Defendant s Analysis of the Profitability of Price Increases and the Detection of Collusion

Transcription

1 Defendant s Analysis of the Profitability of Price Increases and the Detection of Collusion Presented to ABA Section of Antitrust Law Joint Conduct and Economics Committees February 16, 2016 I. Introduction As Scott mentioned, one of the motivators for this program was the Text Messaging Antitrust Litigation, in which the defendants financial analyses of the price increases that were at issue played a large role in the defendants successful summary judgment motion. So we thought we d describe how those financial analyses were used by the defendants, the attacks that plaintiffs made on that use, and how the courts reacted. II. Text Message Facts Let s start with the facts. The case concerned just one pricing option for buying text messages the socalled single use price, also known as the pay per use or PPU price. Most of us pay for text messages under a bundled plan, where we buy either a specific number or an unlimited number of messages for, say, $20 a month. The case did not directly concern the pricing of those plans. Instead, it concerned what a customer pays if he doesn t have a plan and so, pays for messages one-by-one. Pay per use. The case then focused primarily on two rounds of allegedly parallel PPU price increases. 1 The first began when Sprint increased its PPU price from 10 to 15 cents in October 2006, and over the next 7 months, AT&T, then Verizon, and finally T-Mobile matched that 15 cents price. In October 2007, Sprint again increased its PPU price, this time from 15 to 20 cents. The others followed over a 10-month period. 1 There was earlier round of increases by Verizon and T-Mobile: Verizon increased its incoming PPU price form $0.02 to $0.10 in April 2005 and T-Mobile increased its price from $0.05 to $0.10 the following March, about a year later. At that point, all carriers charged $0.10 for both incoming and outgoing text messages.

2 Discovery showed that each company had done financial analyses of each increase. For example, before AT&T followed Sprint s move to 15 cents, it prepared an analysis that estimated the revenue increase, adjusted by (1) how many customers would migrate to a bundled plan in response to a PPU price increase, (2) how many customers would simply turn off the text message feature, and (3) the likely reduction in PPU messages as a result of the price increase, i.e., usage reductions due to price elasticity. Importantly, defendants emphasized in their briefs that each analysis showed that the increase was in the company s interest, regardless of what other companies did. That was fundamentally because the analyses assumed that customers would not change carriers because of a PPU price increase. The vast majority of PPU customers sent very few text messages, and so even a 50% increase, from 10 to 15 cents, would increase their monthly bill less than $1.00. And if they did send enough messages to care about a nickel increase, they would switch to a bundled plan before they would go to the time and bother to switch carriers. In the industry, when a customer changes carriers, it s called churn, and the financial analyses assumed little or no churn. Now, these analyses were considered in the context of plaintiffs allegation of a CEO-to-CEO conspiracy among the four national wireless carriers. However, plaintiffs conceded that they had no direct evidence of the alleged conspiracy, with one exception that became a focus in the case. III. Financial Analyses / Matsushita As most of you probably know, when the defendants moved for summary judgment, the plaintiffs argued they had direct evidence of a conspiracy in the form of an by a T-Mobile employee that called the last T-Mobile increase collusive and opportunistic. That s a discussion for another day, but ultimately, the district court and Seventh Circuit found that the was not direct evidence. A. Matsushita Standard and Financial Analyses And so, in the absence of direct evidence, the summary judgment motion was decided under the Matsushita standard, which requires a plaintiff resisting summary judgment to present evidence that tends to exclude the possibility that the alleged conspirators acted independently. 2

3 Defendants argued that the financial analyses vividly demonstrated each company s independent business judgment, so that plaintiffs could not meet their Matsushita burden. Defendants made some related arguments. For example, citing the 7 th Circuit law, we argued that the analyses were so strongly suggestive of independent action that Plaintiffs burden to show that the inference of conspiracy is reasonable in light of the competing inference of independent action is correspondingly great. Then, we argued that the analyses made the inference of conspiracy unreasonable because after all, why would the defendants spend the time and effort doing these analyses if the conclusion was preordained by a conspiracy? Finally, anticipating plaintiffs attacks on the analyses, we cited the law that says that defendants invoking the Matsushita standard don t need to show that their judgment about their independent economic interest was in fact correct; they need only show that they acted in a manner consistent with their perceived unilateral interest. In the words of the Citric Acid case, business judgments should not be second guessed. B. Plaintiffs Attacks Plaintiffs had three main responses to defendants reliance on the financial analyses. 1. Unexplained Reversals of Position First, they pointed out that some of the defendants (including AT&T) had considered increasing their PPU prices earlier, decided not to, but then changed their mind only after Sprint made the first move. This, they said, was inconsistent with the argument that PPU price increases would be profitable regardless of what other carriers did, thereby undercutting the analyses as evidence of independent action. Defendants responded that there were no unexplained reversals of position. In AT&T s case, for example, the documents showed that it didn t take the price increase earlier because it was concerned about bad PR, and when Sprint took an increase, AT&T waited to see if there was any PR blowback; when there wasn t, it 3

