Andy Ho, Straits Times 15 Nov 07;
TRADE and Industry Minister Lim Hng Kiang said in Parliament this week that competitive markets, which have kept power prices low, will remain a cornerstone of Singapore's energy policy.
But with skyrocketing oil prices, should we go nuclear? Nuclear plants generate huge amounts of power using relatively small plots of land and tiny amounts of a plentiful natural resource, uranium.
This mineral is found in 21 countries, with Australia and Canada making up half of the world's output. With no other major civilian use for uranium, its price depends only on the nuclear industry's needs and does not fluctuate much.
In addition, over the whole life cycle from uranium mining to plant construction, the amount of carbon emitted is 20 to 75 times less than the natural-gas equivalent, the cleanest fossil fuel. Once installed and powered up, nuclear plants also emit no carbon.
Yet, no investors are proposing to build nuclear plants here. Why?
Because in a liberalised electricity market, there are no long-term contracts to buy power from any provider, so nuclear plants must be as good an investment risk as the most economical option right now, which is the combined-cycle gas turbines (CCGT).
Investors will want a power project to be short-term and flexible - like CCGTs which, being modular plants, can be put up in small units from about 300MW upwards.
By contrast, to minimise overall costs with nuclear power generation, several identical reactors must be built. For example, British Energy, the UK's privatised nuclear power provider, argues that a programme of at least 10 new reactors of 1,000MW each (or 10,000MW) would be needed to become economically feasible.
However, peak demand in Singapore is only 5,000MW. Anyway, a 10,000MW programme would likely crowd out private sector investment in power generation.
The construction costs of CCGTs are not only much smaller but also fixed and contractually guaranteed. Moreover, they can be constructed rapidly, so investors see returns on investment quickly.
Industry experience in the United States suggests that it takes about seven years to construct a nuclear plant compared with two for a CCGT, thus postponing returns to investors.
Clearly, a nuclear project cannot match the cost and performance guarantees that investors routinely get with CCGTs, so investors will demand a much higher rate of return, which will then drive up its capital costs even more.
Nuclear plants cost a lot to design and actually build. Not including interest accrued while construction is ongoing, current nuclear plants already incur capital costs four times those of CCGTs, according to a 2003 OECD study. Also, because capital costs are financed, there are interest charges to be spread out or amortised over 20 to 25 years which, in turn, get factored into generation costs.
According to the study, capital costs account for 60 to 75 per cent of total costs to generate nuclear power compared with just 25 per cent for CCGTs. Thus any delays - whether technical or political in nature - in construction and commissioning will raise interest charges more for nuclear plants than for CCGTs - and thus the prices they need to charge electricity users - just to break even.
Somewhat counter-intuitively, there is another even more important component to capital costs at the end of a nuclear plant's life. These are the costs to decommission the plant when it is eventually shut down and to manage the waste it has produced. Waste management must continue over the subsequent decades as the plant remains radioactive for a long time. So funds must be set aside upfront - over the plant's active years - to carry out these activities in the future.
These costs represent probably the biggest economic liabilities for nuclear power providers in a competitive electricity market. For example, liberalising the United Kingdom electricity market saw power prices fall from 2001. Mr James Hewlett of the Energy Information Administration, US Department of Energy, says that, by 2003, British Energy had to be rescued by the UK government to the tune of £3.5 billion (S$10.5 billion) since revenues could not meet its financial commitments for end-of-life costs.
Finally, consider operating costs. The bulk of these with CCGTs is made up of fuel costs. However, only 20 per cent of a nuclear plant's operating costs come from uranium costs, according to the 2003 OECD study.
By contrast, the bulk of a nuclear plant's operating costs are fixed, which means they are incurred whether the plant is generating power or not because its safety systems must be maintained even when it is not active.
Since its operating costs are largely fixed, a nuclear plant's profits depend directly on the power prices it can charge users, where a $1 price hike means almost a $1 rise in profits. But so too with a $1 price drop. This means that, in a liberalised market like Singapore, how profitable it can be will depend on the price its competitors charge, which turns on the prices of natural gas and fuel oil used in CCGTs (and interest rates).
For all these reasons, even as oil nudges up to and even beyond US$100 (S$144) a barrel, investors are unlikely to pile into nuclear power just yet. At some astronomical oil price - and provided the Government is willing to act as insurer of last resort for decommissioning and waste management liabilities - perhaps nuclear power might make economic sense.
Even then, political viability would be a different matter altogether. I don't want a nuclear plant in my backward.
Why nuclear power is viable, sensible and profitable
Letter from Paul Chan Poh Hoi, Straits Times Forum 22 Nov 07
'GO NUCLEAR? No sense or profit in it,' says senior writer Dr Andy Ho (ST, Nov 15). I beg to differ and wish to share my views why nuclear power electricity generation in Singapore is viable.
We are totally dependent on fossil fuel import and should not lose fragile grip on reality with regard to price and supply. Going nuclear makes better sense for posterity.
Just look at the sensitivity analysis of impacts on electricity-generation costs - investment, fuel, interest-rate changes and economic lifetime. The economic lifetime of modern nuclear plants has increased to 60 years with proper maintenance programmes while that of coal, gas and oil-fired generators is less than 25 years.
It may cost US$2 million to build a one-megawatt nuclear plant facility. Good returns justify the huge investments. Currently there are 30 nuclear power plants (NPPs) under construction and 70 more are waiting for construction approval.
According to Professor Risto Tarjanne in his presentation on baseload electricity in Finland, he cited Olkiluoto NPP as a case study. The Finnish company decided to add one 1,600-MW reactor at the cost of three billion euros purely on economic considerations.
The return on investments is more than 18 per cent based on electricity sold at the same price as coal-generated plants. The payback period is less than 20 years if oil and gas prices increased further.
The electricity-generation cost calculated from the historical records of the Olkiluoto NPP amounts to 18 euros/MWh (equivalent to 3.8 Singapore cent/kWh) as a constant price during the past operating period of 1981 to 1999. The investment has been highly profitable.
It is not inconceivable that Singapore could install the nuclear power plant in the middle of the sea if space is a constraint and safety is a concern. The proposition to go nuclear is viable, sensible and profitable.
The beauty is that nuclear electricity cost is quite insensitive to fuel price changes. The nuclear choice would make a major contribution to achieving in 2010 the greenhouse gas emission level in accordance with the Kyoto protocol.
