The Economics of Avoiding Dangerous Climate Change. An Editorial Essay on The Stern Review

Abstract

The problem of avoiding dangerous climate change requires analysis from many disciplines. Mainstream economic thinking about
the problem has shifted with the Stern Review from a single-discipline focus on cost-benefit analysis to a new inter-disciplinary
and multi-disciplinary risk analysis, already evident in the IPCC Third Assessment Report. This shift is more evidence of
the failure of the traditional, equilibrium approach in general to provide an adequate understanding of observed behaviour,
either at the micro or macro scale. The economics of the Stern Review has been accepted by governments and the public as mainstream
economic thinking on climate change, when in some critical respects it represents a radical departure from the traditional
treatment. The conclusions regarding economic policy for climate change have shifted from “do little, later” to “take strong
action urgently, before it is too late”. This editorial sets out four issues of critical importance to the new conclusions
about avoiding dangerous climate change, each of which have been either ignored by the traditional literature or treated in
a misleading way that discounts the insights from other disciplines: the complexity of the global energy-economy system (including
the poverty and sustainability aspects of development), the ethics of intergenerational equity, the understanding from engineering
and history about path dependence and induced technological change, and finally the politics of climate policy.

Full text

EDITORIAL
The economics of avoiding dangerous climate change.
An editorial essay on The Stern Review
Terry Barker
Received: 9 February 2008 /Accepted: 16 April 2008
#Springer Science + Business Media B.V. 2008
Abstract The problem of avoiding dangerous climate change requires analysis from many
disciplines. Mainstream economic thinking about the problem has shifted with the Stern
Review from a single-discipline focus on cost-benefit analysis to a new inter-disciplinary
and multi-disciplinary risk analysis, already evident in the IPCC Third Assessment Report.
This shift is more evidence of the failure of the traditional, equilibrium approach in general
to provide an adequate understanding of observed behaviour, either at the micro or macro
scale. The economics of the Stern Review has been accepted by governments and the public
as mainstream economic thinking on climate change, when in some critical respects it
represents a radical departure from the traditional treatment. The conclusions regarding
economic policy for climate change have shifted from “do little, later”to “take strong
action urgently, before it is too late”. This editorial sets out four issues of critical importance
to the new conclusions about avoiding dangerous climate change, each of which have been
either ignored by the traditional literature or treated in a misleading way that discounts the
insights from other disciplines: the complexity of the global energy-economy system
(including the poverty and sustainability aspects of development), the ethics of
intergenerational equity, the understanding from engineering and history about path
dependence and induced technological change, and finally the politics of climate policy.
1 Introduction
Since the early 1990s, it has been internationally recognized that one consequence of
economic activity has been the accumulation of greenhouse gases and that this may lead to
climate change. This already threatens development in poor countries that are most
vulnerable to climate variability. If unchecked, it will threaten future generations with
unknown but potentially catastrophic climate events, given the availability of fossil carbon
at current prices relative to carbon-free alternatives, which could raise concentrations to
levels not seen for millions of years. At the same time the costs of reducing the emissions
Climatic Change
DOI 10.1007/s10584-008-9433-x
T. Barker (*)
Department of Land Economy, University of Cambridge, 18 Silver Street, Cambridge CB3 9EP, UK
e-mail: tsb1@cam.ac.uk

have been agreed as negligible in relation to expected growth in incomes. Yet, after
17 years, global action has been limited and emissions have continued to grow. Indeed, the
political recognition of the urgency of the problem has only become evident at a global
scale since the publication of the Stern Review of the economics of climate change in 2006.
One reason for this delay has been the response by a clique of economists, dating from
the publication of the first IPCC Report in 1990, to the concern and calls for action by the
climate scientists. The response came in the form of the mis-application of a tool developed
by equilibrium economists for prescriptive public policy: the cost-benefit analysis. This
editorial addresses the question of how and why the focus of the economics of climate
change has shifted from the single-discipline cost-benefit analysis, as in the IPCC Second
Assessment Report in 1995, to the multi-disciplinary uncertainty analyses in the subsequent
IPCC Reports of 2001 and 2007 and the radically different policy prescriptions of the 2006
Stern Review.
The application of traditional cost-benefit analysis has yielded, with some exceptions,
policy prescriptions of insignificant carbon taxes and delayed action until more information
is available on the problem and more R&D has been done to lower the costs of any
response. The new mainstream uncertainty analysis, in contrast, suggests that a political
global decision needs to be taken urgently on targets to avoid dangerous climate change and
that cost-effective and equitable policies and measures should be implemented strongly
without delay to accelerate progress towards a complete decarbonisation of the world
economy.
Defining dangerous in this context is a social and political task. Implicitly some progress
has been made with the European governments’2°C target for average long-term global
temperatures above pre-industrial levels, the G8’s 50% reduction in global GHG emissions
below 1990 levels by 2050 as agreed by the meeting in Heiligendamm, Germany, June
2007, and the “deep cuts”of the UNFCCC Bali Action Plan of December 2007. The AR4
WG3 Summary for Policymakers gives an indication of just how deep the cuts will have to
be. It presents six scenarios from the literature on the scale of action required. For a chance
less than 50:50 that the target will be met, the scenarios suggest that global CO
2
emissions
will have to be between 50% to 85% below 2000 levels by 2050, and becoming negative
(through sequestration and storage) by 2070 and beyond. Therefore to be reasonably sure of
avoiding dangerous climate change defined as a 2°C rise or less, the world should be
aiming for complete decarbonisation by 2050 or earlier, to be safer. All sectors in all
countries should be aiming to stop emitting GHG into the atmosphere as soon as possible
without excessive cost
1
. Without urgent action, the risks of losing coral reefs and pristine
tropical rainforests appear significant, e.g. for a rise above 1°C to 3°C, the AR4 WG2
Summary has “very high confidence”of widespread mortality of coral reefs.
In this editorial I argue that this shift in thinking is more evidence of the failure of
equilibrium economics in general to provide an adequate and coherent explanation of
observed behaviour, from the personal (micro) to the global (macro) scales, with the aim
of guiding policy. Equilibrium economics underlies most textbooks of economics and
journals of economic theory. It is theory largely unsupported by formal scientific observation
and empirical data. Over the past 50 years, it has been increasingly recognised as dependent
1
The carbon price associated with this target is not the topic of this paper, but it is of great interest. Without
sufficient literature to assess such a price, it is necessary to extrapolate from the AR4 literature on less
stringent targets (Barker et al. 2007, pp. 636, 648 and 659–661). A precautionary, provisional estimate for a
global price for the 2°C target appears to start in the range $US80 to $US100/tCO
2
by 2020 and earlier if
possible, rising after 2020 and in year 2000 prices.
Climatic Change

on false assumptions about human behaviour and physical systems, and as based on a rigid
and ill-informed interpretation of utilitarian ethics. The continued use of the assumptions,
most pertinently in the application of cost-benefit analysis and computable general
equilibrium models for climate policy, strongly suggests that their only justification is that
they are required for the mathematics and computation to be tractable. Any empirical support
for the theory has been generally incoherent, ad hoc and rhetorical, with the most outstanding
example the fact that the multi-sectoral equilibrium modelling of GHG mitigation policy
over the century, which dominates the literature, is based typically on one year’s data (and
this is simply to calibrate the model to yield results of the right magnitude, rather than to
provide empirical validation of the results).
In contrast, the Stern Review considers cost-benefit analysis as a marginal analysis
inappropriately applied to a non-marginal multi-disciplinary systemic problem (p. 50). Both
Stern (p. 163) and the IPCC Reports after 1995 take a multi-criteria approach rather than
a narrowly monetary one and question cost-benefit analysis. This is one reason for the
intemperate response from some traditional economists to the Stern Review
2
. The effect of
the more general approach is to criticise, qualify, and generally undermine such equilibrium
thinking in the assessment of the costs and benefits of climate change and its mitigation.
This is not new
3
: government rejection of the cost-benefit valuation of human lives goes
back to the IPCC Second Assessment Report. What is new is that the economics of the
Stern Review has been accepted by governments and the public as mainstream and
consensus economic thinking, when in critical respects it represents a radical departure
from a traditional deterministic treatment and its messages for economic policy.
This editorial takes an interdisciplinary approach to the economics of dangerous climate
change, contrasting the cost-benefit analysis (CBA) with the new economics of risk that
acknowledges and respects the insights and analysis from other disciplines, namely the
other social sciences, climate and geographical sciences, ethics, history, engineering as well
as complexity and evolutionary theory
4
. This editorial introduces the other papers in this
issue on the economics of climate change following the publications in 2006 of the Stern
Review, and in 2007 of the critiques of the Review and the IPCC Fourth Assessment Report
(AR4). The paper by Quiggin consolidates the arguments relating to Stern’s use of the
discount rate, claimed to be the main reason for differences between the conclusions of
Stern “take strong action urgently”and those of the critics. The paper by Jaeger,
Schellnhuber and Brovkin promotes the case of other values than purely economic ones in
social choices for the future climate system. The paper by Hasselmann and Barker proposes
a new international body to develop data and modelling for mitigation policies to avoid
2
Tol and Yohe (2007, p. 233–4) accuse Stern of “substandard analysis”and “dubious economics”; Nordhaus
(2007a, p. 5) claims that Stern is “political”and the discount rates used are “extreme”(p. 6); Dasgupta (2007,
2008, p. 7) accuses Stern of choosing parameter values so “that they yield desired answers.”Such charges are
perhaps more appropriately made against neoclassical economics as in (Mirowski 1989,2002; Barker 1996b,
1998; Nelson 2001; DeCanio 2003; Ackerman and Nadal 2004; Foley 2006; Nelson 2006; Beinhocker
2006). The charges also contrast with the praise by leading mainstream economists and Nobel Laureates
quoted in the Review, such as Mirrlees, Sen, Stiglitz, and Solow, and support after publication from Arrow,
DeLong and Deaton among others. There are far too many references to include in the space permitted, so I
have only included examples of the mainstream literature and the last or later references in the new literature.
3
Previous academic critiques of the optimizing (Barker 1996a,b) and discounting (Hasselmann 1999)
approaches of traditional cost-benefit analysis have had no discernible effect on the methodology or
conclusions.
4
Other disciplines are just as important, including the humanities, but I am restricting the discussion to a few
issues. In particular, I do not consider in any depth the analysis of development, which is crucial for
adaptation to climate change.
Climatic Change

