Predatory Pricing in the Cigarette Industry: The Liggett Case, Study notes of Economics

Download Predatory Pricing in the Cigarette Industry: The Liggett Case and more Study notes Economics in PDF only on Docsity! CASE 10 Predation by a Nondominant Firm: The Liggett Case (1993) William B. Burnett INTRODUCTION In July 1984, Liggett Group, Inc., parent of Liggett & Myers Tobacco Company (“Liggett”), filed suit alleging predatory behavior by Brown & Williamson Tobacco Corporation (“B&W”) for its conduct surrounding the introduction and marketing of generic cigarettes. Key allegations in- volved predatory pricing. Trial began five years later, in July 1989. In March 1990, after 115 days of hearings and 10 days of deliberations, the jury found that B&W had acted predatorily in its pricing of (and in other business conduct related to) generic cigarettes, and awarded damages of $49.6 million—$148.8 million after trebling. In August 1990, the trial judge granted a motion by B&W for judgment notwithstanding the ver- dict, set aside the jury verdict, and entered a judgment in favor of Brown & Williamson. Liggett appealed to the U.S. Fourth Circuit Court of Ap- peals, but that court affirmed the trial judge’s opinion. Liggett appealed again, this time to the U.S. Supreme Court. In June 1993, the Supreme Court, while rejecting the reasoning of the Court of Appeals, upheld the dismissal of the case, relying heavily on the analysis of the District Court in granting judgment notwithstanding the verdict.1 William B. Burnett was an expert for Liggett Group, Inc. in this case. This article represents the views of the author and does not necessarily represent the views of Liggett Group Inc., its succes- sors, or its counsel. While this chapter represents an overview of Liggett v. Brown & Williamson, it does not examine all of the major economic issues that were raised during the case, including the testimony of the author. 1 Liggett Group, Inc. v. Brown & Williamson Tobacco Corporation, 748 F. Supp. 344 (M.D.N.C. 1990), aff’d 964 F.2d 335 (4th Cir. 1992). On appeal to the Supreme Court, the case was recap- tioned to reflect a change in the ownership of Liggett Group, Inc.: Brooke Group, Ltd. v. Brown & Williamson Tobacco Corporation, 61 U.S.L.W. 4699 (1993). 239 In its starkest form, this is a case about the rationality and fact of preda- tory pricing in a highly concentrated, oligopolistic industry by a firm that was not the largest in its industry and did not have a dominant market share. In the 1970s and 1980s, there were only six domestic producers of ciga- rettes, and market concentration was very high. Liggett, the smallest firm in the industry and on the verge of failure in the late 1970s, introduced a no- name, generic line of cigarette products at a price significantly below that of any other product. The dramatic success of Liggett’s generics provoked competitive reaction by B&W, the third-largest firm in the industry with a share of less than 12 percent (see Table 10-1). Liggett alleged that B&W priced its generic cigarettes below variable cost with the intention of either disciplining Liggett’s generic pricing or driving the firm from the market. Evaluating Predation Claims Identifying instances of predatory pricing generally requires a two-step process. The analyst must isolate those market facts that demonstrate (1) that predatory pricing is rational or plausible behavior for the aggressor, and (2) that predatory pricing actually occurred. If predation is not ratio- nal, evaluation of claims of predatory behavior usually need go no further. If predation is plausible, however, analysis of the facts must determine whether it actually occurred. The following definition of predatory pricing is adopted to isolate facts or conditions useful for distinguishing predatory conduct from ag- gressive but welfare-enhancing competitive behavior: Predatory pricing involves incurring losses by pricing below some mea- sure of cost in the short run in order to exclude or discipline rivals in antici- pation of recouping those losses in the future. Losses may be recouped through either the maintenance or establishment, and successful exercise, of market power. TABLE 10-1 Cigarette Market Shares and Concentration 1980 1984 1988 Philip Moris 30.9% 35.3% 39.3% R. J. Reynolds 32.6 31.6 31.8 Brown & Williamson 13.7 11.3 10.9 American Tobacco 10.8 7.9 7.0 Lorillard 9.7 8.2 8.2 Liggett & Myers 2.3 5.7 2.8 Four-firm concentration ratio: 88.0 86.4 90.2 Herfindahl-Hirschman Index: 2421 2534 2799 Source: Plaintiff’s exhibit 4285. 