Jump to main content.

Correction on Global Warming Cost Benefit Conflict

Quick Links




powered by Google

Volume: III # 1 January, 1996
Source Reports:
Chakravarthi Raghavan, Chief Editor of SUNS (South-North Development Monitor and Geneva representative of the Third World Network)

David Pearce, CSERGE, University College London and IPCC III Lead Author
1. Benefits Analysis
2. Cost-Benefit/Cost Effectiveness Analysis
3. Cost and Economic Impacts Analyses
1. Benefits Analysis - Valuation
Environmental Media/Problem:
a. Air
a. Air - Mobile Source
a. Air - Stationary Source
a. Air - Stratospheric
Newsletter Article Text:
A short article on page 11 of the November 1995 EDV&CBN failed to state that the information was based solely on claims from the South-North Development Monitor. The information was presented merely to illustrate a particular position, not as a statement of fact. Below we have reprinted the original article, and a response from David Pearce, CSERGE, University College London and IPCC III Lead Author. We apologize for the oversight regarding what may be the most important damage valuation work in the world. We thank Mr. Pearce for taking the time to respond, and will attempt to obtain a rebuttal in a future issue.

A report which concluded that lives in the developed "North" were 15 times more valuable than lives in the less developed "South" was rejected by several Less Developed Countries, including India, China, Cuba, Brazil and Peru at a recent meeting of the Working Group 3 of the Intergovernmental Council on Climate Change (IPCC) on July 28, 1995. The economists were told to go back and do their work again.

The group is charged with comparing the costs of curbing Green House Gas emissions with the losses that would result from global warming. The result showed that the former would probably be greater than 2% of Gross World Product, while the latter would amount to 1.5% - 2.0% of Gross World Product. Valuing all deaths at the rich country value of $1.5 million per death would raise total damages.

Chakravarthi Raghavan, Chief Editor of SUNS (South-North Development Monitor and Geneva representative of the Third World Network, 228 Macalister Road, 10400 Penang, Malaysia twn@igc.apc.org (+604) 229-3511,2,3 Fax (+604) 229-8106 from ELAN Environment In Latin America internet list October 10, 1995 16:20:09.73


Your November 1995 issue carried an erroneous report on the conflict over the 'valuation of life' in IPCC III. The relevant chapter (chapter 6) reports values of statistical life based on actual studies in different countries. Whether the values used remain as in Chapter 6 or whether a common global average is used makes no difference to the results. What the authors of Chapter 6 did not accept, and still do not accept, was the call from a few delegates for a common valuation based on the highest number for willingness to pay. The attached notes explain one of the reasons why. There are others. You report that 'the economists were told to go back and do their work again': no such request was made nor could it have been honoured anyway since (a) the chapter had already undergone two peer reviews, (b) IPCC delegates (mainly political appointments) have no power to request authors to change already accepted chapters, and (c) ownership of the chapters rests with the authors, not the IPCC.


Critics of IPCC's Working Group III (WGIII) Assessment have focused on Chapter 6 which concerns the monetary estimates of the benefits of global warming control. The apparent belief is that the general tone of WGIII casts doubt on whether it is worthwhile to make significant cuts in greenhouse emissions over and above those that were agreed for the developed economies under Climate Change Convention (and which few countries will actually achieve it seems). This has fostered a search for bigger damages than those contained in Chapter 6 and a campaign to discredit the authors of Chapter 6 for allegedly, playing down the benefits. But the campaigns and the criticisms reflect profound confusions, so much so that the effect is likely to be one of lowering the significance attached to global warming rather than raising it. The critics have effectively shot themselves in both feet.

There are two relevant documents: the full report of around 1000 pages and containing all the chapters, and the 'Policy Makers' Summary' (PMS) of about 30 pages, supposedly summarizing the chapters.

The July 1995 version of the IPCC Working Group III Policy Makers' Summary refers to damages of 1-2% of GDP for developed economies and 2-9% for the developing countries. The overall
'net damages' are put at 1.5-2.0% of GDP. It refers to marginal damage estimates of $5-125 per tc for carbon emitted now, i.e. the damage done by an extra tonne of carbon. These numbers are taken from Chapter 6. The PMS makes no reference to the benefits of control.

The October 1995 version (after Montreal) deletes the numbers for 'per GDP' damage and refers to them as being 'a few percent of world GDP'. It retains the $5-125tc marginal damage estimates. Again, no reference is made to benefits.

In the July 1995 PMS costs of control 'may ultimately exceed 1-2% of GDP' for the OECD countries, who would be the ones having to pay, initially anyway.

In October 1995 the 'costs of substantial reductions below 1990 levels [of emissions] could be as high as several per cent of GDP'. The range of annual costs for stabilizing emissions at 1990 levels is minus 0.5% to plus 2% of GDP at some point in 'the next several decades'.

Why were the damage numbers deleted in Montreal? Perhaps to avoid the very confusion that both the IPCC and the critics had generated for themselves, i.e. the comparison of costs of control 'ultimately exceeding 1-2%' (elsewhere in the PMS) and the 1.5-2% of damages in the July version (from Chapter 6). But that comparison was false anyway, as shown shortly. It should be noted that Chapter 6 is not concerned with the costs of control at all.

