The final scoping report for Eskom's proposed new nuclear plant, the PBMR, has been completed and is now open to the public for comment. This is an important stage in the Environmental Impact Assessment being undertaken.
 
 
Reading Material


The Steve Thomas Report (part 3)

The Steve Thomas Report part 1                              The Steve Thomas Report part 2
7.2 The likely world market for the PBMR

    PBMR (Pty) Ltd and Eskom have always been very vague about target markets and countries as wide-ranging as Chile, Cyprus, Turkey, Saudi Arabia and Egypt have all been mentioned as possible targets. There appears to be little basis for this speculation and these markets should be discounted until there is some substantive evidence to back them up.
     The DFR (PBMR (Pty) Ltd, 2002a, p 50) is ludicrously over-optimistic, given the absence of anything remotely close to a firm order, suggesting that: ‘the sale of PBMR plants and fuel is more likely to be constrained by supply capacity limitations than by demand.’ It backs this up saying:
The market analysis shows that the potential exists for the market to conservatively absorb up to 235 five-pack plants (1 175 modules) over the two decades following the start-up of the demonstration plant. This represents only 3.3 per cent of the world demand for new generation capacity. Notwithstanding this excellent potential, the base-case sales scenario adopted in the enterprise business plan forecasts the sale of only 258 modules over the evaluation period of 25 years, and is therefore conservative.
     Despite the fact that Exelon had already withdrawn from the project when it was published, the FEIR (PBMR, 2002b) still anticipated commercial sales beginning in 2006 with 15 units going to Exelon in the period 2006-8 and a total of 44 units by 2017. Eskom sales were expected to be at a much slower rate, starting in 2007, completing the 10-unit order by 2012 and ordering a total of 20 units by 2017. Other customers were expected to buy 76 units by 2017. So in the first 12 years of the commercial phase, the FEIR forecast sales of 140 units, a slightly faster rate of sales than the DFR.
     Given that over the past decade, the volume of nuclear plant ordered has been only one or two 1000MW units a year, this seems far from conservative. In fact, it seems clear that PBMR (Pty) Ltd has carried out no detailed market analysis on a country-by-country basis and projections are simply an arbitrary percentage of an overall market for power plants. This issue was raised by LRC as Comments on the DFR (Register of Comments, 2002, 28.137) but the response does not make much sense and does not answer the question. It states;
The market studies were based on 53 plants, only one of which is to be sold to Eskom.   Thorough market studies were done as part of the business case.  We are not sure on what the statement “it seems likely that the world market for nuclear power may be no more than 1 or 2 units per year” is based, especially since the world market for new power stations is about $70 million per year.
     No mention is made elsewhere of ‘the market studies of 53 plants’. Since $70 million would only, on PBMR (Pty) Ltd’s figures, cover about half the cost of one PBMR module, it is not clear what the response means.
The fact that a significant percentage of the market is effectively closed to nuclear power by political decision is not taken into account. Even so, it should be noted the DFR represents a significant downgrading of sales forecasts to about 10 units a year from earlier when Nicholls (Nicholls, 2000) forecast 30 units per year.
This weakness was acknowledged by the new CEO of PBMR (Pty) Ltd in September 2004 when he said there was a need for ‘a "much more detailed marketing strategy" with "a strong focus on customers' needs. He said marketing strategies would be tailored to a given country or customer, versus a more generic strategy followed in the past.’
