It all comes down to the assumptions made in the calculations. The Stern Review calculations have been criticised for loading too much of the cost onto the current generation, and for making some unusual assumptions about risk-aversion. David Anthoff, from the Economic and Social Research Institute in Dublin, Ireland, and his colleagues have re-evaluated the Stern calculations and reanalysed the costs, taking into account people's behaviour, and income differences around the world. They come up with a new figure of $206 per tonne of carbon.

Economists usually recommend that carbon is taxed at a price equal to the social cost of carbon (SCC): the value of the damage caused by climate change impacts that would be avoided by reducing emissions by one tonne.

Anthoff and his colleagues explored the sensitivity of the SCC to two variables: the pure rate of time preference and the rate of risk aversion.

Pure rate of time preference is a measure of how the future is valued relative to the present. A low pure rate of time preference is associated with high levels of altruism – accepting high costs now, to ensure that future generations are not burdened with problems the current generation have created. However, most people expect to become richer in the future, and are happy to offset some of the costs until a later date. "Philosophers and religious leaders say that we should have a zero pure time preference, but people tend not to listen," Richard Tol, also from the Economic and Social Research Institute in Dublin, told environmentalresearchweb.

The Stern review assumed a pure rate of time preference of 0.1% per year. "This value is out of step with empirical values, and even the guidance of the HM Treasury, the sponsor of the Stern Review," said Tol. Other studies have tended to use a pure rate of time preference of between 1 and 3% per year.

Meanwhile, risk aversion was set at a relatively low rate of 1 for the Stern Review calculations, compared to between 1 and 3 used in other studies. The low rate of risk aversion assumes that people are willing to accept relatively high levels of risk, rather than paying a cost to reduce the risk.

In addition the Stern Review assumed that people's vulnerability to climate change would remain constant. "For example the Stern Review assumes that people in Africa will be rich enough in the second half of the century to drive SUVs, but not rich enough to buy bed nets to protect their children against malaria," said Tol.

Anthoff and his colleagues recalculated values of SCC, this time using observed behaviour to estimate the pure rate of time preference and rate of risk aversion, and taking into account how vulnerability changes with development. Using these new figures they estimate the SCC to be around $206 per tonne of carbon. The results are published in Environmental Research Letters.

"This value is lower than Stern's $314 per tonne of carbon, but still considerably higher than the price imposed by current climate policy: 10 Euros per tonne in Europe and zero elsewhere," said Tol.