About Public Offering

Contact us:

Subscribe to Public Offering Public Offering RSS Feed


April 22, 2008

The Costs of Climate Change

Geoff Heal
Paul Garrett Professor of Public Policy and Business Responsibility
Print this post

Should the U.S. join the Kyoto Protocol, or at least play a positive role in the search for a successor? Or is this too costly, or otherwise “fatally flawed,” as our president has suggested?

Humanity as a whole needs to address climate change: it has the potential to alter the world around us dramatically, and for the worse. Preventing or at least minimizing climate change will not be cheap, but it will still be a good buy.

Roughly speaking, reducing greenhouse gas emissions to the level consistent with limiting climate change to around 2 degrees Celsius (about 4 degrees Fahrenheit) will cost in the range of 1-2 percent of world income.¹

We need to compare this cost with that of allowing climate change to continue — a harder number to compute. The most thorough and most recent estimate put the cost at “at least 5 percent” of world GDP, with “at least” here indicating that this assumes a rather conservative estimate of temperature increase and includes in the calculation only the costs of climate change that are captured in market transactions.

So this estimate would include the costs of land lost due to increases in sea level, or of reduced agricultural output, but not of species driven extinct by changing environments, or of the spread of disease vectors in a warmer world and many other non-market costs.

My own estimates are that the nonmarket costs are probably going to be bigger than the market costs, so the total could be 10 percent or more of world income.

If you can spend 2 to save 10, that’s a good investment — if the 2 and the 10 are both spent and saved today. A problem in the climate change case is that we spend the 2 now, well before we save the 10: we invest 2 to make 10, but we invest the 2 now and get the 10 spread over many years starting in 2040 or thereabouts. So we have to worry about the discount rate: over long periods it makes a big difference. I argue that the right discount rate for society to use here is very low, so investing 2 now to make 10 in the future still makes sense. But some of my colleagues disagree.

Another element in the calculations is that there is always a small but non-zero risk that the extent and impacts of climate change will be far greater than the central estimates of the IPCC. They admit this and give quite large error bars around their figures. These error bars are asymmetric: the mean increase is in the range 2-4 degrees C with no chance at all of less than 1 degree (we are already at 0.75) and some chance of 6 to 8 degrees. The upper end of this range would not just be costly: it would be disastrous. So we have to see climate policies as insurance policies, insuring against the small risk of a disastrous outcome. We can apply techniques from risk management to work out how big a premium it is worth paying: it could be several percent of GDP.

Back to the Kyoto Protocol. If we want to reduce emissions, then there are three ways of doing this:

  • Order firms to emit less (command and control)
  • Tax emissions
  • Introduce a cap-and-trade system, as the U.S. did for SO2 under the 1990 amendments to the Clean Air Act.

Kyoto chose the last of these three, and the European Union has embedded this in its internal approach to reducing emissions. This also seems to be the preferred method in the U.S. states that are moving to reduce greenhouse gas emissions.

To economists cap-and-trade or taxes are clearly the preferred options, and politically it seems that cap-and-trade is way ahead. Taxes are a political liability, and cap-and-trade has the merit of generating new tradable securities, which the investment banking community loves. If we follow this route, then the CO2 market could become a multi-trillion dollar market within ten to fifteen years.


¹Here’s the calculation for the U.S.:

We currently emit about 7 billion tons of CO2 annually and have a GDP of around $13 trillion. At a very rough guess, real GDP will increase by a factor of 2 to 2.5 by 2050 and without action emissions will rise to about 12 to 14 billion tons — they are rising less fast than GDP. To be reasonably certain that temperature increases are in the range of 2 degrees C we would need to reduce U.S. emissions by around 80 percent by 2050, which is a reduction of 9 to 10 billion tons.

There have been many studies of the cost of reducing emissions, the most notable being a study last year by McKinsey. They suggest that the average cost of reducing emissions will not exceed $40 per ton, and may be less. An easy calculation shows that reducing by 9 to 10 billion tons at $40 per ton will take between 1 and 2 percent of 2050 GDP.

Photo Credit: Mila Zinkova