Conference PaperPDF Available

The Social Cost of Carbon and Competing Decision Frameworks for Climate Policy

Authors:

Abstract

Most economists advocate the use of benefit-cost analysis (BCA) to guide climate policy. However, the application of conventional BCA methodologies and assumptions typically leads to analyses concluding that the benefits of climate mitigation are relatively low on a “present value” basis, and that little action to reduce greenhouse gas emissions is justified in the near-term. Most climate scientists call for more aggressive near-term mitigation and often advocate risk management as the appropriate framework for guiding climate policy decisions. This paper compares the basis for these two, competing decision frameworks: BCA and risk management, and then examines the controversies surrounding the estimation of the benefits of mitigating climate change. Those benefits are embodied in the concept of the Social Cost of Carbon1 (SCC). The relatively low estimates of these benefits that economists typically produce are largely the result of methodologies and assumptions related to: what counts as a benefit; which people “count” in estimating benefits; how to aggregate benefits over time and among people with very different incomes; and whether to account for people’s attitudes toward catastrophic risk. The issues surrounding use of BCA and the SCC may seem obscure, but they are very relevant to current EPA rulemakings to limit CO2 emissions and the Regulatory Impact Analysis that accompanies such rulemakings. They are also critical to a much larger debate over the resources society should devote to mitigating climate change now and in the future.
The Social Cost of Carbon and Competing Decision Frameworks for Climate Policy
Paper presented at the Fall Meeting of the ABA Section on the Environment,
Energy, and Resources
Oct 13, 2011, Indianapolis
Karl Hausker, Ph.D.
ICF International
Washington DC
Most economists advocate the use of benefit-cost analysis (BCA) to guide climate policy.
However, the application of conventional BCA methodologies and assumptions typically leads to
analyses concluding that the benefits of climate mitigation are relatively low on a “present value”
basis, and that little action to reduce greenhouse gas emissions is justified in the near-term. Most
climate scientists call for more aggressive near-term mitigation and often advocate risk
management as the appropriate framework for guiding climate policy decisions. This paper
compares the basis for these two, competing decision frameworks: BCA and risk management,
and then examines the controversies surrounding the estimation of the benefits of mitigating
climate change. Those benefits are embodied in the concept of the Social Cost of Carbon
1
(SCC).
The relatively low estimates of these benefits that economists typically produce are largely the
result of methodologies and assumptions related to: what counts as a benefit; which people
“count” in estimating benefits; how to aggregate benefits over time and among people with very
different incomes; and whether to account for people’s attitudes toward catastrophic risk.
The issues surrounding use of BCA and the SCC may seem obscure, but they are very relevant to
current EPA rulemakings to limit CO2 emissions and the Regulatory Impact Analysis that
accompanies such rulemakings. They are also critical to a much larger debate over the resources
society should devote to mitigating climate change now and in the future.
Two Competing Decision Frameworks
We turn now to the fundamental conflict over the proper policy paradigm or “decision
framework” to apply to the climate change problem. Most scientists (along with some
economists) advocate a risk management framework. They argue that many impacts of climate
change are foreseeable and costly, and that some truly catastrophic outcomes are quite possible,
though our scientific understanding does not yet allow firm estimates of the associated
probabilities and damages. Indeed, the original 1992 U.N. climate treaty uses the language of risk
management, with signatories aiming to reduce atmospheric concentrations of greenhouse gases
with the goal of "preventing dangerous anthropogenic interference with Earth's climate system."
2
MIT’s Joint Program on the Science and Policy of Global Change states it this way:
Human-induced climate change is a problem of risk management. It cannot be proved
that the outcome will be dire or shown with certainty that it will not. As with other
1
Climate change, of course, is driven by emissions of 6 greenhouse gases (GHG) and other climate-forcing
agents. This paper will use “carbon” or “CO2” as a shorthand for GHG emissions. The 6 gases have
differing impacts (global warming potential) and are calculated in “CO2 equivalence.”
2
United Nations Framework Convention on Climate Change, 1992,
http://unfccc.int/resource/docs/convkp/conveng.pdf.
problems we face in private life and public policy, from controlling your cholesterol level
to defense against epidemic disease, uncertain dangers can warrant reasonable measures
to reduce risk.
