Dr Andrew Blakers
Engineering Department
Australian National University
Canberra, 0200
February 1995
In 1983 the Commonwealth blocked the construction of the Gordon-below-Franklin hydroelectric scheme in the western Tasmanian wilderness. In a fit of pique the Hydroelectric Commission of Tasmania (HECT) immediately began the construction of two new hydro schemes, the King scheme and the Anthony scheme. In addition to being environmentally destructive these two new schemes have turned out to be financial disasters. They cost a total of $1,200 million ($2,500 per Tasmanian) to construct and can deliver 112 MW (average) power output. Interest is costing $140 million per year, with no return from energy sales. The main problem is that the HECT grossly overestimated demand growth. This paper examines the Tasmanian electricity system over the past decade, and points out the low risk nature of energy conservation and wind energy as `new' sources of electricity.
In 1983, after a protracted campaign by conservationists, the construction of the Gordon-below-Franklin dam was halted by federal intervention. The Hydro-Electric Commission of Tasmania (HECT) instead proceeded with the construction of two new hydro-electric schemes, the King scheme and the Anthony scheme. It is interesting to re-examine the controversy after 10 years, and see what lessons can be learn.
The issues at the centre of the economic arguments over the Gordon scheme were the likely level of future energy consumption and the relative cost of alternatives such as energy conservation and wind energy. On both of these points HECT predictions proved to be seriously in error and those of conservationists remarkably accurate.
In this paper historical dollar values have been converted to 1993 dollar values using Bureau of Statistics figures for consumer price index movements.
The Tasmanian electricity system is almost entirely based on hydroelectricity, with the exception of the 240 MW Bell Bay oil fired power station. Bell Bay produces electricity at a marginal cost of about 6 cents per kilowatt-hour (c/kWh). For many decades the HECT pursued a program of damming Tasmanian rivers. The Gordon-below-Franklin dam was to be followed by several more on the Gordon and Franklin rivers, as well as on other rivers.
Alternatives to the Gordon scheme put forward by conservationists included the vigorous use of energy conservation measures and the construction of large numbers of windgenerators. Energy conservation is equivalent to a new energy source since energy saved is available for use by others. Energy audits are now common, and usually lead to substantial energy savings that pay for themselves in a few years. This is equivalent to saying that electricity conservation costs only a few cents per kWh saved, which is less than the cost of electricity from new hydroelectric schemes (or other sources).
The cost of wind energy has been steadily falling as mass production commences in a number of countries. The present cost of large (50 MW) windfarms is around $1,500/kW [Blakers, 1991]. In a windy site in Tasmania wind energy on a large scale would cost about 6 c/kWh. This assumes a 20 year lifetime and a 5% real discount rate (a way of relating the value of money in the future to the value of money today). Further improvements in wind energy economics over the next 10-20 years may reduce wind energy costs to 4-5 c/kWh. A study performed by Greenwood [1985] concluded that it would be technically possible to generate 100,000 Gigawatt hours (GWh) per year from wind energy in Tasmania, which is 11 times Tasmania's current electricity consumption. Unfortunately, despite Tasmania's excellent wind regime, the HECT has yet to invest in a demonstration windgenerator.
In 1983 the HECT [HEC, 1983] estimated the cost of wind energy to be about 15 c/kWh (1993$). Outhred and Blakers [1983] estimated at the time that the cost of wind energy in Tasmania would be about 6 c/kWh (1993$). Recently the Tasmanian Energy Minister, Robin Gray, was quoted as saying that wind energy in Tasmania would cost 5-6 c/kWh. It is unfortunate that when he was premier in 1983 he did not hold the same view. An important point is that windgenerators would not be required until demand exceeds supply, allowing further cost improvements. In hindsight it can be seen that the HECT overestimated the cost of wind energy by 2-3 times.
Hydroelectric schemes are characterised by having long construction times - typically 10 years. This means that energy demand needs to be accurately estimated far in advance. If an underestimate is made then demand may outstrip supply. On the other hand, if an overestimate is made then the finished scheme may not be needed for some years after completion, leading to idle capacity. The interest bill on the scheme mounts, with no income from energy sales to pay for it. Unlike thermal generators, hydroelectric schemes cannot be `mothballed' to save fuel costs.