dangerous climate change, based on risk analysis and a systematic treatment of global
energy-environment-economy time-series data.
This editorial sets out four issues of critical importance to climate policy, each of which
have been either ignored by traditional economic modelling of the problem, or treated in a
misleading way that discounts the insights from heterodox economics and other disciplines.
1. The global economy, with its use of energy and emissions of GHGs, is a complex, non-
linear dynamic system with technological change inherent in economic growth.
Climate effects and responses to climate policy are potentially non-marginal changes to
the system in the context of strong uncertainty.
2. The problem of intergeneration equity from climate change and mitigation is primarily
an ethical one, and should be informed by moral philosophy rather than economics in
isolation. Traditional economic models adopt an extreme form of utilitarianism, with a
questionable choice and use of discount rates, ignoring the philosophical literature and
the concept of justice.
3. Engineering and history inform GHG mitigation policy through studies of the
production processes involving the supply and demand of energy, in particular the
technical possibilities of accelerated decarbonisation of the energy system. Economic
history is critical in understanding the relationship between mitigation policies,
economic development and technological change because the time scales are long in
relation to the life-times of most energy-using capital, so that the technologies involved
can respond to carbon-price signals over the policy period. Traditional models assume
continuity and path independence.
4. The politics of mitigation implies unstable alliances and trade-offs between governments
and political parties. By the use of the social welfare function (required for the calculus),
traditional economists simplify social choices and pre-empt political negotiation,
claiming an optimality for their subjective assumptions and market interpretations.
This is an ambitious agenda and the issues can only be summarised, relying on the
underlying literature. The implications for new economic thinking are often unclear,
provisional and tentative in the face of the certainties of some cherished traditional beliefs.
This editorial covers the issues one by one, but first the economics needs to be further
explained and the Stern Review and IPCC Reports interpreted in the light of the shift in
economic thinking.
2“Traditional”and “new”economics approach to climate change
Understanding and solving the climate problem requires the synthesis and co-development
of many disciplines with different traditions and links between them. Economics is relevant
because it explains why human behaviour might lead to climate change (via economic
choices and the use of the atmosphere for free waste disposal) and it provides systematic
methods for assessing (and monetizing) costs and benefits of different activities and
policies. However, it is important to distinguish between a general definition of economics
5
5
Economics is the study of social activity undertaken with its primary purpose the expectation of reward,
which usually involves money, the motivations of such activity and its consequences both good and bad, e.g.
for equity and the environment. In contrast the neoclassical economist Robbins (1932) defined economics as
“the science which studies human behaviour as a relationship between ends and scarce means which have
alternative uses”(p. 16), asserting that economics is a value-free science.
Climatic Change

and the particular approach used (pre-Stern) in most of the literature on the economics of
climate change. Following Beinhocker (2006) and Maréchal (2007) I shall call this
particular approach “traditional”economics defined as
…the set of concepts and theories articulated in…textbooks. It also includes concepts
and theories that peer-reviewed surveys claim, or assume, that the field generally
agrees on. (Beinhocker 2006, p. 24)
6
.
This traditional economics is epitomized by Samuelson’sEconomics, now co-authored
by Nordhaus
7
, and based on the neoclassical mathematical synthesis promoted by
Samuelson that came to dominate mainstream economics thinking in the late twentieth
century. I shall contrast the traditional economics with a “new”economics, as in the title of
Boulding’s1992 book, including complexity, evolutionary and Post Keynesian theory and
emphasising institutions, non-linear dynamics, and deep uncertainty.
Neoclassical economics is defined as being characterized by an emphasis on rationality,
via the use of utility maximization, an emphasis on equilibrium, and by neglect of strong
kinds of uncertainty, particularly of fundamental uncertainty (Dequech 2007, p. 290). The
traditional economic approach to modelling climate change has been almost exclusively
and narrowly neoclassical, adopting a version of expected utility theory with human welfare
usually translating into private market consumption per head in the applied models. The
theory is applied to utility across countries with huge differences in consumption and over
100 years or more into the future, when consumption can rise perhaps 20 or 30 times over.
The formal approach to the problem has been cost-benefit analysis beginning in the 1990s
(e.g. Cline 1991,1992; Nordhaus 1991a,b,1994,2007b; Nordhaus and Boyer 2000). In
simplified terms, the costs of climate change are set against the benefits of taking action in a
way that allows comparison with other potential uses of public revenues. The price of
carbon (the CBA literature is focused on the specific “social cost of carbon”) is calculated
6
It has become debatable whether neoclassical equilibrium economics is mainstream anymore (Colander et
al. 2004; Dequech 2007). This paper argues that Stern has shifted the mainstream away from the traditional
neoclassical approach to climate-change economics. See (Maréchal 2007) for a supporting view. For a more
general view see (Hodgson 2007). Prominent economists are acknowledging that for macroeconomic growth
“The right way to think about this complex set of issues is not clear, but it is clear that the competitive
paradigm cannot be fully appropriate.”(DeLong and Summers 2001). Arrow (2007) accepts the Stern
estimates of costs and benefits, quoting the range of GDP mitigation costs (3.4% to −3.9%) from the meta-
analysis in (Barker et al. 2006). Akerlof (2007), the 2007 President of the American Economic Association,
exposes the weaknesses of neoclassical macroeconomics and suggests that early Keynesian macroeconomic
theory has more explanatory power. Deaton (2007), the 2008 President of the AEA, supports the Stern
treatment of discount rates: “Whatever it is that is generating market behaviour, it is not the outcome of an
infinitely lived and infinitely far-sighted representative agent whose market and moral behaviours are perfectly
aligned, and who we can use as some sort of infallible guide to our own decisions and policies.”(p. 4)
7
Samuelson and Nordhaus (2001) is the 17th edition of a textbook originally published by Samuelson in
1948. Yohe (1989), another contributor to the traditional literature, has published a study guide to this
textbook. Nordhaus is taken as the exemplar in this editorial (rather than a straw neoclassical man) because of
his economics, methods and distinction in the field. He provides full details of equations, exposing all his
assumptions. His approach is followed by many others and his models are widely used for climate policy
analyses. It is characteristic of his work, and of his school, that qualifications to the economics or results (e.g.
extreme events, recycling of carbon tax revenues or induced technological change) are introduced
independently, but subsequently there is usually a reversion to the core model even when the qualifications
produce radical effects on the results (Nordhaus 1993,2007b). Other problematic features one would expect
in a CBA, e.g. environmental co-benefits of GHG mitigation, are ignored. Eventually, if the qualification
becomes mainstream, it is accommodated in the model or analysis (Mirowski 1989,2002).
Climatic Change

in a supply-demand framework, such that the costs of climate change arising from the
marginal addition of CO
2
into the atmosphere are matched in equilibrium with the benefits
of not making the addition, i.e. the mitigation or marginal abatement, giving the “maximum
benefit”for humanity, discounted over all future generations. The treatment of uncertainty
and risk has been to ignore deep uncertainty and convert the asymmetric risks of long-term
irreversible damages into certainty-equivalent damages, which are then discounted when
compared to short-term costs of mitigation. The outcome is an “optimal”price of carbon,
an indefinite “optimal”rise in global temperatures, and very modest prescriptions for
action.
This method rests on the idea that individual preferences are fixed and utilities can be
aggregated and converted into well-behaved mathematical equations in a “social welfare
function”, and differentiated to give stable marginal properties, as the basis for climate
policy. It also crucially assumes that all natural services can be converted to money and
back again at any time, i.e. that there are no irreversible effects of climate change. These
and other assumptions render CBA inadequate and misleading for climate policy analysis
(Azar and Lindgren 2003; van den Bergh 2004; Ackerman and Heinzerling 2004). In
summary, CBA “does not yield transparent or objective evaluations of benefits; rather, it
renders the discussion of benefits obscurely technical, excluding all but specialists from
participation. At the same time, political debate continues behind the veil of technicalities,
as rival experts battle over esoteric valuation problems.”(Ackerman 2004). The approach
(Arrow et al. 1996; Pearce et al. 1996) became highly controversial in the IPCC Second
Assessment Report, when the governments rejected an assumption used in the analysis that
the costs of human life should be different for different countries, depending on their
income levels. Since the CBA literature is voluminous, it is covered in subsequent IPCC
reports, but any implications for policy are heavily qualified and extensive lists are given of
damages that are not or cannot be monetised (IPCC 2007, Table SPM3).
However CBA continues to be the theoretical basis for those advocating the traditional
climate policies. A leading example of the post-Stern CBA approach to the climate problem
is Nordhaus (2007b), who concludes that “the Gore and Stern proposals…are more costly
than [doing] nothing.”(p. 177) A peer-reviewed survey of the costs of “doing nothing”in
the CBA is to be found in (Tol 2005), which covers 80 peer-reviewed estimates, i.e.
fulfilling Beinhocker’s definition of traditional economics. Tol concludes “One can
therefore safely say that, for all practical purposes, climate change impacts may be very
uncertain but [it] is unlikely that the marginal damage costs of carbon dioxide emissions
exceed [$14/tCO
2
]
8
and [they] are likely to be substantially smaller than that.”(p. 2073,
italics added). An update (Tol 2007) covering 125 such studies confirms the median
estimate of $4/tCO
2
. For comparison, a carbon tax of $5/tCO
2
in 2005 dollars (arguably
close to the starting rates of the “optimal”tax from this mainly 1990s literature before any
extra allowance for risks of dangerous climate change was introduced in response to the
issues raised by the Stern Review) converts to an increase of $2.5/bbl in US crude oil prices
8
To avoid confusion and conform to IPCC AR4 WG3 usage, all estimates have been converted to $/tCO
2
in
this editorial. The price base is given where known. Despite their apparent precision, Tol’s estimates cover
different base years and price bases, because he does not appear to have converted the estimates to a
consistent basis. They are all in real terms in his original sources, but typically adopt different base years for
the discounting and prices. This lack of comparability in the underlying data renders his averages
undetermined, a problem not mentioned by Tol.
Climatic Change

