Table of Contents

The future of nuclear power is uncertain. The structure of many countries’ electricity sectors initially favored investments in nuclear power but has evolved to discourage them. Given these current conditions, it is not certain that the world’s nuclear industry will be economically viable through the middle of this century. More than any other power-generation technology option, nuclear power requires a long-term commitment from governments, investors, and publics for it to be used safely and in a sustainable manner. In part to recover rising capital costs, future nuclear power plants will be designed to have operating lifetimes of sixty years or more, and expenses related to planning, licensing, decommissioning, and waste management will extend the life cycle of a nuclear project to a century or even longer.

In general, greater uncertainty about the future of nuclear power implies greater perception of risk. This may deter governments and industries from investing in more advanced and expensive nuclear technologies, particularly those requiring long lead times for R&D and industrial demonstration, and especially if decisionmakers are not confident that these investments will result in commercially exploitable assets based on sound technology that can be further developed.

When most of the world’s nuclear power plants were built, governments considered electricity supply, transmission, and distribution to be a natural monopoly, due to the lack of competition resulting from fixed costs to new entrants. They viewed the high capital investment costs prevailing in the electric power sector as an insurmountable barrier that would prevent new entrants from competing with an established single provider that would enjoy a lower average cost and economies of scale for the production of a public good. Companies in the business of producing electricity invested in nuclear power because the cost would be borne by consumers and reflected in the rate base.

Since the 1990s, many governments have deregulated their electricity sectors to encourage competition. This sometimes implied that the price for nuclear power must be increased compared to other sources for producers to make a profit, and, in most markets, it meant that new nuclear power investments would be more expensive than alternative technologies. In addition, some governments have introduced policy measures that subsidize the development and deployment of selected non-nuclear technologies to generate electricity.

China quickly and impressively assimilated proven foreign know-how to build initial LWRs and then replicate them under tight construction schedules and with few delays. But how quickly will China move forward this century with the advanced nuclear technologies discussed in the last chapter? The realization of the CEFR pilot project required nearly twenty-five years. On the basis of their previous experience, experts who were involved in fast reactor programs in Europe, Japan, and Russia cautioned during workshops in May 2015 and June 2016 that China should not expect to design, build, and operate a commercial demonstration fast reactor without significant delays.201 More broadly, these experts underscored that current multilateral international efforts in support of fast reactor development are beset with the formidable challenge of transitioning from fast reactor R&D programs to deploying commercially viable nuclear systems.202

Given this background, it is important for policymakers and investors alike to consider whether the economic and electricity policy environments in China will indefinitely favor nuclear power and, beyond that, support efforts to deploy more advanced nuclear systems with a different risk profile that may require greater financial and political commitment.

Nuclear Economics in China Before 2005

Beginning in the late 1970s, China planned to introduce nuclear electricity to its mix of power sources having concluded that the cost would be justified by the value of the investment. This conclusion was backed by a number of Chinese studies, which argued that nuclear power would be cost-competitive with other sources.203 From the beginning of China’s nuclear power program, the central government—like governments in other countries that had decided in favor of nuclear power a decade or two before—firmly controlled its electricity sector and was responsible for making all investment and most pricing decisions.

A few skeptics argued then that the high capital cost of nuclear power plants would deter China from investing in renewable energy sources and that nuclear power would not compete with the cheap and available coal resources that supplied about three-quarters of China’s power.204 But the central government’s decision to build a small number of nuclear power plants was based on political, energy security, and technology policy rationales, largely independent of cost considerations, that emerged out of China’s post–Cultural Revolution reform process.

As Deng Xiaoping’s economic modernization led to higher growth, Beijing began reforming China’s electricity sector to incentivize greater power production, including making allowance for both foreign capital financing and broader Chinese investment in nuclear power projects. During the 1980s and 1990s, whether China could afford nuclear power mattered less because most of the nuclear power plants were built under contracts concluded for equipment and services provided by foreign vendor companies. The financing for these projects was underwritten by foreign governments on behalf of their industries. Indeed, between 1980 and 1996, the Chinese state’s contribution to the financing of power sector investments fell from 60 percent to 0.2 percent, while foreign contributions increased from zero to 12 percent.205 Diversification of investment and decentralization of production was encouraged by spectacular economic growth along China’s seaboard—particularly in Guangdong Province, where restive political and industrial leaders were keen to challenge Beijing’s claim to sole decisionmaking authority.206

Economic considerations about nuclear power arose more distinctly during the 2000s, after China’s leaders decided to dramatically accelerate the pace of nuclear power plant investment, in the shadow of coal supply bottlenecks and as demand for electricity continued to increase. Coal shortfalls were accompanied by an increase in coal prices and Chinese coal imports that the government interpreted as a long-term future trend, leading planners to predict that nuclear power would by itself become comparatively more competitive against coal in the future.207 As coal prices and imports rose, state-owned power companies piled up debt and ran losses, and they successfully lobbied the government to permit them to invest in nuclear power. Beijing agreed. In the aftermath of the central government’s 2005 decision to speed up nuclear power development, a nuclear gold rush occurred: SOEs carved out of the former state energy ministry as part of ongoing electricity sector reform placed their bets on government planners’ long-term projections of electricity demand growth, and they formed partnerships with local and provincial governments to propose scores of nuclear power plant projects all over China.208

Chinese Government Assistance for Nuclear Power

The nuclear power plants China built over the ensuing decade turned out to be very profitable. During the period 2002 through 2012, China’s two nuclear power generators, CGN and CNNC, recorded annual returns against assets of 7.1 percent, more than double the levels attained during those years by thermal power plant companies.209 The concerns of these firms’ government shareholders about project risk were assuaged by expectations that nuclear projects would hold their own against coal and renewables because they would be located close to their markets: densely populated seaboard areas that were experiencing high economic growth. Provincial governments’ political resistance against transboundary power sales, and the daunting task facing China’s grid companies to move electricity across long distances, provided still more insurance for nuclear investment projects located along China’s eastern coastline.

