Structural Change and Market Performance in Agriculture:

Critical Issues and Concerns about Concentration in the Pork Industry

Testimony of

Chairman Lugar and Members of the Committee:

The U.S. agricultural industry is in the midst of major structural change. Production is changing from an industry dominated by family-based, small-scale, relatively independent firms to one of larger firms that are more tightly aligned. The sector is becoming more industrialized, more specialized, more integrated, and the pace of change is increasing.

Evidence of increased concentration in hog slaughter

Like many industries, the hog slaughter industry is characterized by a small number of firms producing most of the pork and a fringe group consisting of a large number of small slaughtering firms contributing little to total pork supply. Measures of market concentration suggest that the hog slaughter (packing) industry has become significantly more concentrated since 1985. Concentration indexes for the top 4 and 8 firms have risen. In 1985, the 4-firm concentration index was 32.2 and the 8-firm index was 50.8. By 1997, the indexes had risen to 54.3 and 75.7. Traditional classification would label an industry with these concentration indexes as a moderately concentrated industry. Another measure of industry concentration is the Herfindahl-Hirschman (HH) index which is the sum of the squared market share of all firms. This index for the hog slaughter industry has more than doubled between 1985 and 1997.

The Herfindahl-Hirschman index can be used to estimate the equivalent symmetric number of firms in an industry. That is, if all firms in an industry were the same size, how many would there be? In the case of hog packing the symmetric firm number for 1985 is 22 which is sufficiently large enough to avoid pricing distortions. By 1997, the symmetric firm number for hog packing had fallen to 10. There is no specific value where we can say serious pricing distortions due to market power arise. However, once an industry drops to a symmetric firm number of 10, concern is warranted.

Economic implications of increased concentration

While every industry is unique, economic theory offers general conclusions about pricing under oligopoly and oligopsony (imperfect competition). One factor is the nature of interaction among firms. How does a firm expect its rivals to react when its behavior changes?

A related issue is whether the goods produced by different firms are different or alike. Product differentiation is linked to how demand changes as price changes. Firms can affect the price and rival behavior. More elasticity reduces the ability to manipulate outcomes.

International trade affects the exercise of market power. Domestic firms with market power cannot raise their prices if they face either the threat of import competition or the loss of export markets.

The cost structure of an industry affects pricing. Large fixed costs and entry barriers limit potential competition and enhance market power. Fixed capacity constrains the reaction of rival firms. Economies of scale drive unit costs lower as output rises, and can reduce firm numbers. Falling unit costs lowers output prices if the mark-up is constant, but falling firm numbers increases mark-ups on output.

Vertical relationships between input suppliers and manufacturers become critical. Vertical integration and coordination may occur for several reasons. They can be used to exercise market power.

There is limited information on these factors for the hog packing industry. There is evidence of product differentiation, yet, the aggregate demand elasticity for pork is estimated to be high compared to other foods. The U.S. and the world pork markets appear linked. It is likely that packers have limited market power for pricing the pork. On the other hand, imports of live hogs for packers is not large relative to slaughter. This suggests an industry which sets meat output with little influence on the pork price, but with potentially substantial influence over the live hog price. The hog packing industry is perceived as having large fixed costs and economies of scale. There is evidence that increasingly large numbers of hogs move under contracts. As concentration and integration increases, problems of volatile spot markets and market foreclosure for independent growers are likely to increase.

An illustration of what may be ocuring

We constructed a model to illustrate how concentration might affect the hog price as the number of packing firms declines. Because much of the information necessary to construct an accurate model of the industry is not available, we made assumptions. It is assumed that pork is a homogenous good, and each packer sets its slaughter believing that rival packers will not change their behavior. The industry consists of identical packing firms varied from 1 to 20. Because the United States trades pork with few barriers, the model assumes the price of pork is given. Hogs are treated as non-traded which gives the packers the ability to mark down the animal price. The extent of economies of scale and vertical coordination are unknown and are excluded.

The model calculates how the mark down, or price gap, from the perfectly competitive hog price changes in response to changes in firm numbers. This mark down represents the potential market power of packers to pay less for inputs than would have been the case under perfect competition. If there are 20 firms, similar to the situation in the late 1980s, the price paid to producers for hogs is 95 percent of the perfectly competitive level. As the number of packers falls, the gap on the price paid to farmers increases. At first the gap remains small. Starting from 20 firms, reduced firm numbers does not initially lead to much larger mark downs on the competitive price. When there are 14 firms instead of 20, the hog price is 90 percent of the competitive level. Going from 14 firms to 8 firms increases the mark down from 10 percent to 18 percent. Once firm numbers drop below 5 the mark down increases sharply.

The Herfindahl-Hirschman index indicates that in 1997 the industry consists of 10 symmetric firms. The illustrative results show that the range of 8-10 equal sized firms is a critical transition between the magnitudes of the mark downs. Above 10 firms a change in firm numbers does not cause much change in the mark down. Below 8 firms, the increases in mark down accelerate. It is critical to identify where the industry lies along this relationship. In the case of the model used here, if the industry is presently at 8-10 firms, policy designed to increase firm numbers may be less critical than policy designed to maintain firm numbers. In other words, it may be that the concentration which has occurred so far has not pushed us much below competitive prices, but further concentration could have serious adverse consequences for producers.

Missing information is critical

Knowing the precise shape and location of the relationship is critical. The relationship can only serve as an illustration because critical pieces of the puzzle are missing. Assumptions are made which may incompletely reflect the hog/pork sector. These pieces must be inserted to accurately analyze the consequences of increased packer concentration. One set of critical information covers the role of vertical coordination. Data on the numbers of hogs contracted and owned, the contract terms, and costs of packing firms is required. A weakness is limited knowledge of the extent of product differentiation and supply-use data by product. With additional data, it will be possible to narrow the uncertainty of what is occurring. Providing that analysis should be a high priority.

Summary

We are witnessing the industrialization of agriculture. The structural changes have been pronounced in the pork sector and raise important questions about competitiveness of both product and input markets.

There is evidence of increased concentration to the point where public vigilance is warranted. Concentration indexes are high and may be reaching a point where mark down pricing on hogs could be significant. It is important to have the required information and analysis as soon as possible to properly assess where we are in this respect.