in 2005, or about 2% in US gasoline prices. Such numbers from this established body of
literature appear unbelievably small if they are taken to be, as intended over the last
16 years, the estimated “optimal”costs of risking climate catastrophe
9
.
In summary, the key policy messages, extremely stylized
10
, from traditional economics
have been: (1) introduce a small carbon tax, rising indefinitely in real terms and, if
technological change is considered, (2) wait rather than act now because the costs of
mitigation will fall as a result of technological change. These policy prescriptions have
come from an approach relying on deterministic cost-benefit analysis and high discount
rates, assuming that technological change is exogenous to the economic system or not
affected by the carbon price, and ignoring deep uncertainty.
Contrast the tone and implications of these statements with Stern’s conclusions: “the
benefits of strong early action far outweigh the economic costs of not acting”and “even at
moderate levels of warming, all the evidence…shows that climate change will have serious
impacts on world output, on human life and on the environment.”(pp. xv and xv1). Stern
acknowledges that technological change is partly driven by economic factors, such as the
price of carbon, implying that the benefits from waiting are replaced by benefits from acting
so as to induce the change and reduce the future costs (Köhler et al. 2006). Stern
commissioned an application of the PAGE model (Hope 2003), which computes a
probabilistic risk-based CBA, and considered the CBA approach as one of three lines of
evidence from the literature
11
. Stern asserts that the economics of climate change are now
more appropriately concerned with risk rather than return, and with the development of
technologies for mitigation, both features of the problem that has been evident from the
early 1990s, when the scientific assessments began in earnest. This in turn implies that the
economic problem is one of achieving political targets and lowest costs compatible with
equity and effectiveness, rather than with the political and scientific problem of choosing
the targets themselves.
Contrast also the pre-Stern traditional conclusion that “optimal”temperatures should rise
indefinitely (Nordhaus 1994, p. 89; 1997, p. 324)
12
with the potential impacts of such
10
Dasgupta (2007, p. 5) summarises the traditional view (incorrectly but reflecting academic and political
perceptions) as: “Nordhaus...has been studying the economics of climate change for over three decades. The
most remarkable conclusion of his studies—conducted on his Dynamic Integrated Model of Climate and the
Economy (DICE)—has been that, despite the serious threats to the global economy posed by climate change,
little should be done to reduce carbon emissions in the near future; that controls on carbon should be put into
effect in an increasing, but gradual manner, starting several decades from now. This conclusion has withstood
the many modifications Nordhaus and others have made to the climate science embodied in DICE. Their idea
is not that climate change should not be taken seriously, but that it would be more equitable (and efficient) to
invest in physical and human capital now, so as to build up the productive base of economies (including,
especially, poor countries), and divert funds to meet the problems of climate change at a later date.”
Nordhaus (2007b p. 237) in fact has carbon taxes starting in 2005. Others using DICE (see note 13), but
altering the assumptions, come up with Stern-like conclusions.
11
It is significant that his critics have mainly criticized technical assumptions about discounting in a CBA
and have asserted that he has based his conclusions on this CBA. In fact the results of the CBA are to be
viewed as “indicative only”and to be “interpreted with great caution”(p. 174) and their discussion takes up
only 16 out of nearly 700 pages of the Review.
9
They are also five times smaller that the carbon prices revealed by the market. Phase 2 of the EU emission
trading scheme has a carbon price over $20/tCO
2
(March 2008), arguably to achieve a much weaker internal
EU target than that implied as optimal by the CBA literature.
12
“...the optimal climate change policy reduces long-run global warming from 6.6°C to 6.2°C.”(Nordhaus
1997, p. 324). His current “optimal”rate is closer to 3˚C, see below.
Climatic Change

warming from the IPCC: “Partial loss of ice sheets on polar land could imply metres of sea
level rise …projected to occur over millennial time scales, but more rapid sea level rise on
century time scales cannot be excluded.”(IPCC 2007, p. 13). Baer and Mastrandrea (2006)
present a risk analysis of such dangers: “a precautionary approach requires near immediate
efforts to ‘bend the curve’of global emissions, and much steeper reductions [80% by 2050]
than are currently contemplated”(p. 8). The traditional approach is unsuitable mainly
because it simplifies or just ignores such deep uncertainties of long-term climate
projections. It requires knowledge about the far distant future under climate conditions
radically different from those of today (van den Bergh 2004), as well as many other
assumptions about human welfare and behaviour needed to make the mathematics tractable
(Nelson 2006 p. 93).
However, the literature coming to the conclusions from traditional economists does not
exist in isolation from that of other disciplines addressing the same problem. Climate
scientists address the likelihood and risks of extreme events, and draw conclusions about
what “one can safely say, for all practical purposes”. Ethics considers issues of inter-
generational equity, when climate damages are uncertain and far in the future. Engineering
and architecture give insights into how the capital stock can be designed to reduce
greenhouse gas emissions. Economic geography and history provides understanding as to
how economies grow and how technologies diffuse and evolve. Political science considers
how societies make decisions regarding public policy. Furthermore, economics is not
confined to the study of equilibrium in various guises, assuming groups of identical
representative agents, with entirely self-interested consumers and known, quantifiable
social welfare functions. All these economic assumptions and more are standard in the
traditional cost-benefit analysis that lies behind the conclusions of Nordhaus and Tol
13
(DeCanio 2003). The weakening of the neoclassical analysis of climate change has been
accompanied by a more general undermining of the ideology and prescriptions of
traditional economics by deconstruction of the origins of the theory in physics and
cybernetics (Mirowski 1989,2002). Behavioural economics going back to (Kahneman and
Tversky 1989) has revealed key relevant empirical findings for risk aversion and utility
maximisation that are inconsistent with traditional treatments. Complexity theory and agent-
based modelling has developed a theory of economic evolution (Arthur 1994; Beinhocker
2006).
Traditional economics has developed an approach to climate change, which has
persistently ignored the conclusions and insights of other disciplines. The new economics,
as it has developed in the literature covered by Stern, is more pluralistic and respectful of
other disciplines. The CBA approach is formally replaced by a Multi-Criteria Analysis
(MCA) developed in management science and applied to sustainable development (Munda
2005) in which socio-economic, ecological, and ethical perspectives are taken into account.
Most of the insights and techniques of CBA can be incorporated into the market
(monetised) criterion in MCA, but as in the Stern Review and IPCC Reports, intrinsic
values (non-monetised) of human suffering, damages to nature, and risks and uncertainties
are also taken into account as criteria for social choice.
13
The literature covered by Tol requires the monetization of the costs of climate change and almost without
exception adopts the standard assumptions. Alternatives approaches are provided by Barker 1996b; Barker
et al. 2006; Ackerman and Heinzerling 2004; Padilla 2004; Maréchal 2007.
Climatic Change