Such studies would quickly reveal that for much of the world, new orders for nuclear plants are not feasible. In Europe, many countries have made a decision not to build nuclear power plants, e.g., Austria, Denmark, and Norway or are phasing out nuclear power, e.g., Germany, Italy, Sweden, Belgium the Netherlands and Switzerland or not expanding existing capacity, e.g., Spain. The UK government carried out a review of nuclear power in 2003 and found no case for new nuclear power orders. France decided in November 2004 to build a new nuclear power plant of a French design, EPR, a 1500MW design based PWR technology, and it seems highly unlikely it would abandon this in favour of the PBMR. The medium-term prospects for PBMR sales in Europe therefore appear minimal.
     In the USA, PBMR (Pty) Ltd’s hopes were based on Exelon getting license approval for the PBMR and launching the commercial programme by ordering 10 units. It is clear this will not happen now and while some utilities offer supportive statements to the technology, as expressions of intent to buy plants, these are essentially worthless.
For example,  the CEO of Exelon (John W Rowe) was reported in May 2005 that:
‘the high price of natural gas is an incentive to build new plants, but that an offsetting factor is the continuing low cost of coal. The lack of a solution for nuclear waste is also a deterrent.’
While the CEO of Dominion, another large US utility often mentioned when new nuclear orders are mooted said
“We aren't going to build a nuclear plant anytime soon. Standard & Poor's and Moody's would have a heart attack," said Mr. Capps referring to the debt-rating agencies. "And my chief financial officer would, too."
     The main expected export market therefore appears to be China, but despite several years of discussions, China has made no commitment to South African PBMR technology. Tsinghua University has the only operating PBMR in the world, a 10MW unit that went critical in 2000 using German fuel technology. Tsinghua University is collaborating with US interests from the Massachusetts Institute of Technology on a competitor to the South African PBMR.  Overall it is far from clear who Chinese companies will choose to collaborate with, but all experience shows that Chinese interests will try to ‘indigenise’ any technology they pursue so even if they do collaborate with PBMR (Pty) Ltd, and orders are placed, South African content to these sales would low and the net benefit of these sales to South Africa small.
     It seems more likely that China will produce its own design of PBMR, similar to that of PBMR (Pty) Ltd, which would supply any sales in China and would compete with the South African design in world markets. Nucleonics Week reported in June 2005 that Tsinghua University's Institute for Nuclear & New Energy Technology (INET) expected to complete the design for a commercial scale of plant (about 195MW) by 2006 and have a plant in operation by 2010.  These forecasts may be no more realistic than those of its South African counterpart but the intention to develop an independent design rather than import technology is clear.
     If a world market for high temperature gas-cooled reactors does develop, as well as competition from a Chinese vendor, the South African PBMR may face competition in international markets from the US vendor General Atomics and from Areva, companies that are both developing designs using prismatic fuel.
General Atomics supplied the demonstration HTGR built in the USA (Fort St Vrain) and has the advantage of being US-based and therefore politically well-placed to receive US government funds. Areva has less experience with HTGRs but its huge experience in reactor design and sales gives it advantages in international markets.
A pre-condition for any international sales appears to be obtaining safety approval from the US NRC. Without a US partner and with no sales in prospect, it is not clear why the USA should spend US taxpayers’ money reviewing the PBMR design. If PBMR (Pty) Ltd is to obtain licensing approval in the USA, it seems a large proportion of the cost will therefore have to be borne by PBMR (Pty) Ltd.

