|
|
Nuclear power has never had a physics problem. It has a cheque-size problem. |
|
One Number
The implied market for small modular reactors (SMRs) between 2030 and 2050. The Economist The Economist expects net nuclear capacity outside China and Russia to grow by more than half, to over 450GW, with SMRs making up 40-60% of it. Yet the first SMR in a G7 country, in Darlington, Ontario, is due to connect to the grid only around 2030. |
|
|
One Argument The question is not whether small reactors are better reactors. It should be about whether they are easier to fund.SMRs make sense not because small reactors are better engineering. They make sense because they turn nuclear from a one-off mega-project into a repeatable product that lenders can underwrite. The International Energy Agency says that a long permitting and construction process can push a new large reactor’s break-even to 20-30 years after the project starts. This rules out the project-finance structures used for most other infrastructure. The longer the build, the more the interest compounds. The reactor is expensive. The waiting is worse. A 300MW unit changes the arithmetic — a shorter build, a smaller loan, earlier revenue. To illustrate the scale, World Nuclear News puts each Darlington reactor at enough power for about 300,000 homes. Thus, one megawatt is enough to supply roughly a thousand households. Vattenfall's head of new nuclear, Desirée Comstedt, argues that a series of smaller units needs less space and far fewer staff, and makes logistics more manageable. The demand is not only for electricity. A recent University of Michigan study found high margins for SMRs in hydrogen production for ammonia and refining. Part of the trillion is an industrial market. SMRs are not only an American story. Estonia’s privately owned Fermi Energia plans a 600MW plant built from two GE Vernova reactors, with construction targeted for 2031. That would cover about a third of Estonia's record peak demand of 1,723MW, set in February. Sweden has chosen three Rolls-Royce SMRs for the Ringhals project, where the state is due to become a majority owner. And just last month the European Investment Bank made its first SMR investment, up to €40 million in Finland’s Steady Energy. However, first units are brutally expensive. World Nuclear News says that Darlington’s four reactors are budgeted at about $15.1 billion, or roughly $12.6 million per megawatt. The International Energy Agency estimates SMRs need to reach about $4.5 million per megawatt by 2040 to see a rapid uptake. The Michigan study also found that SMRs cannot compete with wholesale electricity prices. And the IAEA’s high-case projection has only around 10% of new nuclear capacity coming from SMRs. But nobody should expect the first unit to be cheap. First-of-a-kind plants pay for everyone’s learning. Ontario Power Generation expects costs at Darlington to fall with each subsequent unit. That can be tested within a decade.
|
|
One Position SMRs make sense as a financing innovation. A trillion-dollar market is credible on one condition — each reactor must cost less than the last. Watch Darlington units two to four. When you read the next SMR announcement, ask which unit number it is and what the previous one cost per megawatt. I am wrong if the later units come close to the first unit’s cost per megawatt. Then the learning curve really is not there. And the trillion-dollar market becomes a subsidy market. Which project that you know about is stalling not because it is a bad idea, but because no one can fund it in one piece? |
|
If this was forwarded to you and you'd like to receive Corporate Financier's Notes every Thursday, you can subscribe here. Reading on your phone? Tap the three dots or overflow menu in the top corner of your email app and select View in browser for the best experience.
|