3
Climate scientists observe that the last ice age ended about 20,000 years ago, and that world’s
climate has been fairly stable during the last 10,000 years, which saw the emergence of
agriculture and civilization. They project a series of damaging and escalating impacts if average
global temperatures continue rising, as they have over the past 50 years. Fig. 1 is a table of likely
impacts of increases of 1 to 5 degrees centigrade across various sectors as set forth by the
Intergovernmental Panel on Climate Change (IPCC) in its 2007 Fourth Assessment Report
(FAR).
4
These include: decreases in food and water supplies; increases in wildfires, floods, and
storms; species extinction; and coastal wetlands loss. In the last several years, new research has
suggested more severe impacts than noted in the FAR, e.g., more rapid sea level rise, sharper
decreases in agricultural production, and ocean acidification.
The IPCC has also examined scenarios for reducing emissions and stabilizing CO2 concentrations
at various levels, and the associated impacts on GDP growth. Taking into the account the body of
climate science, impact studies, and cost studies, many scientists and policymakers advocate that
managing the risks of climate changes requires that nations should act to limit average global
warming to no more than 2 degrees C. Indeed, world leaders endorsed this goal in 2009 at the
climate treaty’s 15th Conference of the Parties in Copenhagen. Risk management goals are
sometimes expressed as limiting CO2 levels, e.g., to no more than double the pre-industrial level.
Unlike BCA, risk management applied to climate change does not explicitly and quantitatively
balance benefits and costs. However, it can incorporate economic considerations. For instance,
the IPCC FAR concluded that limiting CO2-equivalent concentrations to 445-590 parts per
million is likely to shave less than 0.12 percent off of average annual global GDP growth rates
over the next 40 years.
5
In contrast, many economists advocate a benefit-cost decision framework for guiding climate
policy. They argue that a large array of problems cry out for solution and must compete for
society’s limited resources. They hold that benefit-cost analysis (BCA) is the only logical
framework to guide these decisions, and they strive to conduct BCA in a value-neutral manner.
Furthermore, resource allocation is not a “yes/no” decision, but a “how much” decision, and thus
requires “marginal” analysis. In the climate context, this means that society should reduce CO2
emissions up to the point where the marginal cost of reducing a ton of CO2 is just equal to the
marginal benefit of keeping that ton out of the atmosphere. This has common sense appeal: it is
similar to how people make consumption decisions, and how firms make business decisions
weighing benefits and costs, pros and cons, and advantages/disadvantages on the margin. Indeed,
BCA emerged and evolved from economists’ desire for a similar yardstick for public sector
decisions.
Climate policy debates have generated long-running debates on how to estimate the marginal cost
of emission reductions, mostly surrounding the extent to which there are low-cost/no-cost options
for reducing CO2 . However, this paper focuses on how economists estimate the benefit side. The
“Social Cost of Carbon” (SCC) is the economist’s label for the benefit of reducing CO2
emissions, i.e., it represents the dollar value of the cost (i.e., damages) avoided by reducing CO2
emissions by one ton.
3
See http://globalchange.mit.edu/resources/gamble/spinning.html.
4
All IPCC reports can be found at http://www.ipcc.ch/.
5
IPCC, Climate Change 2007: Synthesis Report, Summary for Policy Makers, 2007, p. 21.
There are long-running debates in estimating the SCC as well, but mainstream economists largely
agree on a set of conventional methodologies (explored in more detail below). The IPCC Fourth
Assessment noted that peer-reviewed SCC estimates averaged about $12 per ton of CO2 ($12
CO2), with a range of $3 to +$95. A more elaborate statistical analysis of SCC estimates
published in 2008 found a median value of less than $6/tCO2.
6
More recently, the Obama
Administration convened an Interagency Work Group on the topic and produced a report
(hereafter, the Interagency Report) with a “central value” SCC estimate of $21/tCO2.
7
Many economists fully accept climate science as set forth in IPCC reports (and further buttressed
by the National Academy of Sciences in its recent review),
8
and they conclude that the SCC is
greater than zero, so some mitigation is justified. However, an SCC in the $6, $12, or even $20
per ton range does not justify much mitigation. Some examples: Danish economist Bjorn
Lomborg and argues that BCA indicates that measures to combat other problems (e.g.,
malnutrition, disease, education) merit large investment now, whereas climate change deserves
much more modest investments, largely devoted to R&D.
