but it's not 3-5 years payback, that's just the payback for the pipes in the ground, you still have to pay for equipment and backhaul, realistically you're looking at more like 7-10 years before break even. Fiber isn't a 100 year investment, fiber installed 20 years ago almost certainly needs at least servicing to bring it up to current standards. I'm not saying it's a bad idea, it's a great idea to build fiber. What I'm saying is that if you can't rustle up the money to build fiber talk to independent ISPs that might be willing to help.
Other posters are right, most ISP's won't consider a build like this for 22 residents. Cost will vary depending on where you live but I'd guesstimate (knowing nothing about your subdivision, so this is a pretty wild guesstimate) that you're looking at about 100k to build out underground conduit, vaults and building entrances. On top of that they've got to bring backhaul to the neighborhood. Now how much are the 22 residences going to be willing to pay per month? I'm guessing $100/home at the high end. Even without working out the cost of pulling the fiber through that conduit and doing 22 installs and pulling backhaul in you're looking at about 46 months payback. Add in those other costs and you're looking at about 5 years payback on build, that's without factoring in the cost to deliver services over that infrastructure. Alternatively you could seek out a local WISP and work with them to bring 100Mbps to the neighborhood and distribute it to the 22 homes, you can probably get yourself 10-30Mbps symmetrical service for about $100 a home with no contracts or commitments. Disclaimer: I am an infrastructure manager for an independent ISP, I have this conversation about 5 times a month with people who "just want fiber"
metered bandwidth doesn't actually make 1 to 1 sense with the cost increases, it just works in the real world as a disincentive for use. Unlimited internet connections were a great marketing item when there was no readily accessible way for users to saturate their home connections for more than a minute or two. The whole oversell design for ISPs depends on the concept that only a fraction of your userbase will be using the full bandwidth available to them at any given time. Unfortunately with the rise of streaming services that's no longer the case and so now the oversell ratios that used to be great for ISPs are no longer working. There are 2 ways to solve this: 1. Build bigger pipes: This is expensive and unless you're building enough capacity so that every user can use their entire connected speed it's not going to be a full solution. It gets you part of the way there though. Unfortunately to do this you'd have to pass heavy costs along to the consumer, this probably won't work as long as the other ISPs are willing to sell the same customer a connection for $30-50/month 2. Get people to use less of the pipe: Go ahead, ask everyone nicely to use less internet, to refrain from streaming netflix, hulu, itunes movies etc.. Good luck with that. The only way to get people to use less bandwidth is to charge for it and make them not want to use it. The goal of these pricing plans isn't to earn money for the bandwidth used, it's to encourage the bandwidth hog users to move to another ISP, thereby making them Comcast or AT&T's problem.