, Azzam (2003) studied the effects of the reduction in uncertainty on the structure, conduct, and performance of livestock markets. Azzam's theoretical findings suggest that, in a setting where cattle feeders and packers are risk averse and packers hold consistent conjectures, increased transparency due to mandatory reporting promotes competition and results in higher livestock prices. However, since increased transparency can make it easier for packers to collude (Wachenheim and DeVuyst 2001), Azzam's assumption of consistent conjectures is restrictive because it rules out the possibility of collusion.

   Njoroge (2003) provides a formal model of collusion that is in line with the arguments in Wachenheim and DeVuyst (2001). In Njoroge's study, collusion arises because risk-neutral oligopsonistic packers, who have divergent priors about the distribution of livestock and meat prices before the Act, have convergent posteriors after updating their priors with the Reports. In that setting, packers are better able to identify a unanimous trigger price and, as a result, find it easier to monitor each other's deviations from a (tacitly) collusive agreement. Risk neutrality, however, rules out gains to firms from the profit variance-reducing effect of increased transparency, and in this context, Njoroge's work focuses on the collusive effect.

   The collusive effect of the Act has been tested empirically by Azzam and Salvador (AS) (2004). The test by AS is based on the theoretical model by Jin (1996), who shows that if the squared sales of risk-averse Cournot firms decline after the introduction of an information-sharing arrangement, collusion is likely. AS found that, of all the regions AMS uses to report the data (Texas-Oklahoma, Kansas, Nebraska, Colorado, and Iowa-Minnesota), the Act was noncollusive only in Nebraska.

   The focus on either the risk effect or the collusive effect of transparency represents a major limitation of the preceding studies. Clearly, by reducing the perceived volatility of livestock prices, transparency generates a risk effect. The risk effect reduces the risk-averse packers' perceived cost of uncertainty and provides economic incentives for increased cattle procurement. Also, by increasing the observation of deviations from collusive behavior among packers, increased transparency may generate a collusive effect. The collusive effect provides economic incentives for packers to restrict livestock procurement toward the monopsony level. Specifically, an increase in transparency can reduce the noise of market signals used by an imperfect cartel to detect deviations from collusive conduct. (6) As a result, transparency can increase the cartel's efficiency in policing the behavior of its members, and, hence, it can increase the expected payoffs from collusion. In this context, the effect of increased transparency on the quantity of procured livestock and social welfare can be determined only by analyzing the trade-off between the risk and collusive effects. (7)

   The objective of this paper is to analyze the impact of market information provided by the Reports on equilibrium livestock slaughter, taking into account both the risk and the collusive effects of increased transparency. In addition, our paper seeks to identify the welfare implications of the Act and the determinants of the socially optimal level of information when the Reports are more informative than the packers' priors and have an influence on the behavior of the members of an imperfect cartel (i.e., a market structure where a small number of firms decide between collusive and noncollusive conduct).

   To analyze the market and welfare effects of the Act, we develop a stylized model of transparency, risk, and collusion that includes elements from the models by Azzam (2003) and Njoroge (2003). This enables us to focus on the analysis of the Act without the added complexity of considering the idiosyncrasies of any particular livestock market. Our model extends the Green and Porter (1984) framework of optimal trigger price strategies for collusion among risk-neutral firms by allowing for packer aversion toward risk. (8) Relaxing the assumption of risk neutrality allows for the explicit consideration of the trade-off between the risk and collusive effects of transparency, facilitating a more comprehensive economic analysis of the Act. In particular, while increasing the transparency in the Green-Porter setting would enhance the effectiveness of trigger price strategies and would reduce the quantity of livestock slaughter, increasing transparency in our setting would also reduce packer uncertainty, prompting an increase in the quantity of procured livestock. Thus, by allowing for packer aversion toward risk, our model explicitly accounts for the counteractive risk and collusive effects of the Act. A key result of our paper is that increasing the information in livestock markets can be social welfare enhancing even if it promotes collusion among packers.

   The rest of the paper is structured as follows. Section 2 determines the equilibrium livestock slaughter under an imperfect cartel market structure. The effect of increased information on the equilibrium quantity procured by the imperfect cartel is examined in section 3. Section 4 identifies the welfare effects of increased transparency and the determinants of a socially optimal level of information. Section 5 summarizes and concludes the paper.

2. The Optimal Livestock Slaughter

   The starting point of the analysis is the same as that considered in Azzam (2003), namely, a homogeneous, risk-averse oligopsony of N symmetric packers, (9) whose objective is to maximize the expected utility of profits. For simplicity, it is assumed that packers face zero processing costs and that the production technology between procured livestock and processed meat is fixed proportions. The livestock supply function facing each packer is given by

   W = [bar.W](Q) + [epsilon], (1)

   where W is the observable stochastic component of the price of livestock and [bar.W](Q) is the nonstochastic component given by

   [bar.W](Q) = a + bQ = a + [bq.sub.i] + [bQ.sub.-i]. (2)

   The parameters a and b are the intercept and the slope of the livestock-supply curve, respectively; [q.sub.i] is the livestock quantity procured by packer i; and [Q.sub.-i] is the aggregate livestock quantity procured by all other packers except for packer i. The parameter e is the random additive noise that consists of both the random shock to the factors of livestock production and the random noise of the market signals. It is assumed that e is identically and independently distributed with zero mean, PDF f(e) and CDF F(e), where f(E) is symmetric around zero,

   F(- [infinity]) = 0, and F(+ [infinity]) = 1.

   The volatility of livestock prices is measured by the variance of the random additive shock, [[sigma].sup.2.sub.w]. Following Azzam (2003), market transparency is defined as the degree to which the volatility of livestock and meat prices is reduced as the Reports become more informative. Given that livestock prices are stochastic and packers are assumed to be risk averse, maximization of expected utility of profits is analogous to maximization of the corresponding certainty equivalent (CE) of stochastic profits. The CE of packer i is given by

   [CE.sub.i]([q.sub.i], [Q.sub.-i]) = [P - a - b([q.sub.i] + [Q.sub.-i])][q.sub.i] - 0.5[lambda]([[sigma].sub.w][[q.sub.i]).sup.2], (3)

   where P is the...