Rod points out:
The shared liability approach [inherent to Price-Anderson], if taken without permission, would violate the anti-trust laws that prevent competitors from cooperating. Price-Anderson’s system rewards the industry for sharing detailed technical information that would normally be carefully protected trade secrets. The nuclear industry’s habit of widely sharing important information and lessons learned from experience is one of the foundations on which its excellent safety record is built.
To those unfamiliar with how the Price-Anderson liability law works, the NRC has a helpful fact sheet which explains the basics. Contrary to common portrayal, Price-Anderson does not simply act as an "escape hatch" for financial liability for nuclear reactor operators. Each reactor operator is required to hold private insurance for $375 million per individual unit; after this limit is exceeded, a second tier of insurance from cross-pooling across each operating reactor kicks in - up to $11.6 billion. In other words, the liability structure of Price-Anderson explicitly makes it such that "an accident anywhere is an accident everywhere" - there is a shared financial liability across every operator.
The dispute arises of course as to what happens above the $11.975 billion liability threshold - yet this has never actually been tested. (In fact, after Three Mile Island, the worst nuclear accident in U.S. history, $175 million in funds came solely from the "first tier" - i.e., private insurance). Critics are of course free to make their case that the cap is too low - but it first involves an honest reckoning of what the actual liability structure actually is.
More important however, arguments like this tend to underscore a common problem with many of some of these so-called "market" critiques of nuclear energy - even cursory inspection tends to reveal how thin a paint job has been applied to what is primarily a vehicle for advancing a dogmatic anti-nuclear agenda. (I say this too as someone who is almost always first disposed toward market-based solutions.)
Such a harsh critique comes primarily from the shallowness of the arguments presented - which, like TCS's "Golden Fleece," focus primarily on the seen (i.e., the DOE's licensing assistance) and less so on the unseen (i.e., the flawed NRC licensing process which makes it incredibly difficult for new market entrants to feasibly seek design certification). In other words, the "market-based" aspect is simply invoked as a shallow pretext for one-sided arguments applied solely to nuclear energy, rather than across the entirety of the energy sector.
Thus why I highlight Rod's point: A fair free-market critique of the energy industry - nuclear included - would also look at the barriers erected by government regulation (which in turn are what spur the calls for so-called "subsidies" which free-market groups generally oppose). The perverse consequences of anti-trust regulation from information-sharing are an insightful example of this; likewise there is the issue of the regulatory standard for long-term disposal of nuclear waste. For example, a geologic repository for permanent disposal of nuclear waste is required to meet a standard of no more than 10 mrem/year exposure to the public over 10,000 years, followed by 100 mrem/year up to 1 million years. (How one evaluates these begins to leave the realm of engineering and move more into the realm of divination...) To give some perspective - this involves a protection standard equivalent to less than 5% of the average background dose one receives from natural sources over the initial post-closure period, and less than 30% in the million-year period. To put it in more familiar terms, this corresponds maximum exposure equivalent to 1 chest x-ray per year in the first 10,000 years, followed by half the dose from a head CT scan (and 1/10th the dose of a whole-body CT scan) over the next 990,000 years. (Discharges from nuclear plants are regulated still more strictly, at 0.3 mrem per year maximum - less than a standard dental x-ray.)
Ask yourself this - when is the last time you've heard of a coal or natural gas facility being required to sequester their toxic wastes from the public in near-perpetuity? (Likewise even with toxic heavy metal wastes incident to the production of solar panels). How many other facilities are required to pay hourly costs for regulators evaluating license applications (much less put in the encyclopedic licensing applications to begin with)?
The point here is not simply to complain or to justify any special treatment on the part of the nuclear industry - but it does provoke a question of why more so-called "market advocates" in energy only look at one side of the coin. A deeper (and more insightful) analysis would consider the inherent barriers erected as well - including the perverse consequences of features like anti-trust regulations and unequal applications of standards for risk exposure across different energy-producing industries - rather than cherry-picking analysis we're typically treated to presently - thin gruel, indeed.