There are serious issues with that approach. If you don't have continuous funding from the projects, you end up with a big overhead of highly paid engineers without work for them to do. Then you have to lay them off, so you lose the institutional knowledge anyway.
We tried this early on with sound transit in Washington state, and because engineering work is boom and bust on a project by project basis, the model just doesn't work. The good people left for better jobs, and we were left with a team that basically couldn't produce, leading to massive delays on the next set of projects.
I think it is for this reason that Switzerland has a fixed budget for their railway construction. And it seems to pay off, Swiss railway is exceptional.
With a small country, stable government, and national level funding, that's perfectly feasible. I wish that would work in the US, but we have hundreds of transit agencies that do capital projects.
Each of those hundreds of transit agencies are small, but many of them cover populations similar to Switzerland. Some of them cover much larger populations. They should be able to do what so soon as doing entirely within their little agency and yet they are not.
Has nothing to do with size, see Japan and China.
What does it have to do with then?
My understanding japan has a surprisingly private train system, and China fits the model of "works when it's constantly expanding"
Japan is mostly using exactly the model I suggested.
What about renting them out to other cities? There are dozens of cities in Europe alone that are planning extensions of their metro systems, or even building ones from scratch (Cluj-Napoca in Romania, recently. Krakow in Poland soon.)
A morbid equivalent from the Middle Ages: bigger medieval cities had their own headsman, and they solved the risk of underemployment by sending him on external "jobs" to smaller towns where executions were rare.
At that point, you're just hiring an engineering firm, so there's no real benefit to government doing it.
There is, because your own projects always get a priority. Plus, the institutional history. If the same group of people projected your metro for the last 40 years, they carry a lot of local know-how in their heads.
Having a single buyer gives the government the ability to force standardization across cities, and squeeze contractors. Knowing the US though, it’ll become a jobs program for bumfuck towns like Plattsburgh NY
I'm not sure I see any mechanism through which government would gain those powers. I'm also not sure standardization across cities is a good idea - that sounds good but it would slow innovation in technology.