240 THE ANTITRUST REVOLUTION The price gap at retail also increased substantially, and generic sales contin- ued to accelerate rapidly. Liggett generics grew from 5.8 billion cigarettes (0.9 percent of all cigarette sales) in 1982 to 23.9 billion cigarettes (4.0 per- cent of all cigarette sales) in 1984. Liggett’s share of total cigarette sales in- creased from 2.9 percent to 4.8 percent between 1982 and 1983, and then to 5.7 percent in 1984. The success of Liggett’s generics thus injected new life into a failing company, in addition to introducing price competition into the industry for the first time in more than forty years. Brown & Wiliamson’s Generic Response The increasing sales of Liggett generics came at a crucial time for B&W and, to varying degrees, other producers. Total cigarette sales in the United States peaked in 1981, at 627 billion “sticks,” and then began a steady de- cline, falling to about 600 billion sticks in 1984 and to 558 billion sticks in 1988, or roughly 1.65 percent per year. Further, B&W’s established, branded products had been gradually declining in share for more than a decade. The firm’s market share fell from about 17.5 percent in 1973 to 14.4 percent in 1979. Liggett’s increasing generic sales accelerated and exacerbated the decline in branded cigarette sales by the larger manufac- turers. B&W thus confronted a declining market share in a shrinking mar- ket, the effect of which was intensified by the shift in sales from branded products to Liggett’s generic cigarettes. Moreover, B&W’s consumer switching studies evaluating the losses of each firm’s branded sales to generics showed that B&W was one of two principal “losers” to Liggett generics, disproportionately contributing about 21 percent of the almost 17.9 billion units gained by Liggett. As B&W’s share of all cigarette sales was then about 12 percent, the firm con- tributed to generics an amount equal to about 1.7 times its market share. For each of the larger producers, such lost sales resulted in very large profit mar- gin losses on branded products that were directly attributable to the growth of Liggett’s generics. For example, estimating the actual and potential unit and dollar losses attributable to generic growth, B&W wrote in March 1984: B&W’s contribution to generics is disproportionately high. Specifically, B&W contributes about 70% more than its fair share volume to gener- ics. B&W losses account for about 21% of generics’ gains. In 1983 B&W lost about 3.7 billion sticks to generics, a variable margin loss of over $50MM. By 1988, this loss could total 18 billion sticks and about $350MM lost variable margin.3 Subsequent to Liggett’s introduction of generics, B&W’s share of all cigarettes continued its historic decline, falling from 14.4 percent in 1979 3 PX-5. PX indicates “Plaintiff’s Exhibit,” or documents introduced into evidence at the trial. 243 Case 10: The Liggett Case (1993) to 11.5 percent in 1983, during which time its unit sales fell even more rapidly, reflecting both declining industry sales and a falling share, from about 88.5 billion units to 68.5 billion units (about 6.2% per year). B&W’s motivation to retain or regain unit sales was further bolstered by the need to utilize fully its new world-class production facility at Macon, Georgia, which was being phased into production. In late 1983, B&W formed a task force of senior executives to evaluate the effects of generic cigarettes on that firm and to make recommendations concerning the possible launch of a generic cigarette product to compete with Liggett’s. After five months of study, B&W’s task force characterized Liggett’s unprecedented and threatening competitive move and the effects it could have on branded cigarette prices and profits as follows: This is the first time that a manufacturer [Liggett] has used pricing as a strategic marketing weapon in the U.S. since the depression era. At that time, when economic conditions were more severe than at present, low price brands captured approximately 23% of the market before the dominant manufacturers dropped their prices on full priced brands, thereby limiting the low price entries to 10%–14% market share during the 1930s. . . . If the economy segment were to grow to 25 to 30 percent of the total mar- ket, manufacturers as a last resort may reduce list prices of full margin products in an attempt to repeat the action of the 1930’s.4 The response of the major tobacco companies to low-priced cigarette in- troductions in the 1930s led to their conviction for attempted monopoliza- tion under sections 1 and 2 of the Sherman Antitrust Act. B&W did not wait until generics’ share reached 25 percent but introduced a generic of its own, the first shipments of which were made in July 1984. In June and July 1984, B&W first offered to sell generic products at a list price equal to Liggett’s ($3.75 per carton for 85-mm cigarettes), with volume rebates (discounts off wholesale list price) of up to $0.30 per car- ton. No sales were transacted at this price. Liggett immediately responded by offering rebates as well, adopting a policy of matching B&W’s rebates to within about $0.07 per carton, usually maintaining a net price (list price less rebates) slightly higher than B&W’s offer. After several rounds of in- creasingly deep rebate offers by both firms in June and July of 1984, B&W was selling generics at rebate levels of between $0.40 and $0.80 per car- ton, depending on the number of cartons purchased per quarter. Discounts increased still further in December 1984, and again several times in 1985. On average, B&W discounts totaled about $0.62 per carton in 1984 and about $0.82 per carton in 1985. These were large and dramatic 4 PX-12. For discussion of cigarette pricing behavior in the 1930s, see Nichols (1949) and Ten- nant (1950). 