Required information

The Applicant should publish the PBMR (Pty) Ltd’s marketing plan and its strategy for gaining license approval from the US NRC in the FEIR

7.3 The South African market for PBMRs

     In the absence of foreign markets, this leaves Eskom as the most likely customer. Eskom has committed to build and operate the Demonstration Plant. It has said it will buy 10 units, but only ‘provided it's the lowest-cost alternative at the time the utility needs to add capacity’.  Note that the DFR (PBMR (Pty) Ltd, 2002, p 50) misleadingly does not include this caveat on cost, saying only: ‘Eskom has provided PBMR (Pty) Ltd with a letter of intent covering the purchase of a demonstration plant and 10 further units.’
     Eskom does not say in the FEIR whether, on current expectations of cost of a commercial unit it expects the condition that it be the ‘lowest-cost alternative’ to be met. Eskom should provide a detailed analysis of the economic conditions that would have to be met, including costs of the alternatives, such as coal, gas and renewables, as well as the cost of the PBMR, for the PBMR to be the cheapest alternative.
Given that commercial orders cannot be placed before about 2013, such calculations are highly speculative. In that time frame, it cannot be assumed that Eskom will exist in anything like its present form and the attractiveness of alternative technologies, such as gas-fired plant and renewables could have changed dramatically.
    In the second half of 2004, pressure on Eskom to commit unconditionally to buy several commercial units increased. In October 2004, Kriek said the PBMR (Pty) Ltd’s business plan ‘envisages Eskom committing up front to some 4,000 MW of PBMR capacity in South Africa, which would allow "economies of scale" and development of a commercially competitive product.’  This plan appeared to be endorsed by the government Minister for Public Enterprises, Alec Erwin, in his mid-term budget statement of November 26, 2004, when he said: ‘plans include the additional generation of 4,000MW to 5,000MW of electricity from pebble bed units located around the country.’ Tom Ferreira, communications manager for PBMR, said that around 4,000MW of electricity could be met by 24 PBMR units each with a generating capacity of 165MW. If the cost of these units was no more than the target cost of US$1000/kW, this would mean that Eskom was being asked to commit to making an investment of at least R25bn before the technology was economically or technologically proven.
     However, the signs are that Eskom itself wishes to distance itself from the project. The forecast time when new generating plant will be urgently needed is difficult to predict because of uncertainties about demand growth rates, the degree to which old plants can be refurbished and mothballed units returned to service. Steve Lennon, Eskom’s MD for resources and strategy suggested that 1000MW of new peaking capacity (power stations only required for times of peak demand) would be needed each year from 2005-09 with base-load capacity (power stations that operate throughout the year) needed from 2010 onwards.  Clearly the PBMR, which cannot be in service as a commercial option before 2015  at the earliest, is of little relevance to this immediate need for new capacity.
     The managerial changes in PBMR (Pty) Ltd in August 2004 when an IDC executive, Jaco Kriek, became CEO and a Department of Trade & Industry Director General, Alastair Ruiters became Chairman, replacing the predecessor from Eskom, Nic Terblanche were reported as being ‘intended to get the project out from under the management of South African utility Eskom, which does not want to be in the business of developing new nuclear technology.’ 
     This very much echoes the position taken by Exelon in 2002 when they withdrew from the project. These changes seem to be supported by the government. Nucleonics Week  reported:
Up to now, the chairman of Eskom Enterprises, Eskom's subsidiary for unregulated industry, has automatically held the PBMR chairmanship, but now it's not even certain that Eskom will be represented on the board. An informed source said the government is "not eager for Eskom to continue as an investor and a potential customer," in part because that would inevitably lead to conflict-of-interest situations.
     The CEO of Eskom confirmed this interpretation in evidence to the South African Parliament Portfolio Committee on Minerals and Energy. He said the IDC was to take over the leadership of the PBMR programme. Eskom would be "playing a lesser role (as a PBMR investor) as we go forward, because we are now going to take the role of customer".  He also seemed to suggest that the PBMR should not go forward without foreign investors. He said more international investors were needed "to be able to advance to the stage where we can construct the demonstration unit and have it commercially proven" and that Eskom would "dilute" its participation as an investor in the PBMR, and allow other investors to be brought in. He also seemed to confirm that PBMR would have to be the cheapest option if Eskom was to buy it: ‘if all of our technical and commercial criteria are met, we'll be taking the first set of units that are produced.’
     The South African government affirmed in October 2004 its commitment to open up the electricity generation sector to foreign investment. The Trade & Industry Minister, Alec Erwin , suggested that about a quarter of the investment needed up to 2009 would come from companies other than Eskom. This effectively removes from Eskom the obligation to ensure there is sufficient generating capacity for the country. It also in effect places Eskom in a competitive market. In this situation, it would be unreasonable to expect Eskom to compete with new generators if it was obliged to buy a number, specified by the government, of PBMRs regardless of whether they were the cheapest option or whether they were even required. The only logical commitment Eskom can be asked to make is that it orders PBMRs when it needs new capacity, provided it is the cheapest option available. In practice, this is a largely empty commitment because, if when it needed new capacity the PBMR was the cheapest option, it is hard to see why Eskom would not order it.
     When the PBMR project was launched, it was expected to be primarily an export project producing about 30 units per year, with two thirds of the units for export. Thomas argued (Thomas, 1999) that the world market forecast was implausible and no more than one or two units per year would be sold. Six years later, the overall world market for nuclear power plants looks no more promising and PBMR (Pty) Ltd has failed to identify any firm prospects export sales.

