Today, heavy rail is extremely expensive to build. But by not having to compete with traffic, it is faster and more reliable than other modes of transit. Light rail is pricier than buses but carry more people and generate more investment.
Will it always be that way? I believe that one day we’ll see automated construction machines that can much more cheaply build a tunnel. We already bore tunnels by machine, instead of blasting by hand as the builders of subways did in the early 1900s. Unlike, say, flying cars or better batteries, there’s no law of physics that says we can’t automate the rest.
Computerized systems could block off a work site, relocate utilities, build the tunnel walls, and package up the debris for removal. It’s just a matter of time and engineering. Plus, underground construction has obvious defense benefits. DARPA should start funding studies into automated tunnel boring.
What will happen if, in the future, we can build tunnels for a fraction the cost today? We could put subways anywhere the ridership would justify the ongoing operating costs (maybe something like this). Heavy rail out to more distant town centers wouldn’t be such a boondoggle.
Of course, if it’s cheaper to build subways, it’s also cheaper to build underground freeways as well. With gas prices what they are, I’d hope we wouldn’t be so foolish, though we’d probably end up with some of each.
Gas prices could radically change with technology as well. Say, a giant solar station in orbit? To me, creating completely new sources of energy is less immediately plausible than construction robots. We already build things with robots (like cars), and the history of computation is a steady march toward automating more tasks that we already do. An orbital solar station or some other major source of cheap power is a whole different undertaking, but who knows?