FTC v Sysco/US Foods
Market definition and the pitfalls of litigating the fix
On 19 February 2015, the afternoon that the Federal Trade Commission filed a lawsuit to block Sysco’s proposed acquisition of rival food-service distributor, US Foods, the agency’s litigation team gathered on the fourth floor of the FTC’s offices at Constitution Center to discuss what was in store over the ensuing weeks and months. Merger challenges are the Holy Grail for any antitrust lawyer who prizes antitrust litigation. For both government lawyers and defence counsel, no kind of antitrust litigation moves as quickly – and no case offers more meaningful opportunities for every member of the litigation team as a government challenge to a high-profile deal.
It was no surprise that the FTC litigation team was excited. It was a young team – only a few had previously been in matters that had ever gone to trial. But the attorneys on the FTC team were as energetic, disciplined and sharp as any I’d ever been around. That afternoon at Constitution Center, we talked about how merger litigations, especially during trial, are like roller coasters, with ups and downs for both sides. We talked about how the FTC’s case will never look better than it does on the day of the filing, but that the team’s superior knowledge of the facts and evidence based on many months of investigation was an advantage that would carry the day if executed properly. And we talked about the importance of “having each other’s back” throughout the case, and the need to stand up to, and not be intimidated by, a defence team that predictably would consist of more than a dozen Big Law partners with courtroom experience. The FTC’s successful challenge to the Sysco/US Foods transaction is really a story about the 15 or so FTC attorneys, working as an integrated, unified team, who did the hard work necessary to win one of the most important merger challenges in years.
The merger involved the two largest food-service distributors in the United States, who also were the two largest food-service distributors selling a broad line of food items to hospitals, restaurant chains and other customers with locations dispersed across the country. The case had everything one could hope for in a merger litigation: hotly contested disputes about market definition (both product and geographic); share calculations; ease of entry; buyer power; and the importance of efficiencies. It had substantial economic testimony on both sides. Neither side conceded much, if anything.
The case also had a twist that, in my view, turned out to be a critical factor in the FTC’s path to victory. The merging parties decided to “litigate the fix” by offering a divestiture of assets, principally a selection of US Foods’ distribution centres dispersed throughout the US, to a large regional competitor, Performance Food Group (PFG). Based on the court’s opinion, PFG ultimately proved to be a much better witness for the government than the parties. More on that below.
What follows is a chronology of what I remember most vividly about the case and the team that made it happen, as well as potential takeaways.
The brewing storm
Late 2014 and early 2015 was a particularly exciting time to be at the FTC. It seemed to many that President Obama’s promise on his way to office in 2009 to invigorate US antitrust enforcement had never really manifested itself, and certainly not in the way it was about to over the next two years. Of course, in the healthcare arena, antitrust enforcement was as strong as ever, especially in challenging reverse-payment patent settlements between pharmaceutical companies. But, for over a decade, merger litigations had been relatively sparse, especially outside the hospital area, for both US antitrust agencies.
That all was about to change. In December 2014, the FTC voted to challenge Verisk Analytics’ proposed acquisition of Eagleview Technology, which would have combined two of the leading providers of aerial imaging technology used by commercial and home property insurers. The parties immediately abandoned the merger after the complaint was filed. In December 2014, there were other major horizontal tie-ups making their way through the commission process: Sysco/US Foods, Zillow/Trulia and RJ Reynolds’ proposed acquisition of Lorillard, among others. And, of course, there were the hospital mergers that the FTC ultimately challenged – Northshore/Advocate, Hershey/Pinnacle and Cabell/St. Mary’s – early in the pipeline. The dynamics were clearly in place for at least a few cases to head to litigation.