In the final PMS, if the two sections are put together, the comparison becomes one of damages of a 'few per cent of GDP' with costs of 'as high as several per cent of GDP'. This is hardly an improvement for the policy maker. Reading the July version a policy maker might have been misled into comparing apparent costs in excess of 2% with benefits of up to 2%, i.e. benefits appeared to be less than (or roughly equal to) costs. In the October version the invited comparison looks at least as unfavourable, if not worse for global warming control. This cannot be what the drafters intended unless, of course, those who wanted such confusion to emerge had got their way. (The authors of Chapter 6 disassociated themselves from the PMS in the Montreal meeting, and continue to do so).

Why can't costs and benefits be compared? They can, but it takes a little analysis to get there.

First, the '%GDP' figures were never comparable anyway. This point is complex and it is easy to see why some readers were confused. Yet Chapter 6 discusses the reasons why the comparison is misleading. The damage figures are for 2 x CO2, i.e. they relate to some year in the future at which time the atmospheric concentrations of CO2 have doubled. The 2 x CO2 figure is a benchmark, nothing else. For if global warming goes unchecked, concentrations will more than double and damage will accordingly get bigger than the levels associated with 2 x CO2. Suppose 2 x CO2 occurs in 2050. Then annual damages before 2050 will be less than 1.5-2% of GDP and damages after 2050 will be more than this. The costs of control figures (the minus 0.5 to 2%) relate to emissions stabilization, not to avoiding 2 x CO2. The 'several per cent' relates to an unspecified 'substantial reduction' below the 1990 levels. The damage figures for 2 x CO2 also have to be interpreted carefully for they relate to the damage from 2 x CO2 as if it occurred now and related to the world economy as it is now. And just to make things worse, even if benefits are the same as avoided damages, all the damages in any year cannot be reduced because of the lags in the system: if global warming is for real, as the IPCC scientists now say, it is too late to do more than reduce some of the damage.

Second, even if the various percentages were comparable, they are total damages and costs. From a comparison of total damages and costs we cannot conclude that nothing should be done, for there may be some level of action less than that addressed by these cost estimates where benefits exceed costs. The proper comparison should be between the marginal costs and benefits, i.e. the cost and benefit of reducing an extra tonne of carbon. If the cost-benefit paradigm is pursued (and no-one believes it is the only way of making comparisons), warming controls should be pursued as long as the marginal cost of abatement is less than the marginal benefit of abatement.

Third, what should be compared is costs and benefits, not costs and avoided damages. Benefits exceed avoided damages because warming control policies yield other benefits besides climate change. Thus, control of CO2 from traffic produces benefits in the form of reduced air pollution, noise, congestion, accidents etc. This point is made crystal clear in Chapter 6. The summary of Chapter 6 does not mention it.

Any policy maker reading the PMS only will get no sense of these confusions because they are simply not addressed in the PMS. One cannot lay the blame for this degree of confusion at the door of the critics alone, although some less obsessive behaviour on their part about trying to inflate the damage figures might have opened their eyes to the massive potential for misinterpretation. A significant part of the problem just 'grew' because of the procedure of using so many different authors to address a very complex problem at such length. The wood got lost because of the trees.

Can anything be rescued? One way to reclaim the need for aggressive action is to compare marginal benefits and costs. Chapter 6 gives estimates of marginal damages ranging from $5-125tc and this range is retained in the PMS. Suppose a fairly low figure of $20 tc is taken as the estimate. This is still not the marginal benefit of control. To get benefits we need to add the 'secondary' benefits of carbon control, and these come to perhaps to another $20 tc, taking a low figure from the ranges reported in Chapter 6. So we have a fairly conservative $40 tc of benefits from preventing one tonne of carbon.

We now need the marginal costs of control. Many of the available estimates are recorded in Chapter 9 of IPCC III, although not always in convenient form. But one thing is clear, there are many, many investments that can be made in carbon sequestration, fuel switching and energy conservation a long way below $40 tc. Even without the inclusion of secondary benefits, there are many options below $20 tc.

Why didn't IPCC say so?

IPCC cannot in fact make policy recommendations. It can only record the 'state of the art' and leave it up to policy makers to infer the policy implications. A careful reading of IPCC III will therefore produce the very result that the critics wanted in so far as they want the cost-benefit implications to favour positive action. But saying benefits exceed costs is tantamount to a policy recommendation and so, if IPCC recognized the inference, they could not make it explicit. Unfortunately, the PMS never even hinted at the inference and, indeed, actually produced the impression that the opposite is the case. The final version of the PMS is no improvement and is arguably worse. The mistake made by the critics was to use up their energy in a misleading campaign to make the damages bigger than Chapter 6 makes them appear to be. Their optional strategy should have been to comb the report for the evidence on the existence of net benefits of control, assemble it and distribute it since they can make appeals for policy where IPCC cannot. Few environmental campaigns have been more ill conceived, and few more counterproductive.

David Pearce, CSERGE, University College London and IPCC III Lead Author e-mail January 3, 1996

Local Navigation

Jump to main content.