I disagree that a traditional ISP would be unable to match the price of Google Fiber in the right circumstances. I've done quite a bit of modeling and research on the issue, and I believe pricing comparable to Google Fiber would be profitable for an ISP, assuming the cost of the build out is known and reasonable. (Note: I assume we are talking about the U.S. here.)
The problem posed to such a venture really is two fold:
(1) There is not a uniform or even well settled authority for ISPs to use public utility right-of-ways. You need to put all your fiber in the easements, conduits, and on the poles owned by telephone companies, cable companies, and electric companies. There are laws which provide for each of these companies to access these resources at (usually) set or ascertainable pricing. There is no such authorities for ISPs. Which means you either have to "sort-of" become a telephone company, or a "cable" company, which exposes you to an entire regulatory regime (including, perhaps required services) that isn't particularly suited for the ISP business. The existing ISP duopoly (cable & dsl) exists because they already have access to the right-of-way by virtue of their original business (telephone and cable tv) and have just "added on" Internet service. How has Google managed to do it then? That is actually interesting. First, as far as the law of at least their primary location (Kansas City, MO & KS) they are in a nebulous legal place where they are an "enhanced video provider", without being a "cable TV operator". This coupled with governments falling all over themselves, and signing agreements they never would for anyone else, at the mention of the word "Google" (and I'm not saying this is a bad thing) has allowed them to accomplish far more from a regulatory standpoint than probably not have been possible for anyone else.
(2) The majority of the capital required for such an endeavor is up front construction costs, which in most places in the United States are either non-ascertainable (due to #1), under the control of your competition (because they control the right-of-way due to #1), or just plain expensive (even without the problems of #1, lots of people digging holes to put fiber in is expensive). The payback period from these upfront construction costs + the regulatory/legal/political uncertainty of #1 makes it a hard sell to whomever provides the capital.
The solution? Enterprising company + visionary city could solve a lot of these problems. The city usually has the authority to grant access to the rights-of-way and might even be able to finance infrastructure projects such as the laying of fiber over the long periods associated with public works projects. However in those instances where those two elements have come together to try to provide this type of arrangement one of two things have happened: (1) gross incompetence by either company or city or political wrangling/red-tape by city send project into death spiral or (2) the incumbent providers have been extraordinarily successful at killing such endeavors at State level. Do not underestimate the State lobbying power of AT&T. It is truly something to behold.
This might not directly answer your question, but the issues with launching a successful residential fiber ISP are much more than just having an alternative business model. You need to have a pretty impressive amount of lobbying prowess, regulatory flexibility, and upfront capital.