The first of these litigations for the commission was Sysco. I won’t get into the specifics of the pre-litigation investigation, but it’s safe to say that I saw this case coming a mile away. The parties were well represented at the commission level, as they were in litigation, and the case proved to be a relatively close call at the agency level, as evidenced by the commission’s 3-2 vote to challenge the deal. But, at least to me, the decision to fight Sysco/US Foods was never a close call. These two companies are great at what they do – and are unique in keeping each other honest – more so than any other food-service distributor. And the proposed fix the parties offered was not only risky in its execution, but fell short of replicating the head-to-head rivalry between the companies. If the commission failed to challenge this deal, we might as well have all gone home.
The 3-2 commission vote
One thing we did not discuss much, if at all, at the 19 February meeting was the potential effect of the commission’s 3-2 vote. I kept telling myself at the time that it didn’t matter and that the court would focus on the evidence instead of the divided commission. But I’d be lying if I said that I wasn’t initially concerned about the optics of the sharp divide. Back in 2004, when the FTC challenged Arch Coal’s acquisition of Triton, then commissioner Thomas Leary had voted against a commission federal court challenge. I was on the defence team representing Arch in that case and we used virtually every filed pleading to remind the court that commissioner Leary had dissented. Who knows what impact that ultimately had in the FTC’s loss, but I could imagine how Sysco and US Foods would do the same in this case, only with greater vigour because of two dissenting votes, instead of the one in Arch Coal.
Happily, my concerns were allayed in short order. We learned the next day that the Sysco case had been assigned to the Honourable Amit Mehta of the DC federal court. The information we immediately gathered about Judge Mehta – razor-sharp, newly minted Obama appointee with a liberal background, who had pledged in his Senate confirmation hearings to faithfully apply precedent – was exactly the draw we wanted. It also put to rest any concerns I had that the court would be influenced in any way by the 3-2 vote along party lines. After reading about Judge Mehta’s background on my iPhone during a bureau of competition meeting about another matter, I passed Debbie Feinstein, the bureau director, a note which said: “This neutralizes the 3-2!”
That Judge Mehta was an Obama appointee and the commission vote was along party lines (the three Democrat commissioners voted to challenge, the two Republicans dissented) was certainly not a dynamic in favour of the defence, although it didn’t favour the agency either. It merely took away the side show of the divided vote at the commissioner level. And it probably influenced the defence team’s decision not to emphasise the 3-2 vote much, if at all, throughout the case.
The pretrial battles
In my experience, the biggest advantage the agencies have in merger litigations is the staff’s superior knowledge of the facts and evidence based on many months of investigation. It certainly isn’t litigation experience: at least it wasn’t back in early 2015 when the FTC had not tried a major merger case in years. The defence litigators are usually playing “catch up” compared to the agency litigation team, especially when the merging parties hire lead litigation counsel that is different from counsel who represented them during the agency review, as was the case in Sysco. We knew we would be ready to put on our case very quickly and we wanted to take advantage of that.
Of course, the merging parties had a strong interest in a timely hearing as well. The proposed merger had been pending for some time and the parties were anxious either to move forward with the deal or to move on to other things. The FTC proposed a hearing beginning in late April 2015 while Sysco and US Foods pushed for a late May start date. The court basically “split the baby” by scheduling the hearing to begin in early May, and gave the parties eight days to try the case. It was certainly a good outcome for the merging parties. They would get an opportunity for ample discovery and plenty of time to prove their case in a courtroom. But we liked the schedule too. We felt ahead of the curve on the facts and evidence, and in my view that dynamic proved critical. I was amazed that, every time Feinstein or I had a question about a witness or other evidence that supported a particular point, the team, led by talented Mergers IV attorney Melissa Hill, showed its mastery of the evidence by sending us information within a matter of hours, if not minutes.
