Environ. Sci. Technol. 2009, 43, 6421–6426
Indicators on Economic Risk from
Global Climate Change
W O L F D . G R O S S M A N N , †,‡
K A R L S T E I N I N G E R , † I R I S G R O S S M A N N , * ,§
AND LORENZ MAGAARD‡
Wegener Center for Global and Climate Change and
Department of Economics, University of Graz, Leechgasse 25,
A-8010 Austria, International Center of Climate and Society,
University of Hawaii at Manoa, 1680 East-West Road,
Honolulu, Hawaii 96822, and Climate Decision Making
Center, Carnegie Mellon University, 5000 Forbes Ave,
Pittsburgh Pennsylvania 15213, phone: 412 268 5489
Received December 17, 2008. Revised manuscript received
June 24, 2009. Accepted June 25, 2009.
Climate change mitigation requires a rapid decrease of
global emissions of greenhouse gases (GHGs) from their
present value of 8.4 GtC/year to, as of current knowledge,
approximately 1 GtC/year by the end of the century. The necessary
decrease of GHG emissions will have large impacts on
existing and new investments with long lifetimes, such as coalfired power plants or buildings. Strategic decision making for
major investments can be facilitated by indicators that express
the likelihood of costly retrofitting or shut-down of carbon
intensive equipment over time. We provide a set of simple
indicators that support assessment and decision making in this
field. Given a certain emissions target, carbon allowance
prices in a cap-and-trade plan will depend on the development
of the global economy and the degree to which the target is
approached on the global and national levels. The indicators
measure the degree to which a given emissions target is
approached nationally and assess risks for long-lived investments
subject to a range of emissions targets. A comparative case
study on existing coal-fired power plants with planned plants and
utility-scale photovoltaic power-plants confirms that high risk
for coal-fired power plants is emerging. New legislation further
confirms this result.
Introduction
Management responsible for emissions-intensive investments has to make major decisions in an environment of
considerable scientific and socio-economic uncertainties.
Rigid limitations of CO2 emissions are now introduced in
political bodies in many countries (e.g., the Waxman-Bill in
the U.S.) or have been implemented in some states. Emissions-intensive investments with long lifetimes such as coalfired power plants or buildings will face the risk of premature
shutdown or demolition. Recently, some companies have
canceled all plans for new carbon-power plants whereas
others go ahead and build (1).
Different schemes to enforce a decrease of carbon dioxide
emissions include the establishment of a cap on emissions.
Allowances are sold or given out for free up to this cap; further
* Corresponding author e-mail: irisg@andrew.cmu.edu.
†
University of Graz.
‡
University of Hawaii at Manoa.
§
Carnegie Mellon University.
10.1021/es8035797 CCC: $40.75
Published on Web 07/17/2009
2009 American Chemical Society
allowances can be bought. Technological fixes, e.g., carbon
capture and storage (CCS) or biological sequestration of
emissions are other methods. Costs and the success of many
of these schemes are highly uncertain (2).
We propose a framework of indicators that project the
dynamics of possible emissions regulations. These indicators
assess investment risks through a systematic investigation
of two central groups of uncertainties, uncertain emissions
policies, and socio-economic-technological factors on which
emissions targets and the costs of emissions reductions
depend.
Various methods, including real options approaches (3),
have been applied to identify price levels of carbon allowances
in a cap and trade system at which different investment
choices become profitable (3-5). The indicators can assess
the changing exposure of investments to retrofitting or shutdown, and can contribute to scenarios of future carbon prices
that are used by methods such as real options approaches.
The indicators are relatively simple, while addressing a variety
of issues:
(a) Given a certain emissions target, allowance prices will
depend on global economic development and the degree to
which an accepted CO2-cap is approached globally or
nationally. As the global economy grows, specific emissions
per unit of GDP have to decrease. Similarly, specific emissions
of a country have to decrease to compensate national
economic growth. Specific emissions of a country might have
to decrease also if the global economy is growing, even if the
economy of that country is stagnant.
(b) Further decrease of anthropogenic emissions is
becoming more complicated and expensive to the degree
that emissions approximate zero.
(c) Given current uncertainties on emissions policies and
climate sensitivity, robust investment decisions need to
consider an appropriately large range of emissions targets.
Indicators must be adaptable to possible downward or
upward revisions of targets.
