Can cost benefit analysis grasp the climate change nettle? And can we justify ambitious targets?
25 Feb 2019
It is easy, for anyone concerned about the future of the planet and our place on it, to assume that formal economic analysis of the case for mitigating climate change is almost redundant or has only limited value in determining the course of action we should take. However, Simon Dietz’s strongly recommended presentation at the Oxford Martin School earlier this month was a timely reminder both that we cannot in practice escape responsibility for some balancing of costs and benefits, and that, in spite of the limitations, trying to answer that question can still yield valuable recommendations and insights.
Weaknesses in the economic calculus
The weaknesses of conventional applied economics, especially cost benefit analysis, in quantifying issues on the scale of climate change are widely recognised.
Some of the weaknesses are of a technical, conceptual or even philosophical nature. In dealing with risk and uncertainty, there is little or no empirical basis for assessing probability distributions. Non-linearities and non-market effects present technical challenges. The policy choices are a long way away from the world of marginal analysis in which cost benefit analysis (CBA) is typically more comfortable. Distributional and inter-generational inequalities bring in philosophical and ethical questions, and economics still lacks a philosophically sound and universally accepted basis for time discounting.
An even bigger issue perhaps has been the inability of conventional macro-economics to capture the complexities, or indeed the potential scale, of major disruptions caused by climate (or, arguably, other non-marginal or disruptive changes such as Brexit). So-called integrated assessment models (IAMs) purport to capture complex feedbacks between climate impacts and economic output, but the judgment from academics on IAMs has been damning. Such models come “close to assuming directly that the impacts and costs will be modest, and close to excluding the possibility of catastrophic outcomes”, according to Nicholas Stern. In other words, they largely assume away the problem they are supposed to be analysing. It is for these reasons that CBA arguments have become the new line of defence for climate sceptics whose refusal to accept the science has clearly become untenable.
In plain language, Robin Harding in the FT claimed that
… one standard model only gives damage greater than 50 per cent of output with 20oC of warming. Combine that with the assumption that the economy will be many times bigger in the future and the problem is clear. Your grandchildren might be cooking in their own fat on the London Underground, but rather than regarding them as dead, these economic models would regard them as wealthier than you. …
… After the financial crisis, the world did not construct vastly complicated models to estimate the chances of another meltdown and the damage it would cause. Policy makers simply recognised that regulations such as the US Dodd-Frank Act are a small price to pay for preventing a repeat performance. It is time to take a similar risk-based approach to the greater problem of climate change.
But we still need to choose between 2 oC and 1.5 oC targets
Simon Dietz’s presentation focused very clearly on a current issue – the case for, and practicality of, adoption of the aspirational 1.5oC target that arose from the COP 21 Paris Agreement. He pointed out, correctly, that the lower target would remove a very substantial part of the “remaining carbon budget”, and that this required unprecedented rates of reduction in carbon emissions, leading us inter alia to early “zero carbon” and the necessity of carbon capture. Other calculations implied a high implicit carbon price of $100 per tonne of CO2, or more.
Defending cost benefit analysis, at least partly on the basis that alternatives faced, and similarly failed to resolve, most of the same difficult issues as CBA, he attempted to answer the 1.5oC question. His analysis, which neatly side-stepped some of the methodological issues, ultimately boiled down to the biggest uncertainty of all – estimating the cost of the damage from climate change.
Successfully avoiding the CBA/ modelling trap described above, of assuming away the problem, he argued that the (wide) range of damage estimates made from an empirical rather than an IAM approach could at the very least make a strong case for the lower target, even if this was not conclusive in CBA terms. Subsequent discussion focused on what was or was not included in the various estimates of the costs of climate change, notably in relation to resource conflicts and migration.
The conclusions may have seemed uncontroversial or even too mild to many in the audience, especially those of us with professional exposure to climate issues, but Simon noted that at least one distinguished economist had estimated 3.5oC of warming, often taken to be catastrophic, as an “optimum” amount. Simon also emphasised the importance of the “options” argument for adopting a lower target. More action now increases the room for manoeuvre as time passes, especially if more knowledge increases the estimated risk of “tipping points” or serious downside and catastrophic outcomes.
