Solar geoengineering (also known as solar radiation management) is a technology in its infancy – and it is controversial. It has the potential to reverse or mitigate some of the global warming caused by greenhouse gases by either releasing cooling particles (for instance sulphur) into the stratosphere, or by modifying clouds over the oceans so that they reflect more heat back into space.

But there are major concerns about how politics could influence research and development, and the deployment of solar geoengineering on a global scale. Last year’s special report by the Intergovernmental Panel on Climate Change (IPCC), Global Warming of 1.5 ºC struck a cautionary note: ‘Although some [solar radiation modification] measures may be theoretically effective […], they face large uncertainties and knowledge gaps as well as substantial risks and institutional and social constraints to deployment related to governance, ethics, and impacts on sustainable development.’ One of these risks could be conflict, should a country use geoengineering without global agreement – an action that cause harm to others.

Here we use game theory to better understand these concerns and find out what could happen if countries were able to move the earth’s thermostat in either direction – by using geoengineering technologies to reduce the temperature and counter-geoengineering to turn it back up again.

Cooling off

Solar geoengineering technologies could be cheap. This creates a problem economists call the ‘free-driver effect’. If the cost is not prohibitive, a single nation (or even a single billionaire) could pay to press the button on a geoengineering action that affects the whole planet.

On first impressions it might sound good for a potential global warming fix to be inexpensive and accessible. But a country with an especially strong incentive to cool the planet – one that is suffering badly due to climate change – could go ahead and deploy a technology that will affect us all, effectively taking a unilateral decision on the optimal temperature for the Earth.

Some like it hot(ter)

One idea to counter this ‘free-driving’ effect is to develop counter-geoengineering. While solar geoengineering would cool temperatures, counter-geoengineering might use similar technology to heat the earth up – for example, by injecting short-lived heat-trapping aerosols into the atmosphere, or using a chemical to counteract a sulphate injection.

The possibility of being able to turn the temperature back up might act as a deterrent to free-drivers. Who would want to risk causing an escalation of opposing climate interventions that would only waste resources? The prospect of counter-geoengineering might reintroduce a willingness to collaborate. We tested this possibility using game theory.

The rules of the game

We set up a two-player game. Each player represents a country (or a bloc of countries) and each has a – potentially different – temperature preference for the planet.

It is a two round game. Round 1 is treaty-making. The players can choose to opt into a treaty and collaborate, or they can opt out. There are two treaty options available: the first is a deployment treaty, where countries jointly decide on the climate intervention that maximises the coalition’s overall payoff. The second treaty option is a moratorium treaty, under which the countries commit to abstain from any climate intervention. Whichever decision they make, they will only enter into a treaty if it is in their best interests – all the players are ‘selfish actors’.

Round 2 is deployment, i.e. modifying the global temperature with a climate intervention that is relatively cheap. If the countries entered into a treaty in Round 1, then they either abstain from a climate intervention (opting for the moratorium treaty) or undertake the intervention cooperatively. If no treaty was formed, the players choose their climate intervention levels non-cooperatively.

We played two versions of the game. In one version only solar geoengineering technology was available to the countries – so they could cool the global temperature but not increase it. In the second version they also had access to counter-geoengineering, so they could also turn the temperature up. Comparing the two versions then sheds light on how counter-geoengineering changes the strategic interaction surrounding climate interventions.

The results: arms race or abstinence

The results of the game reveal the importance of the level of agreement over what countries consider the ‘best’ temperature for the planet.

If countries have similar preferred temperatures but do not choose to enter into a treaty, there is a free-rider outcome – countries would benefit from the temperature reduction caused by another country’s geoengineering actions without themselves contributing much to the cost of deployment.

Where countries differ greatly in their preferred temperature, and if counter-geoengineering is not available (which could be because it has not yet been developed), the result is a free-driver outcome, as predicted. The country with the strongest preference for cooling (the free-driver) turns the temperature right down – even if the other prefers it warmer.

In both of these cases, incentives to cooperate are weak.

However, with counter-geoengineering technology on the table the strategic interaction changes, with two outcomes. A country that views the free-driver’s deployment of cooling as excessive now has a tool to counteract it – and will use it. Without the opportunity to cooperate, this results in a ‘climate clash’, an escalation of cooling by geoengineering and warming by counter-geoengineering that has no winners and is very harmful.

However, if cooperation is an option, this bleak outlook may be enough to encourage countries to work together. In particular, the free-driver may be ready to compromise on the amount of climate intervention it makes.

Cooperation is not guaranteed, though, and the outcome might still be a destructive climate clash. Even if countries do cooperate, they may take the moratorium route – and this could be worse than the free-driver outcome if it means the world misses the opportunity to potentially reduce the damage from climate change by using solar geoengineering.


How solar geoengineering and counter-measures could and should be used to adjust the planet’s temperature is subject to widely differing opinions and intense debate. Certainly our study emphasises the crucial need to focus on how any geoengineering interventions could be governed, with the welfare of the majority a central goal. Cooperative decisions including a broad set of actors typically are welcome, but our results also point to the importance of getting the content of a treaty right.

Of course, there are limitations to our analysis, not least the fact that the paper’s main analysis was undertaken in a two-player game, when in reality we could face complex negotiations between many countries. Countries may also want to modify aspects of the climate beyond temperature – especially rainfall patterns. And geoengineering could affect human and ecosystem health, by causing acid rain or ozone depletion – further effects that could cause tensions if one country forged ahead at the expense of others.



Daniel Heyen is a postdoctoral researcher at ETH Zurich. He is an applied theorist working at the interface of decision theory and environmental economics. Daniel’s main research interest is in societal decision-making under uncertainty and learning. Key topics of his work are the description of scientific uncertainty, the design of decision rules, and the analysis of active learning and the value of information. Prior to his position at ETH Zurich, Daniel was a postdoctoral researcher at the Grantham Research Institute, funded through a Fellowship from the German Research Foundation. Daniel completed his PhD in economics at Heidelberg University. His background is in Mathematics and Physics.

Joshua Horton is research director of geoengineering at the Keith Group. Josh conducts research on geoengineering policy and governance issues, including the regulation of research, liability and compensation, and geopolitics. Josh previously worked as a clean energy consultant for a global energy consulting firm. He holds a Ph.D. in political science from Johns Hopkins University.


Juan Moreno-Cruz is an associate professor at the School of Environment, Enterprise and Development and the Canada research chair in energy transitions at the University of Waterloo. He is also a CESifo research affiliate. He has a Ph.D. (2010) from the University of Calgary and a B.A. and M.S. in electrical engineering from the Universidad de Los Andes. Previously, he was an associate professor in the School of Economics at the Georgia Institute of Technology (2011-2017), were he remains as an adjunct professor. He is a visiting researcher in the department of global ecology of the Carnegie Institution for Science at Stanford University, an advisor for Carnegie Energy Innovation, and a research associate of Harvard University’s solar geoengineering research programme.