Many courses in economics begin with the first welfare theorem. This theorem states that a decentralized market, if left to its own devices, will reach an allocation that cannot be changed to further improve the economic welfare of all market participants. A proof of this result is as simple as it is intuitive: If there exists a desirable exchange of resources, the presence of market-clearing prices alone will provide sufficient incentive for this exchange to occur. A frequent retort to this insight has long been to criticize the concept of a decentralized market as one without practical merit. Evidently prices are the outcome of strategic deliberation conspicuously absent from the theory. If only we understood markets through the lens of game theory and accounted for the presence of market power, our idealization of free markets would be grounded, and a more realistic understanding of the economy revolving around market power would take hold. What should we make of this critique?
I will use this post to introduce you to a thesis by Hart and Tirole (1990), which is also one of my favorite exercises in teaching. The point is that, in quite a few instances, market power can be self-limiting, and the deviations from the welfare theorem are not as significant as its sceptics make them out to be.
Willy Wonka as an Upstream Monopolist
To illustrate Hart and Tirole’s model, let’s consider the monopolist Willy Wonka, an upstream firm supplying inputs, i.e., chocolate, to downstream competitors. Willy Wonka’s objective is clear: limit chocolate sales to increase prices and find a way to appropriate the downstream firms’ revenue. can achieve this by making a take-it-or-leave-it offer to downstream firms—buy the desired number of bars at the monopoly price in the consumer market or none. In this scenario, Willy Wonka’s market power is a true evil, depriving the world of delicious chocolate through engineered scarcity. But is this realistic? Hart and Tirole do not think so. If the price of chocolate bars is high, Willy Wonka has every incentive to secretly sell yet another bar of chocolate to another downstream firm. The additional output will lower the market-clearing price of chocolate bars. However, that loss accrues to the incumbent downstream firms, as they remain bound by the initial take-it-or-leave-it offer, while the additional revenue goes straight into Willy Wonka’s pockets. The catch is this: rational downstream firms anticipate Willy Wonka’s incentive to deceive. They adjust their output expectations up to the point where downstream firms’ output is the same as under standard quantity competition when each firm maintains full control over inputs: Cournot competition. In effect, Willy Wonka’s temptation to deceive under secret contracting takes away his ability to exercise market power on the downstream market. Consumer prices are solely determined by the degree of competition between downstream firms; whether there is an upstream monopolist or a perfectly competitive market for inputs is immaterial as far as consumer welfare is concerned.
A Counterpoint? The Case of Bloomberg LP
The model shows that secret contracting creates a commitment problem for the upstream monopolist. This takes away the upstream monopolist’s ability to exercise market power. Since most non-consumer prices are negotiated in secret, a policy stance could be to be unsuspecting of market power in upstream markets. In my own reading, I would not go as far. Instead, I suspect that in many instances where commitment to prices may be in doubt, upstream monopolists go to great lengths to re-establish price transparency.
Bloomberg LP is a most famous example of an upstream monopolist with a plausibly transparent commitment to high upstream prices. The vast majority of its profits come from renting the Bloomberg terminal, a data and analytics portal, to banks and investment companies (the downstream firms). If Bloomberg could not commit to high fees, the extent of bank competition alone would determine the profit Bloomberg makes. In this scenario, only an increase in banking consolidation could diminish Bloomberg LP’s temptation to further undercut its own fees at the expense of its existing customers. Incidentally, Bloomberg increasing the fee of its terminal following the 2008 financial crisis is evidence in favor of this more benign view that upstream market power is self-limiting. What speaks against this benign view is that the company maintains, by Bloomberg’s own admission, one the of most transparent pricing policies. And transparent pricing, following the logic of Hart and Tirole’s paper, is exactly what allows upstream monopolists to exercise market power in the upstream market:
“Want a special deal? If we give it to you, how could you be sure we’re not giving your competitor an even better one? To publish one price and then negotiate secretly with those you want to favor or those who complain the loudest—that just encourages confusion, dissent, uncertainty, and unpleasantness, not to mention what it says about the seller’s ethics. Favor one client over the other and, when the roles reverse, the favored client forgets and the initially disadvantaged one remembers. At Bloomberg, we have a published price and generally stick to it.” — From Bloomberg by Bloomberg
Questions for Regulators and Businesses Alike
Many more points can be raised. Let’s not. You can read Hart and Tirole’s insightful paper. The comments are open. Let me know what you would like to read about. And also whether you can deduce the answers to the questions below (apologies for the academic tone of this ending…):
A Speaking of welfare, should Willy Wonka be allowed to sell chocolate directly to consumers? Does Willy Wonka have an incentive to do so?
B What does the fact that Bloomberg LP never entered the investment banking industry say about the extent of its commitment problem over the pricing of the terminal?
C If the benign view of the self-limiting role upstream market power is accurate, what kind of regulatory interventions are harmful to consumer welfare?