Required information

The FEIR should specify what obligation Eskom has to purchase commercial PBMRs.

7.4 Benefits to the South African economy
The PBMR programme has always been sold to the South African public as a generator of jobs and wealth. Nicholls (Nicholls, 2000) suggested that the programme would generate 204,546 jobs and additional annual GDP of R18331m (the apparent precision of these inevitably highly speculative forecasts is grotesque). This was on the basis of a total market of 30 units per year, 20 of which were for export a local content of 50 per cent and 10 of which were for South Africa with local content of 81 per cent. The DFR (PBMR (Pty) Ltd, 2002a, p 55) projects annual sales of 10 units with local content for South African units of 69 per cent (48 per cent for the Demonstration Plant) and for export units, the South African content would be 43-65 per cent depending on the market (developed or developing country) and on how many units were sold. These are no more than targets and the actual percentage would be negotiated on an individual basis. If the market for PBMRs was disappointing or a large market was opening up, it may well be necessary to accept lower percentages rather than jeopardising sales. For example, China would be likely to require a very high local content.
     Clearly the lower forecast sales volume and local content figures will dramatically reduce the jobs and economic effects forecast by Nicholls in 2000, perhaps by 75 per cent and the DFR showed figures of 63,719 jobs and GDP of R8522m (again grotesquely over-precise).
     However, it is necessary to look at how these figures were generated. The DFR projects a unit cost for commercial units of about R180m. It forecasts that 40 permanent jobs will be created at the Demonstration Plant site plus about 1400 local construction jobs for about two years. The number of people working in manufacturing plants is forecast to be about 450 (PBMR (Pty) Ltd, 2002b, p 191). If we assume local content is on average about 60 per cent, this means the direct value to South Africa of 10 orders per year would be about R1000m. The number of direct jobs created would be of the order 1000.
     It is therefore clear that projections of 60,000 jobs and GDP increase of R8.5bn must be based on ‘second round’ effects of jobs created in the companies servicing the PBMR programme, for example the steel industry might be able to sell some more steel and in jobs created servicing the needs of the workers employed. Complex computer models of the economy as a whole are used to model these effects but the results should be treated with care (see PBMR (Pty) Ltd, 2002a, p 55-62). Any large programme of spending, if fed into this type of computer model, would produce large numbers of extra jobs and a large amount of extra GDP. For example, if the South African government embarked on a large programme of construction of pyramids, this would generate new wealth and jobs perhaps in the cement and construction sector, but the money would be entirely wasted because the pyramids would be useless. The export orders for the PBMR would generate no permanent jobs in South Africa for operators, and few if any temporary jobs for construction workers, while the pressure from customers would be to maximise their local content, so factory jobs (and second round effects on supplying industries such as the steel industry) would be much less than forecast.

Required information

Eskom should specify how many jobs will be directly created by the programme, for example as plant operators and manufacturing plant employees, specifying the assumptions that lie beneath these forecasts.

7.5 Risk analysis

The risk has always been that if international orders did not materialise, the South African public would be required to bail out the project by placing uneconomic orders. Thomas in 2000 wrote (Thomas, 2000):
However, what will happen if Eskom does go ahead without major international collaborators and the stream of orders does not materialise? Will South African politicians have the nerve to write off the project or will plants be built ahead of need in South Africa just to keep the capability in existence? National flagship projects have a tendency to live long after they should have been killed off and South African consumers will end up paying for a series of expensive white elephants.
     Even if the Demonstration Plant appears to be technologically successful (it will take several years of reliable operation before risk-averse foreign utilities will be convinced of this), that is no guarantee of international sales. PBMR (Pty) Ltd’s cost projections for the commercial units are based on very large and still entirely speculative scale economies. If these are not realised, the commercial design would not be competitive.
     The government appears to be acting to take control of the PBMR project away from Eskom, with IDC taking the lead role, while attempting to oblige Eskom to buy the plants. To some extent, these changes will be of limited interest to the South African public. From a theoretical point of view, if the government is going to oblige Eskom to build more PBMRs than would be economically optimal, it should reimburse Eskom from taxes. However, the public may be largely indifferent whether they pay extra to subsidise PBMRs through their taxes or through their electricity bills. It will be much more concerned about the potential huge loss of public money.
 
