Once the project is done, the money stops. The bigger the project is, the more backers it needs. Many of these have to be bought. The people initiating the pro

Once the project is done, the money stops. The bigger the project is, the more backers it needs. Many of these have to be bought. The people initiating the project don't have money to buy them, but the project budget does, or can. Adding that to the budget would make it a bigger project, needing more backing. So, projects are under-budgeted, and overruns pay off backers. Nobody involved wants that money flow ever to end, so it goes over schedule. Some projects avoid this fate. Solar and wind farms have transparent costs (N units x $M per unit), and start producing revenue almost immediately. Nukes suffer it most, because nobody knows how much any of it ought to cost, it takes forever, and it produces nothing until it is finished several changes of administration later.