In the Matter of McWane

The FTC’s winding pursuit of an alleged monopolist

I joined the FTC in 2011 excited to try antitrust cases. Soon after I arrived, I was approached by a team that was eager to litigate anticompetitive conduct they had uncovered in the ductile iron pipe fittings (DIPF) industry. Their investigation began with an allegation that McWane, the only domestic DIPF supplier, was exploiting its dominant position in order to exclude new entrants. At a time when federal stimulus money required contractors to use American-made DIPF, McWane had enacted a “full support policy” that prevented foreign manufacturers from developing a domestic source of supply. As the FTC’s investigation of McWane’s exclusive dealing policy unfolded, the team uncovered other questionable conduct that caused it to expand its investigation.

McWane was the last US manufacturer in a steadily declining industry. McWane makes the DIPF used in waterworks projects. They are the elbows, couplers, reducers and other fittings that join pipes to one another, and guide water to its destination. By 2006, many US foundries had shuttered and competition from importers was steep. The two principal importers were Star Pipe Products and Sigma Corporation. Together with McWane, the three companies supplied 90% of the DIPF sold in the US. They sell to a variety of purchasers large and small, but a very significant share of all DIPF is sold through distributors that provide not just DIPF, but the pipes they connect and many of the other products used on a typical waterworks project. Distributors vary in size and geographic coverage. Two national distributors, HD Supply and Ferguson, together accounted for approximately 60% of the overall waterworks distribution market.

When bidding for waterworks projects, McWane had to compete with the importers’ ever shrinking prices. McWane analogised its plight to “wrapping dollar bills around the fittings” they sold when competing with importers. But McWane had one advantage: some projects specified that the DIPF (and other materials) had to be American-made. Although DIPF are interchangeable commodity products, if a project were on an air force base, for example, American-made fittings had to be used. For such projects, McWane had no competition and would charge a significant premium. If the contract did not specify that the DIPF needed to be American-made, McWane offered a substantially lower price for the exact same fitting to compete with the importers, Star and Sigma.

Then, in February 2009, the American Recovery and Reinvestment Act gave McWane a substantial boost. The stimulus bill had a “Buy American” provision that required the $6 billion in funds it allocated to waterworks projects to be built with American-made goods. As the only domestic DIPF manufacturer, McWane was in a unique position to reap the benefits.

Star and Sigma did not intend to sit idly by and allow McWane to capture all the newly funded stimulus jobs. Both companies explored making DIPF in the US. When it heard rumblings that Star and Sigma were contesting “their” turf, McWane issued a “full support” policy to its distributors. If distributors did not “fully support McWane-branded products for their domestic fitting and accessory requirements”, the company said, they may forgo participation in rebate schemes and shipping for months. McWane invoked the policy unilaterally and offered nothing in return.

The policy had the desired effect. Sigma abandoned its entry efforts and entered into an agreement to distribute McWane DIPF, essentially at prices set by McWane. The FTC and the administrative law judge, Judge D Michael Chappell, who heard the McWane case also found that the full support programme was a “significant reason” that another distributor, Serampore Industries Private, decided not to enter the domestic fittings market. And while Star boldly went forward, orders were cancelled, calls went unanswered and, after two years, Star had managed to capture a mere 10% of the market. Critically, it failed to generate enough business to support a dedicated foundry. Instead it suffered the higher costs associated with working with jobber foundries.

The FTC launched its investigation in May 2010, and McWane quietly rescinded its policy shortly after. Indeed, McWane never actually told customers that it had rescinded its policy and it still achieved the desired effect. Most of the stimulus money was committed by then. Sigma had abandoned its entry, and Star was struggling.

Although the investigation was initially focused on McWane’s exclusivity policy and its effects, it expanded when the FTC learned that the three suppliers – who together represented 90% of the market – had created a trade association whose sole purpose was to exchange recent shipment (but not price) information. Although the information was aggregated before it was shared with the three competitors, the data clearly played an important role in the three competitors’ pricing decisions.

