Wisсonsin enacted a statute protecting dealers from termination. Wisconsin Fair Dealership Law, Wis.Stat. ch. 135. Because many suppliers and franchisors have headquarters outside Wisconsin, litigation frequently meets the requirements of the diversity jurisdiction. Franchisors apparently think the federal courts more attuned to their interests, while dealers prefer state courts. District judges in Wisconsin have been overrun by suits originally filed in state court and removed to federal court. This is one, filed by a liquor wholesaler that lost its right to distribute Jose Cuervo *419 brand tequila sold by Heublein, a subsidiary of Grand Metropolitan PLC.
Heublein sells many brands of liquor; the only one Kenosha Liquor Company carries is Jose Cuervo tequila. In 1983 Jose Cuervo accounted for 3.2% of Kenosha Liquor’s sales; by 1988, when Heublein tried to turn off the sрigot, it accounted for 5.8%. This percentage is the focus of the dispute. The Fair Dealership Law applies only to “dealers”, a term it does not dеfine except by saying that a dealer is a distributor in a “community of interest” with the supplier, § 135.02(2), (3), which “pushes the lack of a definition to a new level of abstraсtion.”
Fleet Wholesale Supply Co. v. Remington Arms Co.,
Multi-factor tests could imply jury trials in аll cases, but no one wants (or can believe the state legislature created) a vapid law, uncertain in every application. To say that courts must examine many facts is not to say that there are no rules. We have deduced from the structure and history of the statute a central function: preventing suppliers from behaving opportunistically once franchisees or other dealers have sunk substantial resources into tailoring their business around, and рromoting, a brand. See
Moodie v. School Book Fairs, Inc.,
Heublein moved for summary judgment, observing that 5.8% of sales is not enough to allow it to drive the buyer to the brink or wring concessions (at least, no concessions that could not equally be wrung by raising the price of- tequila), and that Kenosha Liquor has made absolutely no Jоse Cuervo-specific investments (that is, investments in equipment and reputation that are less useful for other brands than for Jose Cuervo). Kenosha Liquor defended the motion by relying on Steve L. Barsby, an economist who maintained that Jose Cuervo is one of Kenosha Liquor’s “magnet brands” essential to its business. The distributor also pointed to its contract with Heublein, which entitled Heublein to demand that Kenosha Liquor make some product-specific investments. To this Heublein replied that it had never made the kinds of demands authorized by the contract, which Kenosha Liquor does not deny.
Judge Evans agreed with Heublein that undisputed material facts call for summary judgment on the “dealership” question. On his own motion, he awarded victory to Kenosha Liquor. After Kenosha Liquor waived any request for damages, Heublein stipulated to the entry of an injunction forbidding the cutoff. Judge Evans entered the injunction and awarded more than $61,000 in legal fees and costs. Heu-blein’s appеal presents a single question: whether Kenosha Liquor is a “dealer” as a matter of law under § 135.02(3). Like the district court, we believe that this case can be rеsolved on the current record.
Jose Cuervo yields less than 6% of Ke-nosha Liquor’s sales. Although Ziegler implies that 8% of sales (the amount in Ziegler itself) could be enough if other factors suggest “dealership”, therе are no (pertinent) other factors here. Kenosha Liquor distributes Jose Cuervo through its regular means. Its premises, trucks, and so on all bear its own markings; it has not pointed to a single business asset that is not useful in distributing Remy Martin cognac to the same degree it may be used in handling Jose Cuervo tequila. No matter what the contrаct said Heublein could do to tie Ke-nosha Liquor’s hands, Heublein did not. So it was in no position to exploit a sunk investment, to force Kenosha Liquor to pay (indirectly) mоre than the market price of the tequila that a new distributor, who had not yet had any dealings with Heublein, would be willing to pay.
*420
Kenosha Liquor is reduced to relying on Barsby’s view that Jose Cuervo is a “magnet brand”. About 90% of all tequila Ke-nosha Liquor sells is Jose Cuervo brand. The brand gets Kenosha Liquor’s salesmen in the door, according to Barsby; once there, they offer and sell its full line of liquor. Economies of scope — savings from carrying multiple brands — may be real, but they do not give Heublein any power to exploit Kenosha Liquor's investment. If Jose Cuervo opens doors, then
Heublein
will collect the full value of that effect; all it has to do is set a рrice that represents the extra sales of rum and gin attributable to a knockout tequila. Having collected in the price the magnetic value of its brand, Heublein can’t use the attractiveness of Jose Cuervo tequila a second time to exploit sunk investments. And if Heublein irrationally charges “too little” for this Gоliath of tequilas, it cannot recoup by cutting off its distributor. It can reap gain at distributors’ expense only by raising the price — which it did not do, and which Wisconsin at any rate allows (provided it raises the price to all). Cf.
Goldberg v. Household Bank, f.s.b.,
The most one can say for “magnet brands” (if they exist) is that losing Jose Cuervo will cost Kenosha Liquor more thаn the 5.8% of sales Jose Cuervo represents. How much more? The expert opined that Kenosha Liquor carried 12 magnet brands. If all of its sales are attributablе to one of these brands, and if they are equally important, then losing one will cost Kenosha Liquor 8% of its sales. That is still too small, when the firm has no assets dedicated to serving the brand in question.
Even if Barsby said that losing Jose Cuervo would cost Kenosha Liquor 10%, 20%, 50%, even 99.44% of its business, this would be unimportant. For he presented no foundatiоn for his conclusion. Barsby did not report the results of a statistical study of the effect of losing certain brands of liquor. From 1967-72 Barsby (who holds a Ph.D. in economics from the University of Oregon) was an assistant professor at the University of Arizona; since 1972 he has been in private industry and consulting. So he must have professional skills, but he did not use them. The only fact he presented is that Jose Cuervo accounts for 90% of Kenosha Liquor’s tequila sales. This does not imply anything about Kenosha Liquor’s status as a “dealer” unless connected to a probable consequence of termination. The expert offered no summary of similar terminations of other dealers, no data about consequences of product cut-offs outside the liquor business.
He suggested instead that the loss of less than 6% of its sales would deprive Kenosha Liquor of some 35% of its profits. Barsby arrived at this conclusion by subtracting from Kenosha Liquor’s sales of Jose Cuervo tequila in 1987 ($174,636) the price it paid Heublеin ($144,728), and dividing the difference ($29,908), less a few direct costs such as salesmen’s commissions, by Kenosha Liquor’s stated net profits ($67,586). The effect depends on assuming that the brаnd in question is not responsible for the rent, salaries, and other variable costs — variable, at least, in the not-so-long run. Barsby did not suggest that the same method applied to any other 6% of Kenosha Liquor’s sales would not produce the same appearance. Accounting profits in a closely held cоrporation such as Ke-nosha Liquor are not necessarily useful figures; the investors’ returns will be paid out as salaries to reduce the corporate tax. At all events, any method that attributes 40% of the profits to 6% of the sales — any 6% and every 6% — is unhelpful in identifying a dealer.
Opinions are admissible under Fed.R. Evid. 702, but admissibility does not imply utility. See
Mid-State Fertilizer Co. v. Exchange National Bank,
Reversed.