We begin with a discussion of current emissions targets,
projections of global economic and population growth, and
associated uncertainties. The approach is to dynamically
project these elements by simple equations. The first indicator
measures the degree to which a given climate target is
approached nationally. This provides a framework for judging
the feasibility and security of an investment with respect to
requests it may meet in the near future.
The next set of indicators assesses risks for long-lived
investments such as coal-fired power plants and buildings.
Projecting and comparing uncertainties and risks for specific
decisions of interest can be a useful tool in an environment
of high uncertainty (6) since it allows dynamic linking of
scientific and socio-economic uncertainties without the need
for probability distributions. We will discuss an application
to risks of shutdown given timetables of required emissions
reduction and illustrate this with a case study which includes
coal-fired plants of the Tennessee Valley Authority and a
new utility-scale power plant with photovoltaics. The case
study applies the indicators and confirms their relevance.
Framework for the Indicators
Unpredictable Changes of Emissions Targets. The EU has
mandated that average global temperature increases should
not exceed 2 °C above preindustrial levels by year 2100 (7).
According to the IPCC fourth Assessment Report, limiting
temperature increases to 2-2.4 °C implies a concentration
level of at most 350-400 ppm (ppm) CO2 (8). This is 10-35%
below the concentration level of 450-550 ppm considered
VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6421
by the third Assessment Report in 2001 (9). Hansen et al. (10)
give 350 ppm CO2 as the long-time maximum.
In the first multithousand member ensemble of simulations (run through climateprediction.net) Stainforth et al.
(11) find climate sensitivities, that is, global surface temperature responses to doubled CO2 conditions, ranging from
2-11°K (Kelvin). The uncertainty on the sensitivity of the
climate system to GHG and other forcings (factors influencing
temperatures) has various scientific roots, foremost among
them the complexity of the climate system (12, 13). Schwartz
et al. (13) find that the 5-95% confidence range in global
mean temperature change as projected by the IPCC is much
smaller than that associated with the forcings considered,
yielding, e.g., a factor of 2, while forcing uncertainties yield
a factor of 4. Besides the aerosol contribution to climate
(14, 15), factors that may be insufficiently understood include
the contribution of land use change to warming (16, 17) and
the climatic effects of large-scale atmosphere-ocean variability (18-21).
If unanticipated and not well understood effects mask
the projected warming from GHG emissions temporarily,
emissions targets may be relaxed or even abandoned
(compare ref 18). Targets may be adjusted upward or
downward as understanding progresses. Robust planning
must evaluate the proposed indicators for a range of possible
emissions targets.
We next examine the socio-economic context within
which emissions reductions will proceed.
Projections of Global Economic Growth. Historically,
during the last century the global economy has grown at an
average of 3.2% y-1 and emissions of GHGs have increased
sublinearly with economic growth, due to improvements in
resource productivity. For instance, between 1990 and 2004,
economic growth has increased GHG emissions by 1.57%
per year, thereafter by 2.7% (8). In agreement with the
scenarios of the IPCC we assume an economic growth
between 2 and 4% per year.
Global Population Growth and National Emissions.
Statements from the two largest countries, China and India,
on GHGs strongly suggest that privileges for developed
countries, for instance in the form of ‘grandfathering’ (22)
would not be politically feasible. A possible global equity
approach to specify maximum national allowable emissions
would derive each country’s share from its fraction of the
global population (23). Maximum allowable emissions will
then change according to how the country’s population
changes relative to the global population. A global equity
scheme would be most unfavorable for the US with the
highest per capita emission. With the present global population of 6.7 billion, the US with a population of 300 million
would be entitled to 4.5% of global emissions or 71 ktC y-1.
Taking into account globally declining birth rates (albeit
from different levels in different countries) and increasing
life expectancy, the current assumption is that the global
population may peak at 8-10 billion (24). The population of
the U.S. is increasing, whereas Western Europe’s population
is growing at a slow rate and expected to decline soon (24).
If global population trends continue, maximum allowable
emissions for Western Europe would have to decline by
almost 1/3 respective to the current value.
The Gross World Product (GWP) was $U.S. 54 trillion in
year 2006 (25) with specific emissions of 0.154 tCy-1 per $U.S.