8. Conclusions

The National Environmental Management Act (NEMA) requires developers to demonstrate that their projects are economically sustainable. To judge economic sustainability, it is necessary to look at the life-cycle costs of the Demonstration Plant for the Pebble Bed Modular Reactor (PBMR). The Final Environmental Impact Report (FEIR) does not provide sufficient data to assess these. However, given that by its nature, a demonstration plant will not be economically viable in isolation, to judge whether the expenditure on the next phase is justified, it is also necessary to look at what the prospects of success for commercial PBMR units are.
     Eskom and PBMR (Pty) Ltd are keen to justify the Demonstration Plant on grounds of forecast benefits of a programme of commercial PBMR orders to the South African economy in the FEIR and the associated Detailed Feasibility Report (DFR). However, the FEIR does not provide any information on the economics of a commercial programme and in the responses to comments on the Draft EIR (Register of Comments, 2002), the consultants refused to answer questions on the programme stating ‘the present EIA is limited to a single demonstration module PBMR’.
     However, it is possible to draw conclusions on the economic sustainability of the Demonstration Plant and on any subsequent commercial programme by drawing together the information supplied by Eskom and PBMR (Pty) Ltd officials to various news media.

8.1 The Demonstration Plant

Conclusion 1:
Regardless of its success or otherwise, the Demonstration Plant will leave a substantial liability that will fall on South African public funds caused by the need to decommission the plant and the associated facilities, and to pay for the disposal of the spent fuel. The FEIR and the DFR do not quantify these liabilities, providing no information on spent fuel disposal and no usable information on expected decommissioning cost. However, experience in other countries suggests that decommissioning costs could be of the same order of magnitude as construction costs.

Conclusion 2:
Since details of the project were made public in 1998, costs of the Demonstration Plant have escalated by a factor of at least five. The project lead-time has slipped so that it is now apparently further away from commercial exploitation than it was in 1998 when commercial orders were forecast to take place from 2003. Now, seven years on, commercial orders are not forecast for about 10 years. This shows that the developers failed to understand the scale and nature of their task. There is still considerable scope in the next phase for further cost escalation and delay due to changes to the design and construction problems. The developers’ poor record to date gives little confidence in their ability to control costs and time schedules in the next, more expensive phase.

Conclusion 3:
Forecasts of other economic parameters, such as operating performance, operating cost and decommissioning cost have not been updated since 1998 and appear implausibly optimistic. It is understandable that developers of a project have an optimistic view of the project’s prospects – ‘appraisal optimism’. However, investment decisions should be taken on the basis of sober, unbiased judgements of the most likely outcomes, not the views of the project’s promoters.

Conclusion 4:
PBMR (Pty) Ltd successfully diversified some of the risk away from the South African public for the feasibility phase with foreign partners, Exelon and BNFL Ltd, sharing the costs. However, the cost of this phase (about R2bn) was far more than forecast and the absolute amount paid for by the South African public was not reduced. PBMR (Pty) Ltd has spoken optimistically over the past three years about the prospects of recruiting new partners to replace Exelon and BNFL (if as seems likely it cannot participate, but nothing has come of these negotiations. Until there is solid evidence of new partners being bought in, it must be assumed that the cost of the demonstration phase will fall substantially on the South African public, through Eskom, IDC, or direct government subsidies.