3 Uncertainty in economic systems: equilibrium versus complexity
A critical issue in the understanding of climate change and its mitigation is the treatment of
uncertainty. The use of the word “dangerous”in the UNFCCC objective
14
is interpreted by
the IPCC (IPCC 2007, p.19) in terms of risks of climate change being balanced against the
risks of threatening economic sustainability by the response measures. The problem is
clearly one entailing unknown risks (uncertainties) of the climate system responding to the
anthropogenic pulse of additional greenhouse gas, primarily CO
2
, into the atmosphere
being compared to the largely known risks associated with mitigation policies and the even
lower risks of the co-benefits of mitigation, such as improved air quality. The treatment of
these risks and uncertainties distinguishes the traditional and the new economic analysis.
The classic text (Knight 1921) defines risk as the property of outcomes with quantifiable
probabilities and uncertainty as that with unknown probabilities. Keynes made a similar
distinction: ‘By uncertain knowledge I do not mean merely to distinguish what is known for
certain from what is only probable. About these matters there is no scientific basis to form
any calculable probability whatever. We simply do not know.’[1973,pp.113–14 quoted
by Holt (2007)]. In the traditional CBA, the form of the probability density function is
simply assumed, deep uncertainties are set aside despite the complexity of the scientific
understanding, the risks are assumed to be symmetric despite the key feature of dis-
proportionately higher risks from higher temperatures, and average discount rates are
used for both costs and benefits, despite the differences in risks (Frederick et al. 2002).
Many CBAs are simply deterministic, neither converting uncertainty into “certainty
equivalence”nor subjecting the final model to a sensitivity analysis
15
. However, the
shape of the damage function is uncertain and that the climate science suggests a
significant (i.e. more than 1%) chance of catastrophe, defined by Weitzman (2007)as
average global temperatures rising by more that 5°C above pre-industrial. Weitzman has
extended the traditional CBA analysis to convert the uncertainty into a second-order risk
and his interpretation of the results is to question the validity of CBA: “Perhaps in the end
the climate-change economist can help most by not presenting a cost-benefit estimate for
what is inherently a fat-tailed situation with potentially unlimited downside exposure as if
it is accurate and objective.”Instead, he argues, economists should be “explaining better
to policy makers that the artificial crispness conveyed by conventional IAM-based CBAs
is especially and unusually misleading”.Andthat“in rare situations like climate
change…we may be deluding ourselves and others with misplaced concreteness if we
think that we are able to deliver anything much more precise than this with even the
biggest and most-detailed climate-change IAMs as currently constructed and deployed.”
(p.42)
The CBA estimates prevalent in the literature on the economics of climate change are
highly misleading because the studies set aside or ignore deep uncertainty in costs and
benefits. A critical example is that the global long-run growth rates are almost entirely
exogenous in the models (coming from labour supply and technology), so that the
uncertainties of the effects of climate change on labour and technologies and then on
growth are ignored. The estimated costs of climate change and the optimal carbon tax rates
14
UNFCCC, 1992, states as its objective: “to achieve stabilization of greenhouse gas concentrations in the
atmosphere at a low enough level to prevent dangerous anthropogenic interference with the climate system.”
15
Dietz, et al. (2007) show how risk analysis can change the conclusions of the CBA.
Climatic Change

from the CBAs are essentially the subjective views of Nordhaus, Tol
16
and others presented
as scientifically precise results. Tol in 2005 presented quantified ranges although he
repeatedly acknowledges the uncertainty of the results; in contrast in 2007, responding to
Weitzman’s analysis, he acknowledges that the upper bounds to the costs may be
unquantifiable.
17
Another example is the conclusion from Nordhaus’DICE study (July
2007) that “The total discounted economic damages with no abatement are in the order
of $23 trillion”(p. 181). Such an estimate is much too precise because the uncertainties
in the projections include possible worlds with temperatures >5°C above pre-industrial
averages. This matters because the economists who have made the calculations have
developed a “scientific”consensus in defending the approach. They have argued that the
“social cost of carbon”can be precisely estimated and ranges given with sufficient
precision to form climate policy. Wietzman’s point is that the upper boundaries in such
ranges e.g. a top 95 percentile of about $100/tCO
2
in Tol’s 2005 study, are subjective and
misleading. With notable exceptions
18
the literature on the social cost of carbon, by
ignoring some uncertain damages, truncating probabilities and discounting risky
outcomes, has been promoting an over-optimistic picture of the uncertainties of climate
damage for the last 16 years.
There are also uncertainties associated with the costs of mitigation. In traditional
economics, the Marginal Abatement Cost (MAC) is the cost of abating the marginal
emission of GHGs and the Marginal Abatement Benefit (MAB) is the benefit, in terms of
the reduction in the damages caused by the emissions and converted into money. By
assumption, as the level of abatement rises, costs go down and benefits go up, so that there
is an equilibrium solution for this level in which the marginal cost equals the marginal
benefit, utility is maximized and the optimal carbon tax rate can be calculated. When used
correctly, marginal means ‘vanishingly small’since it is calculated by differentiating cost
curves that are required by the theory to be continuous, but in equilibrium models of
climate change ‘marginal’is used, loosely, to refer to discrete, system-wide changes, e.g. a
shift from fossil fuel to hydrogen in transportation. The benefits in the MAB are the avoided
costs of “doing nothing”as discussed above, including the monetary estimates of avoiding
long-term catastrophe. The costs in MACs can be a large range of disparate, but largely
shorter-term, costs both private and public (or social) with and without market-based
valuations, but all associated with the abatement. It is clear that these costs do not normally
include any political costs of introducing abatement policies and measures. These economic
costs may be offset by ancillary environmental benefits or improvements in efficiency from
the use of tax revenues, but the CBA literature often either ignores these associated
16
Tol ignores the value judgments underlying all cost-benefit analysis, namely that the equilibrium outcome
of rational self interest is in some sense optimal for society. He presents his survey of a set of essentially
subjective estimates, often done by the same clique of authors, as observations of probability density
functions, when they are basically undetermined (see note above), not properly sampled, or even
independent.
17
“… the uncertainty is so large that a considerable risk premium is warranted. …More importantly, there is
a 1% probability that the social cost of carbon is greater than $78/tC. This number rapidly increases if we use
a lower discount rate –as may well be appropriate for a problem with such a long time horizon –and if we
allow for the possibility that there is some truth in the scare-mongering of the gray literature.”Tol, 2007
(conclusion). Tol (2005) does not mention a risk premium or that the damages may be unbounded, although
the fat tail of damages is emphasised as a feature of the analysis.
18
Examples, using Nordhaus’DICE model are the studies by Mastrandrea and Schneider (2001,2004), Azar
and Lindgren (2003) and Ackerman and Findlayson (2007).
Climatic Change

benefits or sets them aside as being too uncertain or assumed to be managed by non-climate
policies.
There are serious problems with the MAC concept and the total costs derived from the
models that use it
19
. The first problem lies in the treatment of uncertainty and technology.
The reasoning assumes that the future schedules of costs are known in advance and
independent of policy. In fact they are uncertain and evidence shows that they are likely to
respond to policy: low-carbon technologies can be expected to develop in response to
higher real prices of carbon, just as energy-saving technologies have responded to energy
prices (Popp 2002). If so, the MAC schedule is not independent of the cost of carbon, so the
schedule is both uncertain and unstable. As noted above, in general the risks and
uncertainties of abatement are much less than those of the damages. If decision-makers are
risk averse, or wish to follow the Precautionary Principle, the fact that abatement costs are
less uncertain is likely to justify action involving higher costs than what otherwise would be
the case. Additionally mitigation reduces the risks of the damages and of future adaptation,
i.e. it reduces the cascade of risks from the emissions. A deterministic equalisation of
estimates of costs and benefits without taking into account these uncertainties ignores
fundamental differences between them.
A second problem is the one Stern identifies as the non-marginal nature of the
economics. As a result of discontinuities and path dependence in the economic system, the
placing of system changes within the apparatus of continuous cost schedules is misleading
because different mixes of old and new technological systems (e.g. a mix of oil-based and
hydrogen-based systems) appear to be highly unlikely because of economies of scale and
specialization and lock-in effects. The complex system effects may be large enough to
achieve significant reductions in costs under new technologies. If there are indivisibilities,
e.g. a global electricity grid for low-cost renewable generation, there is no longer a unique
solution for the equilibrium carbon tax. There is evidence for such system properties from
the investigation of future costs of energy systems undertaken by researchers at IIASA
(Gritsevskyi and Nakicenovic 2000). It is also obvious at the micro scale that the
technologies and costs of mitigation are not continuous. This appears to be the case at the
macro scale because of network economies and technological lock-in. Not only are there
significant discontinuities in the abatement cost schedule, the costs are likely to go up or
down for different levels of abatement depending on the technological system under
study.
The Stern Review’s assessment (p. 338) of escalating macroeconomic costs of mitigation
as targets become more stringent (450 ppm CO
2-
eq and below) is also open to question,
since the underlying literature largely assumes continuity and exogenous technological
progress. Macroeconomic costs may not escalate when policies lead to decarbonisation,
although carbon prices, energy investments and the policy “effort”are all likely to rise,
perhaps disproportionately and the macroeconomic costs do become more uncertain at
higher carbon prices
20
. One of the key findings of the AR4 is that the literature on stringent
19
This is not to undermine the usefulness of the static incremental abatement schedule, showing abatement
options as a set of non-linear steps rising as the carbon price rises, and usually referred to as the MAC.
20
Tol ( 2007) asserts that high carbon taxes would bankrupt some countries. He seems to be confusing tax
revenues with tax payments. The tax revenues accrue to governments and benefit their finances; they benefit
the population if used wisely. One such use is to ensure that all home owners adopt low-GHG technologies.
The tax payments may not be a problem if safeguards to protect vulnerable social groups predate or
accompany the introduction of a carbon tax.
Climatic Change