9
His book Cool It advocates a go-slow
approach:
[T]he major peer-reviewed economic cost-benefit studies show that climate change is real
and that we should do something, but that our cuts should be rather small…
10
Similarly, Prof. Rob Mendelsohn of Yale characterizes many economists’ view:
When economists go to measure the damages of climate change, they are not finding
them to be as large as the climate scientists believe they are. So when you actually go to
measure these damages, they’re not that large. That’s very upsetting. What is true is that
when they go to measure the costs of trying to stop climate change… they are very large,
much larger than the climate scientists would like people to believe. [This] suggests that
you should do moderate amounts of mitigation. And most of the people in this field …
want to do aggressive near-term mitigation.
11
As a final example, Dutch economist Richard Tol’s article “Why Worry About Climate Change?”
has this bottom line:
6
R. Tol, The Social Cost of Carbon: Trends, Outliers, and Castastrophes, Economics E-Journal,
2(2008):25, p.9. Tol offers the median figure $20 per ton of carbon at conventional discount rates and
without equity weighting (see sections below). $20 per ton of carbon is equivalent to $6/tCO2.
7
Interagency Working Group on Social Cost of Carbon, Social Cost of Carbon for Regulatory Impact
Analysis under Executive Order 12866 (February 2010), www.epa.gov/OTAQ/climate/regulations/scc-
tsd.pdf. Page numbers cited below are from this document. The report appears in several other regulatory
proceedings, e.g., www1.eere.energy.gov/buildings/appliance_standards/commercial/
pdfs/mhlf_preanalysis_appendix15a.pdf .
8
National Research Council, America's Climate Choices: Panel on Advancing the
Science of Climate Change; Advancing the Science of Climate Change, National Academies Press,
Washington DC, 2010.
9
Lomborg leads the “Copenhagen Consensus”, a panel of economists that issues a ranked list of the most
promising solutions to the most pressing challenges facing the world.
http://www.copenhagenconsensus.com/Projects/Copenhagen%20Consensus%202008-1.aspx
10
B. Lomborg, Cool It: The Skeptical Environmentalist's Guide to Global Warming, Knopf, 2007.
11
R. Mendelsohn, Speech at AEI Workshop: The Bloody Crossroads of Science and Policy, September 24,
2010, http://www.aei.org/media/energy-and-the-environment/climate-change/the-bloody-crossroads-of-
science-and-policy-video/.
The impact of climate change is relatively small. The average impact on welfare is
equivalent to losing a few percent of income. That is, the impact of a century worth of
climate change is comparable to the impact of one or two years of economic growth.
12
Figure 2 presents results from the Interagency Report to illustrate the magnitude of damages that
economists estimate. The figure shows economic losses as a fraction of global GDP in 2100 due
to increases in average global temperature as estimated by three prominent models using their
default assumptions. Up to 3 degrees C, those losses are between 0 and 2.5% of global GDP
across the three models. One model even shows gains in global GDP up to just below 3 degrees
C. Even at 6 degrees C, GDP losses are still 10% or less.
One is tempted to think that climate scientists and economists are living on different planets, but
they are relying on the same underlying body of climate science and assessment of physical
impacts. Climate scientists are saying climate change poses profound risks to human society and
requires prompt action, and many economists are asking, “Why worry?”
Estimating The Social Cost Of Carbon
Benefit Cost Analysis is a tool economists use to evaluate public policies. BCA is grounded in the
theory of welfare economics. The theory holds that society should maximize the utility (or
satisfaction or welfare) of its citizens by producing various goods and services in “optimal”
quantities such that the marginal benefit equals the marginal cost for each good or service.
Economists believe that private markets for goods and services can perform this function well in
the absence of significant “market failures” (e.g., excessive concentration among sellers or
buyers, excessive information and transaction costs, significant externalities, etc.).
13
Economists
recommend BCA to guide policymakers in achieving the same equation of “marginal benefit =
marginal cost” in providing public goods (e.g., national defense) or in making market
interventions (e.g., policies or programs to account for large externalities such as climate change).
Economists argue that BCA should guide climate policy design, e.g., the stringency of
regulations, the level of carbon taxes, or the size of a cap on emissions.
As noted earlier, economists quantify the benefit side of the BCA ledger by estimating the SCC --
the dollar value of damages avoided by reducing CO2 emissions by one ton. They use Integrated
Assessment Models (IAMs) to arrive at these estimates of the SCC. IAMs link simulations of the
physical climate with simulations of the world economy and model the feedbacks between the
two into a single integrated modeling framework. The models can be relatively simple or
extremely complex. The Interagency Report noted above provides a good overview of the three
IAMs used in that study, which are also the most prominent in the field: DICE, PAGE, and
FUND.
