Image for post
Image for post
The FTC’s case hinges almost entirely on theory, with little empirical evidence

FTC Witness Withers Under Cross Exam in Qualcomm Antitrust Case

In one of the stranger antitrust cases to arise in recent years, the instigator isn’t even one of the adversaries. Apple, not long ago the most valuable company on earth, is in the courtroom as a witness, but the heavy lifting is being done by the Federal Trade Commission (FTC). The target, Qualcomm, has made its living for years licensing intellectual property to phone makers as well as selling both modem chips and application processors. In FTC vs. Qualcomm, now being heard in United States District Court, Northern California District, San Jose Division, before Judge Lucy M. Koh, Apple’s role is merely to help the FTC makes its case. And yet Apple is the reason there is a case at all.

Koh goes back far enough to have presided over a number of cases in which Apple was involved directly. To name several, there was Apple vs. Samsung in a patent-infringement case, which dragged on for seven years and is still under appeal, and an anti-trust and collusion case involving a number of Silicon Valley firms’ agreeing not to hire each others’ technical employees. Koh fingered Steve Jobs as a ringleader in the non-hire case.

Koh has a reputation for making quick decisions and writing her opinions in double time. But in this case, after 50 hours of testimony by both sides, which is piled into a transcript of more than 2,000 pages, she admitted she will need some time to go over it all.

How did we get here? Because Apple was unhappy at having pay patent royalties to Qualcomm, the iPhone maker prodded the FTC to take up its cause. Apple claimed that Qualcomm used its chip supply (it makes the best cellular modems in the business) as a lever to force its customers to pay extraordinary rents, not only for modem patents but for its entire portfolio, which covers other systems on the phone, technologies in base stations, and many other related areas both on and off the phone itself. Apple wanted to base its payments on the price of the modem chips rather than the price of the phones, even though it makes use of a broad array of Qualcomm’s technology, and despite the fact that the latter method has been standard industry practice since before Qualcomm was even in the cellular licensing business.

Making things more bizarre, Apple was never even a direct licensee of Qualcomm’s patents, even though the latter had offered to negotiate such an arrangement. Qualcomm’s license fees were paid by the contract manufacturers (Compal, Foxconn, Pegatron, and Wistron) that made the phones. They had been paying fees under the exact same terms since before Apple entered the phone business. Apple enforced its will by withholding payments to the contract manufacturers, which then withheld them from Qualcomm. That’s quite a bank shot!

To put things in context, Apple has been dining out quite nicely thank you since 2007 on the iPhone, which established the smartphone market and took Apple from an also-ran PC company to the top of the world’s great corporations in 10 years. But there was weakness in this strength. Apple has been overly reliant on the iPhone for its gargantuan profits. The smartphone market is reaching saturation (currently, top OEMs like Apple, Huawei, Samsung, and Xiaomi are fighting it out to dominate in India, perhaps the last large market with some growth left in it). And so, to show the kind of profit growth that would justify its tremendous valuation, Apple has turned to raising prices (hoping for a degree of inelasticity among its customers) and lowering costs, often by squeezing suppliers. This change in tactic is highlighted by Apple’s recent decision to stop reporting iPhone unit sales, which Wall Street interpreted as a signal that iPhone unit growth was dead. Apple is relentless in reporting the good and burying the bad.

Thus, Apple began eyeballing the $7.50 (according to Apple Chief Operating Officer Jeff Williams’s testimony) it had been paying per unit to its contract manufacturers for the Qualcomm patents and decided to try to eliminate the fee. It is ironic that a company with gross margin percentages in the high 30s, far more than any other company in the industry, is whining about somebody else’s profits. The mere presence of such a cash geyser is an indication of market power. Apple makes more money on smartphones than all other companies in the business combined. Another measure of absurdity: the difference between an iPod and an iPhone is mostly the latter’s cellular capability. The difference in price between the two is more than $300. Thus, Apple is making hundreds of dollars on Qualcomm’s cellular patents but declining to pay $7.50 for the privilege. In negotiations, Apple has offered to pay $1.50.

The FTC’s case against Qualcomm hangs of a few concepts: that Qualcomm withheld or threatened to withhold its chip supply from firms unwilling to license its comprehensive cellular patent portfolio, that Qualcomm receives extraordinary rents from this situation, that Qualcomm refuses to license to other component makers (like Huawei, Intel, and MediaTek, which also make cellular modem chips).

As we await Judge Koh’s decision, the FTC’s case looks weak and is not helped by the fact that it bears the burden of proof. Much of its case is built on a theory promoted by its expert witness, Carl Shapiro, a professor at Berkeley’s Haas School of Business. He tried to poke holes in the statements of Qualcomm’s witnesses while Qualcomm spent a good deal of its time poking holes in Shapiro’s testimony. The problem with Shapiro’s theory is that it doesn’t stand up to a raft of real-world data, an abundance of which Qualcomm has at its fingertips, things like how much customers actually pay and what the terms of their contracts are.

For example, under cross-examination by Qualcomm attorney Robert Van Nest, Shapiro admitted that his “tax” theory would predict that even if rival chips makers’ costs were to rise from having to pay, even indirectly, Qualcomm’s licensing fees, their market share would go down. Van Nest pointed out that, in fact, Intel’s share rose during the period in question. But rival chip makers don’t have to pay Qualcomm. Only smartphone vendors do. Shapiro’s “tax” theory also predicted that such cost rises would restrict demand and therefore the total market, but, again, Van Nest, using public market data, showed that the total market, in this case modem chips for premium LTE phones, had exploded. If Qualcomm had been exercising monopoly power, these results would have been unlikely. Any anticompetitive practice would have restricted market growth.