mitigations is insufficient to draw reliable conclusions (WG3, SPM, p. 16). All the available
mitigation options have not been investigated and included in the models in terms of the
speed at which they need to be implemented or their eventual scale. Extrapolation of the
costs in the literature (Barker and Jenkins 2007) suggests that, depending on policies,
macroeconomic costs for more stringent mitigation will remain negligible, but risks of
policy mistakes increase.
The modelling of economic risks in the context of climate change has been taken
forward with the post-Stern work of Dietz et al. (2007), in which the effects of adding risk
to the CBA are shown to increase the costs of climate change significantly. The key features
of a risk analysis of the problem are that the risks and uncertainties associated with the
climate damages are much greater, because of systemic irreversibilities and non-linearities,
compare to those of mitigation, which are largely known from past experience, and that the
air quality benefits of mitigation are even less risky, more immediate, and well-documented
(Barker et al. 2007). A more flexible “new economics”alternative modelling approach to
CBA (Barker et al. 2006) is based on the economic history of institutional structures. It
emphasizes the importance of accounting and economies of specialization and allows for
increasing returns to scale in the factor demand equations. In critical sectors, technology
is modelled to allow for reductions in unit costs as the scale of production increases and
the markets develop. Scenarios incorporating system-wide changes in technology, e.g.
those involving the hydrogen economy, can be developed consistently. This approach
does not impose costs of mitigation by assumption, unlike general equilibrium, so that
an alternative low-carbon world economy may be less costly than business as usual,
depending on the reductions in costs that emerge when new technologies come into
widespread use.
4 Economic ethics, intergenerational equity and the discount rate
Neoclassical economists claim that their work is value-free (Robbins 1932), scientific
(Nordhaus 2007b) or purely descriptive (Pearce et al. 1996; Nordhaus 2007b). In doing so,
it has been plausibly argued that they are drawing on nineteenth century science to promote
a secular, rationalist religion (Nelson 2001, p. 133). Their faith is in the path-independence
of consumer preferences and producer technologies, a faith shown to be empirically false in
psychology, physics and history. Their thinking, apparently logical, is based on the fallacy
that “the pursuit of self interest is guided by objective laws to a socially beneficent
outcome”when instead this pursuit involves moral choices, at both personal and social
levels (Foley, p. xiii).
Nordhaus (2007b, p. 140–1) characterises economics as scientific in being peer-
reviewed and reproducible; he derives the discount rate from a pure description of the
market rather than from a consideration of ethics and moral philosophy. He contrasts his
approach with that of the Stern Review, which he finds unscientific
21
. The many critiques
21
There is a literature devoted to the issue of whether economics is a science or not. See (Mirowski 1989,
2002; Weintraub 2002; Katzner 2003). It is a science in that theory and observation are considered together
when and where possible or in that mathematics is a science (Samuelson’s position). However, neoclassical
path-independent economics as a mathematical science is strictly a branch of mathematics rather than
economics, since it violates a basic law of physics, the Second Law of Thermodynamics. Nordhaus (2007b)
himself is ambiguous about whether economics is a science or not, since he repeatedly distinguishes
economics from science.
Climatic Change

of the Stern Review have been dominated by the discussion of the discount rate
22
(e.g.
Nordhaus 2007a; Dasgupta 2007,2008; Tol and Yohe 2007). The “pure rate of time
preference”is one component of the discount rate used in calculating the costs of doing
nothing in relation to climate change (see Quiggin below for an explanation). Stern adopts a
pure rate of time preference of near zero, drawing on moral arguments, compared with the
rate adopted by the traditional literature, e.g. 1.5%pa in (Nordhaus 2007b) down from 3%pa
in earlier applications of his DICE model. The difference in discount rates between Stern
and the traditional literature is one of the reasons (cited by several of the critics as the main
reason) for the much higher costs of 5 to 20% of global GDP “now and forever”estimated
by Stern (p. xv)
23
for business-as-usual of 2°C to 5°C warming or more, compared with the
cost of 6% of GDP for a 10°C warming from Cline (1992), who also used a near-zero
discount rate as quoted in the IPCC Second Assessment Report (Pearce et al. 1996, p. 208),
but had a different approach to risk, yielding lower costs.
The detailed deconstruction of this difference in the “costs of doing nothing”is covered
by the paper in this issue by Quiggin, so the discussion here can be brief. The first point is
that moral philosophers have long debated the relative weighting to be given in utility
theory between social groups living at different times. The Stern Review commissioned a
review of the ethics of climate change from Broome (2006), who had written earlier on the
issue (1992). He makes uncomfortable reading for traditional economists, partly because he
insists, rightly, that economics is not ethics-free, that basing economics on the ethics of
individuals assumed to be entirely self interested can go badly wrong, and that “willingness
to pay”is invalid as a means of valuation (Broome 2008). This is in direct contradiction to
the analysis of Pearce et al. (1996, p. 196–197) and Nordhaus
24
, when they contrast
prescriptive with descriptive valuations of human life. In considering the ethics of climate
change, Broome positions justice centre stage, arguing that those who cause climate change
should cease to do so because it is unjust, and if they cannot cease, then they should
compensate those who suffer.
Justice as a theory of ethics (Rawls 1971) deserves serious attention as an alternative to
utilitarianism in climate-change analysis. Consider two population groups: a well-off urban
majority, burning fossil fuels, and a subsistence rural minority, dependent on the weather
for food and water. Assume that the costs of mitigation are negligible as the literature
22
The use of a zero discount rate specifically for climate damages in a cost-benefit analysis of climate
change (Hasselmann et al. 1997) anticipated in some respects Stern’s use of low discount rates and also set off
a fierce debate with those supporting an aggregate discount rate for all types of damage (including loss of
human life) or sectors, which they justified by the traditional neoclassical treatment (Heal 1997; Nordhaus
1997) relying on the assumption of social groups being identical representative agents having full
information and foresight. This traditional approach also denies any significance to the empirical finding of
differences between sectors in the discount rates actually used (e.g. private rates being several times public
discount rates). Hasselmann’s reply (1999) also anticipates the emerging resolution of the post-Stern debate
discussed here, specifically the conclusions in (Hoel and Sterner 2007) from a two-sector model.
24
Nordhaus (2007b) claims that his 1.5%pa pure rate of time preference is “designed to provide the most
accurate projections rather than to be normative in nature.”(p. 40).
23
It is not the only reason. Dietz et al. ( 2007) provide four reasons for the higher costs in the Stern Report.
Previous studies, with important exceptions, have “(i)...mostly omitted to adequately employ the probabilistic
results of recent science; (ii)...tended to consider a narrow range of impacts, a product of focusing largely on
2–3°C warming, whereas we now know that there is a possibility of far higher temperatures; (iii)...not used
the economics of risk to the extent appropriate; (iv)...not paid adequate attention to the underlying ethics.”
The overall effect has been to give “on average, strong downward bias on damage estimates in the previous
literature”. (pp. 156–157)
Climatic Change

suggests. Assume also that the rural minority do not share in average global growth; they
can be said, in Rawls’words, to be the ‘least advantaged group’. In his theory, the standard
of living of the most advantaged would be justified only if their privileges maximised the
welfare of the least advantaged group, for example through the general effect of incentives
on the economy. Let us assume there is no such Rawlsian theory of justice in place. If
policy were to be formulated according to the traditional assumptions with the results
described above, the outcome would be a triple injustice:
(1) The rural minority have not been responsible for today’s GHG concentrations causing
climate change, and have not benefited from the comfort and power given by the
fossil energy services, yet being dependent on the weather suffer most of the
consequences.
(2) The minority will suffer much more from future climate change because droughts and
floods threaten their subsistence income, and they cannot buy their way out of the
problems.
(3) The minority’s future income is discounted by an average dominated by the well-off
majority’s income growth, so their future utility counts for much less in the CBA of
global climate-change policy. This outcome is a direct consequence of the discounting
of average consumption net of climate damages (including mortality) by Nordhaus
and others, supported by Dasgupta (2007)
25
.
Since global inequalities over the last century have been increasing (Maddison, 2001)anda
subsistence minority of countries (and social groups within countries) may continue into the far
future, the assumptions may well be more realistic than those of the traditional model. Rawlsian
ethics would focus social policy on preventing the climate change and caring for the subsistence
minority. Instead, traditional models have been used to justify weak policies and inaction.
Broome (2006) also considers expected-utility theory alongside justice as a guide to
social policy. Importantly he distinguishes “value”from “utility”and allows for intrinsic
value in human life and nature. He considers the utilitarian view of climate change, arguing
that (1) lives should not be valued by the method of willingness-to-pay, which makes the
value of people's lives depend on how much they can afford to spend on prolonging them
and (2) future lives should not be discounted in value relative to present lives of similar
quality
26
. The argument that because people in the future are expected to be better off in
real money terms, so that we can then discount a monetary value of their lives (or their
health) runs into serious logical and moral problems, which are not solved by recourse to
the term “statistical lives”. Nordhaus (2007b, p.47) is discounting the quantity and quality
of human life when he includes valuations of mortality and morbidity in the damages from
catastrophe (which are stated to include health damages)
27
. Implicitly, those who discount
25
Like Broome, Dasgupta (2005) is an authority on economics and ethics, but he argues that traditional
economics has solved the ethical problems. However, Dasgupta does not mention the ethical problems
involved with the averaging done in CBA, when the assumed monetary estimates of health and lives are
discounted.
26
Broome’s view on discounting is supported by the utilitarian philosopher R M Hare, who likewise argued
that a discount rate above zero cannot be justified ethically (Hare 1981, p.100–101).
27
Dasgupta (2007) supported Nordhaus’s approach, but not his adopted pure rate of time preference. In
contrast, Stern argues that human lives and environmental quality should be treated separately (p. 165),
although the PAGE model (used to calculate the 5–20% range on costs) appears to include valuations of
human life and health implicitly in its damage functions. Dasgupta (2008) no longer supports Nordhaus, and
concludes that “an optimum policy may not exist”, and (implicitly) that CBA is “an overly formal analysis”
leading to “misplaced concreteness”in its conclusions (since 1991) on climate change.
Climatic Change

