A kWh bond to finance nuclear power plants
January 9th, 2008 by David Newbery, University of CambridgeNuclear power has high capital costs and low variable costs, so that its commercial viability depends critically on the cost of capital – the rate of return it must pay investors – and the price of electricity. Pessimists have claimed that liberalised markets are too risky for new nuclear investment without special support. But is it necessarily correct that nuclear power is a risky and therefore financially costly choice?
On the face of it, nuclear power looks very risky, as its product, wholesale electricity, is sold in very volatile markets. Looking ahead, these price risks are set to increase, for at least three reasons. First, if Europe is serious about the 20% renewables energy target, wind generation will need to rise sharply, to perhaps 30-40% of total electricity output by 2020 (which will require as much wind capacity as all our current generation capacity). If so, when the wind blows there will often be more output than current demand, crashing the spot market with zero prices. In other hours prices will have to be much higher to give an annual average price high enough to pay the full cost of other generation. Price volatility will thus increase dramatically.
Second, the future carbon price, which will be included in the electricity price whenever fossil plant runs, is uncertain and volatile. The carbon credit for nuclear power is therefore risky.
Third, the price of gas is similarly less predictable and more volatile, because of evolving geopolitics and expected oil price instability. British spot gas and electricity prices move closely together, so riskier gas prices mean riskier electricity prices.
On the other hand, equipment suppliers such as Areva and Toshiba-Westinghouse claim that their new plant (EPR, AP1000) is very reliable and that standardised designs (no local changes to fit the country’s peculiar safety demands) can be delivered on time and to budget. If governments offered guarantees against costs resulting from any changes in e.g. safety rules, decommissioning and waste management requirements, political and regulatory risk could also be reduced.
Suppose all this is true – that construction, operating, and regulatory risk can all be insured, leaving only market price risk, does it follow that the required rate of return must be high for new nuclear? Perhaps not, if the company issues a new financial instrument, the kWh bond. This bond will promise to pay the holder the average British pre-tax price of electricity for each kWh (unit) specified on the bond. For example, if the average UK retail electricity price (the average unit cost for a 3,300 kWh/year direct debit customer) is £0.09/kWh, and the bond is for 100kWh, the annual dividend would be £9, of that, the wholesale price might be £4, and the transmission, distribution and retailing margin £5.
Such a bond would be attractive to retail consumers for several reasons: it has an indexed element (the regulated price of transmission and distribution is linked directly to the retail price index) and it provides an insurance against electricity price risk. If the retail electricity price rises because the wholesale price rises, then the bond pays out more and compensates the consumer for the rise in bills. Conversely, if the electricity price falls, the bond pays less but the consumer’s electricity bill is also reduced, leaving him or her no worse off.
How might such a bond be valued? The real UK Government indexed bond rate pays less than 2%. Allowing for default risk a higher rate would be required – perhaps 3% real for a strong company like EdF or E.On. Second, the risk premium on the electricity component is negative, as it reduces the risk of future purchases of electricity. Consumers value fixed price contracts more highly than risky variable price contracts, and would be willing to pay a higher price (i.e. accept a lower dividend) for such a bond. A bond with a specified real return of 5% should trade considerably above par, and might be very popular with life insurance companies.
Clearly the bond would be attractive to the nuclear company, for a considerable part of the bond would be an indexed loan on the value of the plant, and the balance would be a claim on the volume of output (at the wholesale price), which, with performance and regulatory guarantees, ought to carry low risk. Nuclear investment is apparently attractive at a real rate of 5%, solving the problem of how to finance the investment. This form of financing should also be attractive for large portfolio companies, for whom the risk of bankruptcy is low (and should be reduced with such instruments), making dominantly nuclear companies such as EdF able to survive in the capital market even without the apron-strings of a de facto French Government financial guarantee. British Energy, with its fleet of unreliable AGRs, would still face considerable output risk, and that might reduce the attractiveness of such bonds, but not to E.On or RWE, the other potential commissioners of new nuclear stations.
Financing new nuclear power stations in liberalised electricity markets like the UK does not necessarily require subsidies or other forms of government support. All that is needed is some financial creativity to make nuclear power look far less risky to retail investors.
David Newbery, Electricity Policy Research Group, University of Cambridge
P.S. A shortened version of this post was published in the Financial Times on 9 January 2008.
January 11th, 2008 at 12:48 pm
The UK government’s Energy White Paper published on 10 January 2008 leaves open two vital questions influencing private sector investment in nuclear power.