Then and now, the profitability of Chinese nuclear investments has benefited from the support of provincial and central government leaders. This assistance is provided in a number of ways, most significant of which are:

Access to Information and Decisionmakers

The number of corporate entities allowed to invest in the nuclear power sector is strictly limited; all firms are owned and controlled by the state. The Communist Party, the most powerful organization in China, is involved in all top-level personnel decisions. Senior SOE managers in the power sector are promoted into central and provincial governments, where they “retain their links to the companies and provide insights from and arguments favorable to the companies.”210 Nearly all high-ranking government officials are members of the Communist Party and subject to its discipline. SOEs have a Communist Party hierarchy parallel to the management hierarchy, in which the party secretary may have more authority than senior managers.211 Management’s access to the Communist Party is influential and can facilitate support from local and provincial industry and governments, including for pre-authorization infrastructural work to support construction of projects in advance of formal approval by the State Council and other regulators. Many or even most of China’s proposed nuclear construction projects may have benefitted in this way.212 More generally, lack of accountability and transparency in decisionmaking on energy-related issues has, for decades and until recently, been fostered by the relatively underdeveloped status of China’s legal system and civil society.213 This has meant, in practice, that price setting and dispatching policies are subject to ad hoc, crisis-management decisionmaking, where opaque political influence and personal relationships may strongly factor.214 That said, it is also true that since the late 1990s, corporatization of China’s electricity industry—including its nuclear SOEs—has to a certain extent challenged the power of the Communist Party and the Chinese state to direct firms to make nuclear asset management and investment decisions. To a greater degree than before, leaders of China’s nuclear SOEs in recent years have encouraged the development of a management culture that distinguishes the interests of the companies from those of the Chinese state and the Communist Party.

Financial Subsidies

Power-generating SOEs have not been obligated until recently to pay dividends to their government shareholders. So-called policy banks, such as the China Development Bank, have provided these companies, above all CNNC, loans at favorable, state-subsidized rates.215 More generally, the SOEs benefit from controls on deposit interest rates that permit state-owned lenders to provide nuclear-project financing at selected low discount rates that give nuclear power a clear and very large financing cost advantage.216

Favorable Price Setting

Perhaps the most effective potential financial perk that the Chinese state affords nuclear power is the arrangement of favorable price conditions to deliver nuclear-generated electricity to the grid. China’s central planning agency, NDRC, controls the price at which power producers sell their output. It sets a different feed-in tariff for each power source: nuclear, solar, wind, hydro, and coal. For as long as the central government has taken this approach, the guaranteed tariff paid to producers for nuclear power—0.43 renminbi per kilowatt-hour (RMB/kWh) in 2016 and 2017—has been higher than the rate for either coal-fired or hydroelectric power, in part reflecting higher capital costs for nuclear plants.217 How important to the welfare of China’s nuclear sector is this remunerative tariff? “We watch this carefully,” said one Chinese nuclear industry executive in 2015. “If the government were to take this away from us, the future of our business would be in a lot of trouble.”218

Electricity Sector Reform and Nuclear Power

Electricity reform in China began modestly nearly fifty years ago but it has been ongoing and it will likely continue. The more profound and effective China’s electricity sector reform is, the more economics and cost considerations will factor in future nuclear policy decisionmaking.

Electricity Market Reform

Following government decentralization initiatives in the 1980s and 1990s, Beijing began commercializing and unbundling the power sector. Many of the steps China took during the early 2000s looked familiar to government officials and company executives who were deregulating power systems in Western countries.

China’s State Planning Commission, the central government’s mammoth planning agency, was rebranded as NDRC in 2003. It established a central government Energy Bureau, which later was upgraded to a vice-ministerial National Energy Administration, or NEA. The state created a Ministry of Environmental Protection (MEP), and then set up the State-Owned Asset Supervision and Administration Commission (SASAC), which assumed the role of central government shareholder for power sector SOEs. The central government also made an attempt to set up a power industry regulatory commission.

In parallel, China’s leadership took actions to expose its electric power system to market forces. This process began in 2002, when Beijing unbundled the gigantic China State Power Corporation (SPC), which owned most of China’s transmission and distribution infrastructure and a large share of China’s generation capacity. In the wake of the breakup, two grid-management companies were set up that would be independent of five new power-generating SOEs that were also hived off of SPC. These five companies acquired nearly half of China’s electric power generating capacity within five years.219 This drive was fueled by the 2005 central government decision to accelerate nuclear power plant construction discussed above. Power-generating SOEs successfully pressured the government to allow them to raise capital in the bond market to finance their forthcoming nuclear capacity investments in line with the central government’s now-higher reactor-building targets.

Power generators were especially inclined to invest in nuclear power plants for two reasons: 1) rising prices for domestic coal, which—beginning in the mid-2000s, in the wake of coal sector reforms—were tracking international market prices; and 2) the absence of an effective mechanism that linked the price of coal to the on-grid price of thermal power.220 In this situation, it was only a matter of time before industry firms began pressing the government for a more fundamental reform of its price-setting system. Today, this reform is still at an early stage and, in the view of Chinese market advocates, it has far to go. Whereas until the 2000s China’s nuclear sector was too small to influence the state’s price setting, that situation changed as more and more reactors were built. Since 2005, NDRC has assured that nuclear power generators benefit from higher tariffs for the power they sell to the grid.