In contrast to hydroelectric schemes, energy conservation and wind energy deliver extra generation capacity in small blocks. Windgenerators can be installed or energy conservation measures taken as needed, to more closely follow demand. The cost of making mistakes in demand forecasting is greatly reduced (although there is still risk - for example, the uncertain success associated with programs designed to encourage consumers to save energy). Another way of saying this is that energy conservation and wind energy are inherently less risky than hydroelectric schemes. The events of the last 10 years in Tasmania dramatically bear this out.
In August 1983 the HECT projected large demand growth rates as part of its report on the King and Anthony schemes. Figure 1 shows the HECT demand growth projections produced in 1983, together with the demand growth actually experienced. The predictions of conservation groups were much closer to reality. The announcement by Comalco in early 1994 of a 30% cut-back in production took the idle capacity in the HECT system to the equivalent of 3,400 GWh/year (390 MWaverage) or 30% of system capacity. As a result the Bell Bay oil fired power station has been `mothballed'. The idle capacity is roughly double the average annual energy output of the Gordon scheme (which flooded Lake Pedder).
Of course, accurately predicting the future is always difficult. The important point is that a less risky decision could have been made to constuct a windfarm instead of the King and Anthony schemes. Actual construction of the windfarm would have been delayed until shortly before the extra power was needed. It is clear from figure 1 this has not yet occurred, and on present growth rates may not occur for many years (if ever).
Figure 1: Demand growth in Tasmania since 1983, showing the predictions of the HECT in 1983 and the predictions of the SWTC [Lambert, 1982] in 1979. Also shown is the supply capacity of the HECT system, assuming average rainfall and full operation of the Bell Bay oil fired power station. The projected drop in demand in 1994 and 1995 as a consequence of Comalco's cut-back is indicated.
The 1983 report by the HECT recommending the construction of the King and Anthony schemes estimated the likely capital cost and energy cost of the two schemes using a 5% real discount rate. The HECT annual reports over the past 10 years contain information about actual capital expenditure on the King and Anthony schemes, the interest-during-construction costs ascribed to each scheme and the prevailing interest rates. This allows the calculation of the total cost of the two schemes, including actual interest costs. The total costs of the King and Anthony schemes were approximately $623 million and $574 million respectively, which sum to $1,200 million [Blakers, 1994]. The costs predicted in 1983 (1993$) were $556 million and $421 million for the King and Anthony schemes respectively, which sum to $977 million. There was a cost overrun of 23% due to higher than expected interest rates and a stretched construction timetable.
The original 1983 HECT proposal envisaged completion of the Anthony scheme in 1990 and the King in 1991. Actual completion dates were 1994 and 1992 respectively. Interestingly, the general Manager of the HECT, Mr Graeme Longbottom, in both the 1992 and 1993 HEC annual reports, claimed that both the King and Anthony schemes would be finished ahead of time and under budget.
The HECT has accumulated a debt of $1,700 million ($3,600 per Tasmanian) [HECT, 1993]. About $1,200 million (70%) of this debt is due to the King and Anthony schemes. If these schemes had not been built the HECT would now have a low debt-to-equity ratio, and yet would not be in serious danger of power rationing. The most likely cause of a future energy shortage would be a drought. The King and Anthony schemes have small storage lakes, which only add about 2% to the combined energy storage capacity of the HECT system. Adding more hydroelectric capacity without large storage is not an effective way to reduce the effects of drought.
The HEC received compensation money from the Commonwealth Government for the blocking of the Gordon-below-Franklin dam. This compensation was claimed because the cost of alternatives was allegedly higher. This compensation money could have been applied to any alternative scheme (such as energy conservation or windgenerators). As it turned out neither the Gordon dam nor any alternative scheme was necessary. The cumulative present value of this compensation money is $500 million, including interest. The HEC believes that further compensation will be due when it builds its next power scheme. If the HEC had built the Gordon dam its debt would now stand at around $2,500 million. The Commonwealth saved the HEC from financial disaster by blocking construction of the dam. Rather than receive further compensation the HEC should be grateful to the Commonwealth for saving itself from its own folly!
Electricity costs are graphed along with selling price in figure 2. The cost of electricity from the King and Anthony schemes would be 8.3 c/kWh and 10.6 c/kWh respectively (including distribution costs and losses) if all of the output could be usefully used immediately after completion of constuction. This is greater than the cost of electricity from Bell Bay.