14
The methodologies and assumptions used in the Interagency Report also reflect the
12
R. Tol, “Why Worry About Climate Change?,” ESRI Research Bulletin, 2009/1/1
13
Externalities” is the term economists use for impacts stemming from production or consumption of goods
or services that are not born by the producers or consumers of those goods or services. Air and water
pollution are classic examples of a negative externality. Producers emit pollution and impose costs on
others which are not reflected in market prices.
14
The three models are: DICE (Dynamic Integrated Climate and Economy) developed by William
Nordhaus; PAGE (Policy Analysis of the Greenhouse Effect) developed by Chris Hope; and FUND
(climate Framework for Uncertainty, Negotiation, and Distribution) developed by Richard Tol. For a
description of each, see the Interagency Working Group on Social Cost of Carbon, Op. cit..
mainstream of economic thinking on estimating the SCC. The report is written to be accessible to
non-economists and is must reading for all who seek to understand the issues central to this paper.
The sections below ask four key questions relating to SCC estimation that aim to illuminate some
of the key assumptions that have the inevitable effect of turning a “big problem” (in scientists’
eyes) into a “small problem” (in economists’ eyes). The questions are:
What “counts” as a benefit?
Which people “count”?
How do we aggregate benefits among many different people and across different times?
How do we account for the fact that most people don’t like to take risks with large
negative outcomes?
What “Counts” As a Benefit?
Benefits of avoided climate change typically fall into one of three categories:
1. Benefits that have readily available market values. Examples include: changes in energy
costs due to changes in heating or cooling loads, and changes in agricultural/forest
output.
2. Benefits that have no explicit market values, but economic values can be inferred.
Examples include: costs associated with disease or death.
3. Benefits that have no explicit market values, and are virtually impossible to value in
dollar terms. Examples include: loss of species, large-scale ecosystem disruption, and
ocean acidification (with potentially huge negative impacts on ocean life).
Critics often point out that benefits in the third category carry little weight in BCA, and SCC
estimates are no exception. Yet, in the case of climate change, the impacts would be huge and
potentially cause large and costly disruptions in food supplies that are not accounted for
elsewhere in the analysis. Benefits that are importantly qualitatively but cannot be “monetized
(i.e., quantified in dollar amounts ) tend to become footnotes and the subject of calls for “further
research.”
15
SCC estimates further shrink the magnitude of damages by making assumptions about adaptation
to climate change. The Interagency Report noted:
Each of the three integrated assessment models used here assumes a certain degree of
low- or no-cost adaptation….[The DICE model] assumes very effective adaptation, and
largely ignores adaptation costs. [The FUND model] assumes a great deal of
adaptation… including widespread reliance on air conditioning; so much so, that the
largest single benefit category in FUND is the reduced electricity costs from not having
to run air conditioning as intensively
16
….[The PAGE model assumes] that the developed
countries can ultimately eliminate up to 90 percent of all economic impacts beyond the
tolerable 2°C increase and that developing countries can eventually eliminate 50 percent
15
For example, see Interagency Working Group on Social Cost of Carbon, Op. cit., p. 29ff.
16
In other words, according to the FUND model, the largest benefit of reducing climate change is that
people will reduce their use of air conditioners.
of their economic impacts. All regions are assumed to be able to mitigate 25 percent of
the non-economic impacts through adaptation.
17
All of these assumptions on adaptation are little more than guesswork. But they substantially
reduce the SCC estimate.
Which People “Count”?
Any BCA must define its geographic and time boundaries. Some SCC estimates have looked only
at national benefits, despite the global nature of the impacts. This highlights one of the most
challenging aspects of climate change: nations must cooperate in order to effectively address the
problem. Should a nation apply BCA only as it affects people within its borders? Or should it take
a global view and encourage all nations to apply BCA with the same global view? Prior to 2010,
some US government agencies estimated the SCC from a national perspective. However, the
Interagency Report did take the global perspective and observed that a US national perspective
would likely reduce the SCC to 7% 23% of the global value. Any SCC estimate at the national
level inherently reduces the number of people that “count.”