The documents produced by the DIPF suppliers to the FTC suggested McWane’s exclusive dealing was part of a wider range of anticompetitive conduct. The DIPF industry is not a document-intensive one, but when the suppliers did write, their emails were colourful. And the key executives at the three suppliers met and spoke by phone frequently. The documents suggested that the DIPF suppliers engaged in collusion, invitation to collude, information exchange, monopolisation and exclusive dealing.

There was a problem, however: because the team believed the companies’ documents evidenced collusion, the investigation had to be handed over to the DOJ. The FTC does not prosecute criminal violations of the Sherman Act; only the DOJ is authorised to prosecute cases involving potentially criminal collusive activity. Once the FTC team decided the documents suggested price fixing, they suspended the investigation and sent the evidence down Pennsylvania Avenue to the Antitrust Division. The DOJ eventually declined to take the case, sending it back to the FTC. By that time, months had passed, and there was pressure to bring staff’s findings to the commission for a vote.

In January 2012, the commission approved a seven-count complaint against McWane. The complaint was approved 4-1 with only Commissioner Thomas Rosch dissenting. Rosch’s litigation sense was superb, and in his dissent he predicted a significant hurdle that would arise in court. Competitors who were motivated to assist the FTC in putting an end to McWane’s exclusive dealing would lose their zeal to help when accused of price fixing, Rosch predicted – and the FTC would struggle to prove that the three companies conspired. But then McWane turned on its co-conspirators with its exclusivity policy. As usual, Rosch was right.

Sigma opted to enter into a consent agreement with the commission without admitting guilt before the complaint was voted out. Star soon followed. But McWane took the position that it was merely trying to preserve American industry and geared up for a fight.

Because our team had limited experience trying a conspiracy case, one of our early tasks was to meet with a pair of veteran DOJ cartel prosecutors for instruction. When we met, their first questions were about our cooperating witnesses. The FTC does not offer leniency and none of the companies involved had shown any interest in cooperating. When we told the DOJ lawyers that we had no cooperating witnesses, silence descended on the room. How could the FTC expect to try a conspiracy case without cooperating witnesses? We learned that at least these veteran trial lawyers did not recommend going to trial without at least two cooperating witnesses to help tell the story of the conspiracy. No matter; we bravely told them that we had “hot” documents; we planned to treat the industry participants as adverse witnesses; and we had that “great legal engine for discovery of truth”, cross-examination, in our corner. I had seen cases tried very effectively through adverse witnesses. The witnesses could either admit to their illicit communications – and the obvious inferences that came from their phone calls followed by pricing decisions – or look
like liars.

The administrative trial

If we had known that we would be denied the opportunity to cross-examine the conspiracy’s participants, and Judge Chappell would draw no inferences from the phone records of their frequent and fortuitously timed calls, we might not have shown such confidence in our collusion claim. The first hint of our troubles came during the pretrial conference. The federal rules of evidence allow counsel to ask leading questions when questioning a witness identified with the opposing party as a co-conspirator. The FTC equivalent rule is narrower. Counsel may only ask leading questions when questioning an adverse party or “any witness who appears to be hostile, unwilling or evasive”.

Because we would be calling witnesses from all three alleged co-conspirators in our case-in-chief, we addressed the issue at the pretrial. We explained to the judge that although Star and Sigma had settled, we would be calling their executives in our case against McWane. We argued that all three companies were hostile to the FTC’s claims and were co-defendants in civil class actions based on the FTC’s complaint. We asked Judge Chappell to allow us to ask leading questions when calling Star and Sigma executives as adverse witnesses. Instead, the judge wanted to wait and see if the witnesses were truly hostile to the FTC. We would have to show that a witness was openly hostile before we could use leading questions, he said, and he would make those decisions on a witness-by-witness basis after hearing the witnesses on the stand. We never found out just how hostile a witness would need to be to allow us to use leading questions because the judge never found a single witness to be hostile. We were left to try and prove Star and Sigma participated in a conspiracy with McWane by asking the key witnesses open questions. McWane could then follow-up by asking leading questions on cross-examination. McWane’s counsel took full advantage of the opportunity. And the Star witnesses, who would not admit to participating in a conspiracy but were opposed to McWane’s exclusive dealing policy, seemed unable to parse McWane’s lengthy, leading questions. Instead, the witness simply said “yes” to just about any question McWane’s counsel would ask.

