Two-tier competition: putting an end to oligarchies

Johan Hombert, Professor of Finance - May 15th, 2012
Communication towers

How do you get competitive markets when firms that sell an end product control the means of production (as is the case with the heavyweights of mobile telephony in France)? For Johan Hombert and his co-authors, the solution cannot be reduced to simply forcing integrated firms to provide the means of production to other isolated firms.

Johan Hombert ©HEC Paris

Johan Hombert joined HEC Paris in 2010 after working for INSEE and teaching at ENSAE (National School of Statistics and Economic Administration), from which he himself graduated. (...)

See CV

Johan Hombert and his co-authors looked at competition in markets where firms are vertically integrated: that is, they control both the production input — the mobile network they own in the case of mobile telephony — and the final product — the mobile plans they propose to consumers. “We had the telecom industry in mind when we began this workaround 2007, when high-speed Internet was developing in France, because many questions about regulation and competition policy were being asked: How do you develop telecoms and ensure that prices are not too high for the end-user?” says Hombert. Back then, and even now, the market was oligopolistic, with a limited number of players. But the same can be said of numerous other industries.


The sector that best illustrates this research remains that of mobile telephony in France. Before the arrival of Free in late 2011, there were just three operators that owned network infrastructure and mobile licenses: SFR, Bouygues Telecom, and Orange. “What the regulator wanted to do for several years, and what the European Commission also pushed hard to get, was to have more competitors. But entry in this type of market is very expensive because you have to develop a new network infrastructure, put up antennas, etc. So the regulator's idea was to bring in new firms without asking them to build their own network. It was about forcing existing operators to rent their network to new firms and thereby create competition without investment.” And so MVNOs (Mobile Virtual Network Operators) like Virgin Mobile, Auchan, and Budget were born.


The work of Hombert and his co-authors shows, however, that this solution does not lead to competition in the true sense. “Existing operators have little incentive to help their competitors produce goods that will compete with theirs,” he says. Vertically integrated firms get their profits from two sources: selling mobile plans to customers, and providing MVNOs with access to part of their network. “There will therefore be permanent tension. On the one hand, they will want to lower prices on plans to capture more market share, but in doing so, they will kill MVNOs and reduce their other source of profit. They can also try to make a lot of profit by renting their network to a maximum number of competitors, but in doing so, they increase competition in market pricing and reduce their profit in that regard. In fact, every time integrated firms want to be more aggressive on one of the two markets, they negatively impact their other source of profit. In the end, they do not want to decrease their prices on either of these two markets.” This partly explains why the 30 French MVNOs have never succeeded in getting more than 6-7% of market share.

Existing operators have little incentive to help their competitors produce goods that will compete with theirs.


What are the alternatives? The most brutal one is imposing a fixed rate on mobile operators for network rental to MVNOs. “It is very difficult to implement in mobile telephony because the regulator is subject to serious pressure from major operators,” says Hombert. It is also possible to put a ceiling on prices—not too low to discourage investment, and not too high to enable MVNOs to enter the market. “This is also difficult because the regulator is not in touch with fees for network maintenance, upkeep, and development, which complicates the matter.” A third solution involves forcing integrated structures to break into two, as was done in other industries, notably rail, where Réseau Ferré de France rents its tracks from SNCF and other companies. “SNCF would never have agreed to rent its network to competitors, or only at a very high price,” adds Hombert. “And it is even more complicated in mobile telephony; there are more synergies between mobile plan sales and network operations.” The latest solution, and it is precisely the one that has been retained on the market, is the arrival of a new competitor that owns a network infrastructure, even a small one. Although it rents part of its network from Orange, Free covers a large part of the territory with its own network, as opposed to 0% for MVNOs, explains Hombert. “We are seeing prices, which were held at a very stable level, drop rapidly,” he says. “What is going on confirms the predictions of our model: Only the presence of an operator that is not entirely dependent on its competitors creates a situation of real market competition.”

Based on an interview with Johan Hombert, professor of finance, and the article “Upsteam competition with vertically integrated firms” (Journal of Industrial Economics,  vol. 59, no. 4, December 2011, pp. 677-713), co-written with Marc Bourreau, Jérôme Pouyet, and Nicolas Schutz.


The work of Hombert and his co-authors shows that isolated analysis of the upstream market (networks in the case of mobile telephony) can be misleading and that states or the European Commission must also take account of downstream market features (in this case, mobile plans) to determine if competition will be effective, and what solution (fixed price or capped, structural separation of integrated firms or entry of new competition) will be the most appropriate.


The researchers developed a model of two-tier competition between vertically integrated firms and unintegrated downstream firms. They thus showed that when integrated firms compete in prices to offer a homogenous product, the Bertrand logic (which regulates situations of competition where the strategic variable is price) may collapse, and that the price may be above marginal cost in equilibrium. These partial closure equilibria are more likely to occur when downstream competition is intense or when competitors are relatively ineffective.