Why economic models of climate change need better science

One standard model suggests an 18-degree rise in temperatures would only result in global economic output reducing 50 per cent. But most of the Earth would be uninhabitable.

Carbon Brief

Economic analysis can help policymakers gauge how much action it's worth taking to tackle future climate change. But some economic models have been criticised for failing to account for the findings of the latest climate science – resulting, critics say, in significant blind spots.

Prominent climate economist, Lord Stern, is one of those critics. He's previously warned that models "grossly underestimate" the risks of climate change, and consequently the cost of failing to act.

In a new paper due to be published in The Economic Journal and released by LSE this week, Stern and co-author Dr Simon Dietz try to redress the balance. They suggest informing models with the latest research strengthens the case for stringent action to tackle climate change.

Estimating costs

There are a number of models that try to work out how much economic damage climate change will cause in the future, and how much we should pay now to avoid that.

But the latest Intergovernmental Panel on Climate Change report warned that such models were incomplete. The IPCC uses multiple 'integrated assessment models', or IAMs, to estimate the future cost of climate change. The models estimate changes in temperature and economic conditions based on future emissions. They then calculate how much additional damage is caused by emitting an extra tonne of carbon dioxide today.

That cost is known as the social cost of carbon and is expressed in dollars per tonne of carbon dioxide – a carbon price, effectively.

But the information you get out of models in only as good as the information you put in. Professor Chris Field, co-chair of one of the groups that wrote the IPCC report, has previously said the numbers produced by economic models are "old fashioned" when you consider "the modern science".

What does that mean? It implies that economic models don't capture all the potential impacts of climate change. That means they produce low social cost of carbon estimates, and underestimate the risk of failing to take action to tackle climate change.

Updating models

To address these gaps, Stern and Dietz tried to incorporate insights from recent climate science into one prominent model – the dynamic integrated climate-economy, aka DICE, model developed by Yale Professor Bill Nordhaus. The thinking goes that by updating what they describe as the models "unrealistic" assumptions, the results should become more useful.

For example, recent research suggests climate change could seriously damage some countries' infrastructure or cause certain assets (such as coastal power stations) to be abandoned. That could have a big impact on economic productivity, and so Stern and Dietz include this in the model.

They also adjust the model's climate sensitivity input – the impact of doubling the concentration of greenhouse gases on global temperatures. The standard DICE model only looks at the economic impact of around three degrees of warming above pre-industrial levels. Stern and Dietz make it consider a range from 1.5 to 6 degrees.

Finally, they adjust the temperature threshold at which the model assumes large economic losses occur. The standard DICE model suggests an 18-degree rise in global average temperatures would only result in a global economic output reducing by about 50 per cent. But such warming would likely render the Earth uninhabitable for most species, including humans. The tweaked model allows such damage to occur at much lower levels of warming, to try and better reflect the likely real world impact of climate change.

The result of the updates – perhaps unsurprisingly – is that the reformed DICE model implies taking much stronger action than previously. The updated model finds that if policymakers are going to avoid warming of more than 2 degrees above pre-industrial levels, they should try to implement a carbon price of $US32 to $US103 today – either through a carbon market or a tax. The price should rise to $US82 to $US260 in 2035 depending on how great the damage from climate change is expected to be.

Useful indication

The carbon prices need to be taken with a pinch of salt, however. It's uncertain how much emissions could reduce as a result of implementing a stringent carbon price. That's why most models – including DICE – give a range of prices dependent on how big the climate impacts are anticipated to be.

Nonetheless, it's notable how the results differ to the model's previous recommendation and current carbon prices. Even in a best case scenario, the updated model's carbon price is considerably more than the $US12 the standard model recommends. It's also much more than current carbon prices. The EU's carbon market currently has a price of around €6 ($US8), and two US markets have prices of $US2 to $US15.

So even if the updated price range doesn't translate into a hard and fast policy recommendation, the discrepancy with current prices will support the argument that policymakers aren't doing enough to curb emissions.

Originally published on Carbon Brief. Reproduced with permission.

Related Articles