I sometimes hear we can't afford to spend R&D money on MSRs. The best reply to that is we can't afford not to spend energy R&D.
85% of the world's reactors are PWRs, which operate with a thermodynamic efficiency of only 32% [only 32% of heat energy is converted to electricity]. MSRs can run at 750°C, giving a potential thermodynamic efficiency of up to 48%. Half as much again. Current power reactors make about 2370 TWh of electricity per year. The wholesale price of British electricity is about £40/MWh; making the world wholesale value of nuclear electricity about £95 billion/year. Compared to what MSRs could give we're throwing away £47 billion each year as heat which could've been made into extra electricity. All for the one time cost of £1 billion/£2 billion R&D we didn't spend to make MSRs real.
The US molten salt reactor experiment, MSRE, cost $5 million per year. Not the billions anti-nuclear power campaigners claim. Adjusted to today's money it works out at $30.52 pa., or £21.48 pa. Over 15 years that would cost Britain a total of $322 million.
Cost of R&D avoided: About half as much again as world electricity sells for, say: £47 billion (at wholesale prices)