Market Structure Issues in the Livestock Industry
Testimony to the
United States Senate Committee on Agriculture, Nutrition and Forestry
Hearing on the Proposed Ban on Packer Ownership of Livestock and USDA's Enforcement of the Packers and Stockyards Act
July 16, 2002
by
C. Robert Taylor
Alfa Eminent Scholar (Distinguished University Professor) of
Agricultural and Public Policy
College of Agriculture
Auburn University
226 Life Sciences Building
Auburn University
Auburn, AL 36849
Phone: 334-844-1957
email: rtaylor@acesag.auburn.edu
Market Structure Issues in the Livestock Industry
C. Robert Taylor
Thank you, Mr. Chairman and members of the Senate Agriculture Committee. It is my pleasure to present testimony today about needed enforcement of the Packers and Stockyards Act (P&S Act).
Horizontal concentration combined with vertical integration through ownership or through supply chain management (SCM) has occurred at an unprecedented pace in the beef and pork industries. The massive restructuring of the livestock industry was initially driven by corporate efforts to capture private economic efficiencies, but in the last few years appears to be driven more by efforts to gain market power and economic power generally than by efforts to capture economic efficiencies.
In the past decade, USDA/Grain Inspection, Packers and Stockyards Agency (GIPSA) commissioned several university and ERS researchers to conduct studies as part of their investigation of concentration in the meatpacking industry. My testimony will first focus on USDA data that provide compelling evidence that market power issues now dominate efficiency issues. Then my testimony will turn to discussion of the USDA studies because they continue to be quoted widely by defenders of captive supply and because they appear to be used as a basis for the lack of P&S Act enforcement by GIPSA.
Overview of Power versus Efficiency
Defenders of consolidation often point to a general decline in inflation adjusted (real) meat prices in support of their argument that consolidation, concentration, and vertical integration are beneficial to consumers. Figure 1 shows USDA estimates of the real retail value for beef by month since 1980. As can be seen, there is a substantial long-term downward trend, to the obvious benefit of beef consumers. However, much of the decline in retail prices is not due to increased efficiency in the beef sector, but due to a decline in the real price of cattle feed. Figure 1 also shows estimates of the cost of feeding steers in Kansas, adjusted for inflation (1). Because the retail value is for selected cuts of meat, while the cost of gain is expressed in terms of dressed weight for the whole animal, the two series are not directly comparable. However, a rough comparison indicates that about 2/3 of the decreased real retail price for beef in the last two decades is due to lower feed prices, and not to efficiency gains in beef production, processing, storage, transportation, and retailing.
Retail prices alone cannot be used to ascertain the effectiveness, efficiency and fairness of the emerging system. The farm-to-wholesale price spread (or margin) and the wholesale-to-retail price spread give insight into whether monopsony (buyer) and monopoly (seller) power are being exerted by meat packers or retailers.
Farm-to-Wholesale Price Spreads
Figure 2 shows the monthly farm-to-wholesale price spread for beef, adjusted for inflation using the consumer price index (CPI-U), over the January, 1980, through May, 2002, time period. Also shown in Figure 2 is a non-linear trend-line, which is highly significant statistically. All data, including the CPI-U, are published periodically in USDA's Agricultural Outlook (2).
From Figure 2, it can be seen that the farm-to-wholesale (F-W) price spread for beef trended dramatically downward throughout the 1980s and into the early 1990s. A decrease in the F-W price spread is consistent with efficiency gains (lower unit slaughter costs) in meatpacking in a competitive market. That is, in a competitive market, we expect the F-W spread, which reflects gross revenue to packers, to decrease when unit slaughter costs decrease. The trend in the F-W price spread for beef during the 1980s and into the first few years of the 1990s is thus consistent with efficiency gains in meatpacking during that period in a competitive market. As can be seen in Figures 2, however, the trend in the F-W price spread has been strongly upward since the early 1990s, which is a dramatic departure from the efficiency-driven downward trend in the 1980s.
The trend-line F-W price spread for beef in June of 2002 was 50.5% higher than at the bottom of the trend in late 1993. This increase in the F-W price spread reflects a significant increase in the gross revenue to beef packers.
