Price signalling and competition

by

Lots of people are getting upset about price signalling. The ACCC has warned the banks about price signalling. The opposition is planning a private members bill to outlaw price signalling. The government have concentrated on bank exit fees but I am sure they will get around to price signalling sometime soon.

What is missing in this rush to change the law is some balance. Communicating prices to customers is a key part of competition. The public statements that are used to ‘signal’ prices to competitors also keep customers informed. Unless laws about price signalling are very carefully drafted, they can reduce competition and hurt consumers.

When businesses inform consumers about their prices they also inform their rivals. Competitors may signal their rivals by announcing their intentions to change prices or even by the way they change prices. For example, firm A raises its price. Its rivals (as well as its customers) see this price rise, and if the rivals don’t respond firm A lowers its price back below the original level. This harms its rivals (but helps the customers). After a short while, firm A again raises its price to the higher level. It doesn’t take long for the rivals to get the message: “Raise your price to the new higher level or I will punish you through a period of deep discounting”.

This is price signaling. But how do we make it illegal? Do we ban advertising of prices? That would be incredibly harmful to consumers. It would make it hard for consumers to comparison shop and will tend to lead to less comeptition and higher prices. So that seems like a bad idea.

Should we ban temporary discounts? This will also tend to hurt consumers. No weekly specials? Again the harm to customers probably outweighs any benefits.

Should it be illegal for firms to publicly announce future price rises? I can also see how this will harm customers. For example, suppose the Aussie dollar has crashed and I am an importer. I want to compete hard by telling consumers to buy from me today because the change in the exchange rate has pushed up my costs so my prices will be rising. Consumers would be harmed if competition of this sort were ruled illegal because of speculation about price signalling.

The problem is that it is generally impossible to separate out harmful public ‘price signalling’ from desirable public price competition. Get the law wrong and you hurt consumers.

So what should the government do? While public information exchange on prices should be legal, private exchange of prices (including existing prices) between competitors strikes me as pretty dubious. It is hard to think of a situation where one business legitimately has to be informed of a competitor’s price in a way that is not available to the public or to customers. So this is probably a good place to start if the government or the opposition want to tighten our laws on facilitating practices. But it will not (as I understand it) effect the banks.

For those interested in further reading, the classic US case on price signalling is the 1984 Ethyl case. The FTC brought an action against four chemical companies that it claimed (among other things) were using press notices of price changes to facilitate tacit collusion. The FTC lost. George Hay has a nice summary of the case in The Antitrust Revolution (although you need an earlier edition. The third edition has it, but it seems to have gone from the latest edition). George has a nice example of the difficulties of separating out public price information and price signalling. I have reproduced it below.

Finally, before I get some comments praising the idea that ‘all price rises must be cost justified’, forget it. That would effectively mean regulating all prices in the economy and would kill all competition. We rely on competition to set prices because we know that price regulation is usually inefficient and costly. So don’t even go there!

Over to George:

The problem can be illustrated with a simple example. Consider two essentially identical gas stations selling private-brand gasoline directly across the road from one another on a long, isolated highway with no rivals for miles around and no likelihood of new entry. Prior to today, both stations have been charging $1 a gallon, which we assume to be the competitive price. Since the two stations are regarded by consumers as offering virtually identical products, we assume that, at the $1 price, each enjoyed 50 percent of the total gasoline sales.
One station (station A) is contemplating raising the price to $1.25, which its resident economist has computed to be the profit-maximizing monopoly price: the price that would maximize total profits if both stations were to charge the identical price. The scheme will be profitable for A only if station B matches the $1.25 price, since if A raises the price and B does not follow, all of A’s customers will switch to B.
At first glance, it appears risky for A to go ahead with the price increase without prior assurance (i.e., agreement) that B will follow, since a possible strategy for B is not to match A’s price increase and to capture all the business. However, if A raises the price to $1.25, it will learn very quickly whether B has followed, since the same signs that communicate B’s price to potential customers can be seen by A. If B has not followed the increase, A can promptly restore its $1 price. In the interim, A will have lost some sales, but as long as it is able to detect B’s price cutting and responds immediately, the losses will not be too great.
Station A’s ability to detect and respond quickly to B’s failure to match A’s price affects B’s optimal strategy. By keeping the price at $1, B would enjoy a brief period when it would capture nearly 100 percent of the business. However, once A retracted its increase, the firms would go back to sharing the market at $1. If instead B were to match A’s increase (which it can do easily by observing A’s prices directly), it forgoes the brief period of extra business in return for a possible long-term equilibrium in which it shares the market with A at $1.25. Hence, B has an incentive to go along with an increase by A, and A, knowing that B’s best strategy is to follow, can initiate the price increase with minimal risk.
If the scene unfolds as described, the firms jointly achieve monopoly profits without the need for any formal agreement of the kind usually involved in a Section 1 price-fixing case. What we want to consider is whether it is feasible to bring the situation within the reach of the Sherman Act by labeling it a tacit or implicit agreement and arguing that the antitrust laws ought to extend to such agreements.
To see the difficulty of applying the Sherman Act to such conduct, consider how one would explain to a judge or jury precisely what A did that is objectionable, or how one would fashion an effective remedy. It seems foolish to argue that A should have ignored the fact that B would likely follow an increase by A. This is essentially requiring A to pretend that it is in perfect competition rather than in a situation of duopoly. What the argument comes down to, then, is that A should have kept its price at $1 even though it knew that $1.25 would be more profitable.
But if that is the argument, what is the criterion for determining when A has acted illegally? Is it simply the fact that B followed A’s price increase? Surely not, since that would essentially make any price increase that is not subsequently rescinded (because rivals failed to follow) illegal. Can A never initiate a price increase without risking a violation of the antitrust laws in the event its rivals follow? (Of course if rivals do not follow, the price increase is useless to A.) Would we interpret the law to mean thatA can increase prices only when and to the extent that costs go up?
Similarly, it is difficult to argue that B committed an antitrust violation when it matched A’s increase, knowing that if it failed to match, A would simply retract the increase. Can B ever follow an increase by A without itself violating the antitrust laws? Can it follow A’s increase only if it is cost justified? (Is the cost justification based on A’s costs or B’s? Or B’s estimate of A’s cost?) Finally, if the relationship between price and costs is to be determinative, are the federal courts equipped to carry out this regulatory function? These questions suggest the great difficulty of attempting to apply the antitrust laws to classic oligopolistic behavior.

