All the while, Cogent undercuts the market of every other carrier who isn't as efficient as they are, leading to massive losses, bankruptcy filing after bankruptcy filing, out of court reorganizations and purchases for foreign companies, etc.
Banks and Venture Capitalists love this. They think in terms of markets, not in terms of companies. Banks and VCs both make more money when a market is not overcrowded with competitors. Obviously, if you can spend a bit of money to fund a company which will kill off all the weaker competitors, and make that money back in a strong company who benefits from larger market share, then it makes sense in the big picture. I don't have any evidence that this is what is happening in this case, however this kind of thing is done to stabilise markets. To the winner go the spoils. All the network assets are not destroyed. For the most part they are aggregated into the networks of companies who buy the bankrupt assets. Sometimes a network operator will buy and integrate a functioning network. Other times, the corporate customer base will buy the boxes and become more likely to buy bandwidth from surviving operators.
To force either one to peer by law or regulation would be even more disruptive to the industry as a whole, which may even lead to a complete collapse of the existing peering system as we know it.
I think that's too harsh. It would certainly lead to change but that is not the same as collapse. If companies were forced by law to peer, then they would have to do so under a settlement system. Since IP network revenue flow is not based on the end-point usage (per minute call charges) there is no need to engage in massive data collection and analysis as the voice business does. Something based simply on netflow data from peering interconnects combined with the the average bandwidth price on each companies top ten customer contracts would add very little cost to computing the settlement amount. I don't advocate one way or another. But I do expect to see things change when there is instability. --Michael Dillon