Peering has come back in the news with the FCC mentioning it in its set of reviews of the telecommunications market in the USA, following its Network Neutrality decision. The peering and interconnect issues are going to the heart of the telecoms matter in relation to competition, innovation and the Open Network. You don’t need Network Neutrality rules, if you have a well functioning, transparent, interoperable and competitive infrastructure environment.

Peering is a voluntary interconnection of administratively separate Internet networks for the purpose of exchanging traffic between the users of each network. The pure definition of peering is settlement-free, “bill-and-keep,” or “sender keeps all,” meaning that neither party pays the other in association with the exchange of traffic; instead, each derives and retains revenue from its own customers.

According to FCC Chairman Tom Wheeler: ”[T]he manner in which networks interconnect to exchange Internet traffic [peering] is a part of what I call the Network Compact, those values that have traditionally governed successful networks. Thus, it is a question that must concern the Commission. ... The bottom line is that interconnection has always been a key component of telecom policy.”

However, as my American colleague Harold Feld points out, peering is one of those places in networks that routinely start to break down when the balance of power changes enough. It happened with railroads, telegraph, radio networks. It also happens in the phone network, where it had a lengthy run before consolidation started to happen followed by regulated interconnection architecture. It is now also happening with wireless networks.

It happens each time when you reach a stage where a party or set of parties hits the point where it becomes clear that foreclosure from interconnection hurts rivals more than you are hurt in return. At which point, the carrier(s) with superior position start putting up the price or are simply excluding rivals. Without regulatory intervention however, you end up with a monopoly. With different kinds of regulatory intervention, you get different outcomes ranging from collapse to regulated monopoly, the break up of the monopoly (telecoms in the USA) to status quo balance (radio, then TV networks and affiliates until ownership-limits get relaxed), or other potential solutions (the current wireless market, but too early to tell yet).

In several of these cases, the shape of the market changed dramatically over the course of just a few years once it hit a tipping point. Ten years ago, the USA had strong regional wireless (including some like Alltel capable of jumping to national) players and 5 national players. In just 5 years, this dropped to duopoly of AT&T and Verizon and everyone else. Why? In this case, because elimination of two structural regulations, the vertical integration ban and the spectrum cap allowed the telephone companies to reach the tipping point. With greater integration into the vertical structure they could improve their cross-subsidy and avoid overpriced special access (which they charged rivals). With relaxation of the spectrum cap, they could move from a universe where everyone needed to do roaming with multiple partners to a situation where they could be largely self-sufficient, could roam with each other as ‘peers,’ and could impose huge costs on the rest of the industry by excluding them from roaming.

There is no reason why internet peering cannot undergo a similar rapid collapse under the wrong circumstances. Maybe it won’t, but it easily could. It is precisely because these markets are dynamic and fluid that they can suddenly shift to an anti-competitive structure.

Harold concludes that for this reason, the peering market bears watching and understanding at a much deeper level, with some set of triggers for when regulatory intervention is necessary to preserve the competitive nature of the market. Now because IP allows for certain sorts of things that traditional circuits have not, it may well be that the triggers here are different and that interconnection is fine. But we may also (especially with Comcast/TWC) be at another one of these tipping points where concentration at the edge allows a handful of giant last-mile providers to extort terms and reshape the market.