In the coming years under Xi Jinping, two important power reform agenda items may profoundly affect the nuclear sector: Beijing’s long term ambition to introduce market mechanisms, and its even-more-ambitious effort to decarbonize China’s electricity generation system. Both these interests strongly figured in Xi’s 2014 proclamation of a “revolution in energy production and consumption” that would cut back waste, incentivize investment in non-fossil energy sources, enlarge the role of market forces, and reform energy sector regulations and governance.221 The future of NDRC’s tariff setting, including for nuclear power, is at the heart of both initiatives.

Environmental Policy-Driven Measures and Growth

As evidenced, deployment of nuclear power generation technology in China was informed from the outset by the need to diversify away from coal. China’s resolve to decarbonize its power sector has intensified in response to certainty among experts worldwide that atmospheric carbon dioxide emissions must be dramatically reduced to avoid adverse global climate change, but, until now, China’s need to reduce particulate air pollution has been the primary environmental policy driver for nuclear power.

In 2009, China announced that it aimed by 2020 to reduce CO2 emissions intensity (average emissions per unit of gross domestic product) by 40–45 percent below the 2005 level and to increase the share of non-fossil fuels in primary energy production by 2020 to 15 percent.222 In 2016, China ratified the agreement forged at the United Nations climate change conference in Paris on carbon reductions that would limit anticipated global temperature increases. In doing this, China extended the horizon of its climate mitigation commitments beyond 2020 to 2030, and agreed to reduce its CO2 emissions intensity by 60–65 percent and increase the non-fossil share of energy production to 20 percent, implying a peaking of China’s CO2 emissions by 2030. These goals were also reflected in climate policy targets included in China’s Thirteenth Five-Year Plan (2016–2020).223 Under five-year plans covering 2011 through 2020, China has been aggressively expanding investment in power production capacity from wind, solar, and nuclear sources. If goals are met, China’s capacity for wind-powered power generation will have increased from 31 GWe in 2010 to 200 GWe in 2020; solar-powered capacity will have increased from about 1 GWe in 2010 to 70 GWe in 2020; and nuclear power capacity will increase to 58 GWe in 2020 with an additional 30 GWe under construction.

What NDRC’s planning targets imply for the period after 2020 is not spelled out, and this is subject to conjecture and wishful thinking by participants and observers who favor (or not) the deployment of specific power generation technologies. The amount of nuclear power that will be allocated in future five-year plans will depend, inter alia, upon: expectations for economic growth, government carbon-reduction goals and commitments, and technology development including for electricity storage and carbon sequestration as well as power generation. It will also depend on the architecture of the power system—in particular, whether Chinese power demand will indefinitely include a substantial base load that would be served by nuclear power plants, as has been the case so far, and how deeply ultra-high-voltage transmission infrastructure in the future penetrates the Chinese power grid.

Especially prior to the Fukushima accident, some quasi-official Chinese government agency estimates for nuclear power installed capacity in 2050 were as high as 400–500 GWe. These estimates corresponded to numbers used by scientists at the Lawrence Berkeley National Laboratory (LBNL) in the United States beginning during the late 2000s, in cooperation with researchers affiliated with NDRC, which used Chinese data to model future projected Chinese electricity demand and greenhouse gas emissions.224

Barring unforeseen developments, Chinese government planners and research organizations are not expecting that the evolution of China’s power system by 2050 will warrant any decisionmaking that would halt the continued incremental growth in nuclear power–generating capacity. But during the course of the 2010s, projections for future nuclear capacity growth in China have, on balance, dipped sharply lower. This, in part, reflects caution after the severe nuclear accident in Japan in 2011, as well as the expectations for lower economic and power demand growth termed China’s “new normal” by Xi and other Chinese officials since 2014.

Current and persisting “new normal” and environmental imperatives in China’s political economy do not imply that China will turn away from nuclear power by mid-century. Notably, a computer model—developed by a non-governmental research organization critical of nuclear energy and that advocates aggressive energy efficiency policies and renewables build-out in China—has projected that, under a scenario highly favorable to carbon emissions reductions, to maximize reductions China will continue increasing its installed nuclear power capacity to about 180 GWe by 2050. This is less than half of some optimistic pre-Fukushima estimates but it is still twice the amount of nuclear capacity that China expects to have in place by 2025 under the Thirteenth Five-Year Plan, and approaching twice the nuclear capacity of the current nuclear capacity in the United States.

Projected Chinese Power Sources to 2050

Projections by these researchers for Chinese power production in the upper right viewgraph thus anticipate that nuclear power will be a significant component in a climate-favorable power generation mix that would be 82 percent non-carbon-emitting in 2050. This projection is very optimistic and, according to its authors, will not come about without a massive cumulative investment of 35 trillion RMB ($5.2 trillion).225 What in fact transpires through 2050 might differ from these projections to the extent that events are adversely (from the point of decarbonization) influenced by a number of factors including: the capabilities of China’s labor force; whether China invests heavily in cheap natural gas; whether China follows its conventional growth model favoring capital goods investment and construction; whether technical barriers prevent further development of low-carbon technologies; how politicians and planners balance high-penetration renewables in the Chinese power planning system; and whether SOEs are subject to reforms.226