To ensure that the it adequately provides for waste and decommissioning costs the government will create a Nuclear Liabilities Financing Assurance Board (NLFAB) which will determine how the Nuclear Liabilities Fund will be administered. It is anticipated that a mixture of an “up front” lump sum and generation levies will be required. British Energy was unable to pay into the fund and its nuclear liabilities were passed to the taxpayer, so to avoid the consequences of a bankruptcy liquidation or a technical breakdown a considerable sum will be payable in advance.
The Energy Review concluded that nuclear power needed carbon credits at a certain level to be viable, earned on the basis of its claims to offer low carbon generation. In his evidence to a Parliamentary Committee M. de Rivaz of EdF requested that the Emissions Trading Scheme (ETS) be complemented by a mechanism to guarantee a minimum carbon price over the economic life of the investment. The scheme ends in 2012 and the EdF timescale is 60 years being the operational life claimed for the Areva EPR.
The ETS relies on continuing supplies of carboniferous fuels being burned so the users can be obliged to purchase the carbon credits issued to the renewable and nuclear generators. If the scheme is effective in reducing overall carbon emissions, it will be to the detriment of the low-carbon generators as the credits will be worthless. The fossil fuel burners faced with rising fuel costs plus the obligation to purchase carbon credits from their competitors will struggle to survive. If CCS technology is also required, then up to 50% more coal will be burned for the same generation, adding to costs but reducing the requirement to purchase carbon credits. EdF foresaw this coming problem and asked for guaranteed carbon prices over its time-scale, thus demanding the subsidy it claims it does not need.
These two vital questions have to be resolved before any projects are able to be financed.
June 18th, 2008 at 6:47 pm
What makes nuclear infrastructures hard to finance is the diversity of the underlying risks and therefore the difficult allocation process of these risks between the actors involved. Over the last decades, public or semi-public utilities have taken all of them on their balance sheet, financing them with debts or corporate bonds. However the such kind of financing implies an important scale effect and then a risk of low competitiveness for there cannot be many actors on the market. What financing in other sectors of the economy tells us, is that mitigating the risk of an activity can be made by splitting it between specialists eager to take the one they know. So, if governments can provide guarantee on political and regulatory issues and utilities on construction and reliability ones, this leaves us with the volatility of electricity prices. This could be tackled through long term contracts between a few entities, but this kind of contracts are under the scope of the European Commission. So we must try to understand why utilities cannot buy long term futures contract on the public electricity market and have no other alternative than doing over the counter transactions.
Today, on the exchange market, futures with a maturity superior to 2 years cannot be bought and the volume traded are so low that you cannot even imagine hedge 10% of a nuclear power plant production without making the prices reach unknown levels. This could be explained by the relative youth of the exchanges but some structural issues cannot be missed. First, the issue of size of the exchange. In Europe, electricity derivatives can be traded in four different places, so why not merge these four exchanges? Today the insufficient interconnection infrastructures between the grids of European countries make the arbitrage between the different exchanges impossible. Second, the information issue. To pass the volatility of prices risk on to the public market, full information on new projects, maintenance schedule, etc. should be delivered to the public.
Eventually in order to build an exchange liquid enough for the big utilities to be able to hedge their nuclear infrastructure program, interconnection infrastructures between the electric grids should be developed giving then the opportunities for the different exchanges (PowerNext, NordPool, etc.) to merge and better and more complete information should be given to the public.
March 13th, 2009 at 8:11 pm
I came across this site when Googling to find a price per KWatt hour for a calculation on the cost of running a particular 20 kW motor.
This led me to considering the question of what electricity generators are charged for transmission into the National Grid. This seems to make it more difficult to generate wave and wind power from Northern Scotland.
The question is provoked by, what seems to me, a peculiar way of calculating this charge being, as I see, based on an fully inclusive model, that is to say not taking into account the marginal cost of transmission of the additional energy.Each additional unit of electricity has to bear the fixed cost of the N Grid which, presumably, has already been absorbed by the existing power generation capacity?
Am I wrong in my belief and if so where and how?
March 31st, 2009 at 7:06 am
The Energy Review concluded that nuclear power needed carbon credits at a certain level to be viable, earned on the basis of its claims to offer low carbon generation. In his evidence to a Parliamentary Committee M. de Rivaz of EdF requested that the Emissions Trading Scheme (ETS) be complemented by a mechanism to guarantee a minimum carbon price over the economic life of the investment. The scheme ends in 2012 and the EdF timescale is 60 years being the operational life claimed for the Areva EPR.