1000. Global CO2 emissions were at 6.16 GtC in 1990 and at
8.4 GtC in 2006 (26). Stabilization at 350-450 ppm implies
a reduction of global emissions to 2.8 GtCy-1 or 55% below
the level of 1990 in 2050, and to about 1 GtC/year or 84%
below the 1990 level in year 2100 (27). Without economic
growth, this would imply maximum specific emission of 0.051
tCy1- per $U.S. 1000 in 2050 and 0.0246 tCy1- in 2100; with
economic growth of 3.2% /year, specific emissions per $U.S.
6422
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 16, 2009
of GDP should decrease to 0.013 tC per $U.S. 1000 in 2050
and to about 0.00094 tC per $U.S. 1000 in year 2100 (see
Supporting Information (SI) section 1.1).
An indicator Showing Success or Failure Given Emissions Targets. The indicator Gyear(t) measures the degree to
which a given emissions target is approached globally in
year t. Using a percent scale, Gyear(t) shows how much of the
reduction from the present level of emissions down to the
level desired in the long term has been achieved. We here
illustrate Gyear(t) for emissions reductions up to 1 GtC y-1,
corresponding to a threshold of 400 ppm.
(
Gyear(t) ) 100 × 1GtCy-1
/ ∑ global_emissions ) (1)
year(t)
We have chosen a very simple, straightforward form for G
with the following required characteristics. The present value
of Gyear(t) is 11.6%. If emissions continue to increase, G will
continue to fall. A decrease by 1 GtCy-1 at current levels will
increase the value of G only slightly; for further decreases,
G will increase more rapidly. The technological and economic
challenge of achieving a decrease from 2 to 1 GtCy-1 is
reflected in indicator values growing from 50 to 100.
The rapid increase of G at higher levels of achieved
emissions reduction illustrates the need for near-term policies
to be oriented toward long-term targets (28). In parallel with
the substitution of current technologies and processes with
already existing more efficient technologies, technology
specific policies are needed to enable large-scale deployment
of technologies that are currently being developed (28). The
behavior of G signals whether and when new technologies
in combination with other approaches such as CCS are
required.
An Indicator for National Climate Sustainability. An
indicator for national climate sustainability should allow the
assessment of investment risks in that country and could
support the media in reporting the degree to which a country
has realized a given emissions target. We adopt the notion
of fair distribution of feasible carbon emissions by assuming
an equal per capita right to emissions. Based on eq 1 and
using population numbers Popj(t) of country j in year t and
global population Popglob(t), we define indicator Gj(t) for
country j at time t:
Gj(t) )
(
100(Popj(t) /Popglob(t)) 1 · GtCy-1
/ ∑ country j
emissions(t)
)
(2)
This indicator takes percentage values between >0 and 100.
Gj(t) will decrease (increase) if a country’s share of the world
population decreases (increases).
We will illustrate these indicators for the U.S. and China.
With its current population of 1.3 billion, China would be
entitled to about 20% of 1 GtCy-1, i.e., 200 ktCy-1. It is now
emitting 1.7 GtC y-1, to which recent economic growth of
10.9% y-1 added about 187 ktCy-1. The U.S. would be entitled
to about 71 ktC y-1. At the average yearly economic growth
rate of the last 20 years, 3.5%, and the recent increase in
specific emissions, the U.S. will add >56 ktCy-1 to its previous
emission level each year. This means that the U.S. is presently
increasing its emissions per year by its total yearly allowable.
An Expression for Feasible Specific Emissions Subject
to Economic Growth. The product of specific emissions
c(t)and GWP K(t) at time t must be less than feasible total
emissionsa(t) in a given year t:
c(t)K(t) e a(t), for t g t1
or
(3)
c(t) e a(t)/K(t) for t ∈ [2009, 2100]
(4)
Inequalities 3 and 4 must hold for all years beginning with
a year t1. By necessity, a(t) starts at the present value (which
is a kind of global grandfathering) and should approximate
1 GtCy-1 in year 2100. Equation 5 describes the increase
of the GWP K(t) as in the last century, i.e., exponential
growth as assumed in several IPCC-scenarios, with growth
rate v:
K(t) ) K(t0)ev(t-to)
(5)
If economic growth follows that pattern, specific emissions
c(t) must follow a negative exponential to meet inequality 3.
