Statement of Commissioner Mary L. Azcuenaga
Concurring in Part and Dissenting in Part
in Cadence Design Systems, Inc., Docket No. C-3761
The acquisition of Cooper & Chyan Technology, Inc. (Cooper & Chyan), by Cadence Design Systems, Inc. (Cadence), combines the only firm currently marketing a constraint-driven, shape-based integrated circuit routing tool with a firm that was, at least until the acquisition, on the verge of entry into this market. I find reason to believe that the proposed merger would violate Section 7 of the Clayton Act under a horizontal, potential competition theory. On this ground, I support the prior notice provision of Paragraph III of the order, which provides a small measure of horizontal relief.(1) I dissent from the allegations in the complaint and the order provisions that address the vertical aspects of the case.
To establish a Section 7 violation based on the actual potential competition theory, the government must show: (1) that the potential entrant "has available feasible means for entering" the relevant market; and (2) that "those means offer a substantial likelihood of ultimately producing deconcentration" of the relevant market. United States v. Marine Bancorporation, 418 U.S. 602, 633 (1974). In addressing the first element, courts have looked to whether a firm has the capacity, interest and economic incentive to enter.(2) The Commission has adopted the view that "clear proof [is required] that independent entry would have occurred but for the merger or acquisition" and has emphasized the importance of concrete investment plans approved by top management and studies done before or contemporaneously with the acquisition demonstrating plans to enter. B.A.T. Industries, 104 F.T.C. 852, 919-20, 926-27 (1984).
It is a close question whether Cadence was a potential entrant or already an entrant in the relevant market.(3) Regardless of the outcome of that question, my review of the confidential file indicates that Cadence's interest and economic incentive to enter the market were clear, even under the strictest legal standard of actual potential competition. In determining whether Cadence had the capacity to enter the relevant market, the Commission should assess the status quo before Cadence agreed to acquire Cooper & Chyan. Claims that the technology had failed made after the parties agreed to the transaction should be discounted because the incentives to justify the transaction are strong. To support a conclusion that "Cadence's efforts to develop such technology had failed"(4) before the Cooper and Chyan transaction, one would expect to have pre-transaction evidence, such as an indication that Cadence had stopped spending money on the project, some efforts by Cadence to dispel any notion that customers may have entertained that they should refrain from buying other products pending the arrival of the Cadence product, or an indication that Cadence's management had included the failure of the product in their business plans. I have not seen such evidence.(5)
Cadence satisfied the criterion of capacity to enter. As the Commission has observed, "capacity to achieve independent entry successfully is always somewhat speculative,"(6) but Cadence was a technological and marketing leader, and it suffered under no apparent impediment to entry. Even if Cadence did not have a "commercially viable" product at the time of the acquisition, the actual potential competition doctrine applies to firms that have not yet perfected a product and completed all the steps necessary to entry.(7)
The second Marine Bancorporation element appears to be satisfied as well. Before the merger, Cooper & Chyan was the only firm selling a constraint driven, shape-based integrated circuit ("IC") router, and entry by Cadence likely would have produced a significant deconcentration of that market.
The vertical theory of violation alleged in the complaint is that the acquisition of Cooper & Chyan by Cadence will make it more difficult for another firm to introduce a constraint driven, shape-based IC router because such an entrant would need its own IC layout environment to enter the market, and that dual level entry is more difficult. Although this is a recognized theory,(8) I question whether it applies in this case and whether a firm needs to enter both the routing and the environment markets simultaneously.
Cooper & Chyan was successful in developing and marketing its routing program before it gained access to Cadence's environment. In a separate statement, Chairman Pitofsky and Commissioners Varney and Steiger assert that Cooper & Chyan's "sales were modest before the merger announcement." I disagree based on Cooper & Chyan's penetration of the market.(9) Cadence's willingness to pay more that $400 million in stock for Cooper & Chyan also suggests a greater competitive significance than the majority concedes.(10) Cooper & Chyan's record indicates that access to a layout environment is not a precondition to successful entry in the market for constraint drive, shape-based integrated circuit routers. It appears, based on the available information, that dual level entry theory does not apply in this market.
