"Prosocial," Output-Reducing Collusion

Thom Lambert —  27 May 2010

One of my antitrust students recently pointed me to a television commercial that could inspire a great exam question. Unfortunately, I didn’t see the ad until I’d finished drafting this semester’s antitrust exam (which I’ve been grading…hence the absence from TOTM).

The T.V. commercial trumpets an agreement among the members of the American Beverage Association — including soft drink giants Coca Cola, Pepsico, and Dr. Pepper Snapple Group — not to sell high-calorie soft drinks in schools. Interspersed among scenes of a basketball game, a track meet, and a math class in which students clamor, Tracy Flick-like, to impress their attractive young teacher, three delivery people from rivals Coke, Pepsi, and Dr. Pepper smile at each other as they cooperatively restock a school’s soda machines. The voice-over announces:

Competition. It pushes us to work harder. To be better. To win.

But sometimes, even rivals realize they share a common goal.

America’s beverage companies have removed full calorie soft drinks from schools, reducing beverage calories by 88%. Together with schools, we’re helping kids make more balanced choices every day.

There’s also a print ad, which shows our three rival delivery folks standing outside a school with loads of diet drinks, juices, teas, and bottled waters. Just above the logos of Coca Cola, Pepsico, and Dr. Pepper Snapple, the text reads:

Together, we’ve replaced full-calorie soft drinks in schools with lower-calorie choices.

America’s beverage companies have teamed up to remove full-calorie soft drinks from schools. And we’ve replaced them with lower-calorie and small portion-sized options like juices, teas and waters — reducing beverage calories by 88%. Together with schools, we’re helping kids make more balanced choices every day. Learn more at Ameribev.org.

The actual agreement among the beverage manufacturers doesn’t just remove sugary sodas from schools. It also limits the sizes of the “healthy” drinks sold at schools. In elementary schools, juice and milk drinks cannot exceed 8 ounces. In middle school, the limit is 10 ounces. In high schools, it rises to 12 ounces. Sixteen ounce offerings are forbidden in all schools.

How can this arrangement not run afoul of the antitrust laws? Section 1 of the Sherman Act forbids any contract, combination, or conspiracy (i.e., any agreement) that unreasonably restrains trade. The Ameribev pact amounts to an agreement among competitors to limit market output — i.e., to “restrain trade” that otherwise would occur. The only legal question, then, is whether that trade restraint is “reasonable.”

It’s not. Or, at least, not in any way that antitrust would credit. The beverage companies’ agreement not to respond to consumer demand involves no integration of research or production, facilitates the creation of no new product or service, and occasions no reduction of transaction costs. Its only apparent function is to reduce market-wide output, and its profitability depends on the power to control the market. It is, in antitrust parlance, a “naked” output-reducing agreement. Antitrust doctrine has long declared such agreements among competitors to be unreasonable per se and thus illegal.

The parties to the agreement would no doubt retort that their trade restraint is reasonable because it addresses a vexing social problem — childhood obesity. But if they took that tack, they’d really be arguing that unbridled competition produces such attractive product/price offerings that consumers (albeit, young ones) consume too much. In other words, they’d be saying that competition, doing its thing (i.e., lowering price and making product offerings maximally attractive), is itself unreasonable, so that it’s “reasonable” to constrain competition’s excesses via agreement.

The Supreme Court has declared that argument a non-starter. In my favorite of Justice Stevens’ many antitrust opinions (the last of which was issued this week and will no doubt generate some discussion on TOTM), the Supreme Court reasoned that competitors cannot render a trade restraint “reasonable” by arguing that unfettered competition expands market output so much that it leads to antisocial results. The case, Professional Engineers, involved an ethical canon prohibiting professional engineers from discussing prices with prospective clients (and thus from competing on price) until the clients had selected their engineers. The engineer defendants argued that their restraint was reasonable because “bidding on engineering services is inherently imprecise, would lead to deceptively low bids, and would thereby tempt individual engineers to do inferior work with consequent risk to public safety and health.” In other words, the engineers maintained, their agreement to limit competition was reasonable because unhindered competition would lower prices to levels that threatened antisocial results.