Equation 6 specifies feasible total emissions for the S400
scenario. S400 is an emissions pathway to achieve 400 ppm
(equivalent) in year 2100 [12]. It implies emissions at 41% of
the year 2000 value in year 2050 and at 15% in year 2100.
Normalizing emissions of 6.75 GtCy-1 in year 2000 to 1 and
applying nonlinear interpolation allows the following expression of S400:
3
3
3
S(t) ) 1 - 0.9(t - 2000) /(40 + (t - 2000) )
(6)
We have chosen this simple form ofS(t)as it has the same
declining logistic shape as S400 and almost the same values
(0.4 in year 2050 and 0.15 in year 2100). With economic growth
as in 5 and S400 as in 6, an explicit form of feasible specific
emissions c(t) can be derived
c(t) e (1 - 0.9(t - t0)3 /(403 + (t - t0)3))/
(K0exp(ln(v)(t - t0)))
(7)
Specific emissions will have to decline dramatically to 0.008
in year 2100, i.e., to 0.8% of the value of year 2000 (see Figure
1). Here, K0 is normalized to 1 so that K(t) gives the factor
of global economic growth.
Alternative scenarios for economic growth and decrease
of emissions can be specified in eqs 5 and 6, respectively, by
changing parameters. We select the general form of 6 (see
SI, section 1.2)
p(t) ) q + utn /(bn + tn)
(8)
Equation 8 approximates its upper limit q + u for t f ∞. For
t ) b its value is q + u/2, for t ) 0 its value is q.
Risk of Premature Shutdown of an Investment
At present there is little reason to assume that the size of
the global economy will eventually stagnate. Thus, the
existing economy will have to decrease emissions to meet
targets and to make room for the additional emissions of
new economic activity. In addition, even investments with
comparatively low emissions in a category with high
specific emissions may be at risk from costly retrofitting
or premature shutdown due to unexpected breakthrough
innovations.
An Indicator for Risk of Shutdown Subject to Policies
of Emissions Decrease. Investments with longer lifetimes
are particularly at risk from increasing emissions standards.
Examples are power stations with lifetimes of at least 40 or
50 years and buildings. In many developed countries the
electricity sector is contributing 25% to >35% to GHG
emissions. In 2050, the emissions of a new power station
(constructed in 2009) should have decreased to about 15%
of their initial value under Scenario S400, an assumed
economic growth rate of 3% y-1, and a constant share of that
country’s population to global population. An investment
would be at risk of premature shutdown if its specific
emissions cannot be decreased at acceptable costs to about
FIGURE 1. Function c(t) subject to economic growth and
declining allowances for emissions.
15%. The risk rt for shutdown in year t increases with the
ratio of specific emissions s to allowable specific emissions
st, i.e., as a monotonically increasing function:
rt ) rt(s/st)
(9)
Allowable specific emissions will depend on the sector of the
industry to which this investment belongs. However, specifications of sector-specific specific emissions are also subject
to change. For instance, considerable improvements are
currently emerging in electricity production and in the
building sector. Sectors with high specific emissions may
also be addressed by global agreements that seek to prevent
them from moving to countries with less stringent emissions
regulations (“leakage” (29)).
Consequently, an indicator for the risk of premature
shutdown needs to show a nonlinear increase of risk in
dependence on time and the ratio given in ref 9. We
standardize this function to values between 0% risk and 100%
risk. The risk of an investment with specific emissions s and
allowable specific emissions st will increase with ratio s/st.
We could assume that at 5 times the allowable emissions,
the risk of shutdown will approximate 100%. For intermediate
values we assume rapid growth of the risk in a shape which
corresponds to the solution of eq 10. The coefficient n of
function (8) determines the slope of the gradient at intermediate values of the argument. With these assumptions,
suitable parameters in function (8) are q ) 0, u ) 100, b )
2.5, n ) 4.
These values will be assessed with a sensitivity analysis.
Using xt ) (s/st) - 1 as the argument in eq 9, the risk is 0 for
s ) st, i.e., xt ) 0, and 100 for xt ) 6. With these definitions,
eq 10 is similar to the solution of the logistic equation (see
SI, section 1.2) but more adaptable. It shows the risk for
shutdown due to approximation of the capacity limit given
by feasible specific emissions.
rt )
100x4t
(10)
2.54 + x4t
Risk of Shutdown Subject to Timetables of Mitigation.
We now incorporate emissions reduction scenarios into eq 10.