244 THE ANTITRUST REVOLUTION figures when compared with 1984 list prices of $3.75 per carton (85-mm cigarettes); the rebate’s significance is further highlighted when the $1.60 per carton FET is deducted from the list price, yielding B&W gross reve- nues after deduction of federal taxes of $2.15 per carton. B&W’s net real- izations from generic sales, after all deductions for discounts and taxes, were $1.53 and $1.41 per carton in 1984 and 1985, respectively. Against this backdrop of aggressive price cutting and other business conduct, Liggett filed suit against B&W in July 1984, alleging predatory pricing and other unfair business practices.5 LIGGETT’S CLAIM OF PREDATION Liggett’s theory of the case, which was largely derived from and supported by B&W documents, alleged that firms in the highly concentrated ciga- rette industry tacitly colluded to raise branded cigarette prices and profits above competitive levels.6 All cigarette manufacturers, including B&W, benefit from the avoidance of competition and have supranormal profits to protect on the sale of branded products. Liggett, the smallest firm and on the verge of failure, took the unprecedented step of introducing a generic product—at a discounted price. Its incentives differed materially from those of the other manufacturers. Liggett generics would take branded sales from all larger manufacturers but would not “cannibalize” or materi- ally reduce its own very small share of branded sales. B&W observed in March 1984: “Generic growth represents volume erosion for all competi- tors. Unchallenged, L&M will continue aggressive segment development since it has virtually no stake in the branded, full price segment.”7 As alleged by Liggett, B&W priced its product below variable cost to reduce Liggett’s competitive initiatives in low-price cigarettes and to slow the growth rate of generics. B&W cut prices by offering volume rebates as opposed to reducing list prices, expecting that price cuts implemented in this manner would not be reflected in reduced prices to consumers, thus increasing the branded-generic price gap. After an interval of below-cost pricing, B&W hoped to reduce the price gap either by imposing such sig- nificant losses that Liggett would raise generic prices, or by seizing con- trol of the generic category and then raising prices itself. B&W also hoped to induce Liggett to reduce its marketing intended to make consumers 5 Predatory pricing claims were filed under Section 2(a) of the Clayton Act, as amended by the Robinson-Patman Act, 15 U.S.C. Section 13(a); claims involving B&W’s copying of Lig- gett’s generic packages were filed under the Lanham Trade-Mark Act, 15 U.S.C. Section 1125(a). Similar claims were made under various state common law and statutory unfair trade practice provisions. 6 There were no allegations of overt or explicit collusion among firms in the industry. All allega- tions and evidence involved tacit collusion. 7 PX-5. 245 Case 10: The Liggett Case (1993) makes recoupment, after the predatory campaign has ended, far less likely. However, by forcing Liggett to respond via volume rebates so as not to rapidly lose its generic volume, B&W’s offer to sell generics below cost forced the great bulk of the losses onto Liggett, the target. Thus, in the pe- riod of recoupment, B&W had less to recoup than would a dominant firm that was targeting a smaller rival. Second, if B&W succeeded in narrowing the price gap, its recoup- ment was likely to be rapid and certain rather than distant and uncertain. B&W’s branded cigarettes were already priced above competitive levels; every branded unit lost to Liggett generics constituted an immediate loss of substantial margins and profits. If the rate of loss of branded sales could be reduced, as a consequence of its conduct with respect to generics, B&W’s recoupment would be rapid. Without price cutting in the generic category, B&W’s losses would have been even greater. This “simultaneous recoupment,” as it was termed at the trial, meant that price-cutting “invest- ments” made at a specific time were recouped very quickly by reducing the number of branded sales that would have been lost absent these invest- ments. From B&W’s perspective, the margin and profit losses at issue were tremendous. In May 1984 B&W estimated that, if it did nothing with respect to generics, it might lose as much as $398.2 million in