1) NYC's (transit) construction costs are almost entirely driven by 1) shit design and 2) awful procurement policies. The work rules/labor contracts aren't driving much...
2) China has a unique property laws (as opposed to a place like the US which has effectively inherited -- and iterated within -- a Common Law-type framework): China separates land ownership (most of it is effectively government-held or else some other collective type of institution) from use of that land.
In a place like, say CA, the Authority must compensate not just for the actual present value of land but the reduced value based on its limitation of use. That is, the Authority could be forced to
a) compensate a farmer for 500 acres or
b) shift to using another property if the farmer can argue that the 10 out of that 500 acres the Authority needs would render the farm useless (ie.
per se takings are a pain). Eminent domain doesn't work nearly as smoothly - nor nearly as much in the government's favor - as is commonly held.
99% of environmental studies are about mitigating these types of impacts.
The point is to solicit public input on a project to ensure the best project is being built, but the farmer asks,
"What's the justification for using these 10 acres when those 10 acres in that brownfield site 1 mi down the road are just as good?" and this gets more weight than the fact that it would mean shifting the entire alignment and necessitating even more property takings and impacts to residents north and south of the farmer (see:
Santa Clarita effectively forcing the Authority to build a tunnel under the San Gabriels and still complaining about it).
These planning processes - a legal requirement - are meant to produce least impacts for best benefit; the public wants no impacts benefits-be-damned.
The only way to have no impacts is to build nothing: That's why the first reaction is to immediately stop development, because that's what these people want.
It's why the frame is always on
"rising costs," "costs overruns (on something that hasn't even occurred as a cost because its still an estimate) are increasing," and my favorite
"$100 billion (based upon a-worst-case, conservative estimate with a rounding error of over 30%) train!" rather than the fact that the public has intervened every step of the way to force concessions that make it more expensive.
It didn't used to be this way, but it is now.
Facile comparisons to legal environments and economic development models that are oranges to our apples just confuse everything.