Feasible emissions a(t) as in eq 6 specify the necessary decrease
in specific emissions to meet given mitigation goals subject to
economic growth. Equation 9 describes the growing misfit of
an emission-intensive investment the longer it exists, implying
increasing risk for premature shutdown. This is shown using
c(t) from eq 7 as the argument in eq 10 (Figure 2).
For the selected parameters, risk becomes very high within
20 years. As this is less than half the life expectancy of a
power plant, the loss from shutdown could be considerable.
This is illustrated in the case study below. This risk is even
higher for buildings for which depreciation is slower than for
VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6423
FIGURE 2. Risk of shutdown of an investment subject to its
specific emissions in comparison with allowable specific
emissions.
FIGURE 3. Changes in the risk of shutdown. The 50% risk
threshold varies between year 2025 and 2040.
power plants. As zero-energy buildings become state-of-the
art in new construction, the pressure on existing buildings
will increase. While power plants can potentially be retrofitted
with CCS, it is very difficult for many types of buildings to
drastically decrease their energy consumption.
With the symbols from eq 8 applied to eq 10 the ranges
for the sensitivity analysis of eq 10 and the parameters
underlying it are exponent n ∈{1,2,..5}, i.e. the shape changes
from a declining ascent to slow sigmoid to rapid sigmoid;
coefficient b ∈[25,75], i.e. the risk is at 50% within 25-75
years, and economic growth rate v ∈[0.02,0.04] between 2 to
4% per year so that the global economy grows by a factor
between 6 and 34 in year 2100 (see SI Figure S1). Figure 3
shows the risks of premature shutdown given variations in
required emissions reduction (see also SI Figure S2) with
economic growth fixed at 3%, Figure 4 overall sensitivity with
additional variations in economic growth between 2 and 4%.
Case Study
The indicators will be illustrated with a case study of coalfired plants of the Tennessee-Valley Authority (TVA), which
produces 2/3 of its electricity from coal. We will examine (1)
the profitability of existing coal-fired power plants subject
to different prices for carbon allowances, (2) the profitability
of more efficient coal-fired plants, and (3) the prospects of
utility-scale power plants using PV. For both TVA’s current
coal-fired power plants and planned more efficient plants
we consider revenues, emissions, age of the plants, depreciation and the financial effect of carbon allowances (see SI
section 2). We then compare these figures with the 550 MW
PV power plant commissioned by the U.S. utility Pacific Gas
and Electricity (PG&E).
Impact of Carbon Allowance Costs on TVA Electricity
Prices. For an investor in the electricity market, the indicator
of U.S. national climate sustainability (eq 2) is an important
6424
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 16, 2009
FIGURE 4. Fast economic growth causes higher emissions,
which in turn affect a tightening of emissions regulations,
whereas slow growth delays the build-up of GHGs allowing
slower actions for emission control.
indicator. This indicator currently has a low value of 2.8%
and is decreasing. The necessary value is 100%. For comparison, China, which has the same emissions per year as
the U.S., has an indicator value of 11.4%. Equation 7 signals
to the investor that in the long run emissions decreases to
about 4% of the present will be necessary if a cap-and-trade
scheme similar to scenario S400 is pursued. Compared to
these 4%, the Waxman climate change bill is only a first step.
It mandates emission reductions by 17% from 2005 levels by
2020. By 2012, President Obama’s budget presumes a price
of about $13.70 for U.S. carbon allowances (30). This is at the
lower end of the current European Union Allowances price
fluctuations between U.S. $12 and 25.
The total cost of 1 kWh produced from coal by the TVA
in 2008 was Cent 5.35 (See SI Table S1). With a price of Cent
6.51/kWh for Kentucky, as indicated by the Energy Information Administration (31), the profit per kWh for the TVA is
Cent 1.21/kWh. An average fuel consumption of 0.374 kg
coal per kWh in the TVA coal plants emits 1.37 kg CO2/kWh.
At a carbon allowance cost of $13.70 per ton of CO2 this
would incur carbon costs of Cent 1.88/kWh (see SI section
2, implying losses of Cent 0.76/kWh for the TVA.
