Nuclear Financing Problem

Nuclear power plants produce the cheapest electricity of any firm generation source once built. The problem is getting them built. A new nuclear plant costs $10-15 billion, takes 7-12 years to construct, and has consistently exceeded budgets. No private lender will finance that risk at rates that make the electricity affordable.

The financing problem is structural, not temporary. Nuclear's capital intensity (roughly 80% of lifetime cost is upfront) means the interest rate dominates the LCOE calculation. At 4% financing, a $12 billion plant produces power at roughly $60-80/MWh. At 8%, the same plant produces at $100-130/MWh. At 12% (typical for merchant risk), $150-200/MWh. The electricity is identical; only the cost of money differs.

The construction risk premium. Vogtle Units 3 and 4 (Georgia) were budgeted at $14 billion and completed at roughly $35 billion, seven years late. Olkiluoto 3 (Finland) was budgeted at 3.2 billion euros and completed at roughly 11 billion, 14 years late. These overruns are not outliers; they are the pattern. Lenders price this risk into interest rates, which raises LCOE, which makes nuclear less competitive, which reduces orders, which prevents learning-curve cost reduction.

How do countries that build nuclear cheaply break this cycle?

Government-backed financing and standardized serial construction. South Korea and China finance nuclear with sovereign-backed debt at 3-4%, build standardized designs in series (not one-offs), and maintain a continuous construction workforce. France did the same in the 1970s-80s. The US and UK build bespoke designs with long gaps between projects, losing workforce expertise between each one. The financing problem is inseparable from the construction management problem.

---