The scenario in the left viewgraph, featuring less conservation and more coal and nuclear power generation, reflects more conventional Chinese reference assumptions and data from NDRC’s Energy Research Institute (ERI), which collaborated with LBNL. In fact, these experts’ upper and lower bounds for projected electricity generation through 2050 are consistent with a number of other recent projections, such as those made by the International Energy Agency.227 To meet this level of demand, some researchers estimate that, by 2030, China’s grid will likely add 1,000–2,000 GWe in generating capacity, and that, in order to also meet targets China set in 2016 for climate mitigation, 900 GWe of that added capacity must be non-fossil generating capacity—beyond investments covered in the Thirteenth Five-Year Plan. If so, Chinese researchers collaborating with LBNL say that most of the non-fossil capacity additions “will be renewables [and/] or nuclear.”228 More generally, new nuclear capacity will continue to be added, a Chinese electricity planning official said in 2016, “so long China is committed to decarbonize and as long as NDRC and NEA anticipate that renewables penetration in China will be limited to between 20 percent and 30 percent of China’s total power supply.”229

Demographic and geographic aspects will also be critical to the future of China’s nuclear power expansion. So far, all of China’s nuclear power plants have been built along the country’s densely populated eastern seaboard. Researchers in 2015 concluded, however, that these high-income areas are experiencing a leveling off of per capita power consumption, and they caution that “the relationship between per capita electricity consumption and economic development is evolving to a new stage, and that per capita electricity demand is growing only moderately in most Chinese provinces, or even plateauing in the most developed regions, with the exception of a few energy extractive provinces.”230 According to Chinese nuclear industry officials in 2015, projections for continued expansion in nuclear power capacity beyond 100 GWe are based on the expectation that previously foreseen projects for construction of nuclear power plants on sites in inland China will be carried out.231 But in the wake of the Fukushima accident, the Chinese government has been reluctant to implement these plans made during the 2000s. Without inland nuclear projects coming on line, a nuclear generating capacity for China greater than 100–150 MWe by 2050 might not materialize.

Provided that recent expectations for future power demand are accurate, China’s national carbon emissions reduction goals and commitment to the Paris agreement may justify adding a few hundred GWe in additional installed nuclear generating capacity between 2020 and 2030 alone. But actual capacity addition would be subject to numerous constraints, including the future development of power demand under the so-called new normal conditions and possible related political constraints.

If China, in the coming years, prioritizes adding non-fossil generating capacity, nuclear additions may be favored if other non-emitting technologies encounter difficulties. Since the early 2010s, uncoordinated investment in wind machines that increased capacity annually by 60 percent has led to high curtailment rates. The addition of large amounts of intermittent power has also severely challenged the government’s dispatching and pricing system, designed for fixed quotas and fixed prices unrelated to supply and demand. Increasing movements of power through the system are putting China’s dispatching and pricing mechanisms under more and more pressure.232 This pressure may increase in response to Xi’s announcements in 2015 and 2016 that the government aims to introduce competitive dispatching.233

Another factor that cannot be ignored by policymakers considering the future prospects for China’s nuclear power sector is the sheer size and clout of China’s coal sector. Thirty-thousand coal mines employ six million workers. Regardless of central government efforts to cut back, coal production increased from 1 billion short tons to nearly 4 billion short tons between 2000 and 2014. Local and provincial politicians who support the coal industry frequently ignore central government directives. When Beijing orders local and provincial governments to shutter plants, closings are delayed and plants are often restarted as soon as coal prices rise. Investment in new projects goes ahead regardless of excess capacity and in defiance of orders to cut coal production. There is pressure along the entire supply chain: coal mines, generating stations, and engineering and construction firms. Local protests by coal workers are routine, and the potential for unrest may temper Beijing’s resolve to wind down China’s coal industry. On the other side of the ledger are perhaps 700 million urban Chinese who expect the state and the Communist Party to deliver clean air. Until and perhaps after coal emissions peak around 2030, the government will be politically challenged by this dilemma.

Future Nuclear Costs

Chinese decisionmakers’ efforts to substitute nuclear for coal-fired power will be still more difficult because nuclear power in China is expensive. In regions where the coal industry is powerful, coal-fired power producers enjoy considerable cost advantages over their nuclear competitors. Regions with big surplus coal-fired generating capacity are putting base load electricity on the grid at discount prices—for example, 0.3538 RMB/kWh in Shanxi and 0.2937 RMB/kWh in Neiming. These prices are, respectively, 20 percent and 33 percent below the guaranteed support price for nuclear power.234

In recent years, the combination of lower growth rates for both GDP and power demand, comparatively high costs for nuclear power, and political pressures from coal and renewables have led to curtailments in the nuclear power sector. The result has been lower load factors for Chinese nuclear power plants, and pressure from the government on nuclear generators to operate reactors under load-following regimes, a state of affairs that would result in still-higher nuclear power production costs.235 These pressures on costs may increase in the coming years, especially if China adds generating capacity to the grid at rates that are considerably higher than current and projected rates of power demand.236 Should this current situation persist, conflict between the nuclear industry and government market reformers will certainly escalate, especially over how the government prioritizes dispatching of renewables, nuclear, and hydroelectric power sources.