There were other momentous points of the pretrial proceedings that, in my view, proved important to the successful outcome for the FTC. Judge Mehta made it clear that he would make himself available to quickly address discovery disputes as they arose. Every litigator knows that judges don’t like to deal with discovery disputes and want the parties to work things out themselves. But we felt that certain issues that arose during discovery were important enough to raise with the court and to demonstrate to the defence that we would not be steamrolled. A few of those disputes had to do with declarations the defendants had secured from restaurants and other third parties after the deadline for exchanging declarations had passed. Others had to do with the timing of disclosure of materials from the expert witnesses. In the grand scheme of the case, these were probably minor issues, but raising and often winning them helped to energise the FTC team, and gave us confidence that our arguments were resonating with Judge Mehta.
The eight-day hearing
Unlike many kinds of commercial or even antitrust litigation, merger challenges are typically short and sweet. The time frame for a merger litigation brought by either the FTC or DOJ is usually four to six months from complaint to end of trial. That means very long days and sometimes longer nights for both sides’ counsel during that entire period, but the end is always in sight. Every day for six months, the FTC team worked gruelling hours and travelled to Omaha, Minneapolis, southwest Virginia and elsewhere to depose or defend witnesses. One of my everlasting memories of the case will be coming home each night, poking my head into my young kids’ bedrooms to see if they were still awake, and listening to my six-year-old routinely ask: “Daddy, has the FTC solved the case yet?” We hadn’t yet “solved the case” but, as early May approached, the time had come for the FTC to prove its case in court.
Week 1: the FTC makes its case
Prior to the hearing, the team met to discuss the witnesses we would call and the sequencing of those witnesses. We wanted to start strong and finish stronger. After much internal deliberation, we made two strategic decisions about our affirmative case that, in my mind, proved pivotal to winning. First, after questioning a few complaining food-service customers, we would call the chief executive of divestiture buyer PFG as a hostile witness. We knew that PFG would be a fervent advocate for the merger and proposed fix, but we also had plenty of evidence about PFG, gathered from both the investigation and litigation discovery, that supported our theory that the fix wouldn’t do enough to repair the harm the deal would cause. Getting these points in front of the court – even through a hostile examination – before the defence had an opportunity to make PFG look like a knight in shining armour, we believed, would raise serious doubts about PFG in the eyes of the court. Second, we would close our affirmative case with our economic expert, Dr Mark Israel who, we believed, could and would tie the evidence together in a way that would make sense to the court. As the court’s decision showed, we were right on both counts.
As for the hostile direct examination of PFG’s chief executive, I doubt I’ll ever have as much fun in a courtroom. It’s really not difficult examining a hostile witness when you’re armed with a beautifully drafted declaration (based on the witness’s own words, of course), a solid deposition transcript and a set of internal PFG documents that leave little room for an interpretation favouring the merging parties. I’m told that one of Sysco’s lawyers commented to an FTC staff lawyer during a break in the PFG’s hostile direct that “it’s not hard to act like Clarence Darrow when you have a declaration like that.” Maybe not, but it certainly was fun. For almost half a day, the PFG CEO had little choice but to agree with what he and his colleagues previously had testified to or written in internal documents about the anticompetitive nature of the underlying merger, and the significant shortcomings in the proposed divestiture. When he tried to back away or retract his earlier testimony, things didn’t seem to get much better for him or Sysco. While Sysco’s counsel did an admirable job trying to rehabilitate PFG’s chief executive on friendly cross, the court ended up citing testimony and other evidence from PFG as a basis for blocking the merger and rejecting the proposed fix.
The FTC fared equally well by concluding with Israel’s testimony. Despite gaps in available bidding data that would impede any economist from performing an air-tight econometric analysis of the deal, young FTC staff lawyers Steve Mohr and Sophia Vandergrift deftly prepared Israel to testify as if it was a walk in the park. I’ll be candid: before trial, all I really hoped for was for the experts on each side to “wash each other out” so that the court would focus mostly on the qualitative evidence in the case, such as evidence from customers and PFG, and the parties’ internal documents. But, as emphasised by Judge Mehta’s decision, we did far better than that. Israel connected well with Judge Mehta, was effective at transforming the economic aspects of the case into easily understandable principles and, as we had hoped, summarised the entire FTC case in a way that aligned the economic evidence with the customer testimony and corporate documents. The team was so pleased with the way Israel’s direct and cross-examination had gone for the FTC that we later decided to call him as our principal rebuttal witness. Ultimately, Judge Mehta found his testimony more credible and more consistent with the qualitative evidence than that of the two defence experts.
