The implication of the latest energy prices is that project costs must be around £750/kW -- or lower. There is probably no single reason for the dramatic reduction -- the low prices are most likely due to a combination of factors, including:
¥ lower wind turbine prices, which have fallen significantly during the past year
¥ high wind speed sites (particularly in Scotland)
¥ reduced profit margins -- in construction, development and operations
¥ lower site acquisition and bank fees
¥ the entry of vertically integrated developers
¥ cheaper finance, including lower expectations of return on equity
¥ the longer lead times for project completion compared with earlier rounds of NFFO (as projects need not be completed until 1999 they can be priced to anticipate further cost reductions in the intervening years and longer bank loan periods.)
In addition, a number of extensions to earlier wind farms are included in the bids and these benefit from the fact that the infrastructure and grid connections are already in place. The low prices are clearly a result of the very competitive atmosphere encouraged by the bidding process and totally eliminate the criticisms levelled at British wind energy in certain quarters, which focused on the £0.11/kWh price paid under the second round of NFFO. It must be remembered that this was the maximum bid price in that round -- although it was paid to all developers. In the current round (NFFO-3), each project will be paid its bid price and contracts will run for 15 years.
The longer contract period accounts for much of the difference between NFFO-2 and NFFO-3 prices, but there have been substantial real improvements as well. However the comparisons are made, the reductions are dramatic. The average price of completed NFFO-2 projects was about £1100/kW and if this price is linked to a site with a wind speed around 8 m/s the 1994 bid price would need to be around £0.072/kWh. The corresponding NFFO-3 average price is £0.043p/kWh -- a 40% reduction. Looked at another way, the cheapest NFFO-2 projects would have been able to bid at around £0.06/kWh this time -- and the cheapest bid was £0.04/kWh -- a 33% reduction.
But the acquisition of a NFFO contract does not necessarily guarantee that a project will be built. The government "expects only 20 or so [out of 55 in England and Wales] of the wind farms will commission." Another undesirable feature of NFFO is that it is relatively easy to abandon projects -- as experience with earlier rounds has shown. It may, therefore, be premature to conclude that the lower prices will actually be realised, but it is interesting to analyse how the low bids may have been achieved.
One key factor is the "cost of money." Finance for wind energy projects is usually provided partly by bank loans, debt, and partly by investors, equity. A lower rate of interest is paid on the debt -- currently around 11% in the UK, but these payments have priority. Equity investors accept a higher risk (they are paid less, or nothing, if things go wrong) in exchange for higher income in the form of dividends. There is a delicate balance, however, and banks look very carefully at the overall financial stability of a project, in particular the cover ratio, which reflects the safety margin on the debt repayments. Cover ratios can be linked to another key indicator of financial stability -- the overall rate of return on the project. If this falls below about 9% in real terms, meaning net of inflation, most banks regard the level of security on their loan as inadequate as it requires only a small perturbation in the performance of the plant to make it difficult to repay the annual loan instalments. This argument applies to any type of investment, not just wind farms. But these are particularly susceptible to variations in income due to the year-by-year variations in wind strength and risk rises as rate of return falls (figure 2). When these risks are taken into account, project rates of return for wind plant are usually pushed towards 10%, in order to make the risks acceptable.
If, nevertheless, it is assumed that the cheapest NFFO and SRO energy prices are linked to a 9% rate of return, it is possible to work backwards and infer that total project costs including development and bank fees are about £750/kW. This is significantly lower than anything which has yet been achieved in the UK, but may be possible -- in 1999 if not in 1995.
Cost cutting factors
One significant factor in bringing costs down is the continuing fall in turbine prices -- NFFO-2 levels were around £630/kW whereas the current level is nearer £500/kW and further reductions are anticipated, provided the recent rise in price of some raw materials does not nullify the downward trend. Cheaper turbines are also due to several factors, including larger machines, which produce more energy per unit weight (Windpower Monthly, October 1994), and the benefits of mass production and more efficient manufacturing.
