VoxEU Column Environment

What a different world. Costs and policy for a low carbon society

If the world wants to stabilise atmospheric greenhouse gases at 550 parts per million, massive changes are required, especially in the energy sector. This article discusses means and costs of drastically reducing carbon emissions.

No longer confined to the roundtables of politicians and scientists, the debate on climate change has become a mounting wave that doesn’t seem to be losing momentum. Both policy and research communities have focused on the need to stabilise atmospheric CO2 concentrations at about 550 ppm (parts per million, all greenhouse gases included). This is generally considered a very ambitious, hardly feasible target with drastic implications for our economies and lifestyles.

Given projected world population dynamics, this objective requires reducing per capita emissions in the second half of this century from about 2 tonnes carbon equivalent (tC) to about 0.3 tC per year. In other words, the world will have to cut emissions to the per capita average of India today – quite a significant reduction for most industrialised countries (US average per capita emissions are about 5tC) and for countries that aim at similar lifestyle standards. For example, 0.3 tC is the amount of greenhouse gases emitted by an individual flying – one way – from the EU to the US East coast!

Clearly, a world with 0.3 tC per capita per year will be a different world. What are the optimal strategies and the related economic costs of achieving this ambitious, but seemingly inevitable, target?

Energy efficiency and de-carbonisation

Let us assume that economic and population growth cannot be targeted by climate policy. It is clearly undesirable to solve the problem by reducing economic growth and likely politically undesirable to focus on population growth. Emission reductions can then be achieved mainly by increasing energy efficiency and by reducing carbon intensity. Energy efficiency improvements beyond the baseline scenario are the first option to endorse; yet, especially for ambitious emission reductions, energy de-carbonisation is eventually essential.

To achieve a low carbon energy supply, power generation is one of the best options, because of the relative weight on global emissions and the availability of alternative technologies. However, to optimally achieve a 550 ppm concentration target, almost all electricity (around 90%) will have to be generated at low, almost zero, carbon rates by 2050. This is a drastic change and currently pursued through three options: carbon capture and sequestration, nuclear energy and renewable sources. Carbon capture and sequestration (CCS) allows the power sector to continue to use coal, the most available and affordable fossil fuel. However, the necessary investments are very large. To achieve the 550 ppm target, between 30 and 40 1 gigawatt (GW) coal-with-CCS power plants need to be built each year from 2015 onwards, a value in line with the historical capacity building of traditional coal plants (that make up for roughly 50% of electricity generated in the world). A number of large-scale pilot plants should thus be put into place in the next ten years to ensure the feasibility of such a massive deployment.

Nuclear power is at the moment the only proven base load generation for large-scale electricity decarbonisation. In addition, it will become extremely competitive for the range of carbon prices implicit in the adoption of a climate policy designed to achieve the 550 ppm target. However, 20 1GW nuclear plants or more would need to be built each year in the next half century, bringing the nuclear industry back to the construction rates of the 1980s. External costs, such as those related to nuclear waste disposal or proliferation risks, could make this scenario undesirable. In any event, some innovation in the technology as well as in the institutions regulating the global implications of a massive deployment of nuclear energy would need to accompany this expansion.

Renewables, especially wind power, have developed at an impressive rate in recent years (up to 10GW per year), but the limited annual operating hours and costs constrain their potentialcontribution. Despite a small absolute potential of renewables, an almost three-fold capacity expansion with respect to a baseline scenario – more than for any other generation technology – and an overall 17-fold expansion of present installed capacity should be achieved by 2050. This is equivalent to about 60,000 new large open-sea wind turbines. What a different world!

Carbon abatement in non-electric energy may be mostly achieved by improvements in energy efficiency – i.e. through measures meant to reduce fossil fuel consumption – because of the dispersion and limited array of carbon free technologies. Decarbonising such sectors as transport, residential, etc. would significantly ease the attainment of the climate stabilisation target; yet, carbon abatement alternatives beyond energy savings are still expensive and unproven at large scales. Innovation in fields such as batteries, bio-energy conversion, and so forth would open up new possibilities, but they are still in their infancy and might come at unexpected costs, as the recent sensitivity of food prices to energy crops has shown.