In a cap-and-trade system, caps on emissions are set and
permits consistent with those limits are initially issued and
can subsequently be bought. The cap may be lowered over
time, for example according to Scenario S400. Lowering the
cap should increase the costs of emissions. Based on a review
of eleven models, Fischer and Morgenstern (32) indicate
marginal abatement costs of between $40 and $250 per ton
carbon in 1990 dollars for a 20% emissions reduction relative
to 1990, the reduction required under S400 in approximately
2025. This translates to $15-97 in 2006 dollars (33). The
highest price would incur additional costs of Cent 13.3 per
kWh for TVA. This would translate into an increase of
electricity prices in Kentucky by 348% from Cent 5.35 to 18.65,
a scenario where it appears very likely that TVA will not be
able to pass on a significant proportion of the additional
costs to consumers (34).
For a more efficient plant, costs per kWh of electricity
decrease about 5% compared to the TVA’s current plants
(see SI Table S1). We further compare total costs of electricity
for a range of allowance costs (table 1). Overall, costs per
kWh are markedly lower for a new power plant, with about
a 14% difference for low carbon prices and up to 32% for
high carbon prices.
While this illustrates that a more efficient plant will be
much more profitable at high carbon prices, recent events
demonstrate a high risk that investment costs for new power
plants cannot be recovered. Several dozen coal-fired power
plant projects have been canceled, delayed, or rejected in
the past few years (1, 34, 35). To investigate this risk, we
TABLE 1. Comparison of Total Costs of Electricity for a Range
of Carbon Allowance Costs
CO2-costs perqa ton ($)
carbon costs TVA old
$ per kWh
electricity costs TVA $/kWh
including carbon costs
carbon costs power
plant new $ per kWh
electricity costs new plant $/kWh
including carbon costs
15
30
60
200
0.021 0.041 0.082 0.27
0.071 0.091 0.132 0.32
0.015 0.030 0.061 0.2
0.062 0.077 0.108 0.247
compare the costs per kWh of both current and planned
more efficient coal plants with the costs per kWh of electricity
produced by utility-scale PV plants.
Comparison of Coal Power Plants with a Utility-Scale
Photovoltaic Power Plant. PG&E has signed agreements to
purchase electricity from two new PV plants with, respectively, 250 and 550 MW. We estimate the price per kWh of
electricity generated by these plants in dependence of the
investment costs per kW installed (SI section 2.5). If existing
landlines are not sufficient, construction of new lines could
increase this price.
First Solar has announced panel production costs in year
2012 of $0.63 Whp, This means costs for the complete
installation utility scale of $1.31 and costs of 5.76 c/kWh in
favorable locations. Stiff competition in the PV sector will
drive down First Solar’s present high profit rate (SI section
2.5). The respective profit rate of 25 or 50% of First Solar
increases the price to between Cent 7.2 or 8.64, respectively,
or 11.52 at the current profit rate.
This compares to cost per kWh of electricity for TVA’s
current plants at Cent 5.35 and Cent 5.1 for a new plant. At
carbon allowance costs of $25 per ton, the price per kWh of
electricity from a new plant will be at Cent 7.2 (Table 1), and
at Cent 11.52 for allowance costs of $65/ton. Thus, PV may
become competitive much sooner than expected for all
scenarios considered, and it is additionally not affected by
the risks of increasing carbon costs or increasing coal prices.
The case study shows that the threat for coal power plants
is multifaceted and massive. Our sensitivity analysis shows
a rapidly rising risk for all coal power plants. These results
are fairly robust even at large variations of input parameters.
The robustness of the risks to coal plants is due to the tight
relationship between revenues and profits. Additional costs
cannot easily be offset by increases of the electricity price,
and price increases may not be permitted by utilities
commissions (34). Even low costs per ton of CO2 threaten
the profitability of current and newer more efficient power
plants, in particular given the rapid emergence of credible
alternatives such as second generation PV. This means that
it is uncertain whether utilities will be able to recover new
investments.
Discussion
We have developed indicators to support decision-making
for large investments, taking into account uncertain climate
mitigation policies and uncertain socio-economic developments. Climate policies are uncertain because of the complexity of the climate system and our current difficulties to
project and understand how different forcings impact
regional and global climate. A further reduction of targets
appears likely if certain undesirable environmental changes
result. We have also shown how rapid technological changes
and surprises can put carbon-intensive investments at risk.
The rather sudden availability of low-cost PV is a major
example.
