I don't see why ISPs are natural monopolies. I live in Russia and ISPs in big cities are highly competitive.
When you serve high-rise apartment buildings, ethernet is a viable alternative to DSL. I don't see a technical reason why an independent ISP can't serve a few central districts of a city with fiber and ethernet as the last mile (in this case, the last few meters).
I imagine in USA it's ridiculously hard to lay your own cables between apartment blocks and host a few boxes in a common utility cellar; and maybe it's also legally impossible to serve a subset of a city or even a state.
But in Russia each apartment block has 2-3 ethernet ISPs offering Internet plus one DSL. These networks grew organically: they were profitable from the day one when they served one large building and they they added a few buildings a month; the industry wasn't regulated so they didn't have fixed regulation costs.
Now there are some stupid regulations and the costs of starting that kind of business went up, but there are still a lot of options to choose from.
So it seems that by promoting anti-market regulations dedicated to delivering predictable and quality service, you are now stuck with clumsy monopolies. So much for violating principles of capitalism.
Additional beauty: Even in smaller towns or remote districts where there is no competition, the de-facto monopolist still has to look up to the "industry best practices". Of course they may still offer comparatively worse service for the money, but on some level the knowledge of "how the things should be" forces them to develop better network than they could get away with.
You made the point yourself -- they smart small and grow with revenues over time. Eventually the market runs out of growth, so you see consolidation.
US cable companies started out as small cooperatives that served individual neighborhoods without good over the air signal and grew into behemoths. The old US Bell telephone system was formed after waves of consolidation of small telephone companies serving a city or town. The current US cellular market is the same story -- lots of little companies getting eaten up by bigger companies.
Post-Soviet Russia has only been around for a couple of decades. I'm sure that you will see consolidation as time goes on -- there are absolutely economies of scale to be had when operating networks.
But why don't new small players arise as small cooperatives that serve individual neighborhoods (under-served and over-demanding; I assume there are a lot of areas like that in the US)? And they they may slowly eat the behemoth from the inside.
Of course consolidation will happen, but we'll already have good intertubes by then. Also they might follow the cellular carriers path, where there are three to five large carriers covering any (inhabited) area with no further sign of consolidation.
(I assume if large cell companies try to merge antitrust won't let them - not counting under extreme economical or political curcumstances which are known to arise from time to time)
"why don't new small players arise as small cooperatives that serve individual neighborhoods"
I am just going to guess that it is a combination of (1) the expense of getting linked to the rest of the Internet, (2) the low density of knowledge i.e. most people who would be interested in a cooperative don't really know how bad American Internet service is, and (3) the difficulty one faces with laying cables. I could be wrong, of course; I only briefly looked at doing this sort of thing, and I was planning to use wifi (once I saw how expensive things would be, I dropped the plan).
1) Why should it be particularily expensive? Data centers do this and they are competitive.
2) It should be obvious that your internet connection sucks shouldn't it? If it isn't, maybe it's not an actual problem?
I think there should be one success story showing it can be done for a profit and then they will appear like mushrooms all over the country. You could crowdfund one I think.
> When you serve high-rise apartment buildings, ethernet is a viable alternative to DSL. I don't see a technical reason why an independent ISP can't serve a few central districts of a city with fiber and ethernet as the last mile (in this case, the last few meters).
That's what's happening in Australia with our national fibre network. Fibre runs to the bottom of apartment blocks, router in the basement and Ethernet to each individual apartment.
When you serve high-rise apartment buildings, ethernet is a viable alternative to DSL. I don't see a technical reason why an independent ISP can't serve a few central districts of a city with fiber and ethernet as the last mile (in this case, the last few meters).
I imagine in USA it's ridiculously hard to lay your own cables between apartment blocks and host a few boxes in a common utility cellar; and maybe it's also legally impossible to serve a subset of a city or even a state. But in Russia each apartment block has 2-3 ethernet ISPs offering Internet plus one DSL. These networks grew organically: they were profitable from the day one when they served one large building and they they added a few buildings a month; the industry wasn't regulated so they didn't have fixed regulation costs.
Now there are some stupid regulations and the costs of starting that kind of business went up, but there are still a lot of options to choose from.
So it seems that by promoting anti-market regulations dedicated to delivering predictable and quality service, you are now stuck with clumsy monopolies. So much for violating principles of capitalism.
Additional beauty: Even in smaller towns or remote districts where there is no competition, the de-facto monopolist still has to look up to the "industry best practices". Of course they may still offer comparatively worse service for the money, but on some level the knowledge of "how the things should be" forces them to develop better network than they could get away with.