Week one was now in the books and the FTC team had serious momentum. But we knew that the Sysco and US Foods witnesses would be well prepared going into week two.
Week 2: the defence case
The merging companies’ witness line-up was heavily weighted toward Sysco and US Foods executives who would discuss how they perceived competition and the merger’s impact. That was no accident. A team of FTC lawyers, led by Ryan Quillian (aka “Decs a Million” for the seemingly millions of declarations he gathered from customers who opposed the deal), Catherine Sanchez, Jeanne Nichols, Chris Abbott and Mike DeRita, took surgical discovery depositions of Sysco’s potential third-party witnesses, effectively taking them out of Sysco’s repertoire of live witnesses. A defence fact witness line-up, consisting mainly of party executives, was not a bad thing for the FTC. In my experience, judges typically discount party executives’ testimony in merger cases because of their inherent bias. That’s not to say that calling the merging parties’ executives is not important for the defence. It certainly is because the court expects to hear from them, including as to the reasons motivating the deal. But, except in mergers involving an argument that one of the parties might go out of business if not for the deal, I know of no modern merger litigation that turned on the testimony of the companies’ executives. The Sysco/US Foods witnesses were good for the parties but we knew their testimony wouldn’t carry the day. The FTC’s co-lead trial counsel, Alexis Gilman and Mark Seidman, supported by Matt MacDonald, helped seal that outcome with their strong cross-examinations of the Sysco executives.
The other centerpiece of week two was the testimony of the defence economic experts, Dr Jerry Hausman and Dr Timothy Bresnahan. Both are fine economists but, as Judge Mehta found, their opinions did not square with the qualitative evidence. For example, Judge Mehta noted in his opinion that Hausman defied credulity when he admitted that, under his broad market definition (“all foodservice distribution”), every single broadline food-service distributor in the US could merge without any ensuing anticompetitive effects. Hausman’s opinion in this regard proved far too much and probably tainted the rest of his testimony.
Key takeaways from the case
Others will decide what, if any, effect the Sysco/US Foods litigation and subsequent litigation victories for the agencies have had on merger enforcement. After Sysco, the FTC has won three other merger litigations and the DOJ won a few as well. A number of other challenged mergers were voluntarily abandoned by the parties without a trial. Each litigated merger case is influential in its own right, but, together, they manifest a clear willingness and preparedness of the two antitrust agencies to go to court to block deals they view as anticompetitive. There can no longer be a myth that the agencies aren’t willing to – let alone, that they can’t – do it as well as private practice lawyers.
In my view, the Sysco case at a minimum helps to illustrate at least three important points about US merger enforcement that came into sharp focus during the second term of the Obama administration.
In litigation, documents determine outcomes
Judge Mehta’s decision in Sysco, as well as the subsequent decisions in Staples II, Aetna and Anthem, highlight that merger cases are won or lost largely on the substance and content of the business documents of the merging parties (and the divestiture buyer, if there is one). Why is that? As Judge Mehta’s opinion illustrates, testimony and opinions of market participants matter, as do economic analyses. But unlike those sources of evidence, most contemporaneous business documents are not generated in the course of a merger investigation, or in litigation. As a result, courts probably tend to view business documents as the evidence that is most reliable and least susceptible to interpretation or manipulation. And generalist judges tend to hinge their decisions on their comfort zones, which rarely include complex econometric analyses or predictive opinions of customers.
