It seems like everyone would be better served by addressing the root problem, which is that ISPs enjoy local monopolies. This would be a free-market solution that ought to satisfy both free-market and privacy advocates would win. Of course, I don't know if there's a good solution for fixing monopolies that doesn't involve additional regulation...
We could eliminate regulation that makes it impractical to build competing networks. DC has three different broadband providers: Comcast, RCN, and Verizon. All three are in the process of rolling out gigabit service. That's possible because DC is pretty liberal about letting providers only build in the locations where they can make a profit (instead of demanding they cover the whole city in one go).
5G technologies in particular are going to make it extremely easy for startups to break into existing markets. I did a speedtest.net run the other day in Annapolis, and I got 150 mbps--what I get on my fiber connection at home. And that's just with LTE-A. Fiber might still be faster, but that doesn't matter. It's fast enough for plenty of people and would be meaningful competition for incumbents. Cities just have to get out of the way and let companies build those networks in an incremental and demand-driven way.
(I'm not free market fundamentalist--I think it would also be fine to just have to government build the networks. But this stuff is expensive and somebody has to write those billion-dollar checks, and you need to make sure they have the incentive to do so.)
If you're building in a place with electricity, the main relevant property right holder is the utility that owns the poles and conduits (and a regime exists for getting access to those, though it could be improved). Siting things like fiber cabinets isn't as big a deal because hold-out property owners have little leverage. If one won't let you build a node on their property, just build it across the street, and run the fiber along the poles and conduits that are already there.
The real hurdles are cable franchise and zoning boards. When AT&T tried to build fiber in SF, it wasn't individual property owners that tried to stop it from building fiber cabinets--it was NIMBY's acting to get the city to stop development. When Verizon tried to build fiber in Baltimore, it wasn't property owners that stopped it; it was the city, which wouldn't give it a TV license.
So DC doesn't have 3 different broadband options. Select parts of DC do. Go to the middle or upper class areas and you might be spoiled for choice, but how much choice is there in the lower income areas, where it probably matters more?
Also, wireless can be nice, but it is in no way a competitor to wired internet.
Broadband deployment is all about math: you calculate the per subscriber cost, and see if how much you can charge for the service is high enough where, after subtracting costs, you make a return Wall Street is happy with. The math doesn't care about social justice. And you cannot "opt out" of this math through municipal regulation. Because most of the cost of build-out is passing the house, your per-subscriber cost is driven by your uptake ratio: actual subscribers divided by houses passed. If you force providers to build in a lot of neighborhoods where few people can afford the service, you drive down the uptake ratio and drive up the per subscriber cost, making the investment unattractive.
Take a city like Baltimore: it has too many poor neighborhoods for competitors to commit to a city-wide build-out. It's also too poor for the government to be able to afford to build its own fiber network. The alternatives are some competition in wealthier areas (with hopefully knock-on effects everywhere else) or no choices at all.
If broadband deployment to poor neighborhoods is an important issue, then raise taxes and pay for it. Trying to get private companies to pay for it just creates a special tax on broadband. You impose extra taxes on products you want less of, like cigarettes or alcohol, not products you want more of.
As to wireless, it is absolutely a competitor to wired. Throughout Verizon's FiOS footprint, its uptake rate is under 40%. That means 60% of people who can get it choose to get cable instead, even though FiOS is vastly superior. (I don't know why--I think people like the TV packages better on cable.) 5G fixed wireless is absolutely as good as cable--hell, even LTE-A has better upload speeds than cable. Wireless doesn't have to be better than fiber to be a meaningful competitor; it just has to be good enough that it's a viable option for many people.
You said recently that the UK model (regulated last-mile monopoly) works well.[1] Is there a reason it wouldn't work well in the US?
5G fixed wireless is just entering trials. Maybe it'll be a game-changer, but relying on new technologies to create competition is why there isn't effective competition now.
FiOS upload speeds are vastly superior, but some Comcast plans offer higher download speeds at cheaper prices. Typical LTE-A upload speeds seem to be worse than cheaper cable plans.[2]
> You said recently that the UK model (regulated last-mile monopoly) works well.[1] Is there a reason it wouldn't work well in the US?
Two reasons. First, BT Openrearch was originally a government-owned monopoly that was later privatized under specific terms. U.S. networks, in contrast, were built almost entirely with private money in the first instance.
Second, the U.K. policy was designed and implemented by smart regulators at a national level, who did a lot of theoretical work on balancing the need to ensure adequate investment with the need to protect consumers. U.S. policy, in contrast, split up between the FCC and state/local regulators. The latter are the same folks responsible for our roads, subways, and other infrastructure being worse than those in comparable European countries.
Some,[1] but by and large the opposite is true: they pay extra taxes. The primary "subsidy" for rural deployment comes from a special 18% tax on telecommunications services. Additionally, there is typically a 5% gross revenue tax that goes to the municipality.
Telcos are expected to pay for building out to lower income neighborhoods in urban or suburban areas through cross subsidies (everyone pays the same price even though customers in neighborhoods with high uptake ratios cost much less to connect than customers in neighborhoods with low uptake ratios).
[1] There is a "$200 billion" number floating around that's completely false.
The only source I've found for 18% is a report from the Tax Foundation.[1] That includes all federal, state, and local taxes and fees. Do you have a source for your numbers?
https://en.wikipedia.org/wiki/Universal_Service_Fund. The 18% rate is paid on some part of the total bill that the FCC deems to be related to interstate voice, with additional state USF fees paid on intrastate services. Historically, telcos didn't pay money into the USF for broadband (except to the extent they're offering telephone service over it), but historically they also couldn't get money out of it for broadband. Under Obama the USF started conversion into a broadband fund, and USF taxes on broadband will likely follow.
In the strip mall near my house, I can get 140 down/20 up/25-30 ms ping. 5G is supposed to bring that into the gigabit range and ping times below 10 ms: http://www.droid-life.com/2017/01/04/att-5g.
Obviously you're not going to get that on your phone in a crowded mall. But for fixed deployments you're talking about having about as much shared bandwidth per subscriber as cable at similar if not better latency. At vastly lower cost.
Even if the free market choice could do it, I would still want regulation. Because there's no reason to allow them to act in this manner, so just bar them from doing so altogether.
