by Michael Cunningham (“Faustino”)
There are many issues of debate about global warming. Has there been warming this century? Will there be further warming? If so, will the cause be anthropogenic or other? What will be the impacts, both positive and negative? Should we take action to reduce emissions? How might we proceed, and what are the costs and benefits of various approaches?
These issues are of enormous economic and political significance, given claims of “catastrophic” outcomes and that many emissions-offsetting policies are very costly, involve large-scale structural change and would divert resources from other priorities.
The UK Government commissioned a study into the Economics of Climate Change by Sir Nicholas Stern, Head of the Government Economic Service and Adviser to the Government on the economics of climate change and development. Stern’s 2006 paper has been severely criticized by both scientists and economists, but continues to be influential in many countries and remains the basis for UK policy. This continuing influence of Stern’s work has prompted a rebuttal by UK MP Peter Lilley. Lilley has degrees in Natural Science and Economics from Cambridge, has worked as a development economist and investment analyst specialising in energy industries and held economic Ministerial posts from 1987-97. His paper is published by the Global Warming Policy Foundation.
The rebuttal is not a “denialist” paper. Richard Tol notes in his Foreword that “there is an economic case for greenhouse gas emission reduction. We cannot be certain that greenhouse gas emissions do not cause climate change. We cannot be certain that climate change is harmless. In fact, most evidence points in the opposite direction.” In order to focus exclusively on the economics, Lilley’s paper – like the Stern Review – takes the IPCC’s assessment as given. Rather, Lilley says,
“This study simply challenges Stern’s economic methods and conclusions – and shows that his Review was an exercise not in evidence-based policy making but in policy-based evidence making.”
This post on Climate Etc is an assessment of Lilley’s paper. As the paper has about 100 pages and is accessible to the non-specialist, I’ll not go into too much depth here – if you want to go deeper into the reasoning and supporting evidence and argument, please check the original then raise questions/comments on CE if necessary. [Page numbers in brackets] [link] http://www.thegwpf.org/wp-content/uploads/2012/09/Lilley-Stern_Rebuttal.pdf
UPDATE: a new slightly modified paper is posted [here].
Lilley argues that “rather than being used to stimulate public debate, [the Review] has been used to silence dissent from the official orthodoxy … The intelligent lay reader and even the professional economist could not readily work out how the Review had reached its [dramatic] conclusions … [which] became established as proven before their rationale was understood or subjected to criticism. …
“Overall, [the conclusions] give the impression that for a modest cost we can prevent damages which are imminent, would cost us five to twenty times as much and would involve large scale loss of lives and livelihoods. However, these conclusions are highly misleading and misrepresent the body of the Review …” 
Estimating the impact of global warming [25-36]
Lilley says that “Estimating the amount of global warming and its impact if no action is taken to reduce emissions involves a series of steps, each requiring a number of assumptions and estimates.” One must project future concentrations of greenhouse gases; estimate the resultant changes in climate; and evaluate the impact climate change will have on the economy, society and human well-being. This step, for example, requires “estimates of how climate change will affect a host of variables, including the prevalence of diseases, crop yields, energy demand, species abundance etc and how people, business, government and markets would respond to that.
“At every stage, a range of different estimates of each factor have been published in the literature which Stern reviews. Stern generally tends to adopt estimates and assumptions towards the pessimistic end of the range.” 
