Is telecommunications a two-sided market? This post argues that it is. It also explains why this characteristic is central to justifying the change in the funding model for future access networks that operators are calling for.
But let us start at the beginning. The concept of “two-sided markets” emerged at the start of this century in the field of Economic Theory. It quickly attracted interest due to the change of perspective it proposed. In contrast to the traditional approach that characterizes an economic transaction as a relationship between two parties, seller and buyer, the focus shifts to a third party whose activity consists in making these transactions possible. The distinctive question that the new economic models seek to answer is which of the parties they bring into contact, or “sides” of the market, should be charged the most to pay for the services of the intermediary. More specifically, the focus is on how the pricing structure will influence the total volume of transactions that the intermediary makes possible and thus the value it is able to create.
In short, theory tells us that the existence of an intermediary is not enough for a market to be considered two-sided. It is also necessary for the intermediary to be able to generate value by changing the “side” of the market to which it charges a price – the seller or the buyer. This is the link between these new economic models and the recent “OTT fair share” debate on whether content providers – one side of the market – should pay to use the telecommunications networks they use to interact with end-users – the other side.
Telecoms have always been associated with two-sided markets
From the beginning, telecommunications were naturally associated with the new theoretical models. Thus, the seminal articles by Rochet and Tirole already included references to the role of network operators as managers of intermediation platforms that can influence the volume of transactions with their pricing structure. First between individuals wishing to communicate via voice calls, and later between providers and consumers of digital content.
Despite the constant change in the information they carry and the players it interrelates, telecommunications networks have not lost the characteristics of a two-sided market. This is acknowledged in a recent article by B. Jullien and M. Bouvard, which explores the effects of changing the current structure of payment flows on the Internet by introducing a remuneration for the delivery of content to end-users.
It is not an easy road to travel. Still, we should not let the difficulty of implementing it prevent us from seeing the potential benefits of regulatory intervention to fix the problems of a flawed model.
The premise proposed by Jullien and Bouvard is therefore interesting. They assume that the operator is free to charge the content provider a price for the delivery of content, which may even vary depending on the provider or the type of content. Starting from that point, their main – and important – conclusion is that under plausible circumstances that reflect the current reality, a new payment for content delivery would improve social welfare.
Impact on prices and consumption of connectivity and content
It would be naïve to think that nothing would change in final prices if digital service providers had to start paying for content delivery. A first effect to consider would be the possible transfer of this increased cost to the content consumer, in the form of higher prices or advertisements and lower traffic per unit of content (and therefore quality). In response, the consumption of content and the number of users willing to connect to the telecommunication networks would decrease.
Simultaneously, there is another force operating in the opposite direction. The network operator would have a new incentive to increase the number of people connected, and willing to demand content. This is because each additional connected customer would bring with it new incremental revenues from content providers. Given this new incentive, the operator will lower the price of connectivity to the end customer, and thereby induce more content consumption.
The relevant issue for the end-user and social welfare is which of the two transmission channels dominates. If the effect on the connectivity side dominates the effect on the content provision side, there will be an increase in consumption of both connectivity and content. The measure will then potentially be positive even for content providers. They would have to make a payment for distributing their content. However, in exchange, they will be able to enjoy a larger base of potential consumers of their content.
It is efficient to incentivise content providers to save traffic
The interesting and innovative aspect of Jullien and Bouvard’s article is the identification of two realities that explain why the impact of the measure is positive. The first one is the preference of end-customers for flat rates and their aversion to variable payments based on traffic. Essentially, for convenience and lack of knowledge or time to investigate the impact of their content consumption on the amount of traffic generated.
However, it is easy to understand the impact of flat-rate tariffs on networks. You only have to browse any website to see how there is increasingly more audio-visual content instead of images and text. Not only in the content itself, but also in the advertisements. It is equally illustrative to see how streaming video providers generally send their content at the highest possible quality. They do not consider that by loading the network with more traffic, they can generate congestion and negatively affect other users.
Jullien and Bouvard argue that under these conditions, it is preferable to establish incentives to minimise traffic by charging a variable price to content providers, rather than end-users. End-users would react by disproportionately reducing their consumption or by demanding a substantial reduction in the price of connectivity from the operator. On the contrary, content providers are well positioned to appreciate the value to the end user of an increase in the number of bits. A variable payment would provide them with the right incentives to adapt picture quality to screen size, or to introduce better compression techniques.
In summary, the new model would result in a more rational use of the network, with cost savings. This would translate to the end-user in the form of lower prices for connectivity and a higher number of households connected. Similarly, it would benefit society as a whole in the form of lower energy consumption per unit of content transmitted.
Indirect use of advertising to finance connectivity
The second key parameter identified by Jullien and Bouvard to conclude that a price for content delivery is positive for society is less obvious, but equally important. It is what they call return on advertising. Intuitively, the idea is that when advertising is very effective and the user is not excessively bothered by ads, it makes sense from both the consumer’s and the advertiser’s point of view to finance with advertising not only content, but also connectivity.
The new price for traffic delivery is a way to channel that subsidy from the advertiser to the end customer, using the content provider as an intermediary. The final result will be an increased consumption of connectivity, content and advertising, which will benefit consumers and potentially all components of the value chain, including the content providers themselves.
Addressing the risk of sub-optimal investment in access network capacity
Beyond the impact on final prices and consumption, there is another relevant mechanism through which the new payment can increase welfare, and whose study would nicely complement Jullien and Bouvard’s work: the creation of incentives to invest in access network capacity.
When calculating the incremental benefit of increased capacity, the operator only takes into account what it estimates end customers will pay for it. It does not consider the greater profits that would accrue to content providers. In economic terms, there is a positive externality that the operator does not incorporate into its investment decision. In other words, it is a plausible paradox that content providers might not be able to sustain their growth due to not being able to find a way to finance the required investment in access network capacity. A payment for content delivery would address this market failure, improve the expected return on investment and induce operators to provide more capacity.
Conclusion
The provision of services by telecommunication operators constitutes a two-sided market. There are strong arguments to explain how charging a price for the content delivery would lead to an increased consumption of connectivity and content. This would benefit not only the end user, but also content providers under realistic conditions.
European operators are far less profitable than their North American or Asian peers, and the ambitious network investments Europe aims for are far from guaranteed. Accepting that telecommunications networks are a two-sided market, and taking this into account when defining the regulatory framework, would incentivise the deployment of telecommunications infrastructures and their increased use, to the benefit of all European consumers and businesses.
