L. Knife & Son, Inc. v. Banfi Products Corp.

118 F.R.D. 269 | D. Mass. | 1987

MEMORANDUM AND ORDER ON DISCOVERY

GARRITY, District Judge.

The plaintiff, L. Knife & Son, Inc. (“Knife”),¹ is a sizeable distributor of alcoholic beverages. Beginning in 1979, it filed several complaints, consolidated in this action, against its supplier of Riunite wine, Banfi Products Corporation (“Banfi”). Knife alleges violations of Massachusetts common law, the state’s unfair trade practices statute, and federal antitrust law. The chief claims are that Banfi breached its distribution agreement with Knife by unilaterally curtailing its deliveries of wine and that Banfi, during the same period, adopted practices, e.g., unlawful “tying arrangements” and price discrimination, prohibited by antitrust law. In an effort to advance the litigation, impeded at times by the complexity of the case and zeal of respective counsel, a Special Master selected by the parties was appointed by the court pursuant to Rule 53, Fed.R.Civ.P., and empowered to recommend rulings on discovery-related disputes. With his assistance, the parties have made sufficient progress that discovery will be wound down by the end of this month and, barring unforeseen events, trial will begin on May 5, 1987 pursuant to the court’s pretrial orders.

Now before the court is an objection filed by Banfi to certain discovery sought by Knife in connection with its claim of unlawful tying. The dispute centers on interrogatory no. 4(c) through 4(g) of plaintiff’s Third Set of Interrogatories in Civil Action No. 82-0741 and interrogatory no. 9 in plaintiff’s Fourth Set of Interrogatories and Fourth Request for Production in the same action. The interrogatories and document requests probe Banfi’s “overall costs” and “desired profit margin” for Ri-unite wine. Specifically, Knife seeks detailed figures showing the cost to Banfi of wine, freight, and all other expenses that go into the production of Riunite wine. The data is sought allegedly in order to compute damages on its claim that defendant unlawfully tied the purchase of Riunite wine, the tying product, to the purchase of transatlantic freight plus insurance (“freight”), the tied product.²

Banfi first objected to the discovery before the Special Master. After the Special Master overruled the objection and ordered responses to the interrogatories and document requests, Banfi renewed its objection in this court. For the reasons set forth post, the court sustains the defendant’s objection to the discovery sought by plaintiff concerning defendant’s costs and desired profit margins.

The conflicting positions of the parties were first aired and considered at oral argument. Since these initial discussions, the court has deemed the discovery in question to be quite sweeping and not easily contained, thus creating a danger of losing valuable time left for discovery, sidetracking the presentation of evidence at trial, and confusing the issues eventually presented to the jury. The matter was taken under advisement in the hope of finding some manner of reducing its breadth. Subsequently the court, together with the parties, endeavored to narrow or possibly resolve the dispute in the course of several pretrial hearings. The court also issued two procedural orders to bring about a resolution: one directed defendant to produce evidence of the wine’s fair market value independent of the alleged tie-in to freight; the second, issued after defendant’s submission of a pertinent commercial document, directed the parties to confer in an effort to devise a stipulation for purposes of computing damages. More recently, the parties filed letters with the court reiterating their disagreement yet expressing a willingness to continue their pursuit of a compromise. While the written submissions and discussions have illuminated the important issues, they have failed to bring the parties closer to an agreement on the scope of this aspect of discovery. At the same time, the process has reinforced the court’s view that plaintiffs requested discovery is unwarranted.

The problems with Knife’s requested discovery can be understood only in light of its method of calculating damages. Assuming, for the sake of argument, that an unlawful tying arrangement was imposed by Banfi, Knife would compute the resulting damages roughly as follows: (1) extract separate costs strictly attributable to wine and freight from Banfi’s overall costs to produce a bottle of Riunite wine; (2) multiply these costs by Banfi’s profit margin(s) to yield the portion of the total package price attributable to wine and freight; (3) subtract the proper market price of freight from the price of freight constructed in steps (1) and (2), with the difference (i.e., overcharge) amounting to Knife’s damages from the alleged tying arrangement.

While Knife’s “allocation of costs” method has superficial appeal, it suffers from serious shortcomings that are evident on closer scrutiny. First, it presupposes that Banfi’s costs of producing a bottle of wine can neatly be broken down into discrete elements. In fact, there are innumerable costs comprising Banfi’s overall costs, and they cannot simply or readily be distinguished and then isolated, e.g., rent, electricity, local transportation, salaries. It is likewise improbable that defendant can single out each of its manifold costs and then apply a predetermined mark-up to each item. Freight is merely one among all the costs going into the overall cost of producing, transporting, and selling a bottle of Riunite wine; and it is from aggregate costs and market analysis that a percentage mark-up can be developed. Knife’s method of breaking up and then allocating costs and price is purely artificial.