A common argument made by packers to justify the increasing F-W price spread is that they are adding more value. This explanation is easily dismissed for two reasons. First, USDA/ERS calculates the spread for a standard animal so that the spread will reflect only price changes (3). Second, even if meat quality improves over time, there should be no long-run trend in the F-W spread for given slaughter costs.
Another argument is that the strong upward trend in the F-W spread is due to higher wage rates in meat packing. However, Bureau of Labor Statistics (BLS) data show that the weekly pay for production workers in meat slaughtering plants, adjusted for inflation, was the same during the 1990s as it was during the 1980s. Moreover, the inflation adjusted pay for workers in slaughter plants has decreased somewhat during the last few years.
Meat packers claim that they are realizing efficiency gains by moving to larger and larger slaughter operations. Moreover, they claim that unit slaughter costs are actually less for the leaner animals now produced. Realization of these efficiency gains in a competitive environment would result in the F-W spread continuing to trend downward, not upward as it has in the mid to late 1990s.
The upward trend for much of the 1990s and into 2002 is too strong and too persistent to be explained by short-term spikes in prices, spreads, production, or competition with other meats.
The F-W price spread for beef was analyzed in an article published by USDA/ERS in the June-July 2000 issue of the USDA periodical, Agricultural Outlook. This report noted the upward spike in the F-W spread in the last half of 1999 and compared it to previous short-term spikes in 1980, 1991, and 1995.
After noting that previous spikes were short-term, the USDA report stated:
"… a long-term increase would be troubling. Increasing concentration in other sectors of the economy has often reflected intense competition and frequently led to falling costs and prices for the concentrating firms. But after an industry consolidates, when few firms face each other in a stable environment, competition may often become less intense."
The USDA report concluded by asking the question:
"As consolidation is completed, will packers successfully limit price competition among themselves and maintain 1999's high spreads?
As can be seen from Figure 2, high price spreads in 1999 have not just been maintained, but they have actually trended upward in the two years since this report was published. Since spikes in the F-W price spread for beef due to exogenous demand and supply shocks normally last only 2-3 months, the answer to the question posed by USDA is that packers apparently limited price competition and maintained the high F-W spreads
In summary, exertion of monopsony or oligopsony power is the only plausible explanation for the strong upward trend in the F-W spread for beef (4).
Wholesale-to-Retail Price Spreads
The Wholesale-to-Retail (W-R) price spread for beef is shown in Figure 3. Unlike the F-W margin for beef, there is no discernible U-shaped long-term trend in the W-R spread, although there is a slight upward trend in the W-R spread.
The W-R spread can be seen to be less volatile in the 1990s than in the 1980s, perhaps due to industry consolidation. In the early 1990s, the W-R spread swung upward somewhat, and then swung downward somewhat in the late 1990s. During the 1990s, the W-R spread for pork (not shown) tended to move in opposite directions from the W-R spread for beef. Such a linkage between the W-R spreads for beef and pork has been explained by joint costing in meat retailing. But even accounting for joint costing of beef and pork, there is no appreciable long-term trend in the W-R spread for beef during the 1980s and 1990s. However, the recent upward spike stands in stark contrast to the
two previous decades. As can be seen, the W-R spread for beef shot up in mid-2001, taking the inflation adjusted unit gross revenue to beef retailers to unprecedented levels.
The strong upward spike in the W-R spread, which has continued for over a year, suggests that monopoly power has been and continues to be exerted by meat retailers.
Figure 4 shows the farm-to-retail (F-R) spread for beef. An efficiency driven downward trend in the F-R spread during the 1980s is apparent from this figure. However, the strong upward movement during the last few years apparently due to exertion of monopsony and monopoly power has negated the efficiency gains in meatpacking and retailing that occurred during the 1980s.
USDA/GIPSA Captive Supply Studies
GIPSA funded several studies of consolidation, with special reference to the effects of captive supplies (5) on the cash market price for beef. The latest of these studies was authored by Schroeter and Azzam, and is usually referred to as the Panhandle study. Although the GIPSA studies continue to be cited by economists who are defending consolidation and concentration in meatpacking (e.g. Fuez, et al; Ward), it is critical to note that even the most recent GIPSA study used data only through mid-May of 1996. The study is thus dated, particularly in light of the strong upward trend in the F-W price spread for beef (Figure 2) which has occurred largely since data for the Panhandle study were collected by USDA.