The Supreme Court would have none of that. The Court conceded that vigorous price competition might lead to the sort of undesirable, antisocial result the engineers envisioned:

It may be, as petitioner argues, that competition tends to force prices down and that an inexpensive item may be inferior to one that is more costly. There is some risk, therefore, that competition will cause some suppliers to market a defective product.

But the Court insisted that the engineers’ attempt to justify their output-limiting agreement on grounds that competition threatens public safety was “nothing less than a frontal assault on the basic policy of the Sherman Act.” It then explained:

The Sherman Act reflects a legislative judgment that ultimately competition will produce not only lower prices, but also better goods and services. “The heart of our national economic policy long has been faith in the value of competition.” … Even assuming occasional exceptions to the presumed consequences of competition, the statutory policy precludes inquiry into the question whether competition is good or bad. … In sum, the Rule of Reason does not support a defense based on the assumption that competition itself is unreasonable.

The beverage companies’ “Competition leads to socially undesirable over-consumption!” argument is therefore likely to fall on deaf ears.

But might the companies have another argument? They could contend that their refusal to sell certain products in schools helps them compete by projecting a socially responsible brand image that creates goodwill among customers. This argument is belied, though, by the coordinated nature of the trade restraint. If foregoing sales in schools could actually enhance a company’s total sales by creating a favorable brand image, then why must the beverage companies collectively undertake this endeavor? Wouldn’t a company make this move on its own, regardless of what its competitors did? Moreover, wouldn’t collective action actually undermine this image-enhancing objective? After all, if your competitors are taking the same “socially responsible” actions as you, then you don’t really gain a competitive advantage from taking those actions. The fact that the Ameribev campaign is coordinated — and proudly so — suggests that any “We’re enhancing output by improving our brand image!” argument is pretextual.

Let’s face it. The real reason the competing beverage companies agreed to this campaign is because they want to preempt more onerous regulation and/or “sin taxes” on sugary drinks. (See these articles from Advertising Age.) Unlike the purported goal of enhancing output by creating goodwill, this goal is best served by concerted, rather than individual, action. That’s because voluntary action to preempt more onerous regulation is subject to a collective action problem. Any firm that voluntarily cuts back its sales to forestall regulatory action will want to sacrifice as few sales as possible. Each firm also knows that in deciding whether to impose restrictions, regulators tend to look at overall industry trends. Each firm therefore wants its rivals to cut back a lot (so that the industry as a whole appears to be acting responsibly) while it cuts back only a little (thereby minimizing the cost of its preemptive strategy). If every firm has this attitude, though, the total voluntary reduction by the industry as a whole won’t be sufficient to prevent regulatory action. Thus, rivals seeking to forestall more onerous regulation need to commit to each other that they will each achieve specified reductions.

There is, then, an output-enhancing business purpose for the rivals’ agreement to cut back on sales: the agreement is necessary to prevent regulation that would mandate an even greater reduction in output. Should antitrust therefore conclude that the agreed-upon trade restraint is “reasonable”?

I’d say no. There are at least two problems with crediting the “We’re agreeing to reduce output in order to forestall more restrictive regulations!” argument. First, doing so encourages backdoor regulation. By threatening to impose harsh restrictions and encouraging the potential regulatees to adopt a “voluntary” output-reduction instead, regulators can effectively regulate without going through formal procedures, collecting public comments, etc. This will lead to sloppier, less-informed regulation. Moreover, insulating output-reducing agreements ostensibly aimed at forestalling regulation can create cover for collusion aimed at reaping supracompetitive profits. This seems to be what happened in Madison, Wisconsin when tavern owners agreed to eliminate drink specials for the purported purpose of warding off more onerous restrictions. If firms can insulate their output-reducing agreements (e.g., pacts to deny discounts) by saying that the arrangements are aimed at forestalling regulation, then collusion becomes much easier.

So what do others think? Can the Ameribev arrangement pass muster under Section 1? Should competing firms be allowed to agree to reduce output for purposes of avoiding more restrictive regulations?