profits on branded cigarettes for the period 1984 through 1988.8 After imposing the bulk of the losses on Liggett, B&W had only to slow the rate of growth in generics by a small amount to justify its modest investment. Finally, the B&W plan reflected its understanding of Liggett’s inability to weather a sustained attack on its generics. Liggett was a subsidiary of the large, diversified British firm Grand Metropolitan, PLC. Liggett, however, did not have strong support from its parent. Grand Met had originally pur- chased Liggett in a hostile takeover in order to acquire its liquor distribution business and had long sought to sell the cigarette division. Although Grand Met financed Liggett’s losses for the first twelve to eighteen months of B&W’s attack, it placed limits on its support. Liggett’s success in generics in the early 1980s had made the firm’s cigarette business a more attractive investment. B&W was well aware that its generic launch and off-list price discounting had prevented Liggett from completing a leveraged buyout and that, if Grand Met were to sell Liggett, the losses on generics had to be lim- ited or stopped. This could have been done by either (1) eliminating the vol- ume rebates (and accompanying losses) and therefore ceding generic vol- ume to B&W after which B&W could raise generic prices (half of win-win for B&W), or (2) raising the list price of generics, which B&W would fol- low (the other half of win-win for B&W). In both cases, Liggett would lose, having given up control of the category either by allowing B&W to sell 8 PX-12. Brown & Williamson’s measure of loss was stated in terms of “trading profit”; trading profit is net revenue (after taxes and discounts) less manufacturing cost, marketing expenditures, manufacturing overhead expenses, and profit sharing. 248 THE ANTITRUST REVOLUTION more generics or by raising generic prices. In fact, Grand Met allowed Liggett to remain an active participant in the price war for about a year, after which Liggett began to raise generic prices. By the end of 1985, the price gap began to narrow. THE PLAUSIBILITY OF PREDATION, AND PRICING BELOW COSTS The theory of predation (and of Liggett’s case) requires that prices charged for and profits earned on the sale of branded cigarettes be above competi- tive levels. In the absence of noncompetitive prices and profits, B&W would have had no interest or motivation to target Liggett to prevent ex- panison of the generic category. If profits were not above competitive lev- els, industry firms could not have hoped to recoup any investments in predatory behavior. In the cigarette industry, characterized by high market concentration but not single-firm market dominance, such recoupment re- quires that the firms successfully coordinate behavior to raise prices above competitive levels and to avoid destabilizing forms of nonprice competi- tion that threaten the maximization of joint profits. Branded Cigarette Prices and Profits In addition to statements in B&W documents and testimony by both B&W and Liggett executives that cigarette industry profits were greater than in most other manufacturing industries, Liggett’s analysis focused on several categories of economic evidence, each of which is addressed vey briefly below. High Market Concentration Firms in highly concentrated markets recognize their mutual interde- pendence and may tacitly coordinate their behavior to maintain prices and profits above competitive levels. While the existence of high concentration does not prove or establish that firms in the industry successfully maintain prices above competitive levels, such concentration increases the proba- bility that firms will be able to succeed in maintaining these prices. In fact, market concentration in the manufacture and sale of cigarettes was and re- mains very high. Four firms sold more than 88 percent of all cigarettes in the United States, and only a handful of consumer product markets exhibit higher levels of concentration. List Price Leadership In highly concentrated markets, successful efforts to avoid competi- tion may require firms to adopt forms of business conduct that maintain 249 Case 10: The Liggett Case (1993) prices above competitive levels in the face of changed supply and demand conditions. Price leadership, in which one firm announces a price increase and others follow the price leader, is one such mechanism. A strong pattern of price leadership has been documented in the cigarette industry since the 1930s.9 In recent years, the manufacturers’ pattern was one of leader- follower, in which one of the larger firms would announce a price increase that, within several weeks, other manufacturers would follow. The pat- tern was not perfect, but it was highly regular. In the ten years preceding 1989, price increases usually occurred twice a year, and list prices for all branded, full-revenue products have been virtually identical. Price