The construction of the King and Anthony schemes over the period 1983-1994 was simply unnecessary. If the Gordon scheme had been built it would have led to even greater idle capacity. Idle hydro-electric capacity accumulates interest costs without a return from energy sales. The HECT's average interest rate on borrowed capital is 11.75% [Tasmanian Treasury, 1993]. The interest bill on the King and Anthony schemes therefore amounts to $140 million per year.
It is likely that the King and Anthony schemes will not be needed until at least the year 2000 (if ever). Interest compounded for the next 7 years at an assumed interest rate of 5% (real) adds about 40% to the overall cost of the King and Anthony schemes. The extra cost of this idle capacity (ie, 40%) is shown in figure 2.
A major advantage of wind energy (and energy conservation) over hydroelectricity schemes is apparent. Construction of each wind farm, which only takes a year or so, can be delayed until its output is really necessary, thus minimising idle capacity. Windfarms are inherently less risky than hydroelectricity. In addition, `wind droughts' do not necessarily coincide with rainfall droughts.
Figure 2: Energy cost for the King and Anthony schemes, together with a notional cost for wind energy (including transmission losses and a proportion of HECT overheads, totalling 2 c/kWh). Also shown is the extra cost of idle capacity (the two schemes are unlikely to be needed until the year 2000 at the earliest) and the selling price of electricity in Tasmania - the major industrial customers take 64% of the electricity at an average price of 2.5 c/kWh, and other customers take 36% of the electricity at an average price of 10 c/kWh. The average price of all electricity sold is 5 c/kWh. As can be seen, the generation costs from new sources exceeds the average selling price.
The HECT has two classes of consumers - the major industrial customers who take 64% of the electricity at an average price of 2.5 c/kWh, and other customers who take 36% of the electricity at an average price of 10 c/kWh. The average price of all electricity sold is about 5 c/kWh. As can be seen in figure 2, the generation costs from the new sources exceeds the average selling price. In this situation new generation units can never pay back their cost unless some consumers are discriminated against by high electricity costs.
The low price paid by the major industrial customers is a direct subsidy. One consequence of the subsidy is that new industries wishing to set up in Tasmania face high and uncompetitive electricity costs, since the marginal cost of supply is 8-10 c/kWh. Raising electricity prices would leave room to offer new companies reasonable tariffs while allowing more rapid reductions in the HECT debt. However, political considerations make this unpalatable.
The oversupply of electricity in Tasmania puts the state government in a very weak bargaining position when negotiating the price of electricity with the major industrial customers. These companies are aware of the idle capacity in the Tasmanian system and the great need to obtain income to service the $1,700 million accumulated debt of the HECT. Pressure can be applied to the Tasmanian government, by way of threats to close factories in Tasmania, in order to induce the HECT to drop prices or to sell generating units to individual companies at a very low price. Comalco, which traditionally takes about a quarter of Tasmania's electricity, has been having discussions with the HECT on both of the above points. In February Comalco cut aluminium production in Tasmania by 30%, increasing the overcapacity in the Tasmanian system to 30%. Media reports strongly suggest that Comalco will cease operations in Tasmania soon. Both Comalco and the HECT are aware of the dramatic effect that this would have on HECT finances.
The HECT has a debt to equity ratio of 55 to 45 [HECT, 1993] despite decades of dam building. This puts the HECT in a difficult position if electricity sales do not rise. Energy conservation measures can save large amounts of electricity at costs far below 8 c/kWh. Unfortunately, the unwise investment of the HECT in two new hydroelectric schemes which are not needed gives the commission a strong disincentive for energy conservation.
The HECT built at least three hydro schemes too many (the King, Anthony and Pieman schemes). The need to service the HECT's debt from energy sales means that the HECT has a strong disincentive to conserve electricity. It is also puts the HECT in a weak bargaining position when negotiating with Comalco and the other large users of electricity. Tasmania is saddled with $1200 million of additional unnecessary debt and several lovely rivers have been drowned needlessly. The only bright spot is that the original Lake Pedder could be restored by lowering the level of Lake Gordon without any threat to security of electricity supply. In fact, if the Gordon scheme was entirely removed it would only reduce the surplus capacity of the HECT by about half.