Far from headlines with official boasts about new records for China’s ever-growing installed nuclear capacity, data made available to Chinese media in March 2017 underscored the potential threat to nuclear power economics posed by reduced plant availability. According to Chinese nuclear utility executives, a nuclear power plant in China must be operated for about 7,000 hours per year to service the loans that financed the project. Beginning in 2015, in some regions where nuclear power capacity has been steadily built up, availability has dropped to 5,000 hours per year.237

Furthermore, buyers of cheap coal-fired power in the future may pay even less because of falling transmission charges (as low as 0.10-0.15 RMB/kWh over distances as long as 2,000 kilometers, or about 1,242 miles), as the highly corporatized China State Grid Corporation (CSGC) moves forward with record ultra-high-voltage power line investments. For 2016 alone, CSGC, the world’s largest and most cash-rich power transmission and distribution company, allocated RMB 543 billion for new investments.238

During most of the last decade, coal-fired power plants in east coast areas where power reactors are located have been producing electricity at prices close to the high nuclear off-take price. But with more and more power transmitted to the coast from the far-flung mine-mouth coal-fired stations and intermittent sources that are favored by Chinese environmental laws, east coast nuclear power prices may come under still greater pressure. Nuclear investments could be stranded if decisionmakers in Beijing and powerful CSGC break down the political barriers across the country that still inhibit the sale and transmission of electricity across provincial boundaries.

Whether recent local and regional trends toward a decline in nuclear power production become generalized will depend on how the state balances the interests of the different participants who sell their power to the grid. Just over a decade after Wen Jiabao ushered in a nuclear construction bonanza, potential investors today—and perhaps for years to come—can no longer assume that new nuclear power plants in China come with a license to print money.

The more China catches up with the West, the more its economy will lose competitive advantage based on low costs of basic factor inputs. Thanks in large part to inefficient dispatching, capacity overbuilding, lack of transparency in pricing, and selective protectionism, electric power in China today may cost 30 percent more on average than in the United States.239 Given that nuclear power in China has always been more expensive than coal-fired power, if the Chinese government aims to defend the economy’s international comparative advantage, it will seek to control or even to lower the cost of nuclear power.

Unless the government is willing to underwrite more expensive nuclear power production for long-term strategic reasons, cost will be a factor in any decisions made by the government and SOEs on whether to shift investment from PWR-based nuclear power plants toward one-off or limited-series commercial-scale fast reactors and related nuclear fuel cycle infrastructure.

What has happened in China’s HTGR program suggests that a decision by the government in favor of such technology-driven investments is not a foregone conclusion. Last decade, China launched a project to build ten twin-unit 105-MWe HTGR power plants, a total of twenty reactors, in series at the Shidaowan site in Shandong Province. Like the fast reactor, the HTGR was designated a strategic technology in 1986 by central planners. But the HTGR project in Shidaowan will be halted after the first pair of units is completed in 2018. According to officials from the project’s consortium, the generation cost (in part based on the project cost) for these units was found during project implementation to be 25 percent higher than for a Chinese PWR-based power station.240 Utility investors are now planning on building a large PWR on the site instead. The HTGR program will be redesigned for lower construction and procurement costs, and it is foreseen that the next HTGR project will be a 655-MWe station consisting of six modules linked to one turbine generator, intended to reap greater economies of scale.241 Even for a reactor model that the government had favored since 1986 for strategic reasons, comparative costs matter to state-owned investors.

Fast Reactor Costs

The lessons from China’s HTGR program should be transferable to other nuclear investment projects. Compared to China’s PWRs, costs for design engineering, licensing, procurement, and installation will be higher for a prototype or demonstration fast reactor because it will be unique. Operations costs are a function of reliability; what can be expected for a new industrial prototype or demonstration fast reactor is difficult to predict. Compared to LWRs, which provide potential investors a database of about 14,000 years of cumulative commercial operation, the world’s fast reactors have compiled a total experience of just a few hundred years.242 Demonstration units in Japan, the UK, and the United States experienced prolonged shutdowns, and the lifetime capacity of France’s first industrial-scale fast reactor over a ten-year period was just 7 percent.243 On the other hand, two units—Phénix in France and BN-600 in Russia—eventually operated for 151 and 165 consecutive days, respectively. A follow-on commercial-scale French reactor operated successfully throughout 1996, generating more electricity than during its previous nine years of operation, before it was ordered shut by a French government programmatically opposed to continuing the project.244 For future prototype or demonstration fast reactors to succeed, investors need to tolerate that they may have to operate long enough to benefit from a learning curve.

In the wake of France’s experience with fast reactor development over half a century, national utility company Électricité de France will not commit itself to the construction of a commercial fast reactor without remedial attention being paid to specific technical issues that previously limited reliability, including aerosol deposits, corrosion, and sodium containment integrity. Engineering changes in some previously operating units permitted these reactors to, at times, attain availability factors of between 50 percent and 80 percent. But, according to a senior expert in France’s fast reactor program, “to be competitive with other generating systems, a reliability factor higher than 90% will be necessary. Innovations in materials must impact favorably on refueling outages, maintenance, and in-service inspections, lengthen component lifetimes… Industrial application of materials advances should make sure that sodium leaks are very rare events; design of circuits and equipment must permit repair or replacement with a very short delay or a few days.”245

Japanese experts likewise concluded, on the basis of experience from Japan’s fast reactor program, that future reactors must address capacity factor, capital cost, and fuel cost. To lower costs, plant life must be extended from forty to sixty years; reactor systems must be simplified; advanced codes and standards must be used during construction; maintenance periods must be shortened; fuel burnup must be increased; and the operating cycle length between inspection and refueling outages must be lengthened to two years.246

China’s nuclear industry executives say they expect that the government will pay the extra cost for fast reactors if their investment and operating costs exceed costs of PWRs.247 Based on China’s experience in cost management for the CEFR, industry also expects that construction and power generation expenses for a demonstration breeder reactor would be higher than for a Chinese PWR. This view is consistent with the experience of fast reactor experts in Western Europe, Japan, Russia, and the United States queried by the author in 2015 and 2016. Most experts concurred that, overall, the costs related to any new demonstration fast reactor would be higher than for a PWR.248 Chinese sources likewise suggested that investors from the Three Gorges hydroelectric project, who may be willing to support construction of a demonstration fast reactor in Fujian Province, would likely be assured in advance by the government that this unique and strategic-designated project would benefit from government subsidies and/or that the effective feed-in tariff governing power sales from the fast reactor to the grid would be higher than for all other nuclear power reactors—regardless of the interest of NDRC planners under Xi to reduce market distortions in China’s price-setting system.