Our indicators on climate sustainability and on the risk
of premature shutdown of investments allow investors and
managers to assess the consequences of unexpectedly fast
economic growth in large parts of the world, including China,
India, or the “Next Eleven” identified by Goldman Sachs (36);
project different scenarios of future carbon prices, given
climate policy and economic uncertainties; and understand
the impact of low-cost PV which, additionally, sets the mark
for state-of-the art of specific emissions per kWh of electricity.
Indicators that express the uncertainties and sophistication of climate change and mitigation through simple graphs
and equations may effectively support understanding and
decision making. The ability to project risk from climate
legislation for planned investments is advantageous for
making decisions and avoiding exposure to risks that may
otherwise be underestimated.
Acknowledgments
This research was supported by the Austrian National Bank
Research Fund (project 12449), which is thankfully acknowledged. The third author was supported by the Climate
Decision Making Center created through a cooperative
agreement between the National Science Foundation (SES0345798) and Carnegie Mellon University.
Supporting Information Available
Additional text and data, figures and tables.This material is
available free of charge via the Internet at http://pubs.acs.org.
Literature Cited
(1) Sautter, J. A.; Switzer, C. A change in climate? How state
regulators are making decisions in response to global warming.
The electricity journal 2008, 21, 38–44.
(2) Bachu, S., CO2 storage in geological media: Role, means, status
and barriers to deployment. Prog Energy Combustion Sc 2008,
34.
(3) Patiño-Echeverri, D.; Morel, B.; Apt, J.; Chen, C. Should a coalfired power plant be replaced or retrofitted. Environ. Sci. Technol.
2007, 41, 7980–7986.
(4) Blyth, W.; Bradley, R.; Bunn, D.; Clarke, C.; Wilson, T.; Yang, M.
Investment risks under uncertain climate change policy. Energ
pol 2007, 35, 5766–5773.
(5) Bergerson, J. A.; Lave, L. B. Baseload coal investment decisions
under uncertain carbon legislation. Environ. Sci. Technol. 2007,
41, 3431–3436.
(6) Lempert, T. J.; Schlesinger, M. E.; Bankes, S. C.; Andronova,
N. G. The Impact of Variability on Near-Term Climate-Change
Policy Choices and the Value of Information. Climatic Change
2000, 45, 129–161.
(7) Höhne, N., Ellermann C. ; Policy, P. D. E. a. S. , Ed.; DG Internal
Policies of the Union, 2008.
(8) ntergovernmental Panel on Climate Change “Mitigation of
Climate Change, Summary for Policymakers,” IPCC, 2007.
(9) Intergovernmental Panel on Climate Change “Climate Change
2001: Synthesis Report, Summary for Policymakers,” Intergovernmental Panel on Climate Change, 2001.
(10) Hansen, J., Sato, M.; Kharecha,P.; Beerling, D.; Masson-Delmotte,
V.; Pagani, M.; Raymo, M.; Royer, D. L.; Zachos, J. C.; ”Target
Atmospheric CO2: Where Should Humanity Aim?” NASA/Goddard
Institute for Space Studies, New York, NY, 2008.
(11) Stainforth, D. A.; Aina, T.; Christensen, C.; Collins, M.; Faull, N.;
Frame, D. J.; Kettleborough, J. A.; Knight, S.; Martin, A.; Murphy,
J. M.; Piani, C.; Sexton, D.; Smith, L. A.; Spicer, R. A.; Thorpe,
A. J.; Allen, M. R. Uncertainty in predictions of the climate
response to rising levels of greenhouse gases. Nature 2005, 433,
403–406.
(12) Pielke, R. P., Sr.; Schellnhuber, H. J.; Sahagian, D. Non-linearities
in the Earth System. Global Change Newslett 2003, 55, 11–15.
(13) Schwartz, S. E.; Charlson, R. J.; Rodhe, H. Quantifying climate
change - too rosy a picture. Nature Reports 2007, 2, 23–24.
(14) Anderson, T. L.; Charlson, R. J.; Schwartz, S. E.; Knutti, R.;
Boucher, O.; Rohde, H.; Heintzenberg, J. Forcing by aerosols a hazy picture. Science 2003, 300, 1103–1104.
(15) Morgan, M. G.; Adams, P. J.; Keith, D. W. Elicitation of expert
judgments of aerosol forcing. Climatic Change 2006, 75, 195–
214.
