False nuclear hope

Plans to construct Bangladesh’s first nuclear power plant are moving forward fast. On 26 July 2016, Mohammad Mejbahuddin, Senior Secretary of the Economic Relations Department of Bangladesh, and Russia’s deputy finance minister, Sergei Anatolievich Storchak, signed an inter-governmental agreement in Moscow for the construction and commissioning of two 1200 megawatt (MW) VVER-1200 nuclear reactors at Rooppur (also spelt Ruppur) at an estimated cost of USD 12.65 billion, with Russia committing to loan 90 percent of the costs to Bangladesh. Earlier, in December 2015, when the two government Cabinet committees approved the proposal, the Bangladeshi Minister for Science and Technology Yeafesh Osman said, “I believe USD 12.65 billion is a good bargain”. But the question is:  For whom?

A brief history

The idea of building nuclear reactors at Rooppur is very old. It goes back to a 1963 plan by the Pakistan Atomic Energy Commission to build one reactor in West Pakistan and one in East Pakistan. This fifty-year quest for constructing a reactor is blind to what has been learned over the same period about nuclear energy. This history suggests there are now good reasons to believe the people of Bangladesh will end up waiting a long time for nuclear electricity from Rooppur, be stuck with big bills, and be forced to live with the constant worry of a nuclear plant accident.

The main player in this quest to build a nuclear plant is the Bangladesh Atomic Energy Commission (BAEC), which was created in 1973. As spelt out by a senior official in the organisation, “BAEC has, since then, been trying to do the needful so as to make the country embark on NP [Nuclear Power] program”. Successive Bangladeshi governments have been attracted to the idea of a nuclear plant as a modern technological solution to energy shortages in the country, with officials thinking of the possession of a nuclear plant as an exclusive privilege. As Finance Minister Abdul Muhit put it in December 2015, “Now, we are on the verge of entering the elite club of the countries who have nuclear power plants.” Prime Minister Sheikh Hasina even termed Rooppur “the nation’s dream”.

Encouraging the BAEC in this quest are the International Atomic Energy Agency (IAEA), whose mandate calls for the agency to “seek to accelerate and enlarge the contribution of atomic energy to peace, health and prosperity throughout the world”, and various nuclear reactor vendors who hope to profit from the sales of nuclear power plants. The IAEA began work in Bangladesh a while back. It carried out a planning study in 1974-75 and projected between approximately1200 and 3000 MW of nuclear capacity in Bangladesh by 1995, with nuclear power constituting 47 percent of the country’s electricity capacity; a high projection that emphasised nuclear power as single biggest source to meet energy needs.

Finally, there are countries like the US, Germany, France and China which have, over the decades, supported their nuclear industries’ efforts to sell reactors. France, for example, signed an agreement with Bangladesh for the “peaceful use of nuclear energy” in 1980 and offered to sell 125 and 300 MW reactors in the 1980s. But in the last few years, Russia has taken the lead, edging out other countries that could have sold Bangladesh a reactor.

Misleading promises

What can we expect for the Russian reactors coming to Rooppur? The pace of contract signing and government approvals should not mislead one into thinking that the reactors will be ready anytime soon. At the time of signing of the general contract between the two countries in December 2015, the claim was that the two Rooppur reactors would start generating electricity within five or six years, by 2021 and 2022. Evidence suggests that it could be twice as long before the reactors actually start producing electricity.

Establishing nuclear power plants is a slow process. In developing countries that have just one or two reactors, the average construction time – between the first pouring of concrete and the reactor starting commercial operation – were 19, 16, 8.5, and 38 years respectively in the cases of Brazil, Mexico, South Africa, and Iran, as can be calculated from the dates given in the IAEA’s Power Reactor Information System (PRIS). These time periods do not include the lengthy preparation period before the actual commencement of construction.

For Bangladesh, a good example to keep in mind is the two 1000 MW VVER reactors constructed by Russia at Koodankulam in southern India. Construction of these reactors started in 2002 and the reactors were expected to start functioning in 2007 and 2008 respectively. In 2004, the project director confidently projected that the first reactor would start generating electricity from March 2007, one year ahead of schedule. In fact, it took 12 years for the first reactor to become commercially operational. The second reactor is yet to supply electricity to the grid, 14 years after construction started.

To make matters worse, the Russian supplied Koodankulam-1 reactor has performed poorly. According to PRIS, it only operated for 4212 hours in 2014 and for 3993 hours in 2015, which means it has remained inoperative for more than half the time since it started running.

How much will the Rooppur reactors actually cost? The history of hundreds of nuclear plants that have been built suggests the true cost of Rooppur may be much higher than the stated figure of USD 12.65 billion. The actual cost will only be known when the project is actually finished.

Delays in construction add greatly to cost increases. The costs of the Koodankulam reactors went up by more than 70 percent, from the original provision of INR 131.71 billion to an anticipated INR 224.62 billion. Three researchers at Aarhus University in Denmark who analysed construction costs for about 400 electricity infrastructure projects found that overruns afflicted “more than 97 percent of nuclear projects and led to a mean cost escalation of 117 percent per project.”

For the Rooppur plant, with construction yet to start, there already has been a five-fold increase in the expected cost. Back in 2011, Yeafesh Osman, then state minister for science and information and communication technology, announced that the 2000 MW nuclear project would be built at a cost of USD 2 billion by 2018. In other words, it would have cost USD 1 million per MW of power. The current deal, on the other hand, calls for spending USD 12.65 billion for 2400 MW of capacity by 2023-24, or roughly USD 5.3 million per MW of electrical power.