The CEFR, according to CIAE, cost $387 million (RMB 2.5 billion)—if so, that was a bargain by today’s nuclear plant costs but nearly four times more than was budgeted a quarter century earlier, when the project got underway with a budget of $106 million (RMB 680 million). CIAE blamed lack of experience in design engineering and construction, especially for pool-type reactors, for a large share of the overrun. CIAE had to make expensive design changes in major components and, because the Chinese government (in CIAE’s view) did not adequately support the project; the 39 percent share of project costs represented by procurement was lower than for China’s PWRs (although low procurement share reduced the project price compared to PWRs).249 CIAE bought and then modified the project to incorporate equipment from Italy’s discontinued fast reactor project, and supply of equipment from Russian partners in the project was delayed by political turmoil after the collapse of the Soviet Union; these developments increased costs.250 CIAE aims to make a number of improvements to lower project costs for its 600-MWe fast reactor. These concern, inter alia, engineering design procedures and core design, fuel burnup, steam generators, overall heat efficiency, and more efficient use of construction materials—all issues that contributed to higher CEFR project costs.251

Reprocessing Costs

In contrast to cost components for future fast reactors, the costs associated with the investment and operation of bulk reprocessing plants based on PUREX chemistry have been studied at length and are better understood, especially because the reprocessing of spent fuel using this technology has been an established commercial activity for several decades.252 Chinese officials in 2013 told the author that media reports asserting that French vendor Areva had offered to build an 800 MTHM/y reprocessing plant in China for about EUR 20 billion were credible.253 This price was also in line with the cost for a similar, nearly completed plant built by Areva in Japan, which Japanese industry stated in 2007 was 2.2 trillion yen.254

No public cost data is available for the 200 MTHM/y reprocessing plant that China is building at Jinta, but it can be assumed that this expense would be considerably less than for an Areva-supplied plant because, as was the case for the CEFR, a large share of the design engineering, construction, and procurement will be domestically sourced.255 Low cost estimates for this project may have afforded China leverage to persuade Areva to reduce its price for the 800 MTHM/y reprocessing plant by about EUR 5 billion.256

In advance of a decision to build and operate an industrial-scale facility for reprocessing large amounts of spent PWR fuel, Chinese executives in 2015 and 2016 told the author that there was then no agreement between China’s two largest reactor owners, CGN and CNNC, about how to proceed. Because CNNC has been delegated overall responsibility for developing and managing the back end of China’s nuclear fuel cycle industry, SPIC and CGN officials have voiced concern that these firms will be commercially disadvantaged and “overcharged” by CNNC for reprocessing services.257 CNNC may view future reprocessing revenue from captive clients as a way to recoup lower margins it faces in some other civilian and defense-related nuclear business activities.258 The government and CNNC would have several options to raise money for a reprocessing plant investment, including direct financing, soft loans, and recourse to a spent fuel management fund derived by tax revenue levied on all operating reactors in China at a rate of 2.6 Chinese cents per kilowatt-hour (U.S. 0.38 cent/kWh). CGN, which is critical of CNNC’s reprocessing monopoly, advised NDRC in 2013 that the proceeds from the fund won’t suffice to pay for reprocessing.259

On the basis of data available for existing and previous reprocessing programs, external researchers have generally concluded that the cost of reprocessing in China would exceed the cost of dry-storing China’s spent fuel.260 Researchers at Harvard University estimate the cost of reprocessing spent fuel for such a plant to be between $1,000/kg and $5,600/kg, relying upon other projections for capital cost, interest, decommissioning, capacity factor, and operation.261 In response, Chinese sources have claimed that the low figure is closer to what has been expressed in official Chinese data through 2013 based on expectations, including for interest rates, about how the government and CNNC would finance a large reprocessing plant project.262 The higher numbers in the Harvard study, Chinese officials assert, are more consistent with industry expectations for higher factor costs in Western nuclear power programs, such as those included as historical references in the study. In any case, should China choose to invest in an industrial closed nuclear fuel cycle, in addition to the above expense for aqueous reprocessing of spent PWR fuel, it would face costs associated with the production of plutonium fuel (MOX fuel and/or metallic fuel) and waste management, including disposal, along with the appropriate costs for uranium procurement, uranium conversion and enrichment, fabrication of uranium-oxide fuel, dry storage of spent fuel and—for fast reactors—higher costs than obtain for PWRs related to fuel fabrication and spent fuel reprocessing.

