I don’t know the answer to either of your questions. Is track more expensive than new high-speed trains?
I don't think flying is convenient but taking a HSR-train is.
The average American lifestyle (cars included) is what it is . . . so decarbonize that and stop expecting everyone to live like Brooklynites or Europeans.
Though another option is to both densify development near stations and to create a well-functioning feeder transit system to service areas further out. If those can operate at, say, 35 mph / 55 kph average speed, and assuming a 5 minute walkable commute to the work endpoint (jobs are easier to cluster near stations than housing), and with well-timed transfers, then the feed-in distance could be up to about 15 miles / 23 km from the actual HSR station.
Note that average speed is usually about half of top speed, so these would be relatively high-speed commuter-rail systems. Note that the typical average speed of an urban bus, say, is about 5-15 mph (8-24 kph).
That some people are forced into super-commutes now and then doesn’t make it a really sustainable most days situation. So you’re up by 6am or earlier and you’re not getting home until 8pm or later depending on your work hours.
The creation of transportation infrastructure can convert land from low-value (as reflected by market prices) to high, and this has very often occurred. The US Transcontinental Railroad, completed in 1869, was seen by many as a boondoggle, and indeed traveled through all but completely unpopulated (by European settlers) lands. But development quickly followed.
One settlement bypassed by the railroad was Denver, Colorado. That city raised funds to create a spur line to the Transcontinental Railroad in Cheyenne, Wyoming, and in the decade following that line's completion, Denver's population exploded tenfold, from 5,000 to over 50,000 inhabitants.
Many commuter streetcar and heavy rail lines were created as, or in partnership with, real estate developments, and profits from land speculation were often a very substantial portion of overall profits from such operations.
The flip side is that once land values have increased, establishing new rights of way becomes extraordinarily expensive. Which means that backfilling high-speed rail into an extant prosperous region is insanely expensive, not only in direct land acquisition costs but in the costs of passing legislation and litigating various objections to such projects.
Of the major high-speed rail projects created since 1950, at least three, Japan's Shinkansen, France's TGV, and China's CRH, all seem to me to have benefitted either from being developed in a post-war period of depressed property values (Japan & France) or emerged as a greenfield development through regions in which property values had not yet appreciated ... and in which private property ownership isn't a well-developed institution in the first place (China).
The very slowly-progressing process of high speed rail in California is in many ways due to both the high land valuations of that state and a political environment in which legal objections to such projects are easy to mount. The easiest places to proceed with the project are in the relatively depopulated and low-land-value Central Valley. Geography also plays a role in constraining potential routes, most especially within the San Francisco Bay Area (and especially especially along the San Francisco Peninsula, some of the highest-valued land in the world), and in crossing the Tehachapi mountains at the southern end of the Central Valley, through which there are very few viable routes. These create choke points by which interests opposed to, or simply looking to benefit disproportionately from, the project can focus obstruction efforts, and they have.
In darker moments, I reflect that it might be opportune to wait for a major San Andreas Fault movement to press forward with rail plans, whilst opposition is both defunded and occupied with other concerns.
I believe this was every other section, for each mile of track laid. One site I'm finding says that this extended 20 miles to either side of the track:
Approximately 16 percent of Nebraska’s total land mass was given to various railroad companies, either by the federal government or by the state. Along the lines of the state’s two major railroads, the Union Pacific and the Burlington, every other square mile of land (called a "section") went to the railroads. This checkerboard of land extended back twenty miles on both sides of the track. So, the railroads owned a total of twenty sections of land for each mile of road constructed.
<https://www.nebraskastudies.org/en/1850-1874/railroads-settl...>
There are some holdings that are still retained (or were until relatively recently). The UCSF extension along the San Francisco bayfront is on land that had belonged to the Santa Fe Railroad, for example.
How US High Speed Rail projects are utilising this approach I don't know. Much of the valuable land is, of course, presently owned.
I believe this was initially specified under the 1862 Pacific Railroad Act, and may have been later amended:
Section 3 [of the Act] granted an additional 10 square miles (26 km²) of public land for every mile of grade except where railroads ran through cities or crossed rivers. The method of apportioning these additional land grants was specified in the Act as being in the form of "five alternate sections per mile on each side of said railroad, on the line thereof, and within the limits of ten miles on each side" which thus provided the companies with a total of 6,400 acres (2,600 ha) for each mile of their railroad.
However, it's also worth noting that land grants were only doled out in US history in an extremely limited time period--from the 1850s to the 1870s. The majority of railroads were not in fact built with land grants.
I was aware of the "checkerboard" land-grant allocation, but not how deep (perpendicular distance from RoW) it went.
I also wasn't aware of how brief the land-grant subsidy era was. Though it would be interesting to see a track-mileage development timeline or animation showing how trackage developed over time.
My general understanding is that track mileage exploded quite quickly, often with competing railroads establishing duplicate connections especially on highly-attractive routes, with the winner then absorbing (or simply competing to death) the loser(s). There is also a distinction between major (national or regional) Class I railroads, and short-line (usually Class III) railroads, where a short-line often served only a single line or small set of lines within a small region.
Overall, US railroad trackage shrank for much of the 20th century as early, redundant, competitive trackage was abandoned. I'm not sure what recent trends have been.