Stern’s own estimates of the cost of warming were much higher than those in other studies. For example, his estimate of increased building and infrastructure damage from storms was 50-100 times that of a 2011 World Bank study, and he has stated that it could be 5% of Gross World Product, 500 times the Bank’s figure.  Stern’s highest cost estimates depend heavily on the occurrence of “catastrophes,” one of which is potentially beneficial, one which did not occur under previous warm periods and one which might materialise only over thousands of years. [64-66]
Given that Stern’s cost and benefits were to be the basis for significant policies, one would expect some rigour in their estimation. In fact, Stern modelled simple assumptions to get his cost figures. Lilley says that “Although the Review enumerates a whole range of alarming estimates of different forms of harm that global warming could cause,” with “disparate times scales, metrics and coverage,” these are not used. Instead,
“the total cost estimate comes from a simple equation embedded in the PAGE2002 Impact Assessment Model. The model is given a range of assumptions of impacts on the GDP of each geographic area for a 2.5C rise in temperature, [which is] deemed to reduce GDP by between 1.5% and 4% – with a median 2% loss. The loss is then set to increase as a power of temperature ranging between linear and cube – averaging 1.3. [35, 65]
“Additional assumptions are made about the effect of temperature on non-economic factors … a proportion of losses … is assumed to be prevented by adaptation. … There is no overarching scientific theory relating damage to temperature changes – only a patchwork of hypotheses about how different aspects of life might be affected.”
Further, the model assumes that “all forms of damage happen simultaneously and immediately the temperature increases.” 
I am sceptical of any attempt to estimate costs on the timescale used by Stern, and would not base policy advice on such a simplistic and arbitrary assessment.
The costs of restricting emissions [37-45]
While Stern’s damage estimates were far above those in other studies, his estimate of the costs of reducing emissions was far below them. He “concluded that the cost of stabilising emissions at his target [550 ppm] could be limited to just 1% of GDP by 2050. That was below the bottom of the range of Stanford University’s Energy Modelling Forum, whose average was 2.2% of GDP. Moreover, that group of 21 model estimates calculated that the cost would rise to 6.9% of GDP by 2100. Stern did not project costs beyond 2050 – ignoring any subsequent escalation as further reductions in CO2 emissions became increasingly costly.”  That is, having based a very high cost of non-action on assumption, Stern compares it with a very low assessment of the costs of action. This comparison is exacerbated by his use of discounting (see below).
That said, although the marginal cost of any activity tends to rise, in the case of emissions reduction, one would expect it to fall as new, more efficient, technologies appear. This is a major argument for deferring action, and I would be cautious about accepting the pattern projected by the Stanford Forum. Of course, as noted I am highly sceptical of all attempts to provide economic estimates for 40 and more years ahead.
In 2008 the UK Labour Government enshrined in law targets to meet Stern’s objectives. However, it emerged that its own estimate of costs exceed Stern’s. “The Impact Assessment estimated a permanent loss of 1.6% (or in the range of 1-2%) of GDP from 2050 onwards. It put a net present value of up to 400 billion pounds (or 18 bn pa before discounting) on the cost of mitigation up to 2050 but admitted that this excluded the transitional cost which it said could average a further 1.3-2% of GDP up to 2020, not to mention the costs of driving British carbon-intensive industries overseas, the risk of which the IPCC found to be “relatively high” and could result in a leakage of 5-20% of carbon savings overseas.”  Ignoring these costs, the Impact Assessment concluded that the costs were “consistent with the range of costs identified by the Stern Review – 1% of GDP by 2050, within a range of +/- 3%.” (I don’t know whether that means 0.97-1.03 or -2 to +4.)
However, the Assessment said that “Short and medium (i.e. to 2020) transition costs could be in the upper end of the range indicated by the Stern Review.”
Discount rate [46-58]
Lilley says that, without emissions reduction action, “The Stern Review predicts that costs of global warming … will rise faster than GDP until 2200 and thereafter continue to rise in line forever and ever,” which, as Lilley points out, is physically impossible. “By contrast, the cost of reducing emissions starts now. We therefore need to compare … damages into the indefinite future with the cost of taking action in the coming decades to prevent it.” 
