Nuclear Expansion or Phase Out?

Costs and Opportunities

by Enrica De Cian, Samuel Carrara and Massimo Tavoni
Appeared originally at 22 December 2013

After the Fukushima incident in 2011, many countries decided to shrink their nuclear power programmes. This article presents recent research on the optimal role of nuclear power in reducing carbon emissions. Phasing out nuclear power would be costly, since it is currently the cheapest low-carbon alternative to fossil fuels. However, these costs would be largely offset by the implicit subsidy to R&D in renewables, which suffers from innovation externalities. Still, carbon pricing and explicit R&D subsidies would be a more efficient way of determining the future of nuclear power.

“We learned from Fukushima that we have to deal differently with risks… We believe we as a country can be a trailblazer for a new age of renewable energy sources… We can be the first major industrialized country that achieves the transition to renewable energy with all the opportunities – for exports, development, technology, jobs – it carries with it.” Angela Merkel (distinct quotes).

In the aftermath of the accident at the nuclear reactors of Fukushima in Japan, this was the view of Angela Merkel and her government backing up the decision to immediately close the oldest eight German nuclear power plants and to phase out the remaining nine by 2022. Unsurprisingly, this decision provoked a reaction from the German nuclear industry, which argued that early shutdowns could hugely damage the industrial base, and thus the entire national economy.

The disaster at the Fukushima Daiichi nuclear power plant in March 2011 sparked a debate about the safety of nuclear power in other countries, mostly in western Europe. Germany, Belgium, Switzerland, Italy, and Japan, among others, decided to revise their nuclear expansion or development programmes (Rogner 2013). After a decade of steady increase, over the past two years the construction of new plants dropped considerably. Moreover, out of the 435 nuclear reactors operating worldwide as of October 2013, about 350 are more than 20 years old. The decommissioning of old plants not fully replaced by new ones – especially in the US and Western Europe, which feature the most numerous and eldest fleets – is likely to cause a short- to medium-term reduction in electric output from nuclear plants.

In clear contrast with this trend, the British government has just approved the construction of a large (3.2 GW) nuclear power plant, to be constructed by EDF and a Chinese company. The strike price has been set at €0.11/kWh, indexed to inflation. This is far above current wholesale electricity prices, and close to that of renewable alternatives such as wind power. It provides an indication of the relatively high costs of building nuclear plants today – even more so if one considers allegations of illegal state support and a likely probe from the European Commission. The deal has been regarded by many energy analysts as economically unsound – not only when compared to the costs of currently available alternatives, but also on the premise that the competitiveness of renewables is set to increase in the coming years, due to continued technological progress and cost reductions. The Japanese government is also pushing to restart its nuclear fleet after passing safety checks.

The evolution of renewables costs in the coming years will depend on the future of nuclear power, as well as on energy and climate policies. In this context of uncertainty, policy must understand the economic consequences of nuclear power scenarios when accounting for its interplay with innovation and cost reduction in renewables. This article summarises the finding of a recent paper of ours (De Cian et al. 2013), which tries to answer the following questions:

  • What is the role of nuclear power in meeting climate mitigation goals?
  • What would be the implications of nuclear phase-out for the development of renewables, including their technological progress?
  • What would be the ultimate costs of nuclear phase-out when taking into account the positive spillovers in terms of innovation and diffusion processes in renewables?

The role of nuclear power for meeting climate protection goals

Nuclear power provides baseload electricity at virtually zero CO2 content, thus representing an important carbon mitigation strategy. Indeed, scenarios of integrated assessment models foresee a growing role for nuclear in the future – more so for more ambitious emissions reductions. For example, future energy scenarios from a specific climate-energy-economic model (Bosetti et al. 2006, 2009) foresee continued use of this technology over the century. In a business-as-usual world, the nuclear share would remain close to current levels (15% in 2005), contributing to 12% of global electricity production at the end of the century. Should countries succeed in enforcing a global climate agreement limiting global warming to between 2.5 and 2 degrees Celsius, the importance of nuclear power could increase considerably, reaching a share of 34% in the global electricity mix. Other models provide varying ranges of penetration of nuclear power, but all foresee a positive relation between nuclear and the stringency of climate policy (Kriegler et al. 2013).

Should nuclear power be partly or completely excluded from the energy portfolio, countries would need to look at other sources and options to satisfy the growing demand for energy. The composition of the resulting energy mix would depend quite significantly on the policy context. As a first choice, countries would expand energy investments in conventional technologies. Renewable sources and clean power R&D would also attract more resources, but the penetration of yet-to-be-proven technologies would take time to occur. According to our study, so-called breakthrough technologies could start to replace nuclear power and fossil-fuel-based options not before 2035.

Coupling the nuclear ban with a price on carbon would strengthen and accelerate the transition toward a more renewable-based energy mix. Fossil-fuel-based technologies could only be used if equipped with carbon capture and storage, while the penetration of innovative renewable technologies would occur five to ten years earlier.