More importantly, Van Nest illustrated that Shapiro’s theories had not really been tested. From the transcript, here’s an exchange particularly damaging to Shapiro’s credibility:

Van Nest: “You told us last time that you’ve done no empirical work whatsoever to determine the size of any supra-FRAND royalty [excessive pricing] you’re assuming existed; correct?

Shapiro: “What I said is I have not quantified it. I explained why we have good reason to believe that it was substantial.”

Shapiro is proposing that Qualcomm charged too much without defining or even knowing what “too much” is.

Here’s another exchange:

Van Nest: “You’ve done no empirical study of Qualcomm’s license rates, royalty rates, or upfront payments over the years at all; correct?”

Shapiro: “Again, I looked at the licenses. But if you mean by empirical study trying to take all of these hundreds of licenses and distill them into simple data, I don’t think you can do that. So, no, I didn’t do it.”

Van Nest: “Similarly, with respect to your tax theory, you’ve done nothing to analyze the R&D spending of any particular chip making rival to Qualcomm; correct?”

Shapiro: “I have not analyzed — I think if you mean by analyzed, look at exactly why they spent this amount of money or when they did it, I have not got into that, that is correct.”

Van Nest: “Your tax theory predicts that spending will go down, but you did no testing to determine whether or not that was the case; correct?”

Shapiro: “I did not — look, I stand by the theory that if you have — if your margins are cut and your quantities of units are cut, you’re going to have smaller operating income and that’s going to be a drag on possible investments. I did not empirically test how that played out with different modem chip suppliers.”

Van Nest: “In other words, you didn’t actually look at anybody’s margins or anybody’s spending or anybody’s research and development. You simply assumed there would be an impact because that’s what your tax theory predicts; correct?”

Shapiro: “I did not look at — I did not do the type of analysis Professor Snyder [another FTC witness] did regarding looking at rival by rival and explained, I believe, according to my analogy, why I don’t think that’s a good methodology.”

So, all theory and no data by the FTC’s key witness.

And this is important because in earlier testimony, Qualcomm was able to show that its rivals (e.g., Intel) had sufficient money in their coffers but spent inefficiently or chose not to spend on the specific areas of R&D that could have led to a patent portfolio comparable to Qualcomm’s.

And then the zinger (edited a bit for brevity):

Van Nest: “Now, professor Shapiro, this is not the first time you’ve been criticized in public for ignoring real-world evidence, is it?

Shapiro: “Oh, probably not. I’ve been doing this a long time.”

Van Nest: “As a matter of fact, you recently testified on behalf of the government as their chief economic expert in the government’s challenge to the AT&T/Time Warner merger; right?

FTC attorney Rajesh James: “Objection. This is outside the scope of the rebuttal testimony.”

Van Nest: “Your Honor, this is impeachment. It goes to his reliability and credibility as an expert. And he raised, himself, in his testimony, that he’s done this a lot, including for the Department of Justice, including in merger contexts, and that’s what I want to ask him about.”

The Court: “It’s overruled. Go ahead. You may continue.”

Van Nest: “Thank you. You were the government’s chief economic expert in the government’s challenge to the AT&T /Time Warner merger; correct?”

Shapiro: “Correct.”

Van Nest: “And you testified just last year, in March or April, in that proceeding; right?”

Shapiro: “That is correct.”

Van Nest: “And you presented a theory of bargaining leverage based on bargaining theory. You said the merger would allow AT&T to exercise too much leverage over content providers in future negotiations; right? Correct?”

Shapiro: “That sounds correct.”

Van Nest: “Um-hum. And the court there completely rejected your opinion as unreliable, not credible, and completely inconsistent with the real-world evidence; right?”

Shapiro: “I don’t — that sounds very strong. Judge Leon had a number of criticisms of mine. If you want to refer to his specific criticisms, let’s look at his opinion.”

Van Nest: “Let’s do it. It’s in your binder, United States versus AT&T is what the heading says, and it’s Judge Leon’s opinion of June 12th, 2018. And I’ll call your attention to page 113 of that opinion where Judge Leon said, ‘Unfortunately for Professor Shapiro, the facts adduced at trial regarding the real-world operation of affiliate negotiations demonstrated that his testimony “rests on assumptions” that are “implausible and inconsistent with record evidence.”’ He said that; correct?”

Shapiro: “I see that.”

Van Nest: “He went on to say, ‘Indeed, this opinion by Professor Shapiro runs contrary to all of the real-world testimony during the trial from those who have actually negotiated on behalf of vertically integrated companies.’ He said that as well?”

(A certain amount of scrambling around to find pages) …

Shapiro: “I see that, yes.”

Van Nest: “And on the next page he said, page 114, ‘one was left to wonder why Professor Shapiro turned a blind eye to such extensive real-world experience.’ He said that as well; right?”

Shapiro: “I see that.”

Van Nest: “And at the end of the day, … he concluded that ‘the evidence at trial showed that Professor Shapiro’s model lacks both “reliability and factual credibility” and thus fails to generate probative predictions of future harm associated with the government’s increased-leverage theory.’ That’s what the judge concluded in your opinion regarding the model; correct?”

Shapiro: “I see that.

Thus, Shapiro was roundly criticized by Judge Leon in the AT&T case for promoting the same types of arguments raised in this case.

One might ask why the government would rely so heavily on such a flawed witness, and the simplest answer is that the FTC chose Shapiro before the AT&T testimony. One could speculate that it would not have made the same choice had the systematic lack of empirical verification in his work been brought to its attention earlier.

The parties have made final arguments. Now, Judge Koh will ruminate, perhaps for an extended period, and render an opinion. Although it’s possible that the losing side will appeal the outcome whatever it is, and that is certainly true for Qualcomm, one wonders whether the FTC will try to carry this weak case any further if Judge Koh decides against it.

Written by

Technology Analyst

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store