Reinforcing the case for early action and tighter targets
Related arguments have previously been advanced by the actuarial profession, who of course have long specialised in the analysis of catastrophic risk. In 2014 Oliver Bettis addressed an event at the Grantham Institute and argued for a “risk of ruin” approach similar to that used in the insurance industry. The conclusion of his analysis, and using “actuarial” risk parameters, was that:
- the CO2 already released (400ppm) produces unacceptable risk of ruin
- emergency decarbonisation may be the correct risk management response.
- allowing for slow feedbacks, the right target might be below 350ppm.
- removal of CO2 from the atmosphere should be investigated.
In qualitative terms this could be seen as a similar set of conclusions to Simon’s, but with even more aggressive support for lower targets. Actuaries are possibly more risk averse than economists.
And assessing the practicality and cost
Perhaps the most important barriers to a lower target are simply those of practicality and cost. A badly managed and extremely disruptive re-engineering of the entire energy sector, and indeed the entire economy, could after all bring with it its own “risks of ruin”. It is often asserted that the costs of action are simply unaffordable.
However, at least in very broad macro-economic terms, the admittedly high costs of decarbonisation appear as eminently manageable when compared to other major disruptive impacts on the world economy. At a national level for example, if we take Simon’s very approximate estimate of $100 per tonne, and apply it to UK emissions of about 500 million tonnes of CO2, we get a cost of c. 1.4% of GDP; this is consistent with the 1-2 % range common in post Stern discussions of the cost of action necessary to curtail emissions to a “safe” level. As often pointed out, this might mean reaching a particular standard of living less than12 months later by 2050.
While in most contexts 1-2 % of GDP is a substantial quantity, it is comparable to other energy and other commodity price shocks that the world economy has absorbed, mostly without ruinous dislocations. Notably these include the multiple oil price shocks of the last 45 years. Compared to the impact of the 2008 financial crash and its consequences, estimated to have cut GDP by as much as 15%, or to current expectations of the possible near term cost to the UK of Brexit, the cost of climate action can be seen as modest and eminently affordable.
Intellectually therefore the case for ambitious climate action is strong. Less well developed are, first, the levels of political support and, second, agreement on, and acceptance of, the actual means to get there. But that, as they say, is a series of stories for another day.
 This can be viewed online and, together with the immediately previous presentation by Joeri Rogelj provides an excellent introduction to understanding the current state of play on carbon budgets and policy dilemmas.
 Pindyck, Robert S. 2013. "Climate Change Policy: What Do the Models Tell Us?" Journal of Economic Literature, 51(3): 860-72. Pindyck is particularly scathing.
A plethora of integrated assessment models (IAMs) have been constructed and used to estimate the social cost of carbon (SCC) and evaluate alternative abatement policies. These models have crucial flaws that make them close to useless as tools for policy analysis: certain inputs (e.g., the discount rate) are arbitrary, but have huge effects on the SCC estimates the models produce; the models' descriptions of the impact of climate change are completely ad hoc, with no theoretical or empirical foundation; and the models can tell us nothing about the most important driver of the SCC, the possibility of a catastrophic climate outcome. IAM-based analyses of climate policy create a perception of knowledge and precision, but that perception is illusory and misleading.
 The Structure of Economic Modeling of the Potential Impacts of Climate change: Grafting Gross Underestimation of Risk onto Already Narrow Science Models / Nicholas Stern. Journal of Economic Literature, Vol. 51, N° 3, pp. 838-859, September 2013
 Robin Harding. 2014. “A high price for ignoring the risks of catastrophe.” Financial Times. 18 February 2014.
 The maintaining of options is also an important part of the case for attaching a higher value to saving current emissions, rather than assuming a rising real price of carbon, addressed by the author, inter alia in Cumulative Carbon Emissions and Climate Change. Has the Economics of Climate Policies Lost Contact with the Physics? Oxford Institute for Energy Studies Working Paper.
 Risk Management and Climate Change: Risk of Ruin. Oliver Bettis. Event at Grantham Research Institute on Climate Change and the Environment London, 14 January 2014.
This opinion piece reflects the views of the author, and does not necessarily reflect the position of the Oxford Martin School or the University of Oxford. Any errors or omissions are those of the author.