Researchers’ conclusions that the cost of nuclear power in China would be higher with a closed fuel cycle are consistent with results of a number of economic studies for other nuclear programs carried out since the 1990s. Depending on assumptions by researchers in these studies, the share of fuel cycle costs associated with reprocessing and recycling of nuclear fuel may be between 14 percent and 66 percent.263 For most nuclear power–generating countries, the prospect of considerably higher costs would deter investors and governments from committing to an industrial closed nuclear fuel cycle. Informed by an understanding that the fuel cycle cost component associated with managing spent nuclear fuel represents a relatively small fraction of the total levelized cost of electricity generation (according to the Organization for Economic Cooperation and Development [OECD]/NEA perhaps between 2 percent and 4 percent), China may develop confidence that these higher costs would be compensated for by strategic benefits.264

Strategic Takeaways

Economics is increasingly significant to China’s nuclear power program. In the future, the more China’s economy resembles the economies of other advanced nuclear countries, the more economics will factor into decisionmaking.

Like many other nuclear power countries, China was initially undeterred by nuclear costs because its electricity sector was fully integrated and its initial nuclear power plants were financed by foreign vendors and their governments. Beijing favored nuclear power in line with policy goals—technology development, energy security, and pollution reduction—that were independent of cost consideration.

Nuclear costs became a factor in decisionmaking beginning in the 2000s, after Beijing launched electricity sector reforms while accelerating nuclear power plant deployment and corporatizing its power industry to raise more funds for electric power investments.

Today, China’s nuclear power industry is challenged by Beijing’s conflicting electricity policy goals of increasing market transparency while maintaining dirigiste control. After Beijing unbundled transmission and distribution from power generation, nuclear SOEs needed protection from the state to sell their electricity to the grid at a price that guaranteed sufficient revenue to service their growing debt. These perquisites are the lifeline for China’s nuclear power industry, and they would be threatened by market forces that China’s leadership might unleash in their continuing efforts to influence decisions on new capacity investment and lower costs.

Downward pressure on favorable nuclear feed-in tariffs may arise from a shift in China’s growth model, national grid unification, falling marginal power demand, and decisionmaking by corporatized SOEs that departs from the state’s strategic interests. The worst-case economic scenario for nuclear power in China is that nuclear power plants become stranded investments because so-called new normal conditions persist and electricity reform continues in the absence of correctives by the state.

Anticipated economic trends in the 2020s will not favor China transitioning to more advanced nuclear technology and a closed nuclear fuel cycle. The more China compels its electricity sector to lower overall system costs, the more the state will be under pressure to monetize and subsidize the cost of nuclear power’s assumed noncommercial strategic benefits.


201 Views expressed by experts at two Carnegie workshops: “Future Technology Options for Generating China’s Nuclear Power,” Xiamen, May 22–23, 2015; and “The Future of Nuclear Power in China and the World: Reprocessing and Fast Reactors;” Berlin, May 31–June 1, 2016.

202 The International Project on Innovative Nuclear Reactors and Fuel Cycles (INPRO), between 2008–2011 developed an analytical framework for assessing transition scenarios from currently deployed power reactors to a future that includes fast reactors and closed fuel cycles for the production of electric power. The scenarios include deployments by individual countries, by small groups of states, and by coordinated multilateral schemes; they consider reactors, nuclear fuel processing, transport and logistics, and nuclear waste management issues; see “International Project on Innovative Nuclear Reactors and Fuel Cycles,” IAEA, January 1, 2017, The problem of commercializing the fast reactor and closed fuel cycles has also been studied in a strategic context by the Nuclear Energy Agency (NEA) of the OECD; see OECD, Strategic and Policy Issues Raised by the Transition From Thermal to Fast Nuclear Systems (Paris: OECD, 2009).

203 Xu, The Politics of Nuclear Energy in China, 100–06.

204 “China’s Nuclear Boom May Soon Go Bust,” New Scientist, February 14, 1985.

205 Xu, The Politics of Nuclear Energy in China, 112.

206Mark Hibbs, “Nuclear Hurdles Are Politics in Guangdong, Economics in Shandong,” Nucleonics Week, May 20, 1999, 12.

207Xu, The Politics of Nuclear Energy in China, 103.

208 Mark Hibbs, “China’s Utilities Considering More Than 70 Sites for New Reactors,” Nucleonics Week, April 30, 2009, 1.

209 Ryan Rutkowski, “The Economics of Nuclear Power in China,” Peterson Institute for International Economics, October 25, 2013,

210 Philip Andrews-Speed, The Governance of Energy in China: Transition to a Low-Carbon Economy (New York: Palgrave Macmillan, 2012), 173.

211 Ibid., 125.

212 Personal communications from Chinese industry officials, 2015 and 2016.

213 Andrews-Speed, The Governance of Energy in China, 130–3.

214 Author communication with a Chinese electricity planning expert, 2016.

215 Author communications with Chinese industry executives and government officials, 2012 and 2015; and Andrews-Speed, The Governance of Energy in China, 90.

216 Rutkowski, “The Economics of Nuclear Power in China.” At a discount rate of 5 percent, the cost of operating a nuclear plant in China can be equivalent to the cost of operating a coal plant; were the discount rate to be increased to 10 percent, the operating cost of the coal plant would only be one-eighth of that for the nuclear plant.

217 One source cites China Nuclear Energy Association (CNEA) [Xu Yuming, “China’s Nuclear Power Development in Post-Fukushima Era,” May, 2013] as asserting that the feed-in tariff for power generated by AP1000 reactors in China yet to be completed would be RMB 0.45/kWh or more, slightly above the Chinese nuclear average, and that this would be reduced to RMB 0.42/kWh for follow-on units if their capital costs were lower: see “Economics of Nuclear Power,” World Nuclear Association, December 2017,

218 Author communication with Chinese industry official, 2016.

219 Andrews-Speed, The Governance of Energy in China, 67.

220 Some Chinese sources said that at this time, NDRC was unwilling to raise electricity prices charged to industry and that accounted for 70% of China’s electricity production.

