Nuclear power issue

http://www.epw.org.in/epw/uploads/articles/1035.pdf

The cost of nuclear power:

Different estimates put it in the range of Rs. 1.01 -1.98 per KWH

A question:

Presently total power ge. = 110 GW

Nuclear = 3.3 GW

Wind 3.6 GW

In 2020, even if we assume 20GW nuclear power, it would be meeting ~6% of energy needs of the country. Why is this so important then?

The earliest public comparison of the unit costs of electricity from Indian heavy water nuclear power plants and coal plants was presented at the Indian Institute of Science, Bangalore during a debate about the Kaiga reactor in 1988 [Reddy 1988]. It was then formally published in 1990 [Reddy 1990]. A more detailed version of the nuclear versus coal comparison was published as part of the comparative costing of 15 technologies of centralised and decentralised generation of electricity and electricity saving [Reddy et al 1990]. It was clear that nuclear power was uneconomical compared to many alternatives.

However, the atomic energy establishment did not react for almost 15 years. Hence, it is a very welcome step that Srinivasan et al and Thakur have published comments on Ramana et al (2005). The basic criticism of Srinivasan et al is that we have chosen specific plants to draw general conclusions. This requires a clarification of the purpose of the costing exercise. If the purpose was an academic study of the costs of coal and nuclear plants, then the focus should be on average costs. Such a study would permit an understanding of historical cost trends or experience/learning curves for the technology. However, the purpose of Ramana et al (2005) was to look at the comparison of nuclear and coal technologies from the point of view of a decision-making agency (for example, the Planning Commission).

Such an agency has to make an investment decision today in order to choose between these two technologies for the next project. For such an objective, what are important are the short-run marginal costs, i e, the cost of the next nuclear plant and the next coal plant. This means that the comparison has to be between the best nuclear plant and the best coal plant that have already been commissioned. In making this choice, the data on variables like the expenditure pattern, fuel consumption, have to be drawn from the latest plants, which in the case of nuclear and coal plants had to be Kaiga I and II and RTPS respectively. Thus, the average costs of nuclear and coal plants are irrelevant.

Plants that are under construction, but not yet commissioned (for example, Kaiga III and IV) should not be included as far the comparison is concerned. But, they must be immediately brought into the comparison as soon as they are commissioned and the actual (rather than expected) data becomes available. This is mainly because actual numbers invariably turn out to be different from expected numbers.

Cost Overruns

This is amply borne out by the department of atomic energy’s (DAE) record of cost overruns at all the reactors it has constructed. In our paper, we have already documented the cost overruns at all the reactors that had been commissioned before our paper was written [Ramana et al 2005:

1765]. Since then the Tarapur III and IV reactors have been commissioned. These reactors were initially estimated to cost Rs 2,427.51 crore, later revised to Rs 6,421 crore. Thus, as the Comptroller and Auditor General (CAG) pointed out, “excluding the interest during construction of Rs 1,580 crore, the estimated cost went up by 99 per cent” [CAG 1999]. The final cost was about Rs 6,200 crore [Subramanian 2006]. Futuristic developments in technologies pertain only to long-run marginal costing rather than to investment decisions today. Much about futuristic technologies is unpredictable, especially costs. As Niels Bohr is reported to have said, “Prediction is very difficult, especially about the future”. Practically all of the predictions made by the DAE so far have come to nought. This means that a comparison between Kaiga I and II and RTPS VII is the right basis for an investment decision today. Even [Thakur 2005] does not, and indeed cannot, argue that electricity from Kaiga is cost competitive with RTPS VII. However, to bend over backwards to favour the nuclear case, the comparison made by Ramana et al (2005) included the yet-to-be commissioned Kaiga III and IV plants even though the values for some (but not all) of the data for these plants have no empirical basis and involved guesses and estimates. In doing so, the projected values for construction time, performance, and input costs were used even though they are not based on actual experience and achievements. For the most part, it is this data that Thakur has called for changing based on “correct” inputs, though all that “correct” really means is currently assumed data. The suggestions made have been accepted, with the one exception of heavy water (HW) price that we will discuss later, and what is presented in this response is the outcome of reworking all the calculations as per the suggestions of Thakur (2005). In particular, the reworked figures have taken:

· The capital expenditure for Kaiga III and IV to be Rs 2,727 crore.

· The auxiliary consumption for all the Kaiga stations to be 9.5 per cent.

