4 ms·
> Particularly at the level of last-mile infrastructure, it's inevitable that there's going to be a monopoly on high-speed service. So it makes sense for it to
by baddox 12y ago
> Particularly at the level of last-mile infrastructure, it's inevitable that there's going to be a monopoly on high-speed service. So it makes sense for it to be built out by the municipalities themselves.
Both sentences are debatable. Regarding the first sentence, it's possible that the optimal firm count in some communities and especially sparsely populated areas is one (i.e. an actual natural monopoly), but in most communities I would expect it top be at least two or three, since broadband can be delivered through phone lines and cable television lines (which most communities are wired for). There are obviously other feasible media, like power lines and wireless. Could you explain why you believe that last-mile infrastructure is inevitably a monopoly?
Regarding the second sentence, it's debatable that government control is a good solution even in situations where the optimal firm count is one. An ideal government solution may be better than the average private solution, but likewise the ideal private solution may be better than the average government solution. We ought to compare the average case for both solutions, which is to say, the actual outcome we can reasonably expect to get given our specific inputs (our system of government, existing infrastructure, market demand, etc.). I won't claim to have any convincing evidence either way, but I'm not so confident that a government solution is obviously going to be better than a private solution, even in a true natural monopoly.
- cvet 12y agoI love the ideal, but the reality is that a lot of people are getting a half-assed effort from one company that has no economic incentive to improve. Even the threat of governments coming in can provide some of that incentive to make service better. Add to this that the cities that have created their own broadband networks have done it extremely well, and it's hard to make abstract private/public arguments. In Chattanooga, TN: Public: 1 gigabit, $70 a month; Comcast: 25 mbps, $45 a month (I just looked it up for a zip code in the city) 25 mbps / 1 gbps = 0.025 Assuming linear scaling: 1gbps at Comcast would cost $1800/mo. Holy shit. This is exactly what economic theory predicts will happen in the absence of competition. What's often ignored is the efficacy of the government getting stuff done is largely dependent on the competency of the officials. But if you have good officials, the government solution to the high speed problem is a fantastic one.
- baddox 12y agoI don't know the financials of Chattanooga's system, but if they're losing money (when you don't count tax revenue) the price comparison isn't very useful from an economic standpoint, at least without introducing another argument like that wealthier people ought to subsidize Internet access for poorer people. > This is exactly what economic theory predicts will happen in the absence of competition. Well, I would expect competition to emerge in the absence of competition, if the existing firm is charging so much that competition can be profitable (and if competition isn't prohibited by force). It's true that in an actual natural monopoly (which, remember, I think is rare for Internet access) the monopoly firm has notable market power (the ability to set prices higher than marginal cost). But that market power is not unlimited, and the government solution to a natural monopoly requires the government to be both able and willing to produce at a favorable marginal cost and set prices accordingly. > What's often ignored is the efficacy of the government getting stuff done is largely dependent on the competency of the officials. But if you have good officials, the government solution to the high speed problem is a fantastic one. Certainly true. That would be one of the "inputs" I mentioned in my previous comment. Of course, "good officials" probably doesn't mean much more than "true Scotsman."
- cvet 12y agoGood points—the only thing I'd say is that we're talking about a really capital intensive industry where the incumbent may be able to drop prices to compete with a newcomer. Example: If company B comes to a town where company A is already operating, company A can just offer new subscribers the same terms as B, making it hard for B to compete. If B can anticipate A's action (not that hard in this case, because I can do it after a couple of beers) then B may choose not to invest because B's road to a profit requires producing at a lower marginal cost than A and undercutting A's prices. They actually have to pull subscribers away from A. If A is already producing at a low marginal cost but overcharging, then A has the flexibility to fight a price war (and potentially win). Ha maybe I'm banking too much on those competent government officials. The variance in quality is certainly high.
- baddox 12y ago
- teacup50 12y ago> Regarding the first sentence, it's possible that the optimal firm count in some communities and especially sparsely populated areas is one (i.e. an actual natural monopoly), but in most communities I would expect it top be at least two or three, since broadband can be delivered through phone lines and cable television lines (which most communities are wired for). There are obviously other feasible media, like power lines and wireless. You literally described what we already have today. We already know how that turns out, too: horribly. > Could you explain why you believe that last-mile infrastructure is inevitably a monopoly? Could you explain why you think we don't already have one? DSL lost against DOCSIS. If you want last-mile internet of a decent speed, your choice almost everywhere is cable, period.
- upofadown 12y agoIf there is a fibre to the building type network in place it will be pretty hard to justify paying the money to maintain ugly old hacks like DSL and cable internet. You might as well add 56k modems to the list of competitive services. Such things are simply not competitive to fibre. In terms of simple economic efficiency having multiple redundant last mile networks doesn't really make any sense. That's why no city has multiple competitive sewer and water systems. You don't need multiple last mile systems to have a competitive ISP/TV/phone market. You just need good policies to allow entities that want to sell data services to get access to the last mile infrastructure. You can even bill for the LM infrastructure separately if you want to prevent bundling.