Changes Inconsistent with Cost and Demand Changes In both competitive markets and in monopolies where the firm or firms maximize short-run profits, prices generally will move with and re- flect changes in costs and demand. For example, when costs rise, firms will generally increase prices, but by less than the increase in costs (re- flecting the magnitude of the elasticity of demand), and profit margins fall. If prices and margins do not reflect cost and demand changes, there is rea- son to believe that the firms have discretion over the prices charged and that effective competition does not prevail. There were several instances when cigarette price changes did not reflect cost or demand shifts. In 1982, for example, when the FET on cigarettes increased by $1.60 per carton, prices (and profits) actually increased by more than the tax increase. Simi- larly, throughout the 1974–1988 period generally, while B&W’s share and sales were declining, and while cigarette sales in general were falling, B&W’s prices and profit margins rose significantly. In a competitive market, declining demand (for both the firms and the market) and rising costs would have placed downward pressure on margins and profits earned by B&W. Moreover, even in single-firm monopolies or oli- gopoly settings where the firms jointly maximized short-run profits, margins would have fallen, not risen. Yet in the cigarette industry, prices, margins, and profits all rose. It is possible that, when faced with declining demand and in- creasing costs, the cigarette companies discovered that, as they raised prices, quantity demanded did not fall by as much as had been anticipated; that is, the realization that the elasticity of demand was even lower than previously be- lieved led the firms to raise prices even further above competitive levels than in the past. The ability to raise prices in the face of rising costs and declining demand is not consistent with the existence of effective competition. Profits Above Competitive Levels If firms in an industry successfully engage in tacit collusion, maintain- ing prices above competitive levels, then properly measured and evaluated 9 See Tennant (1950) and Nichols (1949). Patterns of price leadership are often very complex and difficult to detect. 250 THE ANTITRUST REVOLUTION TABLE 10-2 Brown & Williamson Generic Cigarette Revenues and Variable Costs, July 1984–December 1985 18 months: July 1984–December 1985 (per carton)a Gross paid sales including excise tax $3.802 Less excise tax 1.600 Net sales $2.202 Less rebates: Trade rebates $0.750 DAIP rebates 0.043 Interest credit on rebates (0.014) Total: Rebates $0.779 Net revenues per carton $1.422 Less variable costs: Variable manufacturing cost $1.039 Leaf LIFO expense (adjusts leaf to current cost) (0.010) SUI revaluation 0.020 Freight, cartage, and insurance 0.071 Total: Manufacturing cost and freight $1.119 Promotional costs $0.028 Direct selling costs 0.017 Displays 0.066 Retailer payments 0.181 Generic leverage program 0.030 Total: Selling costs $0.322 Variable overhead $0.100 Carrying costs $0.140 Manufacturing variances $0.040 Total: Overhead, carrying costs and variances $0.280 Average variable costs per carton $1.721 Excess of average variable costs per carton over net revenues per carton (loss) ($0.298) aNumbers do not add due to rounding. Source: Plaintiff’s exhibit 3967A. purchased during the year and were paid only if the customer bought B&W generic cigarettes. The DAIP rebates were accrued throughout the year but paid annually. The “interest credit” entry in Table 10-2 reflects the interest B&W earned on the rebate monies before rebate payments were made to customers. Table 10-2 also reflects assessment of the following costs of produc- tion and distribution. Variable manufacturing costs included the value of leaf tobacco, cigarette paper, flavorings, packaging, and factory labor cost used to produce generics. “Leaf LIFO expense” is an inventory valuation 253 Case 10: The Liggett Case (1993) entry that adjusted the cost of raw materials from their acquisition expense to replacement or economic cost. The entry “SUI revaluation” was a techni- cal inventory revaluation implemented in spring 1995 subsequent to a for- mula change. Freight, cartage, and insurance reflected the cost of outbound freight to move cigarettes from the manufacturing facility to customers. There were five categories of selling costs. Promotional costs were associated with selling aids such as “battle kits” that were used by sales- people in making presentations to customers. Direct selling costs reflected the salaries and related expenses of six regional sales managers who had no selling responsibilities except for generics. Display expenses reflected the cost to B&W of the racks that hold cigarettes in retail outlets. Retailer payments were costs associated with payments to retailers made in the form of off-price “stickers” and coupons affixed to generic packages and redeemed by the retailer at the checkout counter. The generic leverage program was a promotional program involving both branded and generic cigarettes that was instituted in November and December 1984. If a customer purchased one carton of generic cigarettes over and above his or her average weekly purchase, he or she would re- ceive two $1.50 stickers that would be placed on B&W branded products. The customer thus received stickers for branded cigarettes, but only if the customer bought a larger quantity of generics. Sales of both branded and generic cigarettes benefited from this program. Since three cartons of ciga- rettes were involved in each transaction (an increase of purchases of one generic carton yielded stickers for two branded cartons), one-third of the cost of the program was allocated to generics. B&W also calculated a category of manufacturing costs termed “manufacturing overheads,” which included both fixed and variable com- ponents. Variable costs in manufacturing overheads included such ex- penses as the hourly wages of a shift supervisor, maintenance expense (a portion of which is dependent on the rate of operation of a machine), and utilities that vary with output. All manufacturing activities involve carrying costs. Money is typically expended before a product is sold, including investments in working capital and accounts receivable that are implicit but real costs of doing business. Most firms do not calculate an interest-carrying charge on the funds tied up in their business activities because, in part, it is not particularly large. For tobacco manufacturers, however, such costs are large because leaf tobacco is a significant component of cost and is usually aged in inventory for a long period (two years or more). Brown & Williamson specifically calculated an inventory “carrying cost” that reflected the implicit interest costs of holding inventories of leaf tobacco and other inputs, as well as finished goods. Fi- nally, Brown & Wiliamson’s data for manufacturing costs are “budgeted” or “standard costs.” Those estimated costs must be adjusted, or “brought back to actual,” by calculating manufacturing variances (reflecting the difference between budgeted and actual costs). 254 THE ANTITRUST REVOLUTION During the eighteen-month period from July 1984 to December 1985, B&W sold 50.1 million cartons of generic cigarettes at an average price of $3.80 per carton, but realized net revenues of only about $1.42 per carton after deducting taxes and rebates. The firm’s estimated variable costs were about $1.72 per carton. B&W priced its generics below variable cost by a total of $14.96 million, or $0.298 per carton. Liggett’s calculations showed that B&W priced below variable cost in every month from July 1984 to December 1985 by amounts ranging from $0.081 to $0.813 per carton. Moreover, the amount by which B&W priced below cost remained substantial throughout the period; in December 1985 it priced below vari- able cost by more than $0.34 per carton, equaling or exceeding its losses in any other month in 1985. B&W’s costs were evaluated in detail for the entire period for which data were available. For some new products, firms lose money during in- troductory periods on such outlays as advertising and promotion. Such losses may have nothing to do with predation. Why, then, was it appropri- ate to evaluate only the first eighteen months of B&W’s generic prices and costs? The first reason is highly pragmatic. Responding to B&W’s argu- ments that producing additional data was both burdensome and unneces- sary, the trial judge limited production of most data to the period ending in December 1985. Limiting attention to this period, however, is appropriate for a product such as generic cigarettes. With regard to many new prod- ucts, the costs that cause losses are those associated with spending on in- tangible assets, such as advertising, that are believed to have long-lived ef- fects; that is, they are expected to generate revenues in future periods. For generic cigarettes, spending on intangibles was insignificant. There was effectively no advertising. Most of the introductory “costs” were really price cuts—either in the form of rebates to distributors or stickers and coupons to consumers. Price cuts are not the sort of costs typically treated as long-lived investments. Moreover, B&W documents specifically stated that none of its introductory expenses was an investment. BROWN & WILLIAMSON’S ANALYSIS AND DEFENSES B&W argued that Ligget had failed to prove several important aspects of its case. Their key objections are summarized below. Evidence of Supracompetitive Pricing B&W witnesses either contested the significance or disputed the suffi- ciency of Liggett’s evidence regarding supracompetitive pricing and prof- its. B&W did not contest the fact that industry concentration was quite high, but