As discussed in a recent CE thread Activate your science, a dollar in the future is worth less than a dollar now. In order to compare future streams of costs and benefits, we need to put them on a common basis. This is normally done by applying a discount rate which gives the present value of each stream. As Lilley points out, “The rate at which future costs and benefits are discounted is crucial.” He calculates the present value of $100 [he uses pounds, but no matter] 100 and 200 years hence for various discount rates. At a discount rate of 0.1%, the present values will be $90 and $82 respectively; at 6.0%, the kind of rate used for government investment assessment, the present values would be 30 cents and zero. That is, costs and benefits in the far distance are irrelevant. You can make the costs relevant only by using an absurdly low discount rate.
“In its 700 pages, the Stern Review does not reveal the discount rate used even though this is its most crucial assumption. It was not until some time after publication and as a result of strenuous inquiries, that it emerged that the Review uses a discount rate of just 1.4%.”  Rather than the 30 cents and zero of our 6.0% example, this gives values of $25 and $6, which skews the cost-benefit analysis towards the reducing-emissions-now option.
“The effect of using the Stern Review’s low discount rate is to give huge weight to events in the distant future which are assumed to be the ineluctable consequences of actions taken by this generation. … Nordhaus has estimated that, under Stern’s methodology, half of all benefits of preventing global warming will accrue to generations living after 2800!”
Commonly, the opportunity cost – the risk-adjusted rate of return which can be obtained on alternative investments – is used as the discount rate by governments and businesses. Lilley says that the US government uses 7% pa, and until 2003 the UK government rate was 6%. 
Treatment of uncertainty and risks [59-78]
I have mentioned “risk-adjusted rates of return.” Lilley points out that “There is an additional reason for discounting the future … to account for uncertainty. The analysis discussed in the previous chapter, from which Stern derives his discount rate, tacitly assumes that we have perfect foresight of the future.” I have long argued that the only certainty is change, and that the future will bring unforeseen significant events – “Black Swans.” There are no Black Swans for Stern.
“In the business world it is common to use a higher discount rate the greater the uncertainty about the future. Stern argues for the reverse.” The reasons for a higher rate are that “the less well we can see the future, the less we can meaningfully say about it. Using a higher discount rate shortens the time span over which our guesstimates of the future have any meaningful present value. … [and] because they assume that, for any investment project, unforeseeable events are more likely to make it less profitable than to boost its profitability. … the very needs which make the investment appear profitable are likely to attract new competitors, stimulate human ingenuity and call forth as yet unforeseeable alternatives.” [59-60]
To me, this is the way the world works, it has been that way throughout history, and Stern’s view seems absurd.
Lilley looks at discounting, uncertainty and risk in some depth. He notes that “The Review tackles uncertainty by weighting all possible outcomes by their probability and then discounting them. But it uses a lower discount rate the less well off the outcome leaves us. The effect is to reduce the average discount rate the wider the range of outcomes. The Review assumes that uncertainty consists of a wider dispersion of possible outcomes the further we look into the future. The greater the dispersion, the lower the weighted average discount rate. Hence the lower discount rate over time.” Stern assumes “that the future will be like what we already know – except for ever wider uncertainty about the value of a number of key variables.” 
I don’t believe that there is any time in human history that a prediction of the world 90-100 ahead would have been accurate. Whatever tools we use to predict, there are too many uncertainties, too many Black Swans. What we do know, unlike Stern, is that it will not be what we would expect based on projections of present trends continuing.
I’ll dismiss Stern at this point, but Lilley also addresses issues such as the range of values for climate sensitivity and possible catastrophic climate change impacts. [61-66]
Key recommendations [15, 92-93]
In considering policy, Lilley says that the UK’s “political parties have measured their virtue by the austerity of the targets they sign up to, rather than the benefits their citizens’ sacrifices will bring, adopted a hodge-podge of individually fashionable policies regardless of cost or coherence, and – rather than pursuing evidence-based policy – they have relied on ‘policy based evidence’ like the Stern review. … [there] has been an almost heroic disregard of costs. It should not need saying – if the costs of a proposed strategy exceed the benefits one should amend the strategy or seek another. Unfortunately it does need saying.” 
