Let's take an average over country then apply it in my back-on-the-envelope. Doesn't matter if areas alike Manhattan would probably have 100x the cost of running coax in the rural area (which obviously drives the average).
Then justify the existence of monopolies when in fact purely local providers are actually better able to reduce the costs associated with broadband installation. Those monopolies subsidise central areas by providing way worse service for rural.
I've actually got this issue in London (I didn't check before moving, that our house is literally in the middle of about 5 exchanges). Went as far as reaching providers relating to Gigabit Voucher Scheme or even thinking of asking a friend living about a mile away to provide a wireless link from his house.
They offered to split the cost with the other neighbors on the street. One are elderly who aren't interested in cable, the other are very religous and don't own a television or use the internet. IMO it'd be worth it for just the resale value of the houses.
Hell, I currently live in an affluent Chicago suburb, and I have exposed wires from both Comcast and AT&T just inches from the sidewalk out my back door. My dog literally takes dumps on the internet/TV cables to the townhouses I live in.
Haven't had AT&T service since they sold AT&T BI to comcast, but Comcast has the balls to tell me when I complain about an outage I should upgrade to business class if I want fewer outages (lower speeds, more money, supposedly service guarantees).
There are no federal level universal service mandates for broadband. In any event, mandates aren’t free money. Consumers end up paying for it, the cost is just hidden and unaccountable. The mandate obscures the political question of: should we give a $5,000+ subsidy to people out of other peoples’ pockets, or are there are needier people who are more deserving?
As to Romania—that country has vastly lower broadband penetration in rural areas (where more than half the population lives) than the US: https://www.broadbandtvnews.com/2018/05/22/romanian-broadban.... Also, costs are not easily comparable across countries. The vast bulk of costs of building fiber in the US is labor costs, both for initial build and for ongoing maintenance. Building infrastructure in a blank slate with developing country labor costs is a very different proposition than doing the same thing in the US. (It’s also why China can build tons of high speed rail while the California HSR project ran out of money before getting out of Fresno.)
> the “subsidy” is really coming out of their own pocket
ISPs lobbied really hard during the 1996 rewrite of the USF so they could add the “Universal Access Fund” line item on your bill separately. So, by and large, it’s not coming out of their pocket.
> The universal service fund only recently started offering subsidies for broadband
Not sure what you’re referencing when you talk about subsidies, but the USAC website says that they offered $200m in 2000, and the number quickly goes into the billions from there.
https://www.usac.org/about/tools/publications/annual-reports...
As to broadband subsidies: prior to 2011, the universal service fund only covered subsidies for broadband to schools and libraries, through the e-rate program. It wasn't until 2011 that the USF started subsidizing broadband generally through the Connect America Fund.
Which brings us to assumption 2. At > $100 per month, most people won’t subscribe. (Especially given that, as a matter of local law, in many places you’re required to offer a basic cable tier for $20-$30.) Atlantic Broadband, which covers rural areas around PA and MD primarily, has about 250,000 subscribers, but is available to almost 2 million people. (At least on a census-block basis.) My 40% assumption was quite generous.
There are more modern laying techniques for rural areas that don't require you to dig up that much. A five person team can lay 3 kilometers of fiber a day in rural areas with modern microtrenching solutions. (looks like this for instance: https://www.youtube.com/watch?v=4bfbh3ILLD0)
They almost certainly would make it back over a long enough time period. Broadband is going to get more important over time. Even if just a few people sign up now, future residents may prioritize it higher.
Then, if you’ve got 50 houses, how many will subscribe? Hint: not many, since folks who buy houses in places without broadband probably don’t prioritize broadband. Say you get 20 houses though, with $200,000 invested. That’s $10,000 per house. That’s an optimistic figure. Will you ever make that money back? No. Charter has a market cap of $5,000 per subscriber. If you blow double that on capital costs alone per subscriber, you’ll never come close to making it back. (Especially in a scenario like the above with high maintenance costs.)