This is not a comparison of "budget and time overruns for fusion reactors to budget and time overruns for skyscrapers". It is a statement about management practices, regardless of the project.
The statement is that a blue-sky project such as a fusion system is easily recognized by people on the lookout for sources of unlimited money with no strings attached. When they get control of the project, the bulk of the money will not end up spent on the extremely difficult problems entailed. If the problems are as difficult as expected, handing the money over to people who benefit by not solving them will reliably fail to solve them.
We know with absolute certainty, already, that there is no "holy grail of energy production" at the end of it. The most favorable imaginable result is a system much more expensive to operate than a fission plant that produces no more power.
