The most overlooked facet of the public transit debate.
Think about your favorite mall, or your college or high school, or your grocery store, or the airport. Look it up on a satellite map online. Lenox Mall's footprint is half the size of its parking lot. Cumberland Mall is about the same. Hartsfield-Jackson Intl Airport has more space set aside for parking than for terminals, and that's ignoring all the third party lots in the surrounding areas.
Now, think about what all of that land is worth.
Really, think seriously about the value of that much land. What is an acre (about 40k square feet) worth in the middle of downtown Atlanta? What are the thousand acres surrounding the ATL airport worth? What is it worth if you could reclaim it for parks? What is it worth in sale value and property taxes if it was developed commercially? We are talking about billions upon billions of dollars worth of real estate, and billions of dollars in increased property tax revenue.
Now, let's consider that as applies to The Public Transit Debate.
The more people who ride public transit from Roswell to Atlanta, the fewer lanes 400 needs. Ditto Marietta and 75, Buford and 85, Conyers and 20, etc. The more people who ride the bus or train to the mall or to school, the fewer lanes city roads need, and the less parking area they need. Bicycle and scooter usage also lead to the same result inside the city, but that's another debate for another day. All of the car traffic on a 4 lane highway could be condensed down to a single lane of buses, or a single rail line. Most of the car traffic in a small suburb could be replaced by PRT vehicles on narrower and much cheaper to maintain roads.
It's a chicken and egg problem, as are most things involving public services to replace private services. Different from "I'll ride the bus when it runs more often"/"The bus will run more often when more people ride it", but the same type of problem. Just as a thought experiment, let's pretend we have a time machine, or at least a source of ridiculously large interest-free loans. If we could, today, somehow have already reaped the profits of reclaiming all of that land, we could then use that money to implement a public transit system so awesome that we wouldn't need all of those roads and parking lots any more. And once we have a public transit system that awesome, we can reclaim that land and put it to better or more profitable use, either loading the profits into our time machine or repaying those loans.
So, the question then becomes, how long will it take us to bootstrap that change, as land gets more expensive, gas gets more expensive, cities get bigger, and all of those other changes continue inexorably into the future? It's a hurdle that we must jump, eventually, and the longer we wait the worse of a shock it is going to be to our economy and society. We can't put it off forever. Unless something amazing happens to energy storage and transportation technology in the next few decades we will have to do something about the automobile problem, and most of the "something"s that don't involve government intervention are not pretty.
Your thoughts on the matter are appreciated.
 A typical 400'x400' block (from the downtown/midtown grid) is surrounded by roads 50' wide. Splitting the roads down the middle, and ignoring sidewalks, that's 42.5k square feet of road for 160k square feet of otherwise usable land.
A quick survey of the area tells me that approximately one out of every 12 blocks is solid parking lot or garage, and one out of every 4 blocks is partially (let's call it 25% on average) parking or garages. So for every 12 blocks, there are 510k square feet of road, one whole block of parking garage (160k square feet), three 1/4 blocks of parking garage (120k square feet), and 10 blocks of building (1.6M square feet).
That puts the split around 1:2, and we haven't even taken the interstates or highways into account yet.