The difference and the complication is that the roads and parking lots are needed first so when they a built it is for current needs and maybe a little overbuilt for projected growth.
As discussed at length in this thread, public transit needs density to be worthwhile. Until that density is reached, it is highly inefficient.
That creates another issue. Once the requisite density is reached it is either too late to go back and retrofit light rail or subway onto the existing infrastructure. I shouldn't say "too late", it isn't impossible, just massively expensive. It would be fairly cheap to build a subway or light rail in the middle of nowhere but that would also be useless. It is vastly more expensive to build a subway or light rail in a high density area where it would actually be useful.
I mean, I agree with many of these points, but also disagree at points.
So, the argument about roads and parking lots being needed first sort of implies a new city or something. Most cities, at least at the start, were designed around roads for limited car usage and without much parking at all. Now, cities expanded out. Some parts had space to stretch in that way, some didn't, and some didn't develop semi-dense centers farther out.
So you end up with sort of a guessing game. Can you build transit through less expensive land before it's gobbled up? Can you guess that a place will grow in that way. Some don't (Phoenix, Houston). But that's the same way with roads. Building bigger, more robust highway systems that allow for better traffic flow is easier and cheaper when it's not needed. But that's infrastructure.
And the takeaway is kinda an odd one. Like, public transit is super pricy. And things that make driving easier are also really pricy and unless you get it fantastically right often don't make stuff better. So I suppose the answer is do nothing a let places choke themselves out, but in the end people have this pesky habit of trying to do things to make the places they live nicer, for better or worse.