Another aspect of “which people count” relates to the fact that a huge swath of the world’s
population is not plugged into the monetized, market economies that most of us in the
industrialized world take for granted. Many people in developing countries produce and consume
outside of market economies, or participate only marginally in those economies. Over a billion
people have incomes of less than a dollar a day; and 2.7 billion people have incomes of less than
$2 per day. These are among the populations most vulnerable to the impacts of climate change,
and they include subsistence farmers and fishers. Ideally, SCC should take into account changes
in their “economic welfare”, but if they barely participate in the market economy, then the
impacts on their livelihoods will be left out of the SCC. This further shrinks the estimated SCC
downward from its theoretical ideal of being based on “global economic welfare.”
The time aspect of “which people count” relates to the fact that BCA aims to account for the
welfare of future generations, given that public policies typically impact current and future years.
Perhaps no issue has more intergenerational consequences than climate change, compounded by
the likelihood that changes to the earth’s climate put in motion in this century may be essentially
irreversible for centuries to come. To the credit of the IAM authors, the models attempt to look
two or even three centuries ahead and bravely model economic damages far into the future across
multiple generations. But a vexing problem arises in how to aggregate damages among these
generations to produce an SCC to guide policy today. This leads to the topic of discounting
benefits to future generations, i.e., reducing future benefits to their “present value” to reflect the
time value of money. (See next section.)
How Do We Aggregate Benefits Among Many Different People and Across Different Times?
After defining its geographic and time boundaries, BCA then aggregates monetary impacts across
many persons over many years to arrive at a bottom line. Two critical issues arise:
How should benefits be aggregated when people have very different incomes?
How should benefits be aggregated across different time periods?
17
Interagency Working Group on Social Cost of Carbon, Op. cit., pp. 6-9.
Aggregating benefits when large disparities in income exist poses problems. Economic damages
of $1,000 create a much larger loss in welfare to a person with an annual income of $25,000 than
to a person with an income of $250,000. Similarly, economists often monetize (i.e. assign a
dollar value to) the damages associated with the loss of life (or disease) as a multiple of the per
capita income in a country, which results in the ethically controversial conclusion that lives are
worth more in rich countries than poor countries:
This is an idea that has had a troubled history. In the IPCC's Second Assessment Report,
in 1995, economists calculating the monetary value of global warming damages decided
to value deaths at $1,500,000 in rich countries, $300,000 in middle-income countries, and
$100,000 in low-income countries. When these numbers came to light, governments of
many countries were naturally outraged to find their citizens were worthonly 1/15 as
much as Europeans or North Americans.
18
Some economists recommend application of differential “equity weighting” when aggregating
benefits across countries or regions, taking into account the relative levels wealth. One recent
study explored this and found that a plausible equity weighting doubled the SCC.
19
However, a
meta-study of 211 estimates of the SCC found that less the half of published studies applied any
form of equity weighting.
20
The Interagency Report noted the challenges in developing the
“appropriate” equity weighting, and declined to apply the concept.
Aggregating benefits across time also poses challenges. Standard BCA methodology requires
“discounting” (i.e., reducing) future benefits to reflect the time value of money. This conforms
with common sense: a dollar today is worth more than a dollar tomorrow for both consumers and
businesses. Financial market behavior reflects this principle with impressive precision; and firms
invest to make profits in the future, incurring costs now to gain benefits later. Nevertheless, the
issue of discounting in BCA has spawned a huge literature on whether, when, and how to apply
this principle, especially to long-term, intergenerational decisions.
Arguments abound over the proper discount rate to use in various circumstances. Some
economists recommend a constant discount rate in all circumstances. Others recommend a lower
discount rate for intergenerational decisions, or a “hyperbolic” discount rate that declines over
time. Both approaches give more weight to benefits that occur in distant years. The federal Office
of Management and Budget (OMB) guidance on regulatory analysis directs agencies to use 3%
and 7% as a general rule, but also allows for lower rates in cases with significant
intergenerational effects.
21
The first-order impacts of discounting are straightforward. At an interest rate of 3%, a dollar of
benefit today is worth 97 cents if it occurs one year from now, 74 cents if ten years from now, 23
cents if in 50 years, and about a nickel if the benefit occurs 100 years forward. At higher discount
18
F. Ackerman and E. Stanton, A comment on “Economy-wide estimates of the implications
of climate change: Human health,” Ecological Economics 66(2008): 8-13.
19
D. Anthoff and R. Tol, On international equity weights and national decision making on climate change,
Journal of Environmental Economics and Management, 60(2010): 14-20.