Our trouble became clear when we questioned the executives about their phone records. The records reflected frequent calls between key sales executives during the time they were considering price increases. But the witnesses claimed not to remember the calls or what was discussed. Judge Chappell made clear he would not draw any inference from the fact that the head of sales from one company had a lengthy call with the head of sales from another. Nor would he allow complaint counsel to juxtapose business documents, such as emails and phone records, to show the suspicious timing of the calls. The phone records were to be used in camera because they contained personal as well as business information; the business records were not. The judge curtailed questioning using the phone records, and ultimately found little troublesome about the many calls between key competing executives.

The administrative trial lasted nearly two months. The judge heard testimony from 16 live witnesses. The trial transcript ran to more than 6,045 pages and the trial record Judge Chappell reviewed in rendering his initial decision included an additional 73 days of deposition or investigative hearing testimony. Nearly 2,000 exhibits were admitted into the record. The parties’ proposed post-trial findings of fact and law topped 3,000 pages.

Judge Chappell reached a split decision. In a 464-page decision, the court found complaint counsel had failed to meet its burden on the first three counts and succeeded on the last four. The judge found the conspiracy claim plausible, but the evidence “at best” showed “interdependent or consciously parallel conduct unaided by agreement . . . .” (counts one and three). Although the fittings market was an oligopoly, he found that the information exchanged between the companies was insufficient to facilitate price coordination (count two).

However, Judge Chappell did find the distribution agreement between McWane and its competitor Sigma lacked pro-competitive justifications and was illegal under the rule of reason (count four); and that the two competitors had thus conspired to monopolise the domestic fittings market (count five). He also ruled that McWane’s full support programme, which required exclusivity from its distributors, stopped Star from winning a substantial share of the domestic DIPF market and unlawfully maintained McWane’s monopoly (count six). Finally, the court found McWane’s efforts constituted attempted monopolisation and thus found for complaint counsel on the final count.

A commission divided

Both parties appealed to the commission, which reviews the administrative law judge’s decision de novo. Complaint counsel opted not to appeal the invitation to collude count, based on the belief that we had proven actual collusion, not just an invitation to collude. But the full commission did not agree.

Because it reviews the administrative law judge’s decisions in cases it originally voted to bring, the FTC has been criticised as a “kangaroo court”. The outcome of the McWane case clearly demonstrates that is not the case. The commission has also typically tried to reach unanimity whenever possible. In McWane, they failed to achieve that goal.

A commission of four, including two of the commissioners who had originally voted out the complaint, declined to rubber-stamp the court’s decision or to reverse the court to find in favour of complaint counsel on counts they had initially voted out. Two commissioners found the unlawful conspiracy and information exchange had been proven and two commissioners did not. Counts one and three that were the subject of the deadlock were dismissed as not “in the public interest”. The commission reversed the judge on count four, and found that complaint counsel failed to establish that McWane broke the law in its distribution relationship that led to Sigma distributing its domestic products. In light of that decision, the commission did not need to reach count five, alleging that McWane and Sigma conspired to monopolise the domestic fittings market through their distribution agreement.

Only count seven, the exclusive dealing count, garnered a majority (3-1), with Commissioner Joshua Wright dissenting. The commission thus did not need to reach count six, attempted monopolisation, which was based on the same facts. The commission held that domestically produced DIPF was a relevant market and McWane had monopoly power in that market. The commission rejected McWane’s contention that complaint counsel’s expert was required to offer econometric evidence to prove a market for domestic pipe fittings. The majority found that by impairing its rival’s access to distribution and keeping it from growing enough to become a competitive threat, McWane had not just harmed its rival Star, but competition generally. The commission also rejected McWane’s purported justifications as unpersuasive and not offering any consumer benefits.