From an enforcement standpoint, it is troubling that GIPSA has openly "studied" and publicly reported on the consolidation and concentration issues for over a decade, apparently oblivious to a body of antitrust literature on collusion detection. This literature points out that if the firms understand the range of econometric techniques available to the antitrust enforcement authority, then those firms can calculate the limits of undetectable collusion (e.g. Phlips, p. 131). So, what GIPSA has done with their highly publicized studies is to inform meatpackers of the issues, the data that will be used for analyses, the econometric techniques that will be used for detection, and what constitutes statistical significance and detectable collusion. Perhaps more troubling is that USDA has not used their authority to obtain potentially richer data sets analyzed with much more sophisticated statistical and econometric techniques.
Incomplete Data
The Panhandle Study has a technical design flaw that severely limits its potential for addressing the captive supply issue. For analytical purposes, the market for slaughter cattle is national in scope, as was recognized in the latest GIPSA report on captive supply reporting (6). Yet, the Schroeter and Azzam study was based on data for only four beef slaughter plants, all of which were located in the Texas panhandle. From an econometric standpoint, a national effect cannot, in general, be estimated with data from only for a few packing plants. Thus, conclusions from this study about the aggregate impacts of captive supply on cash market price are questionable.
Overlooked Finding in the Panhandle Study
A critical finding in the Panhandle study, which is easily overlooked, is that (7):
"… the four Texas plants paid significant quality-adjusted price premia for marketing agreement cattle relative to cattle purchased on the spot market (8)."
Quality-adjusted premia for marketing agreement cattle were found to be statistically significant for each of the four plants (two Excel, one IBP and one Monfort). Since marketing agreement prices are generally tied to the cash market price, this indicates that some or all of the captive supply producers may be receiving preferential treatment. A steady stream of unverified information also suggests preferential treatment for some producers.
In economic jargon, the implication of preferential treatment of some of the contractors by packers is that the aggregate supply curve for slaughter cattle is shifted outward, lowering cash market price and thereby harming (discriminating against) independent producers.
The preferential treatment of some producers may allow them to maintain profitability even with lower cash prices. However, independents will be eventually forced out of the business due to sustained losses. The profitability pressure on producers who do not have preferential deals is evident from Figure 5, which shows inflation adjusted monthly cattle feeding return estimates (based on cash market prices) calculated by Kansas State University economists (9).
Figure 5 also shows a linear trend-line. Note that trend-line returns became negative in 1994, which is approximately when the F-W spread (Figure 2) turned upward. Returns to finishing steers in Kansas averaged $41.40/head during the 1980s, while returns averaged a negative (loss) $14.60/head. Premia identified in the Panhandle study are large enough to suggest that contract producers who receive preferential treatment may be making a profit, while independent producers have sustained losses for about eight years.
The long-run implication of preferential treatment of some contract producers is that independents will eventually be forced to exit cattle feeding due to sustained losses. If the beef and pork sectors evolve like the poultry industry we can expect that once the independents are gone, the premia for the remaining cattle feeders will evaporate, leaving them at the mercy of the monopsonistic or oligopsonistic contractors.
Anticompetitive Vertical Integration
Many of the studies of consolidation and concentration in the livestock industry consider potential anticompetitive effects of horizontal concentration, but appear to have the implicit presumption that vertical integration (or VSC) is inherently good because transaction costs are reduced. This narrow view overlooks important literature in industrial economics that establishes conditions under which vertical integration and VSC are anticompetitive.
Hildred and Pinto state:
"… the thrust of SCM (supply chain management) is accomplished through contractual arrangements that leave intact the independent status of the firms involved. However, several models exist in which firms bend others to their wills and lessen competition through various SCM practices."
Contract poultry production is a clear case in point where firms (integrators) bend others (so-called independent contract producers) to their wills (Taylor, May, 2002).
There is substantive theoretical literature that established conditions under which either subcontracting or vertical integration may be privately preferred yet socially inefficient, even in the absence of externalities (e.g. Lyons and Sekkat). In particular, tight vertical relationships, such as those now developing between the concentrated food retailers and the meat packers, have antitrust implications that have not been adequately explored in the Economic Research Service (ERS) or GIPSA studies (10).