7 Responses to "Price signalling and competition"
  1. I should think that the ACCC is objecting to price signalling between banks on the grounds that it is facilitating something like cartel behaviour. I don’t understand from what you’ve written what causes signalling information to sometimes encourage competition and other times encourage collusion, though the paragraph about publicly and privately traded information was clarifying.
    While this seems a very persuasive case for treating price signalling with less heavy hands than government regulation, it would be very helpful to have an explicit (and, if one might be so bold, concise) alternative suggestion—unless you really do believe that banning (however that might be enforced) privately traded information is the only remedy needed.

  2. On the alternative, there is no explicit and concise alternative that covers every situation. There will always be a gray area. But this is common under the Trade Practices Act. For example, low pricing is competitive and helps customers. Predatory pricing is illegal. But how do you distinguish the two?
    So I take the conservative stance of not pushing the law to the point where it is likely to do more harm than good. I think that point is the distinction between public and private communication of price information. This will also lead to ambiguity and hard court battles. And I am not going to try and draft up a suggested version of the section for the Act. However, I suspect it is the best we could do.

  3. As a non-economist, the answer to George’s problem seems simple: We need a C, and a D.
    The problem is that the two Service Stations share a duopoly – They can afford to raise prices because they operate like an effective monopoly. Introducing more petrol stations would make the collusion harder and harder to rely on, until we’re back at a market optimum price.
    The question we need to ask is: What are the structural barriers to entry in the Australian mortgage market? How can Government policy help to reduce them? Let a thousand flowers bloom and you’ll find those 2-2.5% spreads head down to an efficient price.
    (This understanding based entirely on a “The Intersection of Economics and Law” course I took 🙂 )
     

  4. @Anthony: Reducing barriers to entry is the answer but not necessarily easy to implement in practice for the banking sector. However all this on emphasis on ‘banking competition’ and ‘price signalling’ goes to show how politicians are more interested in ‘sexy’ (i.e. electoral friendly) but peripheral than more substantive issues like the implicit guarantee of the banking sector.

  5. I’d really like to see Australia Post enter banking itself. Its a drastic step but it would easily provide competition across the board and keep some degree of fairness in personal & small business banking.
    But then again I’d like to see the ABC publish a weekly or monthly newspaper focusing on policy detail and real news, maybe I’m some crazy socialist.

  6. Stephen,
    The issue that the ACCC has with the banks is signalling prospective price increases.  I can’t see that George Hay’s analysis relates to prospective price increases.  Indeed, if Station A could signal that it was considering a price increase and then Station B signalled that it was considering one also, this would make collusion much easier.
     
    There may be benefits from continuing to allow retailers to announce limited (in stock or time) discounted sale prices.  Presumably, your importer example could come under this exemption.  I don’t think banks could utilise that exemption as a loophole.
     

  7. Banks know who their customers are, so they can just notify them directly of prospective increases, if that is indeed the purpose of these announcements. My own view is that banks couldn’t care less about giving customers a heads-up on where interest rates are going (at least if they are going up). There is simply nothing in it for the banks. It is pretty clear to me the banks are talking to each other with these pronouncements.

%d bloggers like this:
PageLines