The costs of a different world

What are the macroeconomic costs of such drastic changes in world energy systems? How much investment and R&D expenditure are required? Assuming a “cap and trade” stabilisation policy with a perfect international market for emission allowances, our work shows that the total direct cost in terms of undiscounted GDP losses in 2030 would be 1.2 - 1.7% of the world GDP (see Table 3 in Bosetti, Carraro, Massetti and Tavoni, 2007). In the Intergovernmental Panel on Climate Change’s most recent report, the median value for the total direct cost of 550 ppm stabilisation is 0.6% (IPCC 2007). Our cost estimate is larger than the typical IPCC estimate because the inefficiencies and market imperfections introduced in our model better mimic the behaviour of markets and policymakers. In particular, we adopt a non-cooperative game-theoretic framework to better describe global strategic and policy interactions.

The macroeconomic cost of stabilising greenhouse gas concentrations is obviously larger if the time horizon is lengthened. According to our calculations, the cost would be between 2.1 and 3.7 % of world GDP over the course of the 21st century. The cost would be even larger if only a subset of countries adopt climate policies. Therefore, the costs of moving to a drastically different energy system are probably substantial.

Accounting for additional emission reduction measures, such as in the agriculture and forestry sectors, could increase the feasibility and decrease the cost of the climate target. The key role of forestry management in contributing to the overall emission reduction effort has been recognised and emphasised in the United Nations’ Bali Action Plan. For the set of carbon prices implicit in the 550 ppm stabilisation scenario, forestry management could save up to 1.5 gigatonnes of carbon per year in the next fifty years, a figure equivalent to 20% of today’s world emissions. This would have a significant impact on the carbon market, decreasing costs by as much as 30-40% (Bosetti and Tavoni, 2007). Nevertheless, this might also delay ultimately crucial investments and innovation in carbon-free technologies.

Technical change is, without any doubt, a key component of any policy designed to stabilise greenhouse gas concentrations. However, investments in energy efficiency R&D declined after 1980 and have remained low in recent years despite their potential to seriously contribute to achieving the stabilisation target. Innovations in carbon capture and sequestration and nuclear energy are crucial. Low carbon alternatives in the non-electric use of energy are also indispensable.

What is the cost of innovating to achieve the 550 ppm stabilisation target? In our work we show that a dedicated long-term research effort is needed, entailing as much as a tenfold expansion in energy R&D investment by 2050, to reach an annual figure in excess of $100 billion. In terms of GDP share, R&D expenditure in the energy sector should increase from about 0.2% of GDP to about 0.6% by 2050. This important and crucial effort cannot entirely rest on the shoulders of the private sector. Public policies, both domestic and international, are necessary. Domestic R&D incentives and a global R&D fund are likely to be plausible options.

Meeting the climate challenge

Climate change is a serious threat to the stability of the world’s economic system. The stabilisation of greenhouse gas concentrations at 550 ppm could prevent most damages from climate change but requires drastic changes in the energy sector. These changes are costly and can be achieved only if large investments in energy infrastructures and in R&D are undertaken in the next forty years. That requires farsighted and well-designed public policies to provide adequate economic incentives and to mobilise sufficient financial resources.


Bosetti, V., C. Carraro, M. Galeotti, E. Massetti and M. Tavoni, (2006). “WITCH: A World Induced Technical Change Hybrid Model.” The Energy Journal, Special Issue on Hybrid Modeling of Energy-Environment Policies: Reconciling Bottom-up and Top-down, 13-38.
Bosetti, V., C. Carraro, E. Massetti and M. Tavoni, (2007). “Optimal Energy Investment and R&D Strategies to Stabilise Greenhouse Gas Atmospheric Concentrations”, CEPR Discussion Paper 6549.
IPCC (2007) “IPCC Fourth Assessment Report, Working Group III”.
Tavoni, M., Sohngen, B., Bosetti, V. “Forestry and the carbon market response to stabilize climate”, Energy Policy, 35 (2007), 5346–5353.




1 This target roughly coincides with IPCC Post-TAR stabilisation scenario B, which is close to the EU objective of keeping future temperature changes below 2 degrees Celsius.
2 See our paper on “Optimal Energy Investment and R&D Strategies to Stabilise Greenhouse Gas Atmospheric Concentrations” CEPR Discussion Paper 6549.



525 Reads