A particularly relevant body of antitrust literature deals with partial vertical integration. Although the poultry industry, which fully integrated in the 1950s, is often considered the poster child (11) for the industrialization of agriculture (Hayenga, Schroeder, Lawrence, Hayes, Vukina, Ward and Purcell; Taylor, May 2002), the meat packers may not attempt to fully integrate initially because it may not be profitable to do so. Under certain conditions firms (e.g. meat packers) integrate backwards (e.g. into beef production) to satisfy their high probability of demand and use the input market to satisfy their low probability demand (Carlton; Lieberman; Perry). Individual packers may have also used contracts to preempt other packers from accessing a particular supplier (Zhang and Sexton). By partially integrating the firm avoids paying a premium for the input (e.g. slaughter cattle) that is induced by the fluctuating demand of other buyers. Thus, expected total cost of the input to the firm is minimized by partial integration, but not by full integration (Lieberman). Partial integration thereby transfers the risk of demand fluctuations from the meat packer to the partial market that exists after partial integration. Of course, there is the question of how long a partial market will exist if independent producers are forced to exit the industry due to cash prices and increased risks transferred to them by integration.
Many giant agribusiness firms are linked through joint ventures, strategic alliances, partial ownership of competing firms (e.g. Smithfield's ownership of IBP stock), and tight vertical alliances. While such business arrangements may lead to efficiency gains, they may also compromise competition, perhaps in markets outside the scope of the ventures. The Federal Trade Commission recently released guidelines for examining such relationships (12).
Perhaps the seemingly unquestioned acceptance of vertical integration can be traced to the training of economists. Hildred and Pinto are generally critical of instructional programs in business colleges, programs in agricultural economics and most textbooks for overlooking the potential negative effects of VSC. They state:
"… understanding of market structures must be drastically modified to emphasize the existence and exercise of great market power within the food system. New understandings of antitrust policy in vertical relationships are required."
Price Determined by Forces of Supply and Demand
Literature on the meat industries is replete with claims that there are no problems with consolidation or captive supplies because "price is still determined by the forces of supply and demand." This is a meaningless statement because even in oligopolistic or oligopsonistic industries, the forces of demand and supply largely determine behavior of market participants and thus determine market price. Standard economic textbooks establish that supply of an oligopolist is generally not the same as competitive supply, and that input demand by oligopsonists is generally not the same as competitive input demand. Thus the statement that "forces of demand and supply determine price" is essentially meaningless in an antitrust sense; nevertheless this assertion continues to be widely used.
The Well-Being of Society
In a book published in the same year--1921--in which the Packers and Stockyards Act (PSA) was signed into law, Professor Frank Knight maintained that the well-being of society depended on maintaining a balance of: (a) economic efficiency, (b) economic power, and (c) economic freedom. He cautioned that the single-minded pursuit of economic efficiency would be at the expense of economic freedom and an economic power imbalance; nevertheless, economic efficiency has come to dominate the thinking of contemporary economists and also come to dominate public policy, including interpretation and enforcement of the PSA and other antitrust legislation.
In the last few years, in particular, the efficiency gains from consolidation and concentration in meatpacking are arguably quite modest, yet the economic power imbalance worsens and participation in production agriculture increasingly comes by invitation only (13), which is a loss in economic freedom.
The Evolution of Markets
Even Adam Smith, patron saint of CEOs, recognized (in 1776) that there is an inherent instability in a market economy; namely that through natural growth of firms (e.g. Wal-Mart) or through mergers and acquisitions (e.g. Tyson/IBP) that the market system could evolve to monopoly. He added that monopoly would lead to poor management and to higher prices for consumers.
Regulations and enforcement of those regulations are thus necessary for preservation of a market economy. Some global agribusiness firms now have economic and political power exceeding that of many governments. With tremendous economic and political power, and the lack of enforcement of predatory pricing laws, the viability of new firms entering and thereby restoring competition is increasingly a fiction.
Food consumers have historically benefited from efficiency gains and quality improvements resulting from horizontal concentration and vertical integration. But the manifestation of monopsony power several years ago (Figure 2) and now the manifestation of monopoly power as well (Figure 3) have negated efficiency gains realized by beef consumers prior to the mid-1990s. Economic power and market power, and not just economic efficiency, must be a vital part of future studies of the structure, conduct and performance of the food industry.