Readers from many other countries will see clear parallels. While Lilley’s key recommendations are for the UK government, they have broader applicability, and similar points have been argued in Australia. Lilley recommends that:
“The government should cease to rely on the flawed Stern Review to justify policy and should commission a new, independent Review.
“The government should prescribe the same discount rate for assessing costs and benefits of climate policies as it uses for all long-term public projects or explain fully why it is not so doing – and show the sensitivity to alternative plausible discount rates and the Internal Rate of Return of alternative pathways for tackling global warming.
“The Review should assess the costs and benefits for scenarios with varying degrees of international cooperation. Meanwhile, Parliament should remove the legal requirement on the UK to act unilaterally.” 
On the last point, Lilley notes that the annual increase in China’s emissions exceeds the UK’s annual emissions, which are only 2% of the global total.  In effect, he echoes the argument of many at CE that current policies have been made without an adequate basis, and that a far more solid basis is needed to determine what policies might be worthwhile.
In the absence of a new review, Lilley says that “government strategy should at most involve:
- gradually ramping up incentives to reduce carbon emissions
- cost effective measures to increase energy efficiency
- greater focus on incentivising Research and Development
- acceptance that developing countries need to develop the cheapest energy sources available to them
- more emphasis on adaptation to climate change as it occurs
- focussing development aid on helping vulnerable countries adapt to climate change, whatever its cause” [15, 92]
I find it impossible to take the Stern Review seriously, given that its findings and advice are based on predictions of how the next 200 years and more will unfold. I don’t believe that there is any time in human history that a prediction of the world 90-100 years ahead would have been accurate. Whatever tools we use to predict, there are too many uncertainties, too many Black Swans. Given that, the Review seems to have done its best to exaggerate the cost of non-action and minimise the costs of taking action, and its estimates have little credence.
There have been many more technical critiques of Stern’s work, several of which Lilley cites. Yet Stern’s Review is still taken as gospel by many decision-makers and advisers. The significance of Lilley’s paper is two-fold: the author has credibility with policy-makers, given that he is a former government economics minister who is well-qualified to assess the Review; and it is written so as to be accessible to policy-makers and others with little or no technical expertise. It will therefore have more chance than technical academic papers of influencing policy.
To me, that would be a good thing. Taking the science as given by the IPCC, Lilley’s views are generally similar to those I’ve expressed at Climate Etc and elsewhere. I’ve also argued for almost 20 years that if it were necessary to reduce emissions to counter global warming, then it should be done in the most cost-effective way – both to minimise the negative impacts and to increase the chance of popular support. Lilley makes good arguments – as have others for some years – for minimising costs now and encouraging R&D aimed at lower-cost solutions later. [His rationale for a gradual approach is in pp 84-91.]
Many proponents of strong action to reduce emissions appear to have little awareness of or concern for the economic and social costs of many proposed policies, and Lilley’s paper is a useful contribution to a more balanced and informed debate. Its impact in the UK might be enhanced because its release coincided with a Cabinet reshuffle which favoured pro-growth over greener ministers.
Of course, many posters on Climate Etc do not take the IPCC’s science as given, and would favour an independent review of the scientific validity of alleged AGW. But that is an argument for other threads.
Biosketch: Michael Cunningham is a former economic policy adviser to the UK, Australian and Queensland governments.
JC note: I invited Faustino to do a guest post on this topic, based upon his comments in the thread Activate your science. As with all guest posts, this reflects the views of the author only.
UPDATE: email from Chris Hope:
I notice that this blog post has attracted a large number of comments
overnight. Unfortunately one of the main quotes in it is wrong, as I
show in my blog
; if you look in the comments to my blog you will see that Peter
Lilley has accepted the quote is wrong and will correct it. I was in
a position to point out the error becuase it relates to my model, PAGE2002.
Dr Chris Hope
Judge Business School
University of Cambridge
Tel: 01223 338194