such damages at 1.5%pa and higher are valuing the next generation’s lives and health at a
fraction of their own. An equal valuation would transform the policy prescriptions towards
urgent action and high carbon tax rates.
Nordhaus and others rely on the market to provide an estimate of the social discount
rate. The preferences underlying the rate are assumed to be fixed and to take into account
far-future climate damages. Such assumptions are not empirically valid and the procedure
short-circuits the political process, in which for example democratically elected politicians
aim to lead and change preferences (see below). The preferences are also assumed to take a
particular form, in which no ethical preferences are allowed, although in fact people might
prefer that natural resources be preserved as a matter of principle, even though they have no
utility for them. Finally these authors are assuming fungibility of natural and man-made
assets, i.e. that they all have monetary values and can be exchanged. Irreversible changes,
e.g. warming of the oceans leading to loss of coral reefs for the indefinite future, means that
such exchange is impossible. Hoel and Sterner (2007) have explored an extension to the
traditional model allowing for human and natural services and the likelihood that as the
natural services become scarcer, their prices will rise. They conclude that under reasonable
assumptions the discount rate could become negative.
It is the implicit assumption on the part of traditional economists of a ‘moral’superiority
of the market that is at the heart of this debate. Utilitarian philosophers will have none of it.
Traditional economists evade this implication of their analysis, claiming that they are being
descriptive rather than prescriptive, but their logic does not stand up to scrutiny
28
. This is
economics as a religion (Nelson 2001), in which society is composed of self-interested
individuals, whose behaviour is to be assumed rational, then to be interpreted and described
by economics as a mathematical science, e.g. in finding and using the pure rate of time
preference, or the value of human life. The underlying fallacy is that market forces lead by
themselves to intrinsically good outcomes (Foley 2006), a fallacy explored in more detail
by Jaeger et al. below. A “new economics”approach is to acknowledge that there are
ethical, aesthetic and other values, and that all life should not be converted into money, with
the exchangeability that money permits (Ackerman and Heinzerling 2004; Gowdy 2005).
The use of the discount rate to account for time preference and risk should be re-thought to
allow for subjective time preference and a risk analysis independent of the return (Price,
2005). Climate change economics should learn from organizational sciences applied to
management of high-risk activities (Nelson 2008). The distribution of rights consistent with
sustainable development should be considered (Padilla 2004). The anti-utilitarian moral
philosopher Bernard Williams has criticised the reductionism of “utilitarian thought”and
“the device of regarding all interests, ideas, aspirations and desires as on the same level, and
all representable as preferences of different degrees of intensity, perhaps, but otherwise to
be treated alike….The assimilation does not give our convictions enough weight in our own
calculations. At the same time, it can give other people’s convictions too much weight”
(Williams 1985, p86). The utilitarian approach has much to offer, but its claims should be
qualified and limited by considerations other than that of utility, such as those of justice,
well-being of future lives, and benevolence.
28
See Broome (2008). Nelson (2008) reveals other hidden value judgments that may underlie the traditional
neoclassical approach to climate change, leading to the tendency to rigidity and blindness to errors evident in
the critical responses to the Stern Review.
Climatic Change

5 Engineering and history: induced technological change and the costs of GHG
mitigation
The costs of GHG mitigation in traditional economics are derived from the production
function, a concept basic to the determination of the allocation and growth of economic
output, conventionally measured as marketed output, i.e. GDP in national accounts. In the
models this function takes special forms, typically Cobb–Douglas or Constant Elasticity of
Substitution (CES) with tractable properties: they are continuous, typically with constant
returns to scale, and reversible in that outputs can be expressed in terms of their inputs of
labour, capital, materials, and vice versa, a feature that contradicts path dependence, i.e.
the second law of thermodynamics
29
. This economics has been derived by analogy with
physical process of the first law of thermodynamics by Walras drawing on 19thC textbooks
of physics (Mirowski 1989; Beinhocker 2006) without an adequate treatment of time or the
later second law with the underlying physical requirement that all processes involve increasing
entropy. In the case of the burning of fossil fuels, this means the return into the atmosphere
of the original CO
2
captured by plants and fossilised over millions of years as fuels.
More striking still, technological change has been assumed in the traditional muti-
industry treatment to be independent of production change, implying no learning by doing
or by researching (Barker et al. 2007, 11.5). If the general equilibrium models are to include
such endogenous technological change it is usually grafted on to the neoclassical
production function by linking it with an engineering model, typically for the energy
supply and demand sectors. The outcome is inconsistent in that endogenous technological
change is allowed for energy output but not other sectors (as well as carbon prices, many
other relative prices will change as an effect of climate policies) nor for other economic
variables, such as exports, employment or even consumption. It is also incomplete, in that it
ignores the potential interaction between the information economy, energy and new low-GHG
technologies, which accelerates their adoption and diffusion throughout the world economy.
The aggregate production functions, used in the equilibrium economic models, have
been subject to detailed and severe criticism over many years, both of the underlying theory
(Mirowski 1989; Felipe and Fisher 2003; DeCanio 2003) and of the validity of the
empirical estimates (Felipe and McCombie 2005). Theoretically, the use of an aggregate
production function requires two (heroic) assumptions: (1) that it is a meaningful exercise
to combine the processes of e.g. furniture-making, oil-refining, and food-retailing, and (2)
that ALL markets are perfectly competitive. Empirically, the use of National Accounts data
on flows in current prices to estimate production functions is usually flawed, because the
dataset imposes an accounting identity on the monetary value of production and the
combined values of the inputs to production (namely materials, labour and capital), when
capital services are measured as residual profits. The estimates in the literature are often
based on accounting identities, not causal relationships, and hence the very good fits
obtained are entirely artefacts of the data
30
.
29
The second law of thermodynamics is an expression of the universal law of increasing entropy, stating that
the entropy of an isolated system which is not in equilibrium will tend to increase over time, approaching a
maximum value at equilibrium. (Wikipedia, 15.01.08)
30
See (Felipe and McCombie 2005). The empirical basis of the functions actually included in the climate-
policy models is even more compromised, being no more than a collection of guess-estimates from an
inconsistent literature (DeCanio 2003).
Climatic Change

The implication of the production function in the traditional models, both the one-sector
models of Nordhaus and others and the multi-industry general-equilibrium models, is that
because the functional form assumes that the economy is at full employment and maximal
efficiency, any climate policy leads to costs in the form of loss of potential output. It is this
feature that leads to the contrast between the energy-engineering, bottom-up models,
providing estimates of some 6 GtCO
2
-eq mitigation potential by 2030 at net negative costs,
i.e. “no regret options”, compared with no such unrealized potential estimated by the top-
down equilibrium models (IPCC 2007,p.14–16). The potential for energy saving assessed
by countless engineering studies is simply ignored in the equilibrium models by assuming
full information, maximum efficiency and full employment, now and forever, in violation
of the facts. New evolutionary economics can provide insights into the non-economic barriers
to energy efficiency and how they may be overcome (Maréchal 2007, pp. 5183–5184).
The traditional treatment of production also normally rules out of court any modelling
outcome that increases the growth rate of the economy as an outcome of climate policy.
There are many conditions under which GDP may increase, e.g. use of carbon tax revenues
to reduce distortionary taxes, the effect of policy in reducing the widespread under-
employment in developing countries, and the possibilities of more productive technological
pathways. Although documented in the theoretical and empirical literature (Barker et al.
2007, 11.4 and 11.5), these conditions are routinely and implicitly set aside by assumption
in the traditional treatment.
However, complexity economists (Arthur 1994) strongly argue for path dependency and
increasing returns and economic historians have long argued that technological change and
economic growth are intimately related (Maddison 2001) and path dependent (David
2001). The scientific requirement to decarbonise the global energy system is in effect
suggesting the need for another technological revolution, implying major structural shifts in
the energy industries and requiring the diffusion of low-carbon technologies, particularly
across the developing world, which holds the greatest potential for adoption, radical
changes and impacts. In contrast to the eighteenth and nineteenth century changes, the
context is now one of the global spread of almost free information, instantaneously. The
potential for global, induced technological change to reduce costs and even increase GDP is
recognised in the Stern Review and AR4, as is the modelling that relaxes the assumption of
constant returns to scale. In contrast to the equilibrium approach, such modelling has the
great advantage that it aims to explore technological and institutional options that give rise
to opportunities rather than costs, making the problem for international negotiations much
more tractable (sharing out the benefits of a technological revolution, to put it crudely, but a
revolution that will only happen if countries co-operate). At the same time it should be
recognized that badly designed policies and regulatory frameworks, as in banking, could
lead to potential inflation or financial collapse of investments and programmes.
6 Social choice
The switch in policy required to address the climate change problem is an issue of social
choice. Traditional economics approaches this problem by the use of the social welfare
function, which is a mathematical equation, or a set of equations, in an economic model,
intended to represent the social good. However, the concept is fundamentally flawed. When
national governments act, it is much more likely to be ‘in the national interest’than in any
formal manner capable of being represented as a ‘criterion function’,an‘objective function’
or a ‘social welfare function’as some key concepts are known in equilibrium modelling of
Climatic Change