New Concentration Measures Needed
New statistical indicators of consolidation and concentration are badly needed. At present, the only statistical indicators reported by USDA generally are crude measures of horizontal concentration, specifically the four or five firm concentration ratios and the Herfindahl-Hirschman Index (HHI). Moreover, these horizontal indicators are computed to reflect ownership, rather than control (Taylor, 1999). New indicators are particularly needed to measure monopsony and monopoly power, economic power, vertical integration, the extent of VSCs, and the extent of collaborations among so-called competitors. Karier has proposed several such measures. USDA could shed considerable light on the extent of consolidation and power by developing and routinely reporting such measures that go beyond the traditional horizontal measures.
Needed Changes
Although there appears to be a fixation on the captive supply issue, there may be a consensus among academic economists about several problems with the beef industry that USDA could provide leadership in changing. Needed changes to the beef industry include: (a) replacement of the antiquated USDA subjective grading system with objective mechanized and instrumented quality assessment, (b) banning bidding practices which are potential instruments for exertion of abusive market power (Sexton; Connor, Carstensen, McEowen,and Harl), (c) elimination of marketing agreements tied to a cash market in which the contractor is an active participant (Purcell), and (d) providing the same (symmetric) information to both buyers and sellers on a timely basis (14).
The structure of the present food system in general, and livestock industries in particular, has become quite complex (Kinsey; Sexton). Analysis and enforcement of the P&S Act and other antitrust laws will therefore require agencies charged with enforcing these laws stepping up to a much higher level of theoretical and empirical detection of practices that are unfair, deceptive, discriminatory, and anticompetitive. Furthermore, analysis and enforcement of the P&S Act and other antitrust legislation should not just look at the past, but also be forward looking to ensure preservation of a market economy.
Although the meat and poultry industries are complex and industry changes have historically provided benefits to consumers through lower priced and higher quality meats, there is compelling evidence that anticompetitive practices have recently harmed independent producers and meat consumers. Enforcement of the P&S Act is imperative to prevent further abuses of market and economic power.
References
Carlton, Dennis W. "Vertical Integration in Competitive Markets Under Uncertainty," Journal of Industrial Economics, 27: 189-209, 1979
Carlton, Dennis W., and Jeffrey M. Perloff, Modern Industrial Organization, Third Edition, Addison-Wesley, 2000
Connor, John, Peter C. Carstensen, Roger McEowen and Neil E. Harl, The Ban on Packer Ownership and Feeding of Livestock: Legal and Economic Implications, 2002.
Federal Trade Commission and U.S. Department of Justice, Antitrust Guidelines for Collaborations Among Competitors, April 2000
Feuz, D., G. Grimes, M. L. Hayenga, S. R. Koontz, J. D. Lawrence, W. D. Purcell, T. C. Schroeder, and C. E. Ward. "Comments on Economic Impacts of Proposed Legislation to Prohibit Beef and Pork Packer Ownership, Feeding, or Control of Livestock," January 14, 2002
Hayenga, M. L., T. C. Schroeder, J. D. Lawrence, D. Hayes, T. Vukina, and C. E. Ward, Meat Packer Vertical Integration and Contract Linkages in the Beef and Pork Industries: An Economic Perspective, Report to the American Meat Institute, May 22, 2000
Hildred, William, and James Pinto. "Impacts of Supply Chain Management on Competition," Working Paper Series 02-10, April 2002, College of Business Administration, Northern Arizona University
Kinsey, Jean. D. "The New Food Economy: Consumers, Farms, Pharms, and Science," Presidential Address to the American Agricultural Economics Association, published in the American Journal of Agricultural Economics, Vol. 83, 2001: 1113-1130.
Karier, Thomas. Beyond Competition: The Economics of Mergers and Monopoly Power, M.E. Sharpe Publishing Co, Armonk, New York, 1993.
Lieberman, Marvin B. "Determinants of Vertical Integration: An Empirical Test," Journal of Industrial Economics, 39: 451-466, 1991
Perry, Martin K., "Vertical Integration: The Monopsony Case," The American Economic Review, 68: 561-570, 1978
Phlips, L. Competition Policy: A Game-Theoretic Perspective, Cambridge University Press, 1995.