observed (correctly) that its mere existence did not prove that 255 Case 10: The Liggett Case (1993) Moreover, as demonstrated by Liggett’s own data and analysis, B&W’s overall cigarette operations were highly profitable; B&W had not priced below variable costs for all cigarettes, and therefore had not violated the Areeda-Turner standard. B&W also argued that applying the Areeda- Turner test only to generics ignored the existence of “branded generic” products—advertised name-brand products sold at generic-level prices. According to B&W, had its predatory campaign succeeded in inducing Liggett to raise generic prices, all that consumers of generics would have needed to do was to switch to the similarly low-priced branded generics, which had not been the target of predatory pricing. Again, according to B&W, Liggett’s attempt to assert that the mar- ket encompassed all cigarettes while evaluating prices and costs only for generics confused issues for yet another important reason. If all cigarettes was the proper market, then Liggett had a competitive option that would have allowed it to avoid losses on generics while continuing to sell ciga- rettes. Because all cigarettes are in the same market, which implies that the products are substitutable, B&W claimed that all Liggett need have done when B&W began to sell below cost was to substitute the sale of other cigarette products in other categories for its sales of generics. B&W sug- gested that Liggett could have abandoned the generic category, let B&W “bathe in its own blood,” and then reentered the generic category when B&W ceased offering to sell at such low prices. B&W’s presentation relied largely on the textbook dominant-firm model of predation. Liggett, however, never asserted that each or all of those features applied to the cigarette industry or to B&W’s conduct in generics. Moreover, Liggett’s analysis specifically explained how and why it differed from the dominant-firm model yet still embodied the necessary elements of a predation model. At the most fundamental level, a predatory firm targets another equally efficient firm in order to raise consumer prices in the long run. Thus, the proper focus should be on pricing that has the potential or actual effect of allowing the aggressor firm(s) to raise prices in the long run. In markets where products are similar or homogeneous, the Areeda-Turner standard would generally apply to all products. B&W’s as- sertion that it was improper to apply the price/variable cost test only to generics was inconsistent with the fact that cigarettes are differentiated products and that changing prices or promotional activities in one category or segment of the market might well have effects on others, although those effects would likely be muted or slow to occur. Just as competitive moves in menthols would yield delayed but pronounced effects on nonmenthols, competitive actions in generics did have material effects on branded prod- ucts, but they did not occur immediately. Cutting prices on generics had isolated short-term effects on Liggett that were not rapidly transmitted to other segments of the industry. Price cutting, however, did induce Liggett to change its pricing behavior and over time resulted in a reduction of the branded-generic price gap. 258 THE ANTITRUST REVOLUTION Recognizing that cigarettes are differentiated products highlights the relevance of applying the Areeda-Turner test only to generics. Because of the differentiation of cigarette products, Liggett’s generic sales could be tar- geted. The effects of charging below-cost prices via rebates on generics could be limited to this one category, without having those low prices spread to and affect prices and sales of branded products. Because Liggett’s generic sales could be targeted, and that behavior had the potential to force Liggett to raise the price of low-priced generics, it was appropriate to apply the Areeda-Turner test only to that category. Moreover, recognizing that cigarettes are differentiated products also highlights the limits of B&W’s “bathe in its own blood” analysis. Had Liggett been able to move to another segment of the market to avoid B&W’s pricing campaign, it might never have introduced generics. Liggett had been unsuccessful in introducing new branded products for many years. There is no reason to believe that, having been targeted in generics, it was any more likely to succeed now than in the past. Although B&W’s assertion that Liggett could have avoided the effects of predatory pricing might make sense in a market where product differentiation was less important, the premise was implausible in the cigarette industry. Finally, B&W argued that Liggett’s claims were illogical because, if Liggett were driven from generics, consumers could simply purchase the widely available, low-priced, branded generic products sold by several of the large producers, and prices would not be