A crucial issue concerns the heavy water

price to be used for the calculation. In our

paper, we used a heavy water price of

Rs 24,880/kg based on the CAG’s estimate

of the cost of HW at the Manuguru plant

Economics of Nuclear

Power: Subsidies and

Competitiveness

170 Economic and Political Weekly January 13, 2007 [CAG 1994]. In contrast, Thakur and Srinivasan et al suggest that the price should be Rs 12,525/kg in 2003. But Rs 12,525 is a government notified price, and it is by no means clear whether this price reflects the actual cost of production, or whether it is an administrated price that is lower than the actual cost, implying an element of subsidy. In a paper submitted to this journal, I have elaborated on why there does seem to be a substantial subsidy being offered to the Nuclear Power Corporation (NPC) through lowered heavy water costs [Ramana forthcoming]. Based on the limited amount of data publicly available, I estimate that the production cost of heavy water should be over Rs 20,000/kg in 2003 rupees and could well be more. There is thus good reason to not take the Rs 12,525/ kg figure at face value.

Cost of Heavy Water

Given this uncertainty about what the cost of heavy water is, we redo the calculations with two values for the heavy water price – Rs 24,880/kg and Rs 12,525/kg. For the first price, the costs of the different options are given as a function of the real discount rates in Table 1. Above a crossover discount rate of 3.85 per cent, the levelised cost per kWh for nuclear electricity even with the projected costs of Kaiga III and IV is more expensive than that for coal. The crossover rate between Kaiga I and II and RTPS VII is only 2.7 per cent. By no means can these be considered competitive.

For a heavy water price of Rs 12,525/kg, the crossover discount rates are naturally higher; the rate at which the cost of electricity from Kaiga III and IV becomes higher than RTPS VII turns out to be 5.45 per cent (Table 2). However, when one compares the actually constructed Kaiga I and II reactors for which the prices are established and RTPS VII, the crossover rate is less than 3.6 per cent.

Leaving aside the issue of a fair cost for heavy water, the question is whether 5.45 per cent is high or low. There is a longstanding debate about what constitutes a reasonable choice of discount rate. Typical values chosen in costing electricity generation (or saving) technologies have ranged from 8 per cent to 10 per cent (real values) [see Shukla et al 2003 and Nouni et al 2006]. If one were to look to stock market returns as a measure of the cost of capital, nominal return rates have been 23 per cent on average between 1981 and 2005 [Varma and Barua 2006]. Since current inflation rates are less than 6 per cent, this means a real rate of return of over 17 per cent. All of these suggest that 5.45 per cent is a low figure.

However, rather than impose our views on what constitutes a fair rate of return, we redo the entire calculation at the nominal discount rate of 12 per cent which is used by various planning bodies in the country [Bose 2000; Public Accounts Committee (1991-92), 1992].1 This also allows for easier comparison with other studies of the relative economics of different sources of electricity which use a nominal discount rate. We have listed the levelised costs at this nominal discount rate for various input values in Table 3. Even with the most optimistic choice of input parameters and assumptions, nuclear power is not cheaper than electricity from RTPS VII. When we wrote our paper, we were aware that some of our assumptions may be wrong. In part, this was because the NPC was not forthcoming with data [see endnote 2 in Ramana et al 2005]. Some of these assumptions are the ones that Thakur and Srinivasan et al have questioned. However, in order to make amends for any assumptions unfavourable to nuclear power, we did make a number of assumptions favourable to nuclear power. Just as variations in the assumption about the price of heavy water results in changes in the cost of electricity, many of these make a difference to the relative economics of nuclear power. For example, we assumed that contrary to actual practice RTPS VII would have to spend Rs 174 to dispose of each tonne of fly ash. In actual fact, the fly ash is used to make saleable products like mosaic tiles and bricks, and hence the power plant does not pay anything for its disposal [Ghanashyam 2006; Staff Correspondent 2006]. Removing this charge lowers the levelised cost of production at RTPS VII significantly (Table 3). Another significant expense that we did not include in the cost of nuclear power is that of reprocessing the highly radioactive spent fuel that comes out of the Kaiga reactors. At the time when we wrote the paper, there were no public estimates of this cost. Since then, based on an analysis of the government budget figures for the Kalpakkam Reprocessing Plant (KARP), we have estimated that the levelised cost of reprocessing each kilogram of spent fuel is about Rs 26,000 [Ramana and Suchitra forthcoming]. Currently, the DAE bears this entire cost [Thakur and Chaurasia 2005]. This is yet another subsidy offered by the DAE, and effectively the taxpayer, to the NPC. If even a reasonable fraction of this, say Rs 10,000/kg, is assigned to and internalised within the cost of producing electricity at heavy water reactors, then it would become far more expensive than thermal power from coal (Table 3). All of this should not be surprising.