A better alternative

The purported reason for Bangladesh to embark on this project is to meet its increasing energy demand. But a sustainable strategy for meeting the energy demand would prioritise economical sources of electricity generation that can be brought online quickly. Renewable sources such as solar energy meet the twin goals of lower cost and timeliness far better than nuclear power. Globally, there is already little doubt that solar energy and wind energy are more attractive investments than nuclear power plants. As the 2016 World Nuclear Industry Status Report points out, “global investment decisions on new nuclear power plants remained an order of magnitude below investments in renewables.”

Both nuclear and solar power are highly capital intensive forms of generating electricity, and they have relatively lower costs associated with fueling (zero in the case of renewables). For the rate of return of a little over 15 percent as assumed by the Bangladesh Energy Regulatory Commission (BERC), we estimate that the average (levelised) cost of nuclear power from the Rooppur project would be roughly 32 taka per (USD 0.41) unit of electricity (kWh), not including the uncertain costs of dealing with the radioactive waste or decommissioning of the reactor once it stops generating power. If the construction cost were to be more than the estimated USD 12.65 billion, the generation cost would go up too. In comparison, BERC’s estimated tariffs for model wind and utility scale solar power projects are around 11 and 13 taka per kWh. In other words, nuclear electricity from Rooppur will be about three times more expensive than wind or solar electricity.

It will likely take a minimum of 10 years for Rooppur to generate electricity, so any fair cost comparison must be based on what solar power might cost a decade from now rather than at today’s costs. The costs of renewable energy, especially solar energy, are rapidly declining. So it is quite likely that solar energy would cost less than 8 taka per kWh by the time Rooppur becomes operational.

The cost of nuclear power from Rooppur does come down dramatically if one were to assign a lower rate of return to account for the cheap loan from Russia. Assuming a rate of return of 7.5 percent, half of what BERC recommends, we estimate that nuclear power from Rooppur would cost 10.33 taka/kWh. Even then, solar energy turns out to be cheaper and with much less financial risk associated with cost and time overruns. Moreover, solar energy projects too can avail cheap loans and this would lower their cost of production as well. For the same 7.5 percent rate of return, solar power would cost around 4.54 taka/kWh.

Finally, unlike nuclear power plants, solar energy installations can be commissioned relatively quickly, in typically one to two years. This means that by choosing to put money into solar power, Bangladesh could start getting electricity many years earlier than if it built the Rooppur plant.  Bangladesh is beginning the process of significant investment in renewables, setting a target of 3168 MW of electricity from renewable energy sources by 2021. The technical potential of solar energy in Bangladesh has been estimated at 50,174 MW.

One common argument by promoters of nuclear energy is that solar and wind energy can be generated only intermittently. However, intermittency becomes a problem only when the share of renewables in electricity generation becomes quite high. In the case of Bangladesh, where a large fraction of electricity is generated by natural gas plants, it should be possible to deal with this intermittency because the electricity output from natural gas plants can be ramped up and down quickly.

What people don’t have to worry about in the case of renewables is the possibility of  calamitous accidents, a hazard that comes with nuclear reactors. In the aftermath of Fukushima, one does not have to explain that such an accident can mean large populations will have to be relocated and significant areas will be contaminated with a range of radioactive materials. Since Rooppur is to be constructed on the banks of the Padma river, there is also the risk that downstream areas will also be contaminated.

Can Russia deliver?

The Russian deal for Rooppur has its specific set of challenges. There are at least two reasons to question Russia’s ability to deliver on its commitments. First, Russia has made so many nuclear deals in recent years that it may not be able to deliver on all of them. The Russian State Atomic Energy Corporation (Rosatom) claims to have orders for 30 nuclear power plant units in 12 different countries at a total value of over USD 300 billion.

The Russian Parliament’s independent Audit Chamber has documented delays and cost increases in reactors that Rosatom is building within Russia. It is likely that Russian reactor projects abroad will also experience delays and cost escalations. A second reason Russia may not be able to deliver on Rooppur is the collapse of its currency, the ruble. In the case of the reactor for Belarus, the Russians made a fixed price deal that was denominated in dollars. Because the ruble has fallen relative to the dollar, costs to Rosatom have reportedly gone up by 71 percent and Belarus has been asked to provide additional financial support to keep the project going.

Since the Rooppur contract, like most nuclear contracts, is not publicly available, one cannot be sure about the specifics of the deal. However, according to media reports, the contract with Bangladesh is not a “fixed price” but a “cost plus” one where “the vendor has the right to come up with any cost escalation (plus their profit margin) to be incorporated into the contract amount”. Russia seems to have prepared for the possibility that the Rooppur project will cost more than expected and to protect its profits.

Bottom line: the Rooppur reactors may be good for status-seeking project for Bangladeshi politicians; for the bureaucrats and technocrats of the Bangladesh Atomic Energy Commission; and for the Russian nuclear complex and the middlemen who will likely profit from the many subcontracts that would be signed. But it does not look like a good bargain for the people of Bangladesh.

~ M V Ramana is with the Nuclear Futures Laboratory and the Program on Science and Global Security at Princeton University, and the author of The Power of Promise: Examining Nuclear Energy in India.

~ Zia Mian co-directs the Program on Science and Global Security at Princeton University, where he also directs the Project on Peace and Security in South Asia. He co-edits Science & Global Security, the international technical journal of arms control, nonproliferation and disarmament.