Bitten by the Apple
It appears surprising at first sight. Apple, after having unilaterally set a 99-cent-a-song price at its iTunes store, to the chagrin of producers and the glee of its financial officer, decided to leave price setting to the providers when it ventured into the book and apps markets. Quixotic? Not quite. As it turns out, this business model, a fairly new thing known as the agency model, brings Apple the greatest profits. And, for us, higher prices.
How that exactly works is beautifully explained by CESifo Fellow Hans Jarle Kind and his colleagues Øystein Foros and Greg Shaffer in their latest CESifo Working Paper.
Under the agency model, the upstream firms (book publishers and developers of apps) choose the retail price of their products and the downstream firms (Apple, Google and Amazon, for instance) specify how the revenue is going to be split. Apple is known to apply a 70-30 split, whereby 70% of the revenue goes to the upstream firm and 30% goes to Apple.
The move is a bit puzzling. Given the market power of the likes of Apple, why would they cede control of something as fundamental as retail prices? Why not squeeze the upstream firms to the bone, leaving them only with the barest profit possible? And, given its success with the iTunes model, what moved Apple to adopt the agency model versus a more standard business format? Finally, how likely is it for other companies to follow suit and also adopt the agency model? The authors set out to answer all these questions from a competition-based perspective.
Assuming there is competition at both the upstream and downstream levels, they consider several known facts: Apple uses the agency model for books and apps, but not for iTunes, where it has remained in control of retail prices; Google and Amazon followed Apple in adopting the agency model so quickly that they raised red flags among competition watchdogs; and, lastly, the agency model is sometimes bundled with such ancillary contract provisions as most-favoured-nation clauses.
The authors then consider a setting in which the downstream firms decide retail prices, as against one in which the upstream firms do so, taking the revenue-sharing splits as given. They find that retail prices will be higher in equilibrium only if competitive pressures are lower upstream than downstream. This immediately answers the question of why downstream firms would cede control over something as fundamental as their retail prices: if competition among providers is weak, they manage to induce higher prices that way. This, by the way, is known as “strategic delegation”.
And that explains as well why Apple has done it for books and apps but not for its iTunes music: it faces far lower competition distributing the latter than it does with books and apps, where the fiercer competition is upstream and not downstream. As a Steve Jobs biographer cited by the authors quotes, “[Jobs] refused to offer the music companies the agency model and allow them to set their own prices. Why? Because he didn’t have to.”
Moving the Industry
At the time of its entry into the e-books market, Apple faced a difficult decision: Amazon was selling e-books at $9.99, in many cases several dollars below the wholesale price it was paying, so Apple could opt for either taking on Amazon directly by setting its own low prices, or attempting to convince the industry to ditch the wholesale model – where publishers sell their wares to retailers, who then set the price – and move to the agency model, in which control of the retail prices was ceded to the publishers, where competitive pressures were lower. Apple opted for the latter – and moved the whole industry in the process.
The authors’ findings may also shed light on Apple’s decision to adopt the agency model when the iPhone was introduced. While ceding control of retail prices to the upstream firms would lead to lower prices, Apple may have felt either that factors were present that outweighed the benefits of dampened competition, or that it would directly benefit from low prices, perhaps by fostering entrepreneurship and innovation among apps developers.
As regards the second question, namely why would upstream firms earn positive surplus in equilibrium instead of being squeezed for profits, the authors suggest that the upstream firms, despite having little or no bargaining power with downstream firms, can punish the latter if they extract too much by charging higher retail prices for the products sold by the higher-extracting retailer than by the rival downstream firm, provided of course that the substitutability of such firms in the eyes of consumers is high. In other words, that consumers have no preference regarding whom they buy books from, for instance, so long as the price is low.
The authors then examine the question of why the agency model was adopted so quickly by other retailers, such as Amazon. Was pressure involved? If industry-wide adoption arises naturally in equilibrium, their model would predict that no pressure was needed. What they found was that there would be no need to pressure firms into adopting the agency model when retail prices and industry profits are higher with the agency model than without, and competitive pressures upstream are strong enough. However, if upstream competition is very weak, then a prisoner’s dilemma situation might arise, such that Amazon, say, is tempted to deviate from the agency model even if this reduces joint industry profit. Collusive behavior or even direct threats towards Amazon could solve the dilemma. The US's Department of Justice looked closely for signs of collusion or undue pressure, and five publishers have, interestingly, decided to settle out of court. A District Judge further ruled last July 10th that Apple had led a conspiracy to raise e-book prices above those charged by Amazon. Apple will appeal.
Lastly, they consider the fact that Apple, when it entered the e-book market, used a retail most-favoured-nation clause (MFN) in its contracts. These clauses required that the publishers not set higher retail prices at Apple than the retail prices at other downstream firms, whether or not the latters’ prices were controlled by the publishers. In other words, book publishers were not allowed to set Apple’s prices higher than Amazon’s, even if they had no control over how Amazon set its own prices.
They find that if the rival firm declines to adopt the agency model, the MFN can lead to uniform prices that resemble the same outcome that would arise under industry-wide adoption of the agency model, making the rival’s decision irrelevant.
So, all in all, Apple’s bold move turned out to be quite a successful strategy. No wonder that Apple under Jobs became the most valuable firm in the world. St. Jobs, it appears, was not only a technical wizard, but a marketing genius as well. If only we weren’t saddled with higher prices as a result…
Hans Jarle Kind, Øystein Foros and Greg Shaffer: "Turning the Page on Business Formats for Digital Platforms: Does Apple's Agency Model Soften Competition?", CESifo Working Paper No. 4362
> Other CESIfo Working Papers by Hans Jarle Kind