Moreover, calculating an overcharge on freight is insufficient by itself to prove damages on a tying claim, as Knife has already conceded. To compute damages properly, Knife would also have to compare the part of the package price attributable solely to the tying product, wine, with its fair market value. This would be necessary to see whether the price attributable to the wine alone had been lowered to offset a higher price that might have been charged for the tied product. See Kypta v. McDonald’s Corp., 11 Cir.1982, 671 F.2d 1282, 1285.³ Even if we assume that the cost and price attributable to wine can accurately be calculated—quite difficult in itself—the problem of calculating the wine’s fair market value remains.⁴ Plaintiff brushes the issue aside by stating that “The prices at which Banfi purchased Riun-ite is [sic] relevant to determining the fair market value of the tying product, Riun-ite.” Yet it does not explain the connection between cost and market value or give any indication as to how that value can be determined from the information sought. The market for oil is a good example of the difficulties involved. That oil is very inexpensive to produce in the Middle East has little bearing on its much higher cost and price in the West, as evidenced by its wild fluctuations in the past decade. In the end, plaintiff’s method of proving damages would lead the parties on a long and tortuous path without any hope of reaching the issues critical to proving damages on its tying claim.

Instead, we now rule that the analysis of damages and any discovery sought to illuminate the issue of damages in this case should be directed solely to the question of fair market value of the tying product, wine. Under the court’s method, a computation of damages would proceed as follows: (1) establish fair market values for wine and freight separately and add them together to make up a fair market price of the products, as if purchased separately; (2) subtract that total fair market price from the actual price for the package of wine and freight charged to plaintiff, with the difference constituting the overcharge, and damages, due to the alleged tying arrangement.⁵ This method, to be sure, is subject to refinement and leaves open the central question: namely, how to establish the wine’s fair market value. But it has the distinct advantages of directness and logic. The “market value” approach is addressed to what the court regards as the legally central, and intuitively correct, question: did plaintiff pay more for the goods in a package than it would have paid had it been able to buy the goods separately, in separate markets? ⁶

There are several more specific reasons to reject plaintiff’s “allocation of costs” method in favor of some sort of “market value” approach. The first one is the weight of caselaw. The leading decision on damages in a tying case remains Siegel v. Chicken Delight, Inc., 9 Cir.1971, 448 F.2d 43, in which the Court of Appeals for the Ninth Circuit held that a franchisor had engaged in an unlawful tying arrangement yet remanded the case for a proper computation of damages.

The question here is whether the plaintiffs have suffered injury by virtue of the unlawful arrangement____ That arrangement of necessity involved both tying and tied products. To ascertain whether an unlawful arrangement for the sale of products has caused injury to the purchaser, the cost [i.e., price] or value of the products involved, free from the unlawful arrangement, must first be ascertained. Id. at 52.

Subsequent cases have not departed from the Siegel formulation. See Midwestern Waffles, Inc. v. Waffle House, Inc., 11 Cir. 1984, 734 F.2d 705, 719; Kypta, supra, at 1285. In upholding a verdict against a franchisee’s tying claim, former Professor and now Judge Easterbrook wrote that plaintiffs had failed “to show [] that the price they paid for the ... package is higher than the price for those two products purchased in separate markets.” Will v. Comprehensive Accounting Corp., 7 Cir. 1985, 776 F.2d 665, 672. It is the separate prices or values in a market, not a reconstruction of internal pricing policy, that is central to the question of damages in a tying claim.

The second basis for denying plaintiff’s requested discovery is its relative unwieldiness by comparison to the focused inquiry dictated by the “market value” approach. The problem stems from plaintiff’s proposed method of isolating costs and constructing prices in an artificial manner. The approach would require the parties and later the jury to comb through all of defendant’s costs of production and comprehend all of the factors that go into establishing a market price. The process seems almost boundless, and the result would still be a highly imperfect construct.

The “market value” approach, in contrast, would isolate a single, well defined question. Undoubtedly there are uncertainties in the construction of a fair market value for wine. They grow out of the absence of an actual market for wine independent of freight. To aid the parties, the court has proposed various methods of computing an acceptable value from available data: e.g., deriving a value for Riunite wine by comparing, perhaps in a graph, its price relative to comparable brands sold independently of freight; or projecting a value of Riunite wine from its price in 1976, when wine and freight were billed separately, adjusted for inflation and other market changes. While these methods have their flaws and limitations, the court does not view its suggestions as binding. Better methods may be devised by the parties. Moreover, the court does not regard the necessary imprecision, even artificiality, of the market value analysis as somehow precluding its use. In Siegel, supra, the court’s rule compelled the establishment of a market value for a franchise even though it had not been sold in the marketplace as a separate item (i.e., free of the unlawful tying arrangement). Thus, despite the problematic elements of the Siegel method, it is in our opinion superior to the plaintiff’s method, which is even more artificial and exhausting to the extent it would require a complete canvassing and recombination of defendant’s cost and pricing data.

Finally, plaintiff’s method and consequent scope of discovery would greatly complicate the task of the jury. The voluminous record and conduct of the parties in this case suggests that factual and legal issues of some complexity will be contested at trial. The operation of the alcoholic beverages industry will be at issue, as will difficult issues of federal antitrust law, just to name a few. To add another body of material on a separate part of the industry—i.e., wine production in addition to its distribution—would divert the jury’s attention, limited as it is, from the heart of the issues in the case. Undoubtedly, the parties may end up contesting the issue of the wine’s fair market value at trial, and evidence will have to be weighed as to the proper method of computing that value. But most importantly, the evidence will not compel the jury to find their way through another murky area in an already dense and tangled case.