In addition, although Cadence initially denied Cooper & Chyan access to its connections program, it subsequently reversed course and granted the access. This suggests that Cadence capitulated to pressure from customers to grant Cooper & Chyan access and that Cadence has little or no power to deny access to its connections program if granting access is the only way to enable its customers to use a product they want to use. Finally, Paragraph II of the order is premised on the allegation in paragraph 16 of the complaint that "Cadence does not, however, have incentives to provide access to a Cadence integrated circuit layout environment to suppliers of integrated circuit layout tools that compete with Cadence products." The incentives appear to be at least as likely to go the other way. If another company develops an innovative, advanced router, one would assume that Cadence would have incentives to welcome the innovative product to its suite of connected design tools, thereby enhancing the suite's utility to customers.
Paragraph II of the order may be counterproductive and may result in substantial enforcement costs for the Commission. Because Paragraph II bars Cadence from charging developers of "Commercial Integrated Circuit Routing Tools" a higher access fee than developers of other design tools, one possible, unintended consequence of the order is that Cadence may reduce or eliminate discounting of access fees. In addition, enforcement of the provision of the order requiring Cadence to provide access to the connections program to developers of "Commercial Integrated Circuit Routing Tools" on terms "no less favorable than the terms applicable to any other participants" may embroil the Commission unnecessarily in complex commercial disputes.
I concur in Paragraph III of the order and dissent from Paragraph II of the order.
1. The prior notice provision gives the Commission the opportunity to review a future horizontal acquisition by Cadence of another supplier of integrated circuit routing tools. Although this is a horizontal remedy, the complaint contains no corresponding allegations of liability under a horizontal theory. The remainder of the order addresses the vertical concerns of the majority and does relate to allegations in the complaint.
2. Mercantile Texas Corp. v. Board of Governors, 638 F.2d 1255, 1268-69 (5th Cir. 1981); Brunswick Corp., 94 F.T.C. 1174, 1269-72 (1979), aff'd and modified sub. nom. Yamaha Motor Co. v. FTC, 657 F.2d 971 (8th Cir. 1981), cert. denied, 456 U.S. 915 (1982).
3. A firm that can begin to supply a product within one year may be considered a market participant. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, Section 1.32 (1992).
4. See Statement of Chairman Robert Pitofsky and Commissioner Janet D. Steiger at 4.
5. In response to my discussion of this point, Chairman Pitofsky and Commissioner Steiger assert that they have relied on "confidential information from potential customers that tested Cadence's products under development," for their conclusion that the technology had failed. Statement of Chairman Pitofsky and Commissioner Steiger, note 8. In so stating, they reveal that Cadence had products under development and that the products were sufficiently advanced for customer testing. Since it is pointless to debate confidential information, suffice it to say that I disagree with this assessment of the project, based on my review of the customer information and on the views of Cadence's technology development partners. Preliminary testing is an ordinary part of the product development process. Software developers commonly seek customer reactions in beta testing and use the customer responses to refine their products.
6. Brunswick Corp., note 2 supra, 94 F.T.C. at 1269.
7. Paragraph 17 of the complaint alleges that before the merger, "Cadence did not have a commercially viable constraint-driven, shape-based integrated circuit routing tool."
8. U.S. Department of Justice Merger Guidelines, Section 4 (June 14, 1984).
9. The public record demonstrates the success of IC Craftsman (the Cooper & Chyan product) before September 12, 1996, when Cooper & Chyan and Cadence agreed to an interface between their products. In a June 3, 1996 press release, Cooper & Chyan said that it had sold the tool to 24 customers, including such familiar firms as AMD, IBM, SGS Thomson, Sun Microsystems, Fujitsu, Motorola, Northern Telecom and Toshiba. Press Release at http://www.cctech.com/new/press/dacqa.htm. This appears to be a substantial percentage of the universe of potential customers. Cooper & Chyan reported record second quarter earnings and revenues on July 23, 1996, and expressed pleasure at "the continued market acceptance of our IC product line." Press Release at http://www.cctech.com/new/press/q296.htm. The company reported continued improvement during the third quarter, which included the Cadence agreement on September 12, 1996. Press Release at http://www.cctech.com/new/press/q396.htm.
10. Richard Goering, "$420m deal shifts balance of power in board/IC CAD--Cadence acquiring CCT," EE Times, November 4, 1996.
The majority states that sales of the IC Craftsman were "only $13 million" in 1996. To put that amount in perspective, it should be observed that IC Craftsman was first introduced in the second half of 1995. Press Release of July 23, 1996 at http://www.cctech.com/new/press/q296.htm.