the economy and the environment. As Arrow (1967, p. 736) remarks about Samuelson’s
neoclassical treatment, ‘Whose behaviour or whose judgement is referred to in the social
welfare function is never clarified.’
In theory, the concept depends on the validity of adding up the welfare of households or
people such that the aggregate social welfare function is stable and predictable over time.
Arrow (1950) showed that for a set of reasonable assumptions (inter alia: a heterogeneous
population, universality, “independence of irrelevant alternatives”) such aggregation is
impossible except under extremely restrictive conditions. Traditional economics has
resorted to assuming that members of the population, or social groups such as households
or firms, are in fact identical “representative agents”, whose welfare and behaviour can be
aggregated. This assumption, required for the macroeconomic equilibrium models to be
theoretically valid in relation to microeconomic behaviour, is ‘both unjustified, and leads to
conclusions that are usually misleading and often wrong.’(Kirman 1992, p.117). The
aggregate approach also often ignores specific issues of equity and the distribution of
wealth, which are especially important for climate change economics because the costs tend
to be met disproportionately by those who cannot afford insurance, re-location or
adaptation investments.
In addition, the social welfare function is not politically viable. The idea that there is a
stable relationship between different policy objectives such as reduction of greenhouse
gases, economic development, growth in consumption, reduction in unemployment or in
the rate of inflation, does not make sense when the actual political process of policy
formation is considered, whatever the political complexion of the government or the
prevailing consensus about sound policy promoted by international organizations such as
the OECD, the IMF or the World Bank. Institutional decision-making (e.g. that by national
governments) is normally characterised by the achievement of consensus between people
and groups with potential conflicts of interest. If this is so, it is quite easy to envisage the
simultaneous pursuit of conflicting goals, and the sudden alteration of policies as different
interest groups gain precedence. There is no escaping the fact that the goals of national,
economic and social policy are different for different interest groups, and that the national
interest cannot be restricted to a fixed formula. In the face of these difficulties, traditional
economists have resorted to another counter-factual assumption (usually implicit, but
required for an optimal solution), that of a global planner, i.e. a policy dictatorship for good
or bad
31
.
Social choice regarding the climate policy involves social groups, “stakeholders”, such
as government, industry, NGOs, and political parties, in a process of consensus (Ostrom
1990). But it also involves information and the law (Heinzerling and Ackerman 2007). A
real choice requires the equal and simultaneous presentation of feasible alternatives. When
a policy is the subject of political debate and possible implementation by government,
policy advisors consider the benefit that such implementation would produce in each of
various mutually exclusive ‘states of nature’that might follow it, the good being considered
for each group affected over space and time.
31
The traditional cost-benefit analysis has been taken up by Lomborg in the “Copenhagen consensus”to
promote the idea that global problems other than climate change (e.g. HIV) are more worthy of funding. It is
inherently unlikely that the national interest can be identified with the functions for aggregate utility in the
equilibrium models and then solved to obtain ‘optimal’set of policies. If an attempt were made to elicit the
function by asking a series of hypothetical questions of governments, rather than Lomborg’s selection of
economists, it would fail because the answers would most likely be inconsistent in terms of the equation. In
any case, politicians would refuse to commit themselves on hypothetical questions.
Climatic Change

The actual process of developing such information for the global community has been
chaotic. Different governments have produced their own analysis, sometimes as in the case
of the US Administration in 2001, selectively choosing scenario results to meet obvious
political requirements. The paper by Hasselmann and Barker below discusses a way
forward to improve the information basis in developing international climate policy under
the UNFCCC and the IPCC. These bodies have arisen out of the international political
process and are in keeping with decentralised and varied political structures. This process
has brought questions of equity to the fore as witnessed by the crisis in the IPCC’s adoption
of the Second Assessment Report in 1995, with the neoclassical economists’insistence on
valuing human life on an insurance basis. The use of values of “statistical lives”came into
conflict with a perception that human life at present and in the future should be valued
equally irrespective of income or circumstance, for the purpose of agreeing international
policy. The governments of the developing countries arguing their case for equality
prevailed over the expert IPCC economists advising them. However, it is perfectly feasible
that a consensus approach in international negotiation can help to establish policies and
social values in difficult and controversial areas, such as abatement of climate change,
where the interests of different countries and future generations are to be taken into account.
For example, the IPCC’s Summaries for Policy Makers are agreed by all governments
explicitly at international meetings.
7 Towards a new economics of climate change
The IPCC’s sceptical approach to the use of cost-benefit analysis (CBA) as the sole basis
for the economics of climate change has been supported by Stern. The CBA of climate
change after Stern has been developed by Weitzman (2007) to the point of destruction. Just
as a central bank, faced with the risk of the collapse of the banking system, will act on
perceived risks rather than a monetary CBA, so governments have eschewed CBAs in
which the “optimal”solution involves risks of dangerous climate change. The intemperate
and rushed reaction by a clique of neoclassical economists criticizing the economics of the
Stern Review illustrates the sensitivity to the implied criticism of their conclusions.
The subsequent development of the literature in supporting Stern’s conclusions
illustrates how radical the shift in mainstream economic thinking has been. It is now
acknowledged that the economics of climate change is now more appropriately concerned
with uncertainty rather than return, a feature of the problem that has been evident from the
early 1990s, when the scientific assessments began in earnest. It can also reasonably be
argued that CBA is useless for the climate problem because of the uncertainty and risks of
catastrophe. The discounting of costs and benefits in which risks are converted into
certainty equivalents and discounted at market rates has been shown to be misleading and
biased. This in turn implies that the economic problem is one of achieving political targets,
based on scientific evidence, at lowest costs compatible with equity and effectiveness,
rather than with the economics of choosing the targets themselves.
The new information and evidence on the increasing risks of climate change has
reinforced earlier perceptions about the dangers, and raised scientific and political alarm,
but the general message has been to strengthen the evidence, and bring it home to the public
through debate on weather-related catastrophes. In summary, the problem is clear and the
solution appears to be almost costless in macroeconomic terms, but requires the long-term
transformation of the global energy system. Decarbonisation of the global energy system by
2050 seems to be feasible at reasonable carbon prices (at least 100$/tCO
2
by 2020 in year
Climatic Change

2000 prices and continuing to rise thereafter), with benefits to health and negligible effects
on economic growth, but it will require unprecedented global co-ordinated action.
Acknowledgements I would like to thank Frank Ackerman, Eric Beinhocker, John Broome, Simon Dietz,
Klaus Hasselmann, Richard Lewney, John McCombie, John Quiggin, Serban Scrieciu, David Taylor and
Rachel Warren for providing helpful comments on earlier versions of this paper. I also acknowledge the support
of the UK Research Councils in funding my work as part of the Tyndall Centre for Climate Change Research.
References
Ackerman F (2004) Priceless benefits, costly mistakes: what’s wrong with cost–benefit analysis? Real-World
Econ Rev 25:2–7
Ackerman F, Finlayson IJ (2007) The economics of inaction on climate change: a sensitivity analysis.
Climate Policy 6(5):509–526
Ackerman F, Heinzerling L (2004) Priceless: on knowing the price of everything and the value of nothing.
New York, The New Press
Ackerman F, Nadal A (eds) (2004) The flawed foundations of general equilibrium: critical essays on
economic theory. Routledge, London
Akerlof G (2007) The missing motivation in macroeconomics. AEA presidential address, January 2007
Arrow KJ (1950) A difficulty in the concept of social welfare. J Polit Econ 58(4):328–346
Arrow KJ (1967) Samuelson Collected. J Polit Econ 75(5):730–737
Arrow KJ (2007) Global climate change: a challenge to policy economists, economists’voice. http://www.
bepress.com/ev. Cited June
Arrow KJ, Cline W, Maler KJ, Munasinghe M, Squitieri R, Stiglitz J (1996) Intertemporal equity,
discounting and economic efficiency. In: Bruce J, Lee H, Haites E (eds) Climate change 1995—
economic and social dimensions of climate change. Cambridge University Press, Cambridge, pp 125–44
Arthur WB (1994) Increasing returns and path dependence in the economy Ann Arbor: University of
Michigan Press, Michigan
Azar C, Lindgren K (2003) Catastrophic events and stochastic cost–benefit analysis of climate change. Clim
Change 56:245–255
Baer P, Mastrandrea M (2006) High stakes designing emissions pathways to reduce the risk of dangerous
climate change. IPPR, London
Barker T (1996a) Essay review: a review of managing the global commons: the economics of global change,
by William D Nordhaus. Energy Environ 7(1):85–88
Barker T (1996b) Space–time economics. Cambridge Econometrics, Cambridge
Barker T (1998) The use of energy-environment-economy models to inform greenhouse gas mitigation
policy. Impact Assess Proj Apprais 16(2):133–131
Barker T, Pan H, Köhler J, Warren R, Winne S (2006) Decarbonizing the global economy with induced
technological change: scenarios to 2100 using E3MG. Energy J 27:241–258
Barker T, Qureshi MS, Koehler J (2006) The costs of greenhouse-gas mitigation with induced technological
change: a meta-analysis of estimates in the literature. 4CMR, Cambridge Centre for Climate Change
Mitigation Research, University of Cambridge, Cambridge, UK
Barker T, Jenkins K (2007) The costs of avoiding dangerous climate change: estimates derived from a meta-
analysis of the literature. A briefing paper for the United Nations Human Development Report 2007
Barker TI, Bashmakov A, Alharthi M, Amann L, Cifuentes J, Drexhage M, Duan O, Edenhofer B, Flannery M,
Grubb M, Hoogwijk FI, Ibitoye CJ, Jepma WA, Pizer K, Yamaji ‘Mitigation from a cross-sectoral
perspective’Chapter 11 in Climate Change 2007: Mitigation. Contribution of Working Group III to the
Fourth Assessment Report of the Intergovernmental Panel on Climate Change [B. Metz, O.R. Davidson, P.
R. Bosch, R. Dave, L.A. Meyer (eds)], Cambridge University Press, Cambridge, United Kingdom and
New York, NY, USA
Beinhocker E (2006) The origin of wealth: evolution, complexity and the radical remaking of economics.
Random House Business Books, London
Boulding K (1992) Towards a new economics. Edward Elgar, Brookfield
Broome J (1992) Counting the cost of global warming. White House Press, Cambridge, UK
Broome J (2006) Valuing policies in response to climate change: some ethical issues. A contribution to the
work of the Stern Review of the Economics of Climate Change
Climatic Change

