Then how are credit unions able to form locally, with relatively small amounts of money, and offer loans that do not pay off for a long time? How are tree farmers able to make money, when their crop takes upwards of 1 lifetime to sell? If the worry is that the investment wont come because the return is risky and may take a while to become profitable, the market has already figured that out. We see investment in many more areas that is far more risky than an ISP. At least with an ISP, you are very likely to start getting a return, but in many other markets, you literally cannot get a return until the product is ready, yet those markets still exist.
The long return on investment is only part of the problem. You're right that the market has figured out the ISP economics though. That's why it doesn't happen, because the risk/return ratio is bad.
While regulation and other things play a role, this is mostly a case of simple economics. It doesn't make sense to invest in additional ISP infrastructure, so people don't.
The system we have now, is that regulation protects profits. If you opened up the possibility of competition, they would have to sustain the lower rates at all time, and not just when local govs start talking about negotiations.
That's what I'm saying we do, we open up the possibility of competition. What you want is already largely there and as I keep saying, the market has spoken. The returns on investment are simply not worth the risks, hence no investment.
What is more likely, that literally almost every municipality in the country has conspired to prevent telecom competition, or that the economics of building new networks is bad? It is simply far too easy for incumbents to crush startups, at great cost to investors. It's common knowledge that this is the tactic ISPs use.
I would argue that under-building infrastructure is even more costly. The amounts of money being spent in the economy on higher prices is likely a very large number.
Those higher prices do not come from a lack of duplicative infrastructure, as we could achieve those same lower prices using the current infrastructure that we have. Duplication of infrastructure is costly both in terms of disruption and in pure dollars. It's inefficient, and we shouldn't do inefficient things.
Why would you just limit to the last mile. If you think that over building is a problem, why would you go for the last mile, when you need far more lines at that point?
Because the 'last mile' is the connection that goes to various hubs. Traffic once it gets to those hubs is far easier to re-route to different ISPs, so you the last mile is where it makes sense to keep things common.
This is a good example of smart regulations that spur competition instead of limit it.