VOL. 43, NO. 16, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6425
(16) Hale, R. C., Gallo, K. P., Owen, T. W., Loveland, T. R. , Land
use/land cover change effects on temperature trends at U.S.
Climate Normals stations. Geophys. Res. Lett. 2006, 33.
(17) Pielke, R. A. Sr., Davey, C. A., Niyogi, D., Fall, S., SteinwegWoods, J., Hubbard, K., Lin, X., Cai, M., Lim, Y.-K., Li, H., NielsenGammon, J., Gallo, K., Hale, R., Mahmood, R., Foster, S.,
McNider, R. T., Blanken, P., Unresolved issues with the
assessment of multidecadal global land surface temperature
trends. J Geophys Res Atmos 2008, 112.
(18) Keenlyside, N. S.; Latif, M.; Jungclaus, J.; Kornblueh, L.; Roeckner,
E. Advancing decadal-scale climate prediction in the North
Atlantic sector. Nature 2008, 453, 84–88.
(19) Tsonis, A. A., Swanson, K., Kravtsov, S., A new dynamical
mechanism for major climate shifts. Geophys Res Lett 2007, 34.
(20) Zhang, R., Delworth, T. L., Held, I. M., Can the Atlantic ocean
drive the observed multidecadal variability in Northern hemisphere mean temperature? Geophys. Res. Lett. 2007, 34.
(21) Wunsch, C. The interpretation of short climate records, with
comments on the North Atlantic and Southern Oscillations.
Bull Amer Meteor Soc 1999, 80, 245–255.
(22) Michaelowa, A. T.; K. Hasselknippe, H. Issues and Options for
the Post-2012 Climate Architecture - An Overview. International
Environmental Agreements 2005, 5, 5–24.
(23) Baer, P.; Harte, J.; Haya, B.; Herzog, A. V.; Holdren, J.; Hultman,
N. E.; Kammen, D. M.; Norgaard, R. B.; Raymond, L. Climate
Change: Equity and Greenhouse Gas Responsibility. Science
2000, 289, 2287.
(24) United Nations “World Population Prospects: The 2006 Revision,” United Nations Population Division, Department of
Economic and Social Affairs, 2006.
(25) International Monetary Fund “World Economic Outlook (WEO):
Financial Stress, Downturns, and Recoveries,” International
Monetary Fund, 2006.
6426
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 16, 2009
(26) Marland, G., Boden, T. A. Andres, R. J. “Global, Regional, and
National Fossil Fuel CO2 Emissions, In: Trends Online - A
Compendium of Data on Global Change,” Carbon Dioxide
Information Analysis Center, Oak Ridge National Laboratory,
2007.
(27) Meinshausen, M. What does a 2°C target mean for greenhouse
gas concentrations? - A brief analysis based on multi-gas emission
pathways and several climate sensitivity uncertainty estimates.
In Avoiding Dangerous Climate Change; Schellnhuber, J. S.,
Cramer, W., Nakicenovic, N., Wigley, T. M. L., Yohe, G. , Ed.;
Cambridge University Press: Cambridge, 2006.
(28) Sanden, B. A.; Azar, C. Near-term policies for long-term climate
targets - economy wide versus technology specific approaches.
Energy Policy 2005, 33, 1557–1576.
(29) Baron, R. “Sectoral approaches to GHG mitigation: Scenarios
for integration,” OECD/IEA, 2006.
(30) Point Carbon “Carbon Market North America,” Point Carbon
News Vol 4 (8), 2009.
(31) Energy Information Administration “Electric Power Monthly:
Electricity Rate Comparison by State,” Energy Information
Administration, Washington D.C.
(32) Fischer, C.; Morgenstern, R. D. Carbon abatement costs: why
the wide range of estimates. The energy journal 2006, 27, 73–86.
(33) Berry, D. “Investment risks of new coal-fired power plants,”
Western Resource Advocates, 16 pp., 2008.
(34) Dworkin, M. Coal-fired power plants: imprudent investments.
Science 2007, 1791–1792.
(35) Synapse Energy Economics “Don’t get burned: the risks of investing
in new coal-fired generating facilities.,” Synapse Energy Economics, 56 pp., 2008.
(36) Wilson, D. ; Stupnytska, A. “The N-11: More than an acronym,”
Goldman Sachs 2007.
ES8035797