Larger wind turbines have also brought down balance of plant costs -- the remaining project costs once the turbines have been bought. Fewer machines brings down the number of foundations needed and also lower infrastructure costs. In NFFO-2, balance of plant costs averaged about £400/kW. In NFFO-3, they seem to have fallen to about £250/kW in order to bring overall project costs down to around £750/kW.
At this price, projects in America and Denmark are perhaps viable today, but in the UK, the more difficult terrain tends to push costs above those possible elsewhere. Only if wind speeds are around 9 m/s is a project cost of £750/kW possible in the UK in 1995 (figure 3), and only assuming operating costs are also pared to the bone. However, even though £750/kW may not be generally achievable in the UK right now, there is every expectation that it will be by 1999.
The availability of high wind speed sites, especially in Scotland, was a contributory factor in keeping prices low. Wind speeds at sites already commissioned in England and Wales range up to 9 m/s and studies by the Energy Technology Support Unit (ETSU) for the Department of Trade and Industry indicate that there are over 1000 square kilometres of land with wind speeds above 10 m/s in Scotland and correspondingly greater areas are available at wind speeds down to 8 m/s. Wind speeds at most of the Scottish sites may be expected to be in the vicinity of 9 m/s in order to achieve the very low level of bids that were achieved there. Even in England and Wales, where costs may be slightly less, the wind farm sites need wind speeds around or higher than 8 m/s to be viable. It is these very high speeds which will push British wind energy prices towards the lowest levels achievable anywhere in Europe.
The real cost of NFFO
As the cost of renewable energy falls, it is converging with that of electricity from conventional fuel sources. This price parity is the stated aim of the UK government's market support for renewables. As such, accurate measurement of the converging prices is essential as it determines the cost to the government of supporting wind and other technologies. Responsibility for this task lies with the Office of Electricity Regulation (OFFER). It, however, is clearly confused about how to advise government, referring to three different parameters before homing in on "pool price" as the marker. A peculiarly British invention, pool price is supposedly the cost of electricity bought from a largely fictitious "pool" of the stuff. Most electricity analysts, however, wrote off pool price as a meaningful yardstick years ago. Only about 12% of electricity is genuinely traded in the pool and even large industrial users pay, on average, 50% more.
Yet OFFER tenaciously defends its use of pool price to measure the cost of renewables support, even though, at £0.0245/kWh, it bears little relationship to the actual costs of generation from new plant in the UK. Based on OFFER's system of measurement, the government believes the annual cost of NFFO-3 for all renewables is about £20 million. If OFFER were to use a generation cost based on reality -- in practice this is around £0.034/kWh with domestic consumers paying around £0.09/kWh -- the annual cost of the government's support for the wind component of NFFO would be nearer zero.
Futhermore, with particular reference to wind, OFFER's calculation of the cost of NFFO is extremely crude; it does not appear to allow for the fact that the value of electricity -- even in the pool -- is much higher in winter when more energy is generated. Recently a temporary power shortage pushed daytime average values to around £0.08/kWh and peak values to over £0.20 on several occasions. This has provoked renewed calls from many quarters for a review of the procedures.
There is another reason, connected to the value of electricity, why pool price is the wrong reference for renewables. The output from renewable plant is mostly fed into the local utility networks at 11 or 33 kV and, by and large, will be used locally. It therefore has a higher value than power from centralised generation, which must be transmitted at high voltage before being sold to the utilities. High voltage transmission costs, alone, add about £0.005/kWh to electricity costs in England and Wales -- and regional distribution utilities will totally avoid these charges as much of their renewable generation will be absorbed locally. The higher value of locally generated electricity -- embedded generation -- is explicitly recognised in the United States and implicitly in Denmark and Germany when setting tariffs for renewable generation. In the UK, however, the arguments are not yet acknowledged. Nonetheless it seems clear that the real value of wind energy -- at least £0.04/kWh, possibly more -- is very close to the generation costs apparent from the NFFO-3 bids -- provided the developers can deliver their promise. It also follows that the actual cost of the wind component of the NFFO is, in reality, much smaller than the official figure.