affected. In the short run, branded generics did provide a ready alternative to Liggett’s “real,” un- branded product. Liggett’s analysis, however, was that the larger manufac- turers would have been better able to raise the price of branded generics than would be possible with generics alone; that is, once Liggett was disci- plined and had raised prices, branded generic prices would simply be “managed” up. This is, in fact, what occurred. By mid-1989, the price gap between generics, branded generics, and full-revenue, branded products had been reduced to about 27 percent. OUTCOME OF THE CASE After 115 days of hearings and 10 days of deliberations, the jury returned a verdict in Liggett’s favor, finding that B&W possessed market power and had engaged in below-cost pricing that likely harmed Liggett and competi- tion. Despite having agreed to send the case to the jury in the first place, the trial judge, relying on evidence and analysis available to him before the case was sent to the jury, reversed the decision and dismissed the case. The judge’s ruling did not dispute the potential for recoupment by a firm with a 12-percent share in a concentrated oligopoly. He did, however, criticize the economic evidence adduced by Liggett to establish that prices were set at supracompetitive levels, in part because that evidence and data 259 Case 10: The Liggett Case (1993) were inconsistent with testimony by Liggett executives, who had denied both that the industry was characterized by tacit collusion and that the firms in the industry earned excessive profits.17 Moreover, the judge be- lieved that the availability of branded generics at prices lower than full- revenue brands negated any effect that B&W’s actions may have had on Liggett.18 Liggett appealed to the Fourth Circuit Court of Appeals, which af- firmed the judge’s dismissal of the case, but for reasons different from those advanced by the District Court. The Appeals Court decision asserted that only an actual monopolist or explicit cartel rationally would engage in predation because a single firm could never be certain that its rivals would not mistake the disciplining, below-cost pricing for aggressive competitive behavior.19 Liggett appealed to the Supreme Court in November 1992, and the Court agreed to hear the case. In June 1993, by a 6-to-3 vote, the Supreme Court dismissed the case.20 In so doing, however, the Court re- jected the certainty standard enunciated by the Court of Appeals and relied on analysis similar to that of the District Court. Thus, despite finding that (1) the cigarette industry was one of the most profitable in the United States; (2) B&W documents provided a sound basis for concluding that B&W entered the generic category with the intent to engage in anticom- petitive behavior; and (3) there was evidence of below-cost pricing on generic cigarettes, the Supreme Court determined that B&W (contrary to the firm’s own beliefs) could not rationally expect to recoup any of its al- leged predatory investments. The Supreme Court erred (in this author’s view) at several key points in its analysis. After correctly stating the proposition that Liggett’s case re- lied on recoupment by B&W of its predatory investment in below-cost pricing on generics (thereby protecting high-profit, branded cigarette sales), the Court asserted that a necessary element of Liggett’s case was that there be recoupment through supracompetitive, coordinated pricing on low-priced cigarettes. This was not so. What was necessary was that Liggett have a greater incentive to maintain a lower price for generics than 17 The judge had before him the testimony both of Liggett executives and of Liggett’s expert economist on at least three occasions prior to this ruling: at a motion for summary judgment, at the time of a motion to dismiss at the end of Liggett’s case, and at the conclusion of the entire trial, before the case went to the jury. In each instance, the judge allowed the case to proceed, and ultimately to reach the jury, before reversing field. Moreover, the questions and testimony on which he based this ruling are highly suspect. Liggett executives were asked questions embodying terms with which they, as businessmen, were unfamiliar, or they interpreted those terms differ- ently from economists. For example, while an economist uses the term “tacit collusion” to reflect the ability of firms to raise prices without overt collusion, business executives focus on the “collu- sion” portion of the phrase and deny that they have engaged in such illegal, “collusive” behavior. 18 For the judge’s complete ruling, see Liggett Group, Inc. v. Brown & Williamson Tobacco Cor- poration, 748 F. Supp. 344 (M.D.N.C. 1990). 19 Liggett Group, Inc. v. Brown & Williamson Tobacco Corporation, 964 F.2d 335 (4th Cir. 1992). 20 Brooke Groupe Ltd. v. Brown & Williamson Tobacco Corporation, 61 U.S.L.W. 4699 (1993). 260 THE ANTITRUST REVOLUTION