In sum, the court sustains the defendant’s objections to the requested discovery and overrules the Special Master’s ruling. We also adopt as a measure of damages the market value approach as described and justified ante. Yet that is the extent of our rulings at this time. We do not order or preclude other discovery consistent with the pretrial scheduling orders. Nor do we establish the sort of proofs that may be adduced in support of or opposition to a particular calculation of market value and, thereby, of damages. For the reasons stated ante, the court narrowly holds that the discovery sought by plaintiff, premised on its stated method of computing damages, is irrelevant and “unduly burdensome” within the meaning of Rule 26(b)(1) and Rule 26(b)(l)(a)(iii), Federal Rules of Civil Procedure.

Addendum

After first drafting this memorandum of decision, the court received a letter from plaintiff’s counsel in support of a more detailed and somewhat more refined version of its “allocation of costs” method. The analysis, though bolstered by use of cost figures supplied by the defendant, actually revealed the serious flaws of plaintiff’s position. First, it left open the problem of determining the fair market value of wine sold alone, an issue the court regards as significant even if plaintiff’s theory of damages be accepted. Second, plaintiff made a completely arbitrary, if initially appealing, breakdown of costs.

The basic difficulty lay in its haphazard treatment of certain expenses as a “recovery of costs” and other expenses as unspecified “overhead and profit.” In its letter plaintiff computed damages as follows: (1) it added up defendant’s costs of wine in Italy, inland freight, overseas freight, and insurance, calling their sum, insofar as it is part of the overall package price (to plaintiff), a “recovery of costs”; (2) it constructed another category of figures by subtracting these costs from the overall package price to plaintiff, calling the result “overhead and profit”; (3) it computed a percentage of the “recovered” costs attributable to overseas freight, the tied good; (4) it computed a portion of the “overhead and profit” attributable to overseas freight according to this percentage; and (5) it added the cost of overseas freight to defendant and the figure yielded at step (4), regarding the sum as the portion of the overall package price attributable to the tied product. The method was straightforward only in appearance.

By identifying some costs and not others, plaintiff in effect chose the percentage of “recovered” costs attributable to ocean freight (the tied product) and, under its method, the amount of “overhead and profit” tacked onto the cost of the ocean freight. Why such costs as insurance and inland freight, singled out by plaintiff, are not simply part of “overhead and profit” is not clear; why rent, electricity, and the like are not extracted, but left undifferentiated, is also not stated; why “overhead and profit” is even proportionately allocated to overseas freight and wine, though defendant’s costs are connected chiefly to the wine business, similarly is not apparent.

These are only some of the unexplained, and inexplicable, steps taken by plaintiff in applying its methodology. The proposed theory of damages is simply too manipulable and without rational foundation. It is not adopted by the court.

1 . The other plaintiff, Seaboard Products, Inc., is under the legal control of Knife and therefore only a nominal party to this action. Consequently, the court will refer strictly to plaintiff Knife in its discussion.

2 . Tying exists, under antitrust law, when sellers of a particular good (the tying product) condition its sale on the purchase of another, separate good (the tied product). See, e.g., Northern Pacific Railway Co. v. United States, 1958, 356 U.S. 1, 5-6, 78 S.Ct. 514, 518-19, 2 L.Ed.2d 545. At first impression, Knife’s tying claim seems novel. It apparently rests solely on Banfi’s decision, in 1976, to change a policy of billing wine and freight separately to its current practice of charging one undifferentiated package price. Before and after the changed billing, Knife purchased freight as well as wine from Banfi and, according to counsel, did not ask to buy the freight separately until after the lawsuit was filed.

3 . In this respect, caselaw seeks to take account of the interdependent pricing of the tying and tied products, a factor emphasized in economic analyses of tying arrangements. See, e.g., P. Areeda & D. Turner, 5 Antitrust Law ¶ 1134b at 204 (1980); R. Posner, Antitrust Law: An Economic Perspective 173 (1976).

4 . The fair market value of the tied product, freight, does not appear to present a serious problem. Since there is a ready market for freight, the parties need not attempt to construct a hypothetical market value by extrapolating or projecting from imperfect data. The only issue, it seems, is which market price of freight among. the numerous prices available is appropriate in this case.

5 . The court does not rule out the possibility that an excess charge resulting from this calculation may be caused by circumstances other than tying—e.g., improved service. If such an excess exists, defendant will certainly be free at trial to dispute its causal connection to the alleged tying arrangement. In the interest of simplicity, however, the court will assume for purposes of resolving the instant dispute that no other factors exist.

6 . The importance of positing and, even more, identifying separate product markets is now apparent in the substantive law of tying arrangements, see Jefferson Parish Hospital Dist. No. 2 v. Hyde, 1984, 466 U.S. 2, 19-22, 104 S.Ct. 1551, 1561-64, 80 L.Ed.2d 2, as well as in the proper computation of damages. Cf. Will v. Comprehensive Accounting Corp., 7 Cir.1985, 776 F.2d 665, 672-73.