Purcell, Wayne. "Contracts and Captive Supplies in Livestock: Why We Are Here, Implications, and Policy Issues," "Testimony at the Denver Captive Supply Forum," dated September 21, 2000, accessed at http://www.usda.gov/gipsa/forum/purcell/html
Schroeter, J. R., and A. Azzam, "Econometric Analysis of Fed Cattle Procurement in the Texas Panhandle," Report to GIPSA in fulfillment of Cooperative Agreement No. 98-PPD-01, November, 1999.
Sexton, Richard J. "Industrialization and Consolidation in the U.S. Food Sector: Implications for Competition and Welfare," American Journal of Agricultural Economics, 82:1087-1104, 2000.
Sexton, Richard J. "Testimony at the Denver Captive Supply Forum," dated September 28, 2000, accessed at http://www.usda.gov/gipsa/forum/sexton/html
Taylor, C. Robert, "Hiding the True Extent of Concentration and Market Power with Partial Ownership and Strategic Alliances", 1999, accessed at http://www.ag.auburn.edu/dept/aec/faculty/rtaylor.html
Taylor, C. Robert, "Restoring Economic Health to Contract Poultry Production," Agricultural and Resource Policy Forum, Auburn University, College of Agriculture, May 2002.
Taylor, C. Robert, "Where's the Beef: Monopoly and Monopsony Power in the Beef Industry," Agricultural and Resource Policy Forum, Auburn University, College of Agriculture, March 2002.
USDA, "Consolidation in Meatpacking: Causes & Concerns," Agricultural Outlook, June-July 2000, pp. 23-26.
USDA/GIPSA, Captive Supply of Cattle and GIPSA's Reporting of Captive Supply, Jan. 11, 2002.
Ward, Clement. E. "A Review of Causes for and Consequences of Economic Concentration in the U. S. Meatpacking Industry," Current Agriculture, Food and Resource Issues, 2: 1-28, 2002.
Zhang, Mingxia, and Richard J. Sexton, "Captive Supplies and the Cash Market Price: A Spatial Markets Approach," Journal of Agricultural and Resource Economics, 25: 88-108, 2000.
1 The Kansas State University data can be accessed at
http://www.agecon.ksu.edu/livestock/Livestock%20Databases/Livestock%20Databases.html
2. 2 3. 3 4 It has been suggested that the F-W price movements are due to factors such as price dynamics, competition
with other meats, production levels, and expansion versus contraction phases of the industry. However, I have
done extensive econometric analyses of beef, pork and poultry spreads, including these factors as candidate
explanatory variables. Although some of these variables are statistically significant in explaining the F-W
spread for beef, the non-linear u-shaped trend remains highly significant in their presence.
5 Captive supplies are generally defined to be slaughter cattle that are packer-owned or controlled 14 days
prior to slaughter.
6 USDA/GIPSA, Captive Supply of Cattle and GIPSA's Reporting of Captive Supply, Jan. 11, 2002.
7 While the data collected for the Panhandle study are inadequate for econometrically analyzing the effects
of captive supply on cash market price, they are appropriate for analyzing the quality-adjusted premia for
marketing agreement cattle at the four plants.
8 Schroeter and Azzam, p. 29.
9 The Kansas State University data can be accessed at
http://www.agecon.ksu.edu/livestock/Livestock%20Databases/Livestock%20Databases.html. The variable
code is KSYRLCOG in the monthly cattle supply database.
10 See, for example, Chapter 12 in Carlton and Perloff.
11 Although the broiler industry by some business executives as a model to be emulated, it should be
recognized that not all is well in the poultry industry. See Taylor for discussion of some of the problems
due to a lack of transparency in contracting and due to an imbalance in economic power favoring the
integrator over the contract producer.
12 Federal Trade Commission and U.S. Department of Justice, Antitrust Guidelines for Collaborations
Among Competitors, April 2000.
13 Kinsey notes that participation in the evolving food system is often by invitation only.
14 The intent of Mandatory Price Reporting was to provide information to all market participants in a timely
and symmetric way. However, as implemented, mandatory reporting has not accomplished this purpose.