The rapid decline in the costs of competitive low-carbon technologies over recent years, most notably renewables, has induced some policymakers to speculate that the decarbonisation of the electricity sector is possible without nuclear power, and hopefully at moderate costs. Provided that climate policy is designed in a flexible enough manner, our study suggests that ambitious emission reduction goals can be met even if currently important carbon-free technologies such as nuclear power are phased out. This finding has been confirmed by large model ensemble studies (Kriegler et al. 2013), though it is important to remark that it assumes carbon capture and storage will be implementable at large scale.

Is phasing out nuclear power costly?

The phase-out of nuclear power provides an implicit subsidy to alternative technologies, including less mature ones. This induces investments in innovation to early stage technologies that feature higher learning potential and international externalities compared to the alternatives that are displaced. Learning-by-doing and international diffusion of knowledge are side effects of R&D processes and of technology deployment, which are only partly appropriated by investors, due to failure in the innovation markets. As a consequence, the economic penalty of meeting a given emission reduction target without the option of nuclear power could be partly compensated by the welfare gains caused by the penetration of technologies with innovation externalities.

Let us consider for example a policy aimed at limiting global warming to 2 to 2.5 degrees Celsius. Phasing out nuclear increases the aggregate economic cost of the climate policy, but very mildly – from 2.74% to 2.78% (see Figure 1). If there were no positive externalities associated with the technologies that replace nuclear, policy costs would have increased more, to 3.17%. Indeed, technology benefits reduce the macroeconomic loss by 0.39%. The technology benefits due to activities incentivised by the implicit subsidy to learning technologies caused by the nuclear phase-out are thus able to almost completely offset the cost of losing an important mitigation option, which otherwise would be substantial. These results are robust to different climate policy scenarios.

Figure 1: Economic costs of achieving a climate goal, measured in net present value consumption losses compared to the baseline (5% discounting)Click to enlarge

Technology benefits take time to materialise and are distributed unevenly across countries. Greater benefits would occur in the regions that in the future would rely more on nuclear power, though secondary effects also play a role. Important factors to consider are the trading position of each region on the carbon and oil markets, and the interaction with the international prices of carbon permits and oil – the former being much more significant. Technology benefits, in fact, reduce the carbon price by about 10%, which represents an additional benefit for permit importers but, conversely, a penalty for permit exporters.

Looking ahead

Coal is the leading contributor to greenhouse gas emissions, which are growing at a rather steady pace of 2% per year and leading to more severe climate change, in addition to local air pollution. Nuclear power is the best available competing technology which does not emit CO2, and is expected to be part of the mitigation portfolio. In this article we have emphasised the innovation benefits of constraining a mature technology like nuclear, which can partly offset the costs of foregoing the latter. However, an even more efficient and desirable solution would be to provide a clear signal to markets in the form of carbon pricing, as a way to internalise the climate externality. This could be complemented by specific policies aimed at internalising the technology externalities in learning technologies such as renewables. For example, R&D subsidies would allow bringing R&D investments closer to the social optimum. (Policies so far have been concentrated on subsidies for installation – a less efficient mechanism). The future of nuclear power and the implications for innovation in renewables could be best determined by these forces alone.


  1. Bosetti V, C Carraro, M Galeotti, E Massetti, and M Tavoni (2006), “WITCH: a world induced technical change hybrid model”, Energy Journal, 27(Special Issue 2), pp. 13–38.
  2. Bosetti V, E De Cian, A Sgobbi, and M Tavoni (2009), “The 2008 WITCH model: new model features and baseline”, Fondazione Eni Enrico Mattei Working Paper 2009.85.
  3. De Cian, E, S Carrara, and M Tavoni (2013), “Innovation benefits from nuclear-phase-out: Can they compensate the costs?”, Climatic Change, doi: 10.1007/s10584-013-0870-9, in press.
  4. International Atomic Energy Agency (2013), “PRIS (Power Reactor Information System), The Database on Nuclear Power Reactors”.
  5. Kriegler, E, J Weyant, G Blanford, L Clarke, M Tavoni, V Krey, K Riahi, A Fawcett, R Richels, and J Edmonds (2013), “Overview of the EMF 27 Study on Energy System Transition Pathways Under Alternative Climate Policy Regimes”, forthcoming in Climatic Change.
  6. Rogner, H H (2013), “World outlook for nuclear power”, Energy Strategies Reviews, pp. 291–295.
  7. World Nuclear Association (2013), “World Nuclear Power Reactors & Uranium Requirements”.
Share this Econintersect Article:
  • Print
  • Digg
  • Facebook
  • Yahoo! Buzz
  • Twitter
  • Google Bookmarks
  • LinkedIn
  • Wikio
  • email
  • RSS
This entry was posted in Economics, Energy, Environment and tagged , , , , , , , , , , , . Bookmark the permalink.

Make a Comment

Econintersect wants your comments, data and opinion on the articles posted. You can also comment using Facebook directly using he comment block below.