4 took the increase. So its reversal of position was explained by its unilateral analysis. 2. Why Just Follow? Second, plaintiffs argued that if PPU prices can be profitably raised unilaterally, as these analyses claimed, why did all the carriers simply match Sprint s increase? When Sprint went to 15 cents without any loss of customers, why didn t AT&T or Verizon go to 17 cents or 18 cents? And if the analyses were right, why hadn t there been any price increases after the lawsuit was filed? Defendants responded that they had only considered matching, and not exceeding, Sprint s increases to avoid bad publicity. As for the absence of post-lawsuit price increases, the number of PPU customers as compared to customers on a bundled plan had shrunk so much that there wasn t much revenue left in a PPU price increase, and the lawsuit itself chilled further increases for such little benefit. 3. Senior Management/Top Down Third, plaintiffs focused on the fact that the analyses assumed little or no churn. They argued that this was circumstantial evidence of a conspiracy. In their view, the analyses assumed the company increasing its price wouldn t lose customers to competitors because it knew in advance that other carriers would follow the increase, giving customers no reason to switch. As a result, they said, the financial analyses were shams, designed to cover up the CEO-to-CEO agreement to raise PPU prices. In response to this sham allegation, defendants pointed to evidence (1) that there was nothing unusual about the analyses they used the same methodology typically used; (2) that there was no evidence that senior management meddled in the analyses or directed a result; and (3) that there were good economic reasons why the analyses assumed no churn. Judge Kennelly in the district court came down hard on plaintiffs sham analysis argument. He pointed out that if the analyses were shams to cover up a CEO-to- CEO conspiracy, you d expect to find some evidence of top-down direction, but 4

5 there wasn t any. Moreover, given the dozens of people involved in conducting, reviewing, and debating the analyses, he said, Plaintiffs theory requires the inference of a vast, multifaceted, highlycoordinated conspiracy dedicated to creating a Potemkin village of seemingly independent analyses to lend paper support to each collusive pricing move. This inference is unreasonable and unsupported. IV. Seventh Circuit Opinion Now, as you probably know, Judge Posner s opinion on appeal focused mostly on the T-Mobile and on the difference between tacit and explicit collusion and why tacit collusion is legal. As for the financial analyses, the plaintiffs on appeal backed away from alleging that the analyses were outright shams and argued with more restraint that the analyses had sufficiently clear flaws or questionable premises that a jury could find they were post-collusion justifications for the increases. I think it s fair to say that Judge Posner focused on the analyses key assumption about churn, to see if it made economic sense in the context of this market. Concluding that it did meaning that the market behaved as one would expect, without a hypothesis of collusion he had no trouble affirming summary judgment. But toward the end of the opinion, he took defendants to task for overstating their case with respect to the financial analyses. First, we had emphasized that each defendant made an independent profitability analysis, and Posner said, in essence, So what? one would expect such independent evaluations even if the firms were expressly colluding, as the independent evaluations would disguise what they were doing. So, when Judge Kennelly was confronted with the argument that the financial analyses were all shams, he focused not so much on the economic merits of the analyses, but on the improbable and factually unsupported story line of a vast, multifaceted, highly-coordinated conspiracy. J. Posner, on the other hand, acknowledged that financial analyses could be cover-ups, but focused on why, in this case, the analyses underlying assumptions made economic sense. 5

6 Second, you ll recall that defendants argued that the analyses showed that it would be profitable for each company to increase its price regardless of whether the other carriers did the same. Posner called that argument overkill, because even if the analyses did assume others would follow, that by itself would be evidence only of tacit collusion, which is perfectly legal, as he spent most of the opinion explaining. So an assumption that competitors will follow an increase does not suggest that the analysis was a sham or that the defendants colluded. 6