Studies in many countries have found

similar results. In the country with the most

nuclear reactors, the US, a detailed study

conducted at the Massachusetts Institute

of Technology compared the constant or

“levelised” price of electricity over the life

of a power plant that would be necessary

to cover all operating expenses and taxes

and provide an acceptable return to investors

for a number of base load electricity

sources [Deutch et al 2003]. The study

found that unless there are dramatic

improvements in nuclear cost factors (and

none in the other technologies), nuclear

power is simply not competitive with the

other technologies. As things stand, the

MIT study’s model estimated a real levelised

cost of 6.7 cents/kWh for nuclear power,

4.2 cents/kWh for pulverised coal, and

3.8-5.6 cents/kWh for gas. It is no wonder

Table 1: Levelised Cost of Electricity with

Heavy Water Price of Rs 24,880/kg

Discount Rate Kaiga I Kaiga III RTPS VII

(Per Cent) and II and IV

2 1.28 1.19 1.36

3 1.43 1.31 1.39

4 1.61 1.43 1.42

5 1.81 1.57 1.45

6 2.04 1.72 1.49

Table 2: Levelised Cost of Electricity with

Heavy Water Price of Rs 12,525/kg

Discount Rate Kaiga I Kaiga III RTPS VII

(Per Cent) and II and IV

2 1.18 1.10 1.36

3 1.31 1.19 1.39

4 1.47 1.29 1.42

5 1.65 1.41 1.45

6 1.86 1.55 1.49

Table 3: Levelised Cost of Electricity at a

Nominal Discount Rate of 12 Per Cent

Heavy Water Kaiga I Kaiga III RTPS VII

Price and II and IV

Rs 12,525/kg 1.97 1.63 1.56

Rs 24,880/kg 2.11 1.77 1.56

Rs 12,525/kg, no

ash disposal cost 1.97 1.63 1.43

Rs 12,525/kg, waste

management cost

of Rs 10,000/kg 2.20 1.86 1.56 Note: All prices are in 2003 rupees. Economic and Political Weekly January 13, 2007 171 that there have been no new nuclear reactor orders in the US for nearly three decades. To conclude, the NPC’s claim that nuclear power is cost competitive can only make sense if one ignores the significant subsidies offered to it through the DAE. Even when those are included, nuclear power is not really economically viable in a competitive environment.

Email: ramana@isec.ac.in

Note

1 The relation between real and nominal discount rates is in principle simple. But if the inflation rate varies, then it is no longer so straightforward. The real (dr) and nominal (dn) discount rates in any given year are related by: (1+dn) = (1+dr)*(1+i(Y)), where i(Y) is the inflation rate for that year. In projects such as the ones we are considering here, expenditures are made over several years with different inflation rates. Thus it is not possible to simultaneously talk about a uniform nominal and a uniform real discount rate.

US positive on India nuclear deal

The Bush administration has said that Washington will continue to work for making the landmark India nuclear deal a reality.

'We are still very supportive of the deal. We still would like to see it move forward, US State Department spokesman Sean McCormack said.

.

The Indian government and the political system is engaged in a debate, and we'll see what the outcome out of that debate is,' he said.

Asked if believed that the deal was 'dead', McCormack replied, ' We continue to support the agreement and would like to move forward with it' refusing to make predictions if the process will go through next year.

'I'm not going to make any predictions. We continue to support it though.' McCormack said.

Refusing to get into the debate, that is currently going in India between the political parties over the deal, he said 'I understand there are a lot of intensive domestic political discussions in India.

The Indian government and the political system will play out those discussions. It's not something that we're going to directly participate in.'

Indian-Americans to persuade BJP, Left to accept deal

http://www.hindu.com/2007/10/27/stories/2007102754041400.htm

Washington: Concerned over the opposition of civilian nuclear deal in India, a group of Indian-American community leaders are planning to travel to New Delhi to convince the BJP and the Left parties to accept the deal as it was in the best interests of the two countries.

“Many of us are planning a trip to New Delhi in November to meet leaders like Atal Bihari Vajpayee, L.K. Advani, Brajesh Mishra, Prakash Karat and others to exchange ideas,” chairman of the Dallas based U.S.-India Forum, Ashok Mago told PTI over telephone.

The comments of the former National Security Adviser, Brajesh Mishra, has ignited a new ray of hope among the supporters of the agreement, he said referring to Mr. Mishra’s recent interview in which he had asked the leader to go ahead with the deal saying “Clinton administration would not have offered such a deal.” The Indian-American community leaders would make the leaders, opposing the deal, aware about how difficult it was to get the bill passed and that the similar opportunity might not be there in the foreseeable future, he said.

“We love the country of our birth, but I doubt if Indian-Americans have the stamina to do it all over again. We would ask BJP leadership that there would never be a better opportunity to show that you care more for the betterment of common people than party politics and are willing and ready to support the government on this issue. Such a step will enhance creditability not only among the masses in India but all over the world,” Mr. Mago said. Mr. Mago and his organisation were instrumental in lining up the Texas Delegation in the House of Representatives and the Senate to vote for the Hyde Act. — PTI