In the Matter of

General Mills, Inc.

Docket No. C-3742

I respectfully dissent from the decision of the majority to issue a consent order against General Mills, Inc. relating to the acquisition of the branded ready-to-eat ("RTE") cereal and snack food businesses of Ralcorp Holdings, Inc. ("Ralcorp"). My dissent rests on two grounds.

As noted in the Commission's complaint, General Mills will not acquire the private label RTE cereal or snack food businesses of Ralcorp. Ralcorp instead will form a new entity, New Ralcorp Holdings, Inc. ("New Ralcorp"), to hold the private label cereal and snack food businesses that General Mills will not acquire. Under the acquisition agreement, New Ralcorp has the right to manufacture and sell a private label version of the Chex RTE cereal products, but is restricted from transferring this right to a third party without permission from General Mills. The acquisition agreement further provides that New Ralcorp may not produce private label Chex products for a period of eighteen months following consummation of the acquisition.

My first reason for voting against issuing the consent order is that the Commission lacks sufficient evidence to support the unilateral effects theory alleged in the complaint. Second, it is completely unnecessary -- and in fact creates inefficiency -- to bar enforcement of the parties' non-compete agreement. Whatever minimal competitive risks this transaction may raise are adequately addressed by eliminating the restrictions on Ralcorp's ability to transfer manufacturing and sales rights for private label Chex to a third party.

General Mills' share of the RTE cereal market will increase by approximately three percent as a result of the acquisition. The number of competitors in the RTE cereal industry will remain the same, and General Mills will remain the second largest RTE cereal producer in the United States.(1) New Ralcorp will immediately assume Ralcorp's position as the largest private label cereal producer in the United States. Moreover, General Mills' post-merger share of the RTE cereal market will be between 25 and 31 percent (depending on whether share is measured in pounds or sales dollars), well below levels suggested by the Horizontal Merger Guidelines as the minimum threshold at which the Commission might reasonably presume market power.(2) It is hard to understand under these simple facts how the majority determined that the acquisition will enable General Mills unilaterally to exercise market power.

Unable to presume market power, the Commission instead relies upon a "close substitutes" theory of unilateral harm, notwithstanding a paucity of empirical evidence demonstrating that Ralcorp's branded Chex products are the closest substitutes to the branded cereals of General Mills. Although Chex products clearly compete with the branded General Mills RTE cereal products, consumers have a preference for variety when they choose RTE cereals and frequently choose among the many branded and private label cereals produced by RTE cereal manufacturers in the United States. Not surprisingly, Judge Wood reached this conclusion in her opinion explaining why she refused to block the acquisition of the Nabisco RTE cereal assets by Kraft General Foods in early 1993.(3) In Kraft General Foods, an empirical analysis of cereal purchasing patterns suggested -- as it does in the present matter -- that consumers have many attractive alternatives from which to choose in the event that one RTE cereal producer tries to raise prices above competitive levels. Overall, the empirical evidence does not support the Commission's claim, under either a "close substitutes" or a dominant firm theory, that General Mills would be able unilaterally to raise the prices of its branded RTE cereals after the acquisition.

Even if I agreed with the majority that this consent order rests upon an empirically sound theory of competitive harm, the order would bar General Mills from enforcing an arguably procompetitive non-compete agreement that is properly limited in scope and duration. Covenants not to compete are often included in contracts for the sale of a business, and generally are enforceable when ancillary to an enforceable agreement and reasonable in geographic coverage, scope of activity, and duration. Lektro-Vend Corp. v. Vendo Co., 660 F.2d 255, 265 (7th Cir. 1981) ("The recognized benefits of reasonably enforced non-competition covenants are now beyond question."), cert. denied, 455 U.S. 921 (1982); United States v. Addyston Pipe & Steel Co., 85 F. 271, 281-82 (6th Cir. 1898), aff'd as modified, 175 U.S. 211 (1899).(4) Judicial inquiry into non-compete provisions generally focuses on whether the restriction is reasonably necessary to protect the legitimate business interests of the party seeking to enforce the provision. United States v. Empire Gas Corp., 537 F.2d 296, 307 (8th Cir. 1976), cert. denied, 429 U.S. 1122 (1977); Sound Ship Bldg. Corp. v. Bethlehem Steel Corp., 387 F. Supp. 252, 255 (D.N.J. 1975), aff'd, 533 F.2d 96 (3d Cir.), cert. denied, 429 U.S. 680 (1976).

The Commission has often recognized that competitive benefits can flow from a non-compete clause in the context of the sale of a business. The Commission's recent issuance of a consent order in Ciba-Geigy, Ltd., et al., Docket No. C-3725 (April 8, 1997), is illustrative. In Ciba-Geigy, the Commission imposed an affirmative obligation on the newly merged entity, Novartis AG, not to compete in the United States and Canada for six years in the sale of animal flea control products.(5) As the Ciba-Geigy order indicates, the Commission clearly recognizes that non-compete clauses -- even when long in duration and broad in scope -- can serve legitimate procompetitive purposes in some circumstances by allowing an acquiring entity a brief period to re-deploy the acquired assets in a manner that increases competition in the marketplace. I am therefore puzzled why the Commission so hastily condemns a non-compete provision here that is only eighteen months in duration, limited to the manufacture and sale of private label Chex products, and arguably necessary to protect the legitimate interests of the contracting parties.(6)

Because I find that the facts do not support the Commission's theory of unilateral competitive harm in this instance, and because in any event I disagree with the Commission's decision to bar enforcement of the non-compete provision contained in the parties' acquisition agreement, I have voted against issuance of the consent order.

1. General Mills' share of branded cereals will of course increase as a result of the transaction, but the complaint does not allege a relevant market consisting of "branded RTE cereal." Indeed, the provisions of the order (which affect the disposition of assets used in the production of nonbranded cereals) make sense only in the context of an "all RTE cereal" product market.

2. See U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines  2.211, 4 Trade Reg. Rep. (CCH)  13,104, at 20,573-79.

3. State of New York v. Kraft General Foods, Inc., 1995-1 Trade Cas. (CCH)  70,911, at 74,039, 74,066 (S.D.N.Y. 1995).

4. See also Business Elecs. Corp. v. Sharp Elecs. Corp., 485 U.S. 717, 729 n.3 ("The classic 'ancillary' restraint is an agreement by the seller of a business not to compete within the market.").

5. See Paragraph VI of the order in Ciba-Geigy.

6. Barring enforcement of the non-compete agreement might undermine adherence by the parties to the supply agreement, an element of the acquisition agreement found acceptable by the majority.