8.2 The commercial plants

Conclusion 5. PBMR (Pty) Ltd’s analysis of the world market for PBMRs is simplistic, taking no account of any of the commercial or political factors that would apply in key export markets. A particular concern is finance for export orders. This is an important issue for developing countries, which are likely to account for a significant proportion of the forecast orders. Such countries frequently have difficulty financing large investments. The World Bank and most other International Financial Institutions do not provide finance for nuclear investments. The South African PBMR could face strong competition from other types of high temperature reactor, notably a very similar Chinese design and models offered by Areva and the US company, General Atomics. Until a rigorous market analysis has been carried out and subjected to independent scrutiny, and arrangements for helping finance export orders made explicit, PBMR (Pty) Ltd’s assumptions on the likely world market have no basis.

Conclusion 6. Pressure is mounting on Eskom to commit to buy large numbers (24) of commercial units even before the technology has been technically and economically proven. Eskom appears, rightly, to be holding to its position of only buying it if the PBMR is the cheapest option available, something that will not be known until the Demonstration plant is in service and has operated for some time. If Eskom is required to make such an advance commitment, it could be forced to purchase uneconomic plants, raising the price of power to consumers, and adversely affecting public welfare and the competitiveness of the South African economy.

Conclusion 7. The future of Eskom is uncertain. The South African government has been considering reforms to Eskom for a number of years, including its privatisation and its break-up into competing units. There can be no guarantee that in 2013 or later, when the first commercial orders for a PBMR might be placed that Eskom will exist in any recognisable form, much less one that can be obliged to order a particular type of power plant, especially if it does not represent the best commercial option.

8.3 Overall conclusions

Conclusion 8: The PBMR project is a highly risky venture. The feasibility phase has cost about R2bn, about two thirds of which has been paid by South African public money. Despite this expenditure, there is still ample scope for the project to fail. The next phase will require a much higher level of expenditure, at least R10bn, with at least half of this again coming from the South African public. If the project fails, there will be significant consequences for the South African public either through higher electricity prices (if Eskom is forced to bear much of the risk) or through taxation if the government has to write-off the costs.

Conclusion 9: The National Environmental Management Act (NEMA) requires developers to demonstrate that their projects are economically sustainable. The FEIR does not provide the data necessary to make such a judgement. This information strongly suggests there is a high risk that the project will not be economically sustainable. On the available evidence, the project does not meet the requirements of the NEMA and the applicants, Eskom, should not be given approval.

Conclusion 10: The current high fossil fuel prices and the measures to reduce greenhouse gas emissions seem to give a new impetus to generation technologies that do not use fossil fuels. However, it should be remembered that previous oil price spikes (1974 and 1980) were short-lived and resulted in little nuclear expansion apart from in France. Investors are unlikely to make multi-million dollar investments in new nuclear power plants on the basis of a short-term oil price spike which could have disappeared long before a nuclear plant could be brought on-line. On greenhouse gas emissions, nuclear power faces competition from renewable technologies and energy efficiency measures, options that generally do not encounter the public acceptability problems that nuclear power suffers from.
 

References

T Auf der Heyde & S Thomas (2002) ‘The PBMR project: An assessment of it economic viability’ South African Journal of Science, 98, January/February, pp 36-42.

D R Nicholls (2000) ‘Status of the pebble bed modular reactor’ Nuclear Energy, 39, 4, pp 231-236.

PBMR (2002a) ‘Report on the Proposed Demonstration Module and Potential Commercialization of the Pebble Bed Modular Reactor’, PBMR (PTY) Ltd, Centurion.

PBMR (2002b) ‘Final environmental impact report for the proposed Pebble Bed Modular Reactor (PBMR) Demonstration Plant at the Eskom Koeberg nuclear power station site in the Western Cape, South Africa’, PBMR (PTY) Ltd, Centurion.

Register of Comments (2002) ‘Register of Comments on draft EIRs – June 2002’

S Thomas (1999) ‘Arguments on the Construction of PBMR Reactors in South Africa’, Briefing Paper, SPRU, University of Sussex. (http://www.nirs.org/intl/SThomasPMBRpaper.htm).

S Thomas (2000) ‘The Pebble Bed Modular Reactor – The latest rabbit out of the nuclear hat?’ Energy & Environment, 11, 2, pp 184-87.