221 Xinhua, “Xi Stresses Efforts to Revolutionize Energy Sector,” China Daily, June 14, 2014,

222 Deborah Seligsohn, “China’s State Council Unveils 40–45% Carbon Intensity Target,” Insights (blog), World Resources Institute, November 26, 2009,  

223 “China’s Legislature Ratifies Paris Agreement on Climate Change,” Xinhuanet, September 3, 2016,

224 Nan Zhou et al., “China’s Energy and Carbon Emissions Outlook to 2050,” Lawrence Berkeley National Laboratory, April 2011,, 53.

225 Energy Research Institute, Lawrence Berkeley National Laboratory, and Rocky Mountain Institute, “Reinventing Fire: China,” Rocky Mountain Institute, September 2016,

226 The Rocky Mountain Institute presented the results of the study in September, 2015: “Decarbonizing China’s Power Sector for Cleaner Air and Climate Smart Cities,” Wilson Center, September 29, 2015,

227 Gang He, Jiang Lin, and Alexandria Yuan, “Economic Rebalancing and Electricity Demand in China,” 2 (see figure 1 “Recent Projections of China’s Electricity Demand”).

228 Ibid., 12.

229 Author communication with Chinese electricity planning expert, 2016.

230 Gang He, Jiang Lin, and Alexandria Yuan, “Economic Rebalancing and Electricity Demand in China,” 11.

231 Author communication with Chinese nuclear industry executive, 2016.

232 Author communication with Chinese electricity planning research official, 2016.

233 Emma Richardson, “The Great Wall Ahead of China’s Integration of Renewables,” Australian Energy Council, August 11, 2016,

234 Thomas Rawski, “Growth, Upgrading, and Excess Cost in China’s Electric Power Sector” (unpublished paper, University of Pittsburgh Department of Economics, 2017), 16–18.

235 C. F. Yu, “Worries Below the Surface in China,” NIW, January 29, 2016, 7; C. F. Yu, “Beijing Drafts Rule to Put Reactors in Load-Following Mode,” NIW, August 19, 2016, 3.

236 Kimfeng Wong, “Coal Loses More Market Share to Nuclear, Renewables,” NIW, February 19, 2016, 6.

237 China Nuclear Energy Association, “Report Issued on China’s 2016 Nuclear Power Operations” [in Chinese], February 3, 2017, ChinaPower,

238 John A. Mathews and Hao Tan, “China’s Continuing Green Shift in the Electric Power Sector: Evidence From 2016 Data,” Asia-Pacific Journal 15, issue 10, no. 4 (May 15, 2007):

239 Thomas Rawski, “Growth, Upgrading, and Excess Cost in China’s Electric Power Sector.”

240 Author communication with Chinese nuclear R&D official, March 2017, and Chinese nuclear industry executive, 2016.

241 Author communication with Chinese industry executive, September 2016.

242 Sauvage, “RNR-Na Prototypes et Industriels,” 4.

243 Thomas B. Cochran et al., “Fast Breeder Reactor Programs: History and Status,” International Panel on Fissile Materials, 2010,, 10.

244 Arthur Jobert and Claire le Renard, “Framing Prototypes: The Fast Breeder Reactor in France (1950s-1990s),” Science and Technology Studies 2 (2014): 21–2.

245 Sauvage, “RNR-Na Prototypes et Industriels,” 10.

246 H. Ohshima and S. Kubo, “Sodium-Cooled Fast Reactor,” Japan Atomic Energy Agency, February 2012,

247 Author communication with Chinese nuclear industry executives, 2016.

248 Participants at Berlin Carnegie Workshop, May 2016; U.S. national laboratory fast reactor experts, 2016.

249 Yang, “Economic Issues of Fast Reactor in China.”

250 Author communication with Chinese nuclear R&D official, 2015.

251 Yang, “Economic Issues of Fast Reactor in China.”

252 Matthew Bunn, Hui Zhang, and Li Kang, “The Cost of Reprocessing in China,” Project on Managing the Atom, Belfer Center for Science and International Affairs, Harvard University, January 2016,

253 Phil Chaffee, “Behind the Areva-CNNC Talks,” NIW, May 6, 2016, 8.

254 Bunn, Zhang, and Kang, “The Cost of Reprocessing in China,” 45.

255 Author communication with Chinese official, 2012.

256 “Areva Says Decision on China Nuclear Reprocessing Plant Expected Soon,” Reuters, May 18, 2017,

257 Similar concerns were raised by foreign utility customers of Areva for costs of reprocessing, since their national laws obligated them to contract for reprocessing services that would be provided by a de facto monopolist.

258 C. F. Yu, “China’s ‘Big Three’ Push Ahead on Fuel Fab,” NIW, March 18, 2016, 3.

259 C.F. Yu, “CGN’s Search for Back-End Alternatives,” NIW, November 28, 2016, 4–5.

260 Bunn, Zhang, and Kang, “The Cost of Reprocessing in China,” 55. For an 800 MTHM/y plant over forty years, the total cost for capital and operation is estimated to be between $27 billion and $80 billion, compared to $6.4 billion for dry storage.

261 Bunn, Zhang, and Kang, “The Cost of Reprocessing in China,” 59. The Harvard estimates are for a reprocessing plant with a throughput of 800 MTHM/y.

262 Author communications with Chinese nuclear experts, 2015 and 2016.

263 NEA, The Economics of the Back-End of the Nuclear Fuel Cycle (Paris: OECD, 2013),, 107.

264 Ibid., 141.