Now, back to those pesky Bus Orgs exams — ugh!

Thom Lambert


I am a law professor at the University of Missouri Law School. I teach antitrust law, business organizations, and contracts. My scholarship focuses on regulatory theory, with a particular emphasis on antitrust.

4 responses to "Prosocial," Output-Reducing Collusion

    Dom Armentano 8 June 2010 at 12:38 pm

    As most of you know, I would, of course, allow an agreement by auto manufacturers to limit horsepower…or limit anything, so long as the agreement was not legally enforceable. But back to the specific matter at hand, is there not a fundamental difference between selling sportscars to consumers and selling soft drinks to (or through) a public school or school district? Is not the consumer in our soft drink example the school; I’m sure it’s the school that’s dictating the demand curve and, perhaps, even paying the bills. So, in fact, this is an agreement totally in compliance with the real demand side of the market; i.e., the SCHOOLS want output limited and the beverage industry trade association is complying with their wishes. How is that welfare reducing?


    I agree with Dom Armentato that it is not likely a court would see the agreement as plausibly pro-competitive. If consumers place higher value on the high-calorie drinks, an agreement to replace them with lower valued goods, in other words, an agreement not to compete in the supply of the high calorie drinks, is not immune from Section 1. And of course, one might expect an agreement by expensive sports car dealers to replace their cars with clunky used cars that sports car consumers want less to reduce output in the relevant market. But I will play the devil’s advocate from a different angle: is there an agreement here at all? I’ve seen a lot of press releases on the Guidelines, but nothing denoting the actual terms. The passing or collective endorsement of a set of “best practices” to which members of the industry can voluntary choose to adhere to or not is not necessarily an actionable antitrust conspiracy. Of course, calling something “voluntary guidelines” won’t immunize an actual agreement if it is there. But it seems that the parties were pretty careful — EXCEPT for in their commercials and in print!!! — to make sure to emphasize that the antitrust-relevant choices were made unilaterally. But I can’t imagine antitrust counsel would have given the thumbs up to the commercials…

    antitrust guy 8 June 2010 at 8:43 am

    Consumer choice is being reduced. Would we feel the same way about an agreement among auto manufacturers that all cars should get certain MPG? What if they agreed instead to limit horsepower? Or what if auto dealers agreed to limit their hours?

    On point 2, while smaller beverages are less expensive, it is rare for a smaller package of any product to be less expensive, ounce for ounce, than a large one. Unit costs generally decrease as package size increases (although, yes, some food producers rely on this consumer assumption and price the largest boxes higher per ounce than the mid-sized box–check cereal prices next time you shop). But ounce for ounce, typically the cheapest soda comes in a 2 liter bottle; the most expensive in a 12 oz (or less) can.

    On point 3, go back to point 1: Such benefits are often touted by cartelists: “We’re stabilizing the market!” “We’re ensuring consistent prices!” “A diamond is forever!”

    If the agreement isn’t naked then there’s sufficient nudity that anything more than a quick look is prurient.

    Don Armentato 8 June 2010 at 8:42 am

    Let me play the devil,s advocate and argue that the “agreement” does not appear “naked” to me, nor should it to the courts. Naked agreements reduce output and raise price and promise higher profits without any consumer benefits or cost savings. First, it’s not obvious that output will be reduced: certain high calorie drinks are being REPLACED with low calorie drinks…no obvious output restriction. Nor does limiting drink size (but not the number of drinks)automatically restrict output. Second, how are prices increased by any of this? Indeed, one would assume that the the smaller drink sizes are associated with lower prices; price per ounce may actually decline. Third, the “consumers” presumably benefit by the healthy switch and so don’t the school district’s (under pressure to do something about obesity). Clearly, then,there are some consumer “benefits.” Finally, we are not privy to the economics of producing the different beverages. There may well be “cost savings” associated with the low cal drinks and the move to smaller cups. In sum, if output is NOT in fact restricted and prices NOT increased, and if there are ANY consumer benefits or ANY cost savings, then the agreements are, by definition, reasonable. Do I think that any court would see it this way? Very unlikely.