20
R. Tol, The Social Cost of Carbon: Trends, Outliers and Catastrophes, Economics E-Journal, 2(2008):
25.
21
OMB, Circular A-4 (guidance to Federal agencies on the development of regulatory analysis as required
under Section 6(a)(3)(c) of Executive Order12866, Regulatory Planning and Review, the Regulatory Right-
to-Know Act, and a variety of related authorities.), September 13, 2003, p. 35-36.
rates, benefits shrink even faster as they move forward in time, and vice versa with lower
discount rates.
The Interagency Report chose 3% as its “central value” discount rate and estimated the SCC at
$21/tCO2, and also estimated the SCC at discount rates of 5% and 2.5%, resulting in $5/tCO2 and
$35/tCO2 respectively.
22
Subsequently sensitivity analysis conducted at 2% and 1% resulted in
estimates of $63/tCO2 and $266/tCO2.
23
These results demonstrate the high sensitivity of the SCC
to the discount rate used. The Interagency Report noted:
The choice of a discount rate, especially over long periods of time, raises highly
contested and exceedingly difficult questions of science, economics, philosophy, and law.
Although it is well understood that the discount rate has a large influence on the current
value of future damages, there is no consensus about what rates to use in this context.
Despite this acknowledgment, the report hewed closely to conventional BCA practice in its
“central value” estimate and choice of a range of sensitivity analysis. Accounts of the interagency
process indicate that disciplines other than economics had little sway in the choice of discount
rates. The result is that welfare of future generations received little weight in the SCC.
How Do We Account for the Fact That Most People Don’t Like to Take Risks With Large
Negative Outcomes?
Economists have shorthand for this question: How do we account for “risk aversion”?
Economists have a certain lexicon for describing how people make choices under conditions of
uncertainty, recognizing that some degree of uncertainty permeates all of our decisions. Common
sense suggests that, in making choices when the future may turn out any of several ways, we
consciously or unconsciously weigh probabilities of these future outcomes and consider the “pay-
off” of each in terms of our well-being. Individuals behave differently based on their tolerance for
risk and the “expected value” of the choice.
A simple insurance example illustrates the concept of expected value and attitude toward risk.
Assume that a driver owns a Lexus worth $50,000 and her chance of an accident that totals the
car is 1 in 1000 each year. The expected value of her loss is $50 per year. She can purchase
insurance to cover the loss. If she is “risk neutral”, she will pay no more than $50 per year. If she
is “risk preferent” (i.e., a gambler who likes risk), she will be willing to pay some lower amount,
less than the expected value of the loss. If she is “risk averse”, she will be willing to pay more
than $50 for insurance.
In the real world, individuals and organizations of all types display risk aversion. They pay more
than the expected value of an event in order to insure themselves against the consequences of an
event. There are a plethora of insurance products along with other risk-hedging actions one can
take (e.g., the use of futures markets).
22
The Interagency Report also included a fourth value at the 3% discount rate to represent “higher-than-
expected impacts” given the probabilistic nature of SCC estimation. See discussion later in this paper.
23
L. Johnson and C. Hope, Revisiting the SCC Estimates Developed by the U.S. Government, Working
Paper, Natural Resources Defense Council, Washington DC, November 17, 2010.
Uncertainty permeates many of the factors determining the impacts of climate change, e.g., the
climate sensitivity to increased CO2 concentrations and the likelihood, magnitude, and timing of
the associated impacts, along with economic growth projections and the CO2 intensity of that
growth. This leads scientists and economists to run their models in a probabilistic manner,
generating multiple scenarios with associated probabilities. Some of these scenarios convey
catastrophic impacts, leading some to compare the design of climate policy to buying the right
amount of insurance to prevent such catastrophes. Of course, there is no insurance policy
available. As a planet, we have no choice: we are “self insuring” against climate change risks. If
people are risk averse to catastrophic climate change, they will want to pay more than the
“expected value” of the damages generated by IAMs (multiple scenarios each weighted by their
respective probabilities). Thus a “risk averse” estimate of the SCC will be higher than a “risk
neutral” estimate.
Standard practice in BCA is to take a risk neutral approach to uncertainties, as reflected in OMB
guidance on the issue.
