Jonathan
There are several reasons for this. First, the US just doesn't need base load right now, hasn't needed it since about 1973. There was a huge glut of building base load plants in the 1970s, in fact there was a lot of over-building due to the belief that the power consumption would grow at 12% a year as it did in the late 1960s. Most utilities need 200-300 MegaWatts at a time, not 1000-1300 MW.
Second, nucs are very expensive to build. If you build a big enough one, you can get the capital cost down to a value that is 'competitive', but the problem then is what do you do with the power? If you need 200-300 MegaWatts to ensure the reliability of your system, for example, why would you want to sink a few billion into 1200 MW?
Third, the industry hasn't built any nucs in twenty some years. After TMI, then Chernobyl, (TMI in particular), there were a lot of mandated upgrades to the designs of plants. Also, inflation in the late 1970s, early 1980s drove interest rates through the roof.
These factors raised the costs of nuclear plants by a great deal, and caused many of the financial brains in the utility industry to be very sour on nuclear plants. One particular example is a plant in the midwest that came on line in the 1980s. It was targeted for 600 million each for two units. Final price was about 3 billion per unit. Just about bankrupted the utility. Seriously. This is a very big deal in the industry. In fact, I personally believe that a new nuc won't be built by anyone who lived through that era. It's going to take some 'young turk' CEO with vision and guts.
By the way, nearly 20 years later, that plant is a 'cash cow', and the utility is raking in revenue 'hand over fist'.........
There is also the issue of the high level waste. Yucca mountain is still far from certain, and there is no viable alternative. That has to be solved in order to give the stability needed for utility directors the financial level of comfort needed to support building a nuc.
Then there are the protests by the 'anti' crowd. Any board of directors with 'half a brain' would be VERY cautious about committing their utility to be 'the first kid on the block' to build a new nuc. The 'anti' zealots would 'swarm to the rescue' of their precious planet. In droves. This would (will) drive up costs. When the first US utility does order a new nuc, there is a whole a bunch of lawyers who are going to get quite wealthy thank you.
As to 'why not build a nuc anyway and sell the power'? The problem here is 'to who'? In order to be profitable in a deregulated market, you have to make the plant, run the plant, and ship the power cheaper than the next guy. You have a very tough time doing that building a new plant. The capital expense is tremendous.
One of the reasons that nuc plants are now 'cash cows', is that their operating record has improved tremendously over the last 20 years. It used to be the 'nuclear fleet' has a capacity factor of roughly 50%-60%. That means if you had ten nuclear reactors with say, 10,000 MW of nameplate power, you could count on having 5,000 MW at any one time. Nowadays, the average capacity factor is roughly 90%-100%. That means that if you have 10,000 MW of power nameplate, you can count on 9,000 -10,000 MW of power available to your customers. That is nearly double what you had available 20 years ago, with the equipment (generating plants) that you had 20 years ago. That is like building 4-5 new plants over that 20 years. It also corresponds to additional revenue generation of millions of dollars every day.
Bottom line. If you are going to build new capacity, you will build a combined cycle gas plant. Cheaper and faster to build, easier to run. But then, if natural gas prices skyrocket,,,,,your'e screwed.
Dewey Oh, by the way. There really isn't a 'nuclear industry'. There is an 'electric industry'. All nuclear plants (except those few owned by 'Uncle Sugar'

owned and operated by investor owned electric utilities. Those financial entities that I spoke about earlier run those.