Blair's nuclear dream faces financial meltdown

On 9 November 2009, energy secretary Ed Miliband announced the location of possible sites for ten new nuclear power stations in Britain. This represented another step in the process instigated by Tony Blair in May 2006 when he told the CBI that ‘nuclear power was back on the agenda with a vengeance’.

What made this commitment politically viable, especially to MPs who were reluctant to accept another ‘blank cheque’ from public funds for nuclear power, was the commitment that no subsidies would be offered to new nuclear power stations. The phrase repeated through various white papers was:

‘It would be for the private sector to fund, develop, and build new nuclear power stations in the UK, including meeting the full costs of decommissioning and their full share of waste management costs.’

Potential builders, notably EDF, the assumed front-runner for building plants, were quick to assure the government that it would not need subsidies to build new nuclear plants. This position of no subsidies has been maintained by both the government and the utilities, although as the expected economics of new nuclear deteriorates, it is becoming clear that what the government and EDF believe constitutes a subsidy is very different to the usual definition.

The UK government has not claimed that nuclear was as cheap as coal or gas-fired generation, but if the carbon price averaged €36/tCO₂, they claim nuclear would be cost-effective. It promised to streamline the planning process so that developers would be sure plants would not be delayed by lengthy public inquiries. It required the UK safety regulator to carry out a generic design assessment (GDA) for two or more new designs (so-called Generation III+) to be completed by 2011. This would mean buyers had a choice of designs that they knew had cleared all major regulatory hurdles.

The position on subsidies is in stark contrast to that of the USA, which launched its programme to revive nuclear power, Nuclear 2010, in 2001. The US government assumed that the new nuclear designs would be economic but that utilities would be reluctant to order them until they had seen them demonstrated.

It therefore offered a range of subsidies to try to persuade the utilities to build about half a dozen demonstration plants. The two key subsidies were loan guarantees so that at least part of the loans needed to build the plant were guaranteed by the federal government and a subsidy of 1.8c/kWh for the power produced for the first handful of units.

It proved to be the loan guarantees that were key. It is usually assumed construction cost and the cost of finance account for 70 per cent or more of the cost of a kWh of nuclear electricity so keeping the cost of borrowing down is vital if nuclear is going to be competitive. Originally, guarantees were offered for up to 80 per cent of the loans needed to build the plant.

It was assumed that about half the cost of these plants would come from loans and half from self-finance, so about 40 per cent of the cost of the plants would be covered by guarantees. Loans backed by government guarantees would be essentially risk-free and the rate would be similar to Treasury rates. At that time, the nuclear industry was confidently predicting that the new Gen III+ plants could be built for about $1,000/kW of capacity, so that a typical nuclear plant with a capacity of 1.5 million kW would cost $1.5bn. The US government was therefore expecting to have to provide guarantees worth about $3-4bn.

However, by 2008/09 the scale of the US programme had risen to up to 15 units and the expected cost of nuclear plants was now $5,000/kW. The banks and utilities had clearly signalled to the government that the original level of loan guarantee coverage was not sufficient to persuade the banks to lend the money and they successfully lobbied to get the coverage up to 100 per cent of finance and 80 per cent of the total cost.

This meant that the required level of loan guarantees had increased to more about $120bn. The Congressional Budget Office estimated that the default rate would be 25 per cent, costing taxpayers about $30bn. It was far from clear that even leaving only 20 per cent of the cost not covered by loan guarantees would be sufficient to ensure orders were placed.

In the meantime, a queue quickly formed for the huge subsidies on offer from the US government and by 2009, utilities had announced plans for 31 new units and cost estimates were beginning to emerge from these utilities based on their discussions with vendors. These were generally in the order of $5,000/kW.

This raises two issues: why is raising finance such a major issue now when it has not been in the past; and why has the estimated cost apparently increased fivefold in only a decade? In the past, when nuclear power plants were ordered, the electric utilities were monopolies and if anything went wrong with a nuclear plant that raised their costs, they simply put their prices up to recover the extra costs.

In effect, the finance was 100 per cent guaranteed by consumers, which made the loans near enough risk-free, cheap and easy to obtain. Now, with the electricity sector open to competition, any company whose power is too expensive will go out of business. This risk was graphically illustrated in 2002 when the privatised nuclear generation company, British Energy, collapsed. The company had no borrowings and the government chose to rescue the company at a cost to taxpayers of more than £10bn, but if British Energy had had any borrowings the lenders would have lost their money.

The second question is more difficult to answer. The promise of the new generation designs was that they would learn all the safety lessons of Three Mile Island and Chernobyl. However, because the designs could be rationalised and simplified to get rid of all the safety ‘bolt-ons’ they had accumulated in the 1980s and 1990s, they would be much cheaper than plants completed in the 1980s and 1990s.

This expectation seems to have been an illusion. The first order for a plant of the new generation, the notorious Olkiluoto plant in Finland, had a price tag in 2004 of nearly $3,000/kW. As has been widely reported, almost everything that could have gone wrong with the construction of this plant has gone wrong and after four years of construction, when it should have been complete, it was still about four years from completion and more than 75 per cent over-budget (around $5,000/kW). There appears to be a significant risk the utility will default on its loans.

All experience of estimated costs in the nuclear sector is that they turn out to be underestimates, so there is a high risk that the actual costs will be even higher. However, while there are factors in the last decade that might, at first glance, partly explain the price rises — such as rising real commodity prices and skills shortages — the truth may be that the prices forecast now are consistent with actual prices from the 1990s and the $1,000/kW promise was a fantasy. For example, the UK’s most recent nuclear power plant Sizewell B, completed in 1995, cost more than £3bn or about $4,000/kW in 1995 money.

So where does this leave the UK programme? The UK government has confirmed in 2009 that it is sticking to the cost estimates on which it based its forecast that nuclear was cost-effective with an average carbon price of €36. It assumes that plants could be built for only £1,250/kW or about $2,000/kW. It also assumes that the real cost of capital would be only ten per cent.

While the generic design assessment (GDA) process is claimed still to be on target for completion in 2011, the conflicts between the regulator and the vendors over what is required to be safe are increasingly intractable. While GDA will be given in 2011, some design issues will remain open and will have to be resolved for individual orders. Yet the main point of the GDA was to reduce the likelihood of design changes being imposed during construction — a major risk for financiers.

Policy makers are now living in the world of make believe. Nuclear finance will only make sense with subsidies; one suggestions is a floor to the carbon price of £40, adding about ten per cent to average electricity bills. That might provide a profitable price for nuclear generated electricity, but it would represent a substantial distortion of the market and is not in line with government policy, confirmed by DECC on 9 July 2009:

‘We agree with the European Commission that there are significant risks in attempting to manage the carbon price. Introducing price caps or floors makes emissions trading a game of betting on the next government intervention in the market. A better approach is to set the right, long-term regulatory framework with a reducing cap on emissions, as we have done with [EU ETS] Phase III (starting in 2013), and allow the market to help achieve these reductions cost-effectively’.

Nuclear cannot proceed on the terms that exist at present and that being so, the United Kingdom is without a credible energy policy.