24
Economists acknowledge that society may not be risk neutral with
respect to climate change, and IAMs make some effort to account for catastrophic risks in
estimating the SCC. However, risk neutrality remains standard practice in dealing with
catastrophe, as reflected in the treatment of this issue in the Interagency Report:
A key question unanswered during this interagency process is what to assume about
relative risk aversion with regard to high-impact outcomes. These calculations do not
take into account the possibility that individuals may have a higher willingness to pay to
reduce the likelihood of low-probability, high-impact damages than they do to reduce the
likelihood of higher-probability but lower-impact damages with the same expected
cost…. If individuals do show such a higher willingness to pay, a further question is
whether that fact should be taken into account for regulatory policy. Even if individuals
are not risk-averse for such scenarios, it is possible that regulatory policy should include
a degree of risk-aversion.
25
The Interagency Report tipped its hat to the risk aversion issue by providing in its SCC estimates
a value of $65/tCO2 representing “…the 95th percentile SCC estimate across all three models at a
3 percent discount rate, …included to represent higher-than-expected impacts from temperature
change further out in the tails of the SCC distribution.”
26
Predictably, this value as well as the
average estimates at discount rates of 2.5 and 5 percent receive little attention relative to the
“central value” estimate of $21/tCO2.
In the text quoted above, the Interagency Report questions the degree to which individuals show
risk aversion to various “high-impact outcomes.” The prevalence of insurance purchases by
individuals to protect against a variety of high-impact outcomes (e.g., death, disability, long-term
medical care, loss of property) would seem to provide the answer.
27
Firms and organizations of
all types show the same risk aversion in their insurance purchases and other related behavior.
There seems little ground to argue that government actions and policies should not reflect the
same kind of risk aversion that the rest of society shows.
Concluding Thoughts
24
OMB, Op. cit., p. 42.
25
Interagency Working Group on Social Cost of Carbon, Op. cit., p. 30.
26
Interagency Working Group on Social Cost of Carbon, Op. cit., p. 1.
27
See also all-purpose “umbrella” insurance policies.
Economists apply BCA broadly to a wide variety of policy issues, though rarely without
generating controversy and/or push back from other disciplines. Regardless of BCA’s
appropriateness in guiding decisions in other issue areas, economists’ conventional
methodologies and assumptions seem ill-suited for application to climate change. As
conventionally applied, BCA:
Ignores various significant global damages that cannot be monetized.
Reduces other benefits with arbitrary assumptions about no-cost or low-cost adaptation
that are little more than guesswork.
Does not adequately account for the changes in welfare to billions of people who
participate only marginally in market economies.
Does not apply any “equity weighting” when aggregating benefits across countries or
regions that have very different levels of wealth, thus giving relatively greater weight to
impacts of rich people relative to poor people.
Applies discount rates to future benefits that result in inadequate weight being given to
the welfare of future generations.
Applies a “risk neutral” perspective to potentially catastrophic impacts when society is
clearly risk averse to such outcomes.
The risk management framework described earlier avoids many of these pitfalls, but needs more
elaboration on how to determine the resources that should be used to mitigate climate change, and
the appropriate timeline. A blending of the two frameworks may also be possible. Getting climate
scientists and economists on the same planet can only bring major benefits.
Figure 1
Source: IPCC Fourth Assessment Report
Figure 2
Source: Interagency Working Group on Social Cost of Carbon, Op. cit., p. 9
... Views vary on the correct discount rate for climate policies as well as the extent to which rates differ between developing and developed countries. 6 Some economists give more weight to environmental benefits that occur in distant years and recommend a lower discount rate for intergenerational decisions or a 'hyperbolic' discount rate that declines over time (Hausker 2011). For example, the Stern Review recommends a declining social discount rate, with rates lower than 3% for investments beyond 30 years (Stern 2007). ...
... d Standard practice in benefit-cost analysis is to take a risk-neutral approach to uncertainties. In the real world, individuals and organisations of all types display risk aversion to catastrophic impacts (Hausker 2011). e EPA (2016). ...
Chapter
Full-text available
The ocean and its resources provide key ecosystem services and benefits that are crucial for human well-being and the prosperity of the global economy, but these services are at risk. The ocean’s wide range of ecosystem services (including food, energy, recreational/ cultural services and trading/transport routes) is vital for the well-being of society. However, climate change, overfishing, pollution and a loss of biodiversity and coastal ecosystems are eroding the ability of the ocean to sustain livelihoods and prosperity.
ResearchGate has not been able to resolve any references for this publication.