Wright disagreed in a lengthy dissent. He believed complaint counsel had fallen short in demonstrating that McWane was successful in harming competition. He assumed that McWane was a monopolist in the domestic-only fittings market, agreed that the full support programme was an exclusive dealing arrangement, and concluded that there was “ample record evidence” that the programme harmed Star. But he contended that the government “failed to carry its burden to demonstrate that the full support programme resulted in cognisable harm to competition”.

According to the dissent, complaint counsel had failed to show that Star’s inability to afford its own foundry was the equivalent of it being unable to achieve minimum efficient scale; failed to link locking Star out of the market to McWane’s alleged maintenance of monopoly power; and miscalculated how much of the market McWane’s behaviour foreclosed. Wright declined to credit evidence from Star as to what would be required to achieve minimum efficient scale. He demanded evidence from an independent expert addressing minimum efficient scale in the DIPF industry. The difference between harm to Star and harm to competition was whether McWane prevented Star from achieving minimum efficient scale. Because exclusionary practices can be pro-competitive, Wright argued that complaint counsel should have to produce “clear evidence of anticompetitive effect”.

The Eleventh Circuit appeal

McWane had successfully pared down the complaint from the original seven counts to a single charge that it had engaged in exclusive dealing. A respondent can choose to appeal an adverse decision by the FTC to any circuit. Alabama-based McWane appealed to the Eleventh Circuit Court of Appeals – one seen as particularly tough on the FTC, having reversed a number of
commission decisions.

The Eleventh Circuit affirmed the commission’s decision. McWane argued that because complaint counsel’s expert did not support its market definition analysis with any econometric analysis, the domestic DIPF market could not be sustained. The court ruled there is no legal requirement for an econometric study; and that courts routinely rely on qualitative evidence to support a market definition finding. The court found McWane’s internal documents revealed its intent was to raise Star’s costs and impede it from becoming a viable competitor. The fact that Star had entered the domestic DIPF market with modest success did not dissuade the court. It recognised that some case law supported McWane’s argument that Star’s limited ability to enter cast doubt on McWane’s alleged monopoly power. But Star’s inability to constrain McWane’s pricing, combined with McWane’s overwhelming market share, proved sufficient to convince the court that the company was indeed dominant.

The court disagreed with Wright’s dissent, and found direct and indirect evidence of harm to competition. Star’s entry did not constrain McWane’s prices. Indeed, not long after Star entered, McWane raised prices, and its profit margins after Star’s entry were higher than they were previously. The McWane exclusivity programme prevented Star from emerging as an effective competitor. The court also relied on the “particularly powerful” evidence of McWane’s intent in finding that its exclusionary practices harmed competition.

Like the commission, the court rejected McWane’s purported pro-­competitive justifications as unpersuasive. McWane’s desire to keep sufficient sales to continue profitably operating its lone remaining domestic foundry was neither illegal, nor pro-competitive. McWane did not increase sales by lowering prices, increasing quality or service, but by taking sales from its only rival; consumers saw no benefit from that. McWane also argued the full support programme was necessary to prevent Star from cherry-picking the most common fittings, causing McWane to lose sales of popular fittings but requiring it to continue to support a full line, including less-often used fittings. But the commission and the court agreed that McWane could have lowered prices on the common fittings at risk of being cherry-picked, and raised prices on the less common ones. McWane’s documents were clear that the programme’s purpose was to “keep Star from driving profitability out of our business” and make sure Star could not “reach a critical mass that will allow them to continue to invest and achieve a profitable return.” The court found McWane’s documents supplied powerful evidence that its pro-competitive justifications were merely pretextual.

Conclusion

The McWane case ended as it began, as an exclusive dealing case. Successful monopolisation cases are rare. And while everyone involved had some sympathy for McWane’s desire to support its American foundries, exerting monopoly power over potential entrants was not the solution. The case also demonstrates that the FTC’s administrative process is not the kangaroo court it is sometimes accused of being. Everyone involved – commissioners, staff and the judge (and his staff) – works hard to arrive at a just result based on the evidence available.

Unlock unlimited access to all Global Competition Review content