A case in point is Judge Mehta’s conclusion – central to our victory – that the FTC had demonstrated that there is a relevant market consisting of national customers of broadline food-service distribution. Sysco and its strategic consultants had documents in their files that classified certain customers as “national” customers, as did US Foods and PFG. Instrumental to Judge Mehta’s decision about market definition were these materials – most notably, a document prepared by Sysco’s strategic consultant McKinsey – which repeatedly referred to “national customers” and defined them in the same way the FTC did in court. I remember vividly when Krisha Cerilli, an FTC attorney who, along with Kit Rogers, mastered the efficiencies part of the case, handed me that Sysco document late in the investigation. It was as close to a “smoking gun” document in the case as there was, and it featured prominently in our opening statement and closing argument, as well as in our written submissions. And, more importantly, Judge Mehta described this document as compelling evidence in support of a national customer market, in his opinion blocking the deal.
Litigating the fix is a risky strategy
The Sysco case, as well as the more recent DOJ litigation against Aetna/Humana, underscore that litigating the fix poses a risk for merging parties. By introducing a proposed divestiture, the parties inject the proposed buyer as the functional equivalent of a party opponent, whose internal documents, statements and deposition testimony are admissible against the merging parties as well. Both PFG and Molina (the proposed divestiture buyer in Aetna/Humana) had many documents that hurt the mergers’ prospects for success, both on the underlying theory of harm and the insufficiency of the proposed fix. Both third-party divestiture buyers turned out to be much better witnesses for the government than for the merging parties.
To date, there has been only one case in which the merging parties have litigated the fix and won in court: Arch Coal. But unlike Sysco, Aetna, Staples II and other cases where the strategy failed, the proposed divestiture in Arch Coal was clean, straightforward and consisted of a stand-alone business. The recent court decisions also suggest that the merging parties, instead of the FTC or DOJ, have the burden of showing that the proposed remedy addresses the underlying merger’s harm – a departure from the holding in Arch Coal. So, litigating the fix may not get any easier for merging companies.
Efficiencies are an afterthought in litigation
Sysco’s claimed efficiencies from the proposed deal were significant, rising to over $1 billion in annual savings, according to McKinsey. These claims were taken very seriously at the agency level; one of the two dissenting commissioners later told me that he viewed them as quite influential in his vote not to challenge. But in the staff and bureau’s view, the magnitude of merger-specific efficiencies would be dwarfed by the transaction’s harm to consumers of the companies’ services, and in particular their national customers.
Once we were in litigation, I personally paid little attention to the efficiencies part of the case. I knew that component was in very good hands with Cerilli and Rogers, and that the court’s decision almost certainly would not hinge on the efficiencies arguments. The law on efficiencies is simply too slanted against the merging parties in court to really make a difference. No court has ever allowed a merger based principally, much less solely, on an efficiencies defence, and the only lower court that tried – the court in FTC v Heinz – was emphatically reversed. Unfortunately for merging parties, the trend in the law continues to work in the wrong direction when it comes to the weight given to efficiencies in court. The recent decision in Anthem/Cigna only reinforces that result.
Despite the court decisions, my experience is that efficiency arguments really do matter at the agency level, and play significantly into the FTC and DOJ’s prosecutorial discretion in bringing cases. A decision to litigate becomes much harder – as it did for some at the FTC in Sysco – when there is a risk of a “false positive”; namely a risk that blocking the merger will not protect competition, but will prevent valuable cost savings. By contrast, a decision to challenge becomes easier when the parties’ efficiency claims are tenuous, or unsubstantiated, because the risk of harm caused by erroneous agency intervention is fairly low. My instinct is that efficiency arguments will have more effect on enforcement decisions in the Trump administration than in prior administrations.
I will always be grateful for my time at the commission, including the experience of FTC v Sysco/US Foods. The case was one that needed to be brought to help fulfil President Obama’s promise to invigorate antitrust enforcement, and it proved to be a decisive win for the agency. Only time will tell whether Sysco and the spate of other merger litigation victories for the government will influence enforcement decisions in the Trump and subsequent administrations. I know that the agency lawyers hope that they do.