Broome J (2008) Why economics needs ethical theory. In: Basu K, Kanbur R (eds) Welfare, development,
philosophy and social science: essays for Amartya Sen’s 75th birthday, vol. 3. Oxford University Press,
Oxford
Cline WR (1991) The economics of the greenhouse effect. Econ J 101(407):920–37
Cline WR (1992) The economics of global warming. Institute for International Economics, Washington, DC
Colander D, Holt R, Rosser Jr JB (2004) The changing face of mainstream economics. Rev Polit Econ 6
(4):485–499
Dasgupta P (2005) What do economists analyze and why: values or facts? Econ Philos 21:221–278
Dasgupta P (2007) Comments on the Stern Review’s Economics of Climate Change. Natl Inst Econ Rev
199:4–7
Dasgupta P (2008) Discounting climate change. Rev Environ Econ Policy (in press)
David P (2001) Path dependence, its critics, and the quest for ‘historical economics’. In Garrouste P,
Ioannides S (eds) Evolution and path dependence in economic ideas: past and present. Edward Elgar, London
Deaton, A (2007) Letter from America—On transatlantic vices, or Stern in America. Royal Economic
Society, Newletter, October
DeCanio S (2003) Economic models of climate change a critique. Macmillan, New York
DeLong JB, Summers LH (2001) The new economy: background, historical perspective, questions, and
speculations. Proceedings, Federal Reserve Bank of Kansas City, pp 11–43
Dequech D (2007) Neoclassical, mainstream, orthodox, and heterodox economics. J Post Keynes Econ 30
(2):279–302
Dietz S, Hope C, Stern N, Zenghelis D (2007) Reflections on the Stern Review (1): a robust case for strong
action to reduce the risks of climate change. World Econ 8(1):121–168
Dietz S, Hope C, Patmore N (2007) Some economics of ‘dangerous’climate change: reflections on the Stern
Review. Glob Environ Change 17:311–325
Felipe J, Fisher FM (2003) Aggregation in production functions: what applied economists should know.
Metroeconomica 54:208–262
Felipe J, McCombie JSL (2005) How sound are the foundations of the aggregate production function (in
Symposium on the Aggregate Production Function). East Econ J 31(3):467–488
Foley D (2006) Adam’s fallacy: a guide to economic theology. Harvard University Press, Massachusetts
Frederick S, Loeenstein G, O’Donohue T (2002) Time discounting and time preference: a critical review.
J Econ Lit XL:351–401
Gowdy J (2005) Towards a new welfare economics for sustainability. Ecol Econ 53:211–222
Gritsevskyi A, Nakicenovic N (2000) Modelling uncertainty of induced technological change. Energy Policy
28(13):907–22
Hare RM (1981) Moral thinking. Oxford University Press, Oxford
Hasselmann KS, Hasselmann R, Giering, Ocaña V, Storch Hv (1997) Sensitivity study of optimal CO2
emission paths using a simplified Structural Integrated Assessment Model (SIAM). Clim Change
37:345–386
Hasselmann K (1999) Intertemporal accounting of climate change: harmonizing economic efficiency and
climate stewardship. Clim Change 41:333–350
Heal G (1997) Discounting and climate change. Clim Change 37:335–343
Heinzerling L, Ackerman F (2007) Law and economics for a changing world. Harv Law Policy Rev 1
(2):331–362
Hodgson GM (2007) Evolutionary and institutional economics as the new mainstream? Evol Inst Econ Rev 4
(1):7–25
Hoel M, Sterner T (2007) Discounting and relative prices. Clim Change 84:265–280
Holt R (2007) What is Post Keynesian economics? In: Forstater M, Mongiovi G, Pressman S (eds) Post
Keynesian macroeconomics. Routledge, London
Hope C (2003): The marginal impacts of CO2, CH4 and SF6 emissions. Judge Institute of Management
Research Paper 2003/10, University of Cambridge, UK
IPCC (2007) Climate Change 2007: Synthesis Report. Geneva: IPCC
Kahneman D, Tversky A (1989) Prospect theory: an analysis of decisions under risk. Econometrica 47
(2):263–291
Katzner D (2003) Why mathematics in economics? J Post Keynes Econ 25(4):561–573
Keynes JM (1973) The general theory and after, collected writings, vol. XIV. Macmillan, London
Kirman AP (1992) Whom or what does the representative individual represent? J of Econ Perspect 6(2):117–136
Knight FH (1921) Risk, uncertainty, and profit. Hart, Schaffner & Marx; Houghton Mifflin Company,
Boston, MA
Köhler J, Grubb M, Popp D, Edenhofer O (2006) The transition to endogenous technical change in climate-
economy models. The Energy Journal 27:17–55
Climatic Change

Maddison A (2001) The world economy: a millennial perspective. OECD, Paris
Maréchal K (2007) The economics of climate change and the change of climate in economics. Energy Policy
35:5181–5194
Mastrandrea MD, Schneider SH (2001) Integrated assessment of abrupt climatic changes. Climate Policy
1:433–449
Mastrandrea MD, Schneider SH (2004) Probabilistic integrated assessment of “dangerous”climate change.
Science 304:571–575
Mirowski P (1989) More heat than light economics as social physics: physics as nature’s economics.
Cambridge University Press, UK
Mirowski P (2002) Machine dreams economics becomes a cyborg science: light economics as social physics:
physics as nature’s economics. Cambridge University Press, UK
Munda G (2005) Multiple criteria decision analysis and sustainable development. In: Figueira J, Greco S,
Ehrogott M (eds) Multiple criteria decision analysis: state of the art surveys. International Series in
Operations Research & Management Science 78. Springer, Berlin, pp 953–986
Nelson JA (2006) Economics for humans. University of Chicago Press, Chicago
Nelson JA (2008) Economists, value judgments, and climate change: a view from feminist economics. Ecol
Econ 65:441–447
Nelson RH (2001) Economics as religion from Samuelson to Chicago and Beyond. State University Press,
Pennsylvania
Nordhaus WD (1991a) A sketch of the economics of the greenhouse effect. Am Econ Rev 81(2):146–50
Nordhaus WD (1991b) To slow or not to slow: the economics of the greenhouse effect. Econ J 101(7):920–
37
Nordhaus WD (1993) Optimal greenhouse-gas reductions and tax policy in the “DICE”Model. Am Econ
Rev 83(2):313–317
Nordhaus WD (1994) Managing the global commons: the economics of climate change. MIT, Cambridge, MA
Nordhaus WD (1997) Discounting in economics and climate change. Clim Change 37:315–328
Nordhaus WD (2007a) A review of the Stern Review on the economics of global warming. J Econ Lit 45
(3):686–702
Nordhaus WD (2007b) The challenge of global warming: economic models and environmental policy, July
24, 2007, Yale
Nordhaus WD Boyer J (2000) Warming the world: economic models of global warming. MIT, Cambridge,
Mass
Ostrom E (1990) Governing the commons: the evolution of institutions for collective action. Cambridge
University Press, Cambridge
Padilla E (2004) Climate change, economic analysis and sustainable development. Environ Values 13:523–544
Pearce DW, Cline WR, Achanta AN, Fankhauser S, Pachauri RK, Tol RSJ, Vellinga P (1996) The social
costs of climate change: greenhouse damages and the benefits of control. In: Bruce J, Lee H, Haites E
(eds) Climate Change 1995—economic and social dimensions of climate change. Cambridge University
Press, Cambridge, pp. 125–44
Popp D (2002) Induced innovation and energy prices. Am Econ Rev 92(1):160–180
Price C (2005) An intergenerational perspective on effects of environmental changes: discounting the future's
viewpoint. In Innes JL, Hickey GM, Hoen HF (eds) Forestry and environmental change: socioeconomic
and political dimensions. Report 5 IUFRO Task Force on environmental change
Rawls J (1971) A theory of justice. Harvard University Press, Massachusetts
Robbins L (1932) An essay on the nature and significance of economic science. Macmillan, London
Samuelson PA, Nordhaus WD (2001) Economics, 17 edition. McGraw-Hill, New York
Stern N (2006/2007) The economics of climate change: The Stern Review. Cambridge University Press,
Cambridge, UK (References are to the book, published in 2007)
Tol RSJ (2005) The marginal damage costs of carbon dioxide emissions: an assessment of the uncertainties.
Energy Policy 33(16):2064–2074
Tol RSJ (2007) The social cost of carbon: trends, cutliers and catastrophes. Economics Discussion Papers,
2007-44. http://www.economics-ejournal.org/economics/discussionpapers/2007-44
Tol RSJ, Yohe GW (2007) A review of the Stern Review. World Econ 7(4):233–250
van den Bergh JCJM (2004) Optimal climate policy is a utopia: from quantitative to qualitative cost-benefit
analysis. Ecol Econ 48:385–393
Weintraub ER (2002) How economics became a mathematical science. Duke University Press, Durham, NC
Weitzman ML (2007) On modeling and interpreting the economics of catastrophic climate change. http://
www.economics.harvard.edu/faculty/weitzman/files/modeling.pdf. Cited Dec 5, 2007
Williams BAO (1985) Ethics and the limits of philosophy. Routledge, London
Yohe GW (1989) A study guide to accompany Samuelson & Nordhaus. Mc-Graw-Hill, New York
Climatic Change