My current understanding is that the economics did not work out because: power company borrows x billion in capital to build a nuclear power plant, interest rates in the 80's are in the double digits, activists tie up the startup of the power plant for 5-10 years doubling the cost due to interest, power company finally drops starting the plant altogether and the public eats the loss.
Of course there are all sorts of complications to what really happened, like power consumption rates going up like crazy while AC is adopted and power companies thinking that the trend would continue forever, which didn't happen.
Have something you want to believe and you can find a narrative to fit it. This is a really hard trap to avoid and I'm sure I fall in it often.
While regulation certainly didn't make anything easier, and can probably be blamed for some percentage of the failures, there's a loooot more to the story. As you point out, incorrect demand predictions, and long term planning set during a time of plenty can require cancellations too.
I wish I had a more authoritative book from a disinterested historian, but the best I've found as a summary is the classic 1985 Forbes article[1], which I found through Wikipedia's list [2] of cancelled projects, the list that initially made me think, "huh, TMI can't be blamed for that one, or that one..."
[1] reprint on a blog, with some weird commentary about Gore on top which can be ignored: http://blowhardwindbag.blogspot.com/2011/04/forbes-article-r...
[2] https://en.m.wikipedia.org/wiki/List_of_cancelled_nuclear_re...
Love to hear of any sources you might have too!
... and fuel-cycle then decommission -related costs.
About those last ones: https://news.ycombinator.com/item?id=17844976
However, after investigating the history more closely, it appears that nuclear construction projects were boondoggles even before Three Mile Island. Economics, not activism, is what killed nuclear in the US.
Nuclear has always been marketed under a lie, that its cheap, and perhaps too cheap to meter. In reality, well engineered projects are insanely complex, requiring huge material and labor costs during construction. And then there's maintenance.
People always talk about regulations driving up costs for nuclear, but they never talk about inappropriate regulations that drive up costs, meaning that the high costs are necessary.
And when you think about the design of the things, there's no reason to think it should be cheap: the nuclear is merely a heat source, which still requires the massive amount of machinery needed for steam turbines. The only part of the thing that's potentially cheaper is fuel, but to handle that fuel you've got to have massively engineering and equipment efforts.
This history of nuclear is always glossed over by its boosters. It's really expensive, and if we didn't have better options now, we should build it anyway. And we should have built a lot during the 80s like France did.
But today, grid storage is far cheaper than nuclear, and it makes no sense at all to ever build a new nuclear plant. Let's keep existing ones running until they're replaced with other carbon free sources, but for new investment, current nuclear tech is dead. New SMR, maybe maybe